Journal of Cell Science partnership with Dryad

Journal of Cell Science makes data accessibility easy with Dryad

Grb2 mediates semaphorin-4D-dependent RhoA inactivation
Tianliang Sun, Rameshkumar Krishnan, Jakub M. Swiercz

Summary

Signaling through the semaphorin 4D (Sema4D) receptor plexin-B1 is modulated by its interaction with tyrosine kinases ErbB-2 and Met. In cells expressing the plexin-B1–ErbB-2 receptor complex, ligand stimulation results in the activation of small GTPase RhoA and stimulation of cellular migration. By contrast, in cells expressing plexin-B1 and Met, ligand stimulation results in an association with the RhoGTPase-activating protein p190 RhoGAP and subsequent RhoA inactivation – a process that involves the tyrosine phosphorylation of plexin-B1 by Met. Inactivation of RhoA is necessary for Sema4D-mediated inhibition of cellular migration. It is, however, unknown how plexin-B1 phosphorylation regulates RhoGAP interaction and activity. Here we show that the activation of plexin-B1 by Sema4D and its subsequent tyrosine phosphorylation by Met creates a docking site for the SH2 domain of growth factor receptor bound-2 (Grb2). Grb2 is thereby recruited into the plexin-B1 receptor complex and, through its SH3 domain, interacts with p190 RhoGAP and mediates RhoA deactivation. Phosphorylation of plexin-B1 by Met and the recruitment of Grb2 have no effect on the R-RasGAP activity of plexin-B1, but are required for Sema4D-induced, RhoA-dependent antimigratory effects of Sema4D on breast cancer cells. These data show Grb2 as a direct link between plexin and p190-RhoGAP-mediated downstream signaling.

Introduction

Semaphorins (Semas) are a large family of evolutionarily conserved secreted or membrane-bound proteins (Tamagnone et al., 1999). Recent studies have implicated these molecules in many processes of neuronal development, including axonal fasciculation, target selection, neuronal migration, and dendritic guidance, as well as in the remodeling and repair of the adult nervous system (Roth et al., 2009; Giger et al., 2010). Moreover, semaphorin expression has been described in many tissues outside the nervous system and their functions include the regulation of cell adhesion and motility in angiogenesis, immune responses, tumor progression and metastasis (Roth et al., 2009; Vadasz et al., 2010).

The majority of the cellular effects of semaphorins are mediated by plexins. Four families of plexins exist in the mammalian system: plexin-A1-4, plexin-B1-3, plexin-C1 and plexin-D1. Four members of the plexin-A family in most cases require neuropilins as ligand-binding partners to respond to semaphorins, whereas the three members of the plexin-B family are directly activated by semaphorins (Takahashi et al., 1999; Tamagnone et al., 1999). Whereas plexin-B1 binds Sema4D, plexin-B2 can be activated by Sema4C and Sema4D, and plexin-B3 has been shown to respond to Sema5A (Tamagnone et al., 1999; Artigiani et al., 2004; Masuda et al., 2004; Deng et al., 2007). All members of the plexin family are characterized by the presence of the Sema domain at their extracellular part and an untypical Ras GTPase activating domain (GAP) at their intracellular portions (Tamagnone et al., 1999). Members of the plexin-B family interact with and are phosphorylated by the receptor tyrosine kinases ErbB-2 and Met (Giordano et al., 2002; Swiercz et al., 2004). The activation of plexins by semaphorins initiates a variety of signaling processes, which involve several small GTPases of the Ras and Rho families. In addition to an R-RasGAP function, activated plexin-B1 and plexin-A1 have been shown to interact with other small GTPases, including GTP-bound Rac1 and RhoD as well as Rnd1, Rnd2 and Rnd3 (Vikis et al., 2000; Oinuma et al., 2004b; Oinuma et al., 2004a; Puschel, 2007; Tong et al., 2007; Saito et al., 2009).

Interestingly, B-family plexins were shown to activate or inactivate small GTPase RhoA, depending on the interaction with the specific receptor tyrosine kinase – ErbB-2 or Met (Swiercz et al., 2008). Activation of the plexin-B1–ErbB-2 complex by semaphorin 4D (Sema4D) results in the activation of the kinase activity of ErbB-2 and subsequent phosphorylation of plexin-B1 on tyrosine residues 1708 and 1732 (Swiercz et al., 2009). Phosphorylated tyrosine residues provide docking sites for SH2 domains of 1-phosphatidylinositol-4,5-bisphosphate phosphodiesterase gamma-1/2 (PLCγ1/2) enzymes. After recruitment of PLCγ into the plexin-B1 receptor complex, PLCγ through its SH3 domain mediates the activation of guanine nucleotide exchange factors (GEFs) – PDZ-RhoGEF or LARG – that stably interact with the C-terminus of B-family plexins (Aurandt et al., 2002; Driessens et al., 2002; Hirotani et al., 2002; Perrot et al., 2002; Swiercz et al., 2002). This PLCγ-dependent signaling mechanism is required for Sema4D-induced, plexin-B1-mediated stimulation of cell migration and neuronal growth cone morphology (Swiercz et al., 2009).

By contrast, Sema4D-mediated activation of the plexin-B1–Met complex results in the inactivation of RhoA (Swiercz et al., 2008). Similar to activation of RhoA, inactivation of RhoA downstream of plexin-B1 also appears to be dependent on the tyrosine phosphorylation of plexin-B1 (Swiercz et al., 2008). In addition, members of the B-plexin family were proposed to interact with p190 RhoGAP, the GTPase-activating protein for RhoA (Barberis et al., 2005). Activation of plexin-B1 in fibroblasts results in an interaction with p190 RhoGAP and subsequent inactivation of RhoA (Barberis et al., 2005). On the cellular level, Sema4D-induced, Met-mediated phosphorylation of plexin-B1 inhibits migration of cancer cells (Swiercz et al., 2008).

Molecules taking part in plexin-B1–ErbB-2-mediated RhoA activation and stimulation of cell migration are well described (Swiercz et al., 2009). By contrast, molecular mechanisms linking Met-mediated phosphorylation of plexin-B1 to activation of p190 RhoGAP and the regulation of RhoA activity are unknown.

Here we report that upon activation by Sema4D, plexin-B1 becomes phosphorylated by Met at a particular tyrosine residue on its intracellular portion. This phosphorylated tyrosine residue serves as docking site for the SH2 domain of Grb2. Grb2 is thereby recruited into the plexin-B1 receptor complex, and through one of its SH3 domains, mediates the interaction and activation of p190 RhoGAP, resulting in RhoA inactivation and downstream cellular effects.

Results

Phosphorylation of specific tyrosine residues of plexin-B1 is required for Sema4D-induced RhoA deactivation

We have previously shown that tyrosine phosphorylation of plexin-B1 by the receptor tyrosine kinase Met is crucial in plexin-B1-mediated deactivation of the small GTPase RhoA. To identify the tyrosine residues of plexin-B1 phosphorylated by Met, we tested tyrosine to phenylalanine (Y/F) mutants of all 26 tyrosine residues (Swiercz et al., 2009) of the cytoplasmic portion of plexin-B1 for their ability to become phosphorylated by Met (Fig. 1A). We observed that the mutation of tyrosine residues 1864 and 2094 led to a significant decrease in Sema4D-induced Met-dependent phosphorylation of plexin-B1. The mutation of tyrosine residues 1708 and 1732, which are phosphorylated by ErbB-2, do not influence Met-dependent phosphorylation of plexin-B1 (Fig. 1A). Mutation of tyrosine residues 1864 or 2094 shows no effect on Sema4D-mediated, ErbB-2-dependent phosphorylation of plexin-B1 (Fig. 1B). The mutation of both tyrosine residues [plexin-B1(Y1864F/Y2094F)] completely abrogated Sema4D-induced phosphorylation of plexin-B1 by Met (Fig. 2A). Consistent with the established role of plexin-B1 tyrosine phosphorylation in Sema4D-induced RhoA deactivation, plexin-B1(Y1864F/Y2094F) was not able to mediate Sema4D-induced deactivation of RhoA (Fig 2A). By contrast, we found no difference in Sema4D-regulated R-RasGAP activity or Rac1 association in cells expressing any of the tyrosine mutants of plexin-B1 (data not shown), or the Y1864F/Y2094F double mutant of plexin-B1 (Fig. 2B,C, respectively). This indicates that the Y1864/Y2094F double mutant of plexin-B1 had not lost its ability to mediate Sema4D signaling and that the tyrosine phosphorylation by Met is required for RhoA deactivation but not for the regulation of other plexin-B1 signaling pathways.

Fig. 1.

Characterization of Met-mediated phosphorylation of plexin-B1. (A) HEK293 cells were transfected with wild-type, VSV-tagged plexin-B1 [(VSV)-plxB1], or the indicated tyrosine to phenylalanine (Y/F) mutants, Met and PDZ-RhoGEF (PRG). Cells were then treated with 1 µM ErbB-2 inhibitor AG1478 for 30 min and incubated without or with 150 nM Sema4D (Sema4D −/+), lysed, and plexin-B1 was immunoprecipitated (IP). Immunoprecipitates were subjected to SDS-PAGE and were then immunoblotted (IB) using anti-VSV and anti-phosphotyrosine antibodies (anti-P-Tyr). (B) HEK293 cells were transfected with wild-type plexin-B1 [(VSV)-plxB1], or the indicated Y/F mutants together with PDZ-RhoGEF (PRG). Cells were preincubated with 150 nM Sema4D (Sema4D −/+). Samples were processed as described above and representative autoluminograms are shown.

Fig. 2.

Effects of the plexin-B1(Y1864F/Y2094F) mutant on plexin-B1 signaling. HEK293 cells were transfected with cDNAs encoding wild-type (WT), VSV-tagged plexin-B1 [(VSV)-plxB1] or its mutant (Y1864F/Y2094F) together with Met and (A) Myc-tagged RhoA [(Myc)-RhoA] and wild-type PDZ-RhoGEF (PRG) or (B) R-Ras, Rnd1 and PDZ-RhoGEF (PRG) or (C) the constitutively active GST-tagged Rac1 mutant [Rac(Q61L)] and wild-type PDZ-RhoGEF (PRG). After 48 hours, cells were incubated without (− Sema4D) or with 150 nM Sema4D (+ Sema4D) for 20 min. Thereafter, cells were lysed, and plexin-B1 was immunoprecipitated (IP) using anti-VSV-sepharose (A,C). (A) Precipitated proteins were analyzed by immunoblotting using anti-phosphotyrosine antibodies. In a parallel experiment, active RhoA was precipitated from cell lysates as described in the Materials and Methods. (B) Activated R-Ras was precipitated from the cell lysate with GST-fused Ras-binding domain of Raf1. Shown are autoluminograms of immunoblots of lysed (lysate) or precipitated samples (IP/GST-RBD or GST-Raf1 pull-down) using the indicated antibodies (IB).

We previously showed that the ErbB-2-phosphorylated tyrosine residues 1708 and 1732 of plexin-B1 provide docking sites for SH2 domains of PLCγ1/2 (Swiercz et al., 2009). To test whether the Met-mediated phosphorylation of plexin-B1 also results in the formation of docking sites for SH2 or PTB-domain-containing proteins, we synthesized peptides corresponding to 15 amino acid sequences of the plexin-B1 cytoplasmic portion containing tyrosine residues 1864 or 2094 in a non-phosphorylated or a phosphorylated version. Peptides were then bound to agarose beads and incubated with lysates of HEK293 cells. After extensive washing, the bound proteins were eluted and separated using SDS-PAGE. Whereas no additional bands were obtained with the peptide containing phosphorylated tyrosine 1864 compared with the non-phosphorylated version (data not shown), one protein specifically interacted with phosphorylated peptide containing residue 2094 (Fig. 3A). The band detected by Coomassie Blue staining was excised and analyzed by matrix-assisted laser desorption ionization mass spectrometry. In two independent experiments, growth factor receptor bound protein-2 (Grb2) was identified as the SH2-domain-containing protein interacting with phosphorylated tyrosine 2094 (Fig. 3A). To test whether Grb2 indeed bound exclusively to the peptide containing phosphorylated tyrosine 2094, the material isolated with the different peptides was subjected to an immunoblot analysis using an antibody against Grb2. We identified Grb2 in the material isolated with the phosphorylated peptide 2094 but not with the unphosphorylated peptide (Fig. 3B).

Fig. 3.

Detection of proteins interacting with pY2094. (A) Peptides representing amino acids 2086–3001 of plexin-B1 containing a non-phosphorylated (N) or phosphorylated (P) tyrosine residue 2094 were bound to agarose beads and incubated with lysates from HEK293 cells for 2 hours. Bound proteins were separated by SDS PAGE and visualized by Coomassie Blue staining. Band appearing only in sample eluted from phosphorylated peptides was cut out and analyzed by mass spectrometry. (B) HEK293 cell lysates (input) were incubated with immobilized peptide containing non-phosphorylated (N) or phosphorylated (P) tyrosine residue 2094 (peptide pulldown). After washing and elution, proteins bound to the peptide were separated by SDS-PAGE, and immunoblotting was performed with the anti-Grb2 antibody (IB). (C) HEK293 cells were transfected with WT plexin-B1 [(VSV)-plxB1], or the indicated Y/F mutants. Cells were starved and incubated without or with Sema4D (Sema4D −/+). Thereafter, WT or mutated plexin-B1 was immunoprecipitated from cell lysates using anti-VSV antibody. Precipitated complexes were separated by SDS-PAGE and proteins were visualized using indicated antibodies (IB). (D) HEK cells were transfected with (VSV)-plexin-B1 with wild-type (WT) or mutated (Y1356F) Met. Immunoprecipitation was carried as described above. Precipitated samples were separated by SDS-PAGE and proteins were visualized using indicated antibodies. (E) HEK cells were transfected with wild-type (WT) or mutated (N2096D), VSV-tagged plexin-B1 together with indicated clones. After treatment with control (−) or Sema4D (+), anti-VSV precipitations were carried out as described above.

To test whether Grb2 can interact with plexin-B1 in a cellular context, we expressed VSV-tagged plexin-B1 and Met in HEK293 cells. We found that Grb2 coimmunoprecipitated with wild-type plexin-B1 in a Sema4D-dependent manner (Fig. 3C). Surprisingly, mutation of the tyrosine residue 1864 shows no effect on plexin-B1–Grb2 interaction, whereas the mutation of the tyrosine residue 2094 abolished binding of Grb2 to plexin-B1 (Fig. 3C). In cellular context, stimulation with Sema4D results in phosphorylation and activation of plexin-B1 and Met. It has been shown that the phosphorylated tyrosine 1356 of Met provides a docking site for Grb2. To exclude the direct involvement of Met in the interaction between the plexin-B1–Met receptor complex and Grb2, we studied this interaction in the presence of wild-type Met or tyrosine 1356 to phenylalanine mutant of the receptor [Met(Y1356F)]. We observed that plexin-B1 coimmunoprecipitates with Grb2 in presence of both wild-type and Y1356F Met (Fig. 3D). It is known that asparagine in the +2 position of the phosphorylated tyrosine is crucial for binding the SH2 domain of Grb2 to its targets (McNemar et al., 1997). To confirm the specificity of the plexin-B1–Grb2 interaction, we analyzed the interaction between Grb2 and plexin-B1 asparagine to an aspartic acid mutant. We observed that, in contrast to the wild-type plexin-B1, plexin-B1 (N2096D) was unable to interact with Grb2 upon Sema4D stimulation (Fig. 3E). Thus, Sema4D-dependent, Met-mediated phosphorylation of tyrosine residue 2094 induces an interaction of plexin-B1 with Grb2.

Endogenous plexin-B1 and plexin-B2 interact with Grb2 upon activation by semaphorins

To further evaluate the Sema4D-dependent interaction of plexin-B1 with Grb2, we studied the human breast cancer cell line MDA-MB-468. MDA-MB-468 cells express both plexin-B1 and Met, and stimulation with Sema4D results in a Met-dependent phosphorylation of plexin-B1 (Swiercz et al., 2008). Upon stimulation of cells with Sema4D, endogenous plexin-B1 could be coimmunoprecipitated with endogenous Grb2 (Fig. 4A,B). Previously we showed that in MCF-7 cells expressing ErbB-2, plexin-B1 coimmunoprecipitates with Grb2 in a PLCγ1/2-dependent manner. In contrast to MCF-7 cells, siRNA-mediated knockdown of PLCγ1 in Met-expressing cells (MDA-MB-468) has no effect on plexin-B1–Grb2 interaction (Fig. 4A,B). This strongly suggests that Met-dependent interaction between plexin-B1 and Grb2 is direct and independent from the presence of PLCγ.

Fig. 4.

Grb2 interacts with B-family plexins. MDA-MB-468 cells or MCF-7 cells transfected with the control siRNA or siRNAs against Grb2 or PLCγ1/2 (PLCγ). 48 hours after transfection, cells were incubated in the absence or presence of 150 nM Sema4D (Sema4D −/+). Cells were then lysed, and the proteins interacting with plexin-B1 (A) or Grb2 (B) were co-immunoprecipitated (IP) using anti-plexin-B1 or anti-Grb2 antibodies. Lysates and immunoprecipitates were analyzed using the indicated antibodies (IB). (C,D) MDA-MB-468 cells were transfected with siRNAs as indicated, 48 hours later cells were incubated in the absence or presence of 150 nM Sema4D for 20 min. Active RhoA (C) or R-Ras (D) was precipitated from cell lysates as described in the Materials and Methods. Shown are autoluminograms of immunoblots analyzed with the indicated antibodies (IB).

Additionally, we tested whether PLCγ-independent interaction between plexin and Grb2 is also true for plexin-B2. Indeed, we found that upon stimulation with Sema4C, plexin-B2 interacts with Grb2 in MDA-MB-68 cells in a PLCγ-independent manner (Fig 4B). This supports the notion that Met-dependent phosphorylation of plexin-B family members regulates a direct interaction with Grb2.

Grb2 is required for Plexin-B1-induced RhoA deactivation but not R-Ras inactivation

Plexin-B-mediated RhoA inactivation requires interaction of plexin-B family members with the GTPase-activating protein p190 RhoGAP (Barberis et al., 2005). Consistent with previous data, we observed that the siRNA-mediated knockdown of p190 abolishes RhoA deactivation downstream of plexin-B1. Interestingly, we also found that knockdown of Grb-2 has an inhibitory effect on plexin-mediated RhoA deactivation (Fig. 4C). In contrast to the Sema4D-induced RhoA deactivation, the regulation of R-RasGAP activity of plexin-B1 did not depend on Grb-2 or p190, as indicated by the lack of any effect of knockdown of those proteins on Sema4D-induced inhibition of R-Ras (Fig. 4D). Consistently with previous data, we did not observe any effects of the knockdown of PLCγ1/2 on RhoA inactivation or R-Ras GAP activity downstream of plexin-B1 (Fig. 4C,D, respectively). Hence, the recruitment of Grb2 and p190 into the plexin-B1 receptor complex is necessary for RhoA deactivation but not for R-Ras function regulated by plexin-B1.

Analysis of the interaction between plexin-B1, p190 RhoGAP and Grb2

To obtain insight into the mechanism underlying the role of Grb2 in the p190-dependent RhoA inactivation through plexin-B1, Grb2 mutants lacking functional domains (Fig. 5A) were expressed in HEK293 cells. The removal of the SH2 domain abolished the interaction between Sema4D-activated plexin-B1 and Grb2, whereas, the deletion of both SH3 domains had no effect (Fig. 5B). In agreement with this result, the SH2 domain of Grb2 interacted with Met-phosphorylated plexin-B1 (Fig. 5B). Seemingly contrary, we observed that the removal of the SH2 domain of Grb2 did not influence the interaction between plexin-B1 and p190 (Fig. 5C), probably because of the presence of the intracellular Grb2 in these cells. In addition, we observed that the mutation of the C-terminal SH3 domain of Grb2 abolishes the interaction between plexin-B1 and p190 but not the interaction between plexin-B1 and Grb2 (Fig 5C and 5B, respectively). Consistent with previous results (Fig 3A), we found that the point mutation of the tyrosine residue 2094 (Y2094F) but not of tyrosine residue 1864 (Y1864F) abolishes the interaction between plexin-B1 and the SH2 domain of Grb2 (Fig 5D). Thus, the SH2 domain of Grb2 mediates the interaction with plexin-B1 phosphorylated on tyrosine residue 2094, whereas none of the other domains appears to be critically involved. Additionally, these data suggest that the C-terminal SH3 domain of Grb2 mediates the interaction between Grb2 and p190, establishing its role as a link between phosphorylated plexin-B1 and a p190 RhoGAP.

Fig. 5.

Grb2 mediates plexin-B1–p190-RhoGAP interaction. (A) Structure of the Grb2 mutants. (B,C,E) HEK293 cells were transfected with plexin-B1 [(VSV)-plxB1], Met, wild-type Grb2 or its mutant [(Myc)-Grb2]. Cells were incubated without or with 150 nM Sema4D, lysed, and the interaction between plexin-B1, p190 and Grb2 and its mutants was analyzed by immunoprecipitation. (D) Wild-type or Y/F mutants of [(VSV)-plxB1], plexin-B1 were co-transfected with the Myc-tagged SH2 domain of PLCγ1. Protein–protein interaction was analyzed by co-immunoprecipitation. Samples were then analyzed by immunoblotting using indicated antibodies (IB). (E) Activated RhoA was detected as described in the Materials and Methods. (F) HEK cells were transfected with Myc-tagged RhoAQ63L alone or together with other cDNAs and siRNAs as indicated. 48 hours after transfection, cells were starved, picked up using EDTA and replated on non-coated (NC) or fibronectin-coated (FI) surface. Transfected HEK293T cells were seeded on uncoated (U) or fibronectin-coated plates. Interaction between RhoA and p190 RhoGAP was studied using immunoprecipitation. Shown are autoluminograms of immunoblots of lysed (lysate) or precipitated (IP) samples.

To study the role of different Grb2 domains in mediating downstream effects of plexin-B1 signaling, we determined RhoA activity in transfected HEK293 cells (Fig. 5E). Whereas the expression of the SH2 domain of Grb2 fully abolishes RhoA inactivation in response to Sema4D, the removal of the SH2 domain did not influence plexin-B1 signaling (Fig. 5E). In addition, the removal of the C-terminal SH3 domain completely blocked plexin-B1-mediated RhoA inactivation (Fig. 5E). Interestingly, the Grb2 mutant lacking the N-terminal SH3 domain was still able to mediate the Sema4D-induced interaction between plexin-B1 and p190 RhoGAP and RhoA deactivation (Fig. 5C,E), thereby indicating the requirement of the C-terminal SH3 domain of Grb2 for the downstream signaling of plexin-B1.

Grb2 activates p190 RhoGAP

Our results suggest that Grb2 mediates the interaction between plexin-B1 and p190 RhoGAP to allow RhoA deactivation downstream of Sema4D. To test the possibility that the interaction between plexin-B and p190 results in an activation of the RhoGAP activity, we examined the interaction between p190 RhoGAP and a dominant-active form of RhoA – RhoAQ63L. It has been proposed that GAP proteins form stable complexes with dominant-active forms of small GTPases (Garcia-Mata et al., 2006) and that this interaction can be used to estimate the level of activation of GAP proteins. Indeed, we observed a stable interaction between p190 and dominant-active RhoA in HEK293 cells. The amount of RhoA(Q63L) precipitated with p190 RhoGAP greatly increased by stimulation with Sema4D in cells expressing plexin-B1 (Fig. 5F). In agreement with the previous data (Garcia-Mata et al., 2006), we saw an increase in p190–RhoA interaction after plating cells on fibronectin (Fig. 5F). Consistent with the important role of Grb2 in plexin-B–p190 signaling, we found that knockdown of Grb2 fully abolished stimulatory effects of Sema4D (Fig. 5F). This indicates that the Grb2 not only mediates the interaction between plexin-B1 and p190 RhoGAP but also increases its enzymatic activity, which is required for Sema4D-induced RhoA inactivation.

The SH2 domain of Grb2 interacts directly with a phosphorylated tyrosine 2094 of plexin-B1

Our data suggest a direct interaction between phosphorylated plexin-B1 and the SH2 domain of Grb2. In agreement with the previous results, we observed that the SH2 domain of Grb2 purified from bacteria interacted directly and specifically with the peptide containing phosphorylated tyrosine 2094 but not with a non-phosphorylated form and not with the peptides representing other tyrosine residue phosphorylated by Met (Fig. 6A). Additionally, we could not detect any interaction between the SH2 domain of Grb2 and peptides containing tyrosine residues that are phosphorylated by ErbB-2 (Swiercz et al., 2009) (Fig. 6A). Consistent with the previous results, we also found that, in contrast to the SH2 domain of Grb2, neither N- nor C-terminal SH2 domain of PLCγ1 are able to directly interact with phosphorylated tyrosine 2094, thereby indicating a specific requirement of the SH2 domain of Grb2 for a direct interaction with Met-phosphorylated plexin-B1.

Fig. 6.

Direct interaction between plexin-B1 and Grb2. (A) Peptides of plexin-B1 containing a non-phosphorylated (N) or phosphorylated (P) tyrosine residues 1707, 1732, 1864 and 2094 were bound to agarose beads and incubated with purified GST-tagged SH2 domain of Grb2 [(GST)-Grb2] for 2 hours. Bound proteins were separated by SDS PAGE and visualized by immunoblot with the anti-GST antibody. (B) Peptide containing a non-phosphorylated (N) or phosphorylated (P) tyrosine residue of plexin-B1 was incubated with bacteria-purified GST-tagged SH2 domains of Grb2 and PLCγ1 (as indicated) for 2 hours. Bound proteins were visualized using anti-GST antibody. (C) Purified wild-type or ΔSH3 Grb2 [(FLAG)-Grb2, (FLAG)-Grb2ΔSH3] were incubated with glutathione-S-Sepharose in the absence (−) or presence (+) of p190 RhoGAP (GST-190). Bound proteins were visualized using an anti-FLAG antibody. (D) MDA-MB-468 cells were transfected with FLAG-tagged Grb2 or its mutant. After stimulation with Sema4D, cells were lysed and protein complexes were precipitated using an anti-FLAG antibody. Bound proteins were visualized using indicated antibodies.

Grb2 interacts directly with p190 RhoGAP

Our data suggest an interaction between plexin-B1-bound Grb2 and p190 RhoGAP. To test whether Grb2 and p190 can directly interact in vitro and in tumor cells, we tested Grb2-p190 interaction using purified proteins and in the MDA-MB-468 cell line. Indeed, we observed that purified p190 RhoGAP could bind to purified Grb2 but not to its mutant lacking the SH3 domain (Fig. 6C). Additionally, we observed that in MDA-MB-468 cells, Sema4D-mediated interaction between p190 RhoGAP and Grb2 is abolished by the removal of the C-terminal SH3 domain of Grb2, thereby indicating a direct, SH3-domain-mediated interaction between p190 RhoGAP and Grb2.

Grb2 regulates plexin-B1-mediated inhibition of migration of MDA-MB-468 cells

The activation of the plexin-B1–Met receptor complex by Sema4D has been shown to exert antimigratory effects in various cells including MDA-MB-468 cells a process requiring the inactivation of RhoA (Swiercz et al., 2008). MDA-MB-468 cells react to stimulation with Sema4D with a decreased migration, a phenotype that is connected to the Met-mediated phosphorylation of plexin-B1 (Swiercz et al., 2008). To test whether Met-mediated interaction and activation between plexin-B1 and Grb2–p190 is a part of this cellular phenomenon, we analyzed the relevance of Grb2 for plexin-B1-mediated regulation of cell migration. The depletion of Grb2 or p190 RhoGAP, which are both expressed in MDA-MB-468 cells, abolished Sema4D effects on wound healing in MDA-MB-468 cells (Fig. 7A). The ability of cells to migrate was not generally affected by the lack of Grb2, as indicated by the insensitivity of FBS-stimulated wound healing to the knockdown of Grb2 or p190 RhoGAP (Fig. 7A).

Fig. 7.

Grb2 regulates Sema4D-mediaded inhibition of migration. (A,B) MDA-MB-468 control cells or cells stably transfected with a virus containing wild-type ErbB-2 were transfected with the indicated siRNAs. 48 hours after transfection, cell migration was analyzed using Transwell migration chambers (B) or a wound-healing assay (A). MDA-MB-468 cells were allowed to migrate in the absence or presence of 150 nM Sema4D (black and white bars, respectively) or in the presence of 10% FBS (grey bars), Migrated cells were counted, and data were expressed as mean values of triplicates. (C) MDA-MB-468 cells were transfected as indicated. Cells were allowed to migrate in the absence or presence of Sema4D. Migrated cells were fixed, and cell migration was calculated as described in the Materials and Methods. *P<0.05; ***P<0.001.

To again establish the link between Met-mediated phosphorylation and Grb2–p190-mediated deactivation of RhoA and the cellular effects mediated by the Sema4D, we stably transfected MDA-MB-468 cells with cDNA encoding wild-type ErbB-2 (see the Materials and Methods). The expression of the recombinant protein and the knockdown efficiency were shown by western blot (data not shown). Sema4D loses its ability to block cell migration in control cells treated with anti-Grb2 or anti-p190 or anti-Met siRNA (Fig. 7B and data not shown), whereas in the cells expressing ErbB-2 knockdown of Met together with Grb2 or p190 RhoGAP did not influence the ability of Sema4D to increase cellular migration.

Finally, we tested whether expression of the plexin-B1 mutant lacking the ability to bind to Grb2(Y2094F) in MDA-MB-468 cells influences Sema4D-mediated deactivation of cell migration. We found that plexin-B1(Y2094F), which lacks the ability to mediate Met-dependent RhoA deactivation, was unable to influence cell migration in a Sema4D-dependent manner. Taken together, these data indicate that the specific phosphorylation of plexin-B1 by Met but not ErbB-2 and subsequent interaction and activation of Grb2–p190 RhoGAP complex is crucial for the Sema4D-induced inhibition of cellular migration.

Discussion

The cellular effects of semaphorins are quite versatile, and in many cases it has been reported that the effects induced by a particular semaphorin might even be opposing, dependent on the cellular context. A possible mechanism underlying the versatility of semaphorin functions is the ability to interact with various receptor tyrosine kinases to form multimolecular receptor complexes. In recent years, it has been shown that the Sema4D receptor plexin-B1 can interact with various receptor tyrosine kinases including OTK/PTK-7, Met and ErbB-2 (Whitford and Ghosh, 2001; Giordano et al., 2002; Conrotto et al., 2004; Swiercz et al., 2004), and that Sema4D-mediated activation of the receptor complex results in phosphorylation of both plexin-B1 and the receptor tyrosine kinase. We have shown that plexin-B1-mediated activation and inactivation of the small GTPase RhoA requires ErbB-2 and Met, respectively, and that the reciprocal regulation of RhoA activity involves the phosphorylation of specific tyrosine residues of plexin-B1 (Swiercz et al., 2008). Furthermore, cellular effects mediated by Sema4D appear to be dependent on the identity of receptor tyrosine kinase present in the plexin-B1 receptor complex. The presence of ErbB-2 resulted in Sema4D-induced stimulation of migratory activity, whereas, opposite effects of Sema4D were observed in cells expressing Met (Swiercz et al., 2004; Swiercz et al., 2008). An additional level of complexity in plexin signaling can be observed at the level of different cell types, whereas in breast cancer cells, promigratory effects of Sema4D are clearly dependent on the presence of ErbB2 in the receptor complex (Worzfeld et al., 2012). The presence of Met in a complex with plexin-B1 has been shown, in addition to its antimigratory effects on breast cancer cells, to mediate promigratory and proinvasive effects in a variety of cell types, including liver progenitor cells, liver hepatocellular carcinoma, colon adenocarcinoma and pancreatic cell lines and, in some cases, neuronal migration (Giordano et al., 2002; Conrotto et al., 2004; Giacobini et al., 2008). These data suggest that an additional, yet unidentified, mechanism might control signaling downstream of plexin-B1.

There is much evidence that Sema4D binding to plexin-B1 induces RhoA activation through the RhoGEF proteins PDZ-RhoGEF and LARG, which stably interact with the C-terminus of plexin-B1 (Aurandt et al., 2002; Driessens et al., 2002; Swiercz et al., 2002). However, it has also been shown that plexin-B1 can mediate inhibition of RhoA via the Sema4D-dependent recruitment of p190 RhoGAP into the semaphorin receptor complex (Barberis et al., 2005). We confirmed that Sema4D-dependent plexin-B1 activation can lead to opposing effects on RhoA activity, and we demonstrate that the identity of the receptor tyrosine kinase determines the net effect on RhoA activity. Whereas ErbB-2 is required for RhoA activation, preferential interaction of plexin-B1 with Met results in the inhibition of RhoA (Swiercz et al., 2004; Swiercz et al., 2008). We have shown that the activation of the plexin-B1–ErbB-2 receptor complex results in the phosphorylation of tyrosine residues 1708 and 1732 of plexin-B1, thereby providing docking sites for the SH2-domains of PLCγ (Swiercz et al., 2009). After recruitment of PLCγ into the plexin-B1 receptor complex, PLCγ, via its SH3 domain, mediates the activation of PDZ–RhoGEF–LARG. This PLCγ-dependent signaling mechanism is required for Sema4D-induced, plexin-B1-mediated regulation of cell migration and neuronal growth cone morphology (Swiercz et al., 2009).

It is currently not known how p190 RhoGAP activity is regulated by plexin-B1. There is, however, evidence suggesting that the ligand-dependent association between plexin-B1 and p190 RhoGAP is an important step (Barberis et al., 2005). Additionally, it has been shown that p190 RhoGAP can be activated by direct tyrosine phosphorylation (Chikumi et al., 2002); however, we did not observe tyrosine phosphorylation of p190 RhoGAP in response to Sema4D stimulation (data not shown). Here we have shown that upon activation by Sema4D, the intracellular tyrosine residue 2094 of plexin-B1 becomes phosphorylated by Met and serves as a docking site for the SH2 domain of Grb2. Grb2 is recruited to a plexin receptor complex and mediates an interaction and activation of p190 RhoGAP. This is a requirement for plexin-B-mediated RhoA inactivation, as well as for downstream cellular effects of Sema4D, such as plexin-B1-mediated antimigratory effects.

The tyrosine residue 2094 of plexin-B1 is conserved among members of the plexin-B family. Our data show that plexin-B2 also interacts with Grb2 upon activation by its ligand, semaphorin 4C. Thus, the observed tyrosine-phosphorylation-dependent recruitment of Grb2 is likely to be a common mechanism of plexin-B family members. The phosphorylation of plexin-B1 by ErbB-2 was not affected by the mutations of Y2094, indicating that Erb-2 and Met regulate the function of plexin-B1 by phosphorylation of different tyrosine residues of plexin-B1. This is consistent with the observation that the phosphorylation of plexin-B1 by ErbB-2 and Met induces different and sometimes opposing downstream signaling and cellular effects (Swiercz et al., 2008). Phosphorylation of tyrosine residues results in the formation of docking sites for SH2 and PTB-domain-containing proteins (Lim and Pawson, 2010). Using a peptide-based approach, we identified the SH2-domain-containing protein Grb2 as an interaction partner of Met-phosphorylated plexin-B1. The analysis of amino acid sequences surrounding tyrosine residue 2094 of plexin-B1, ALHELYKYINKYYDQ, shows that the region surrounding tyrosine 2094 fits the consensus sequence recognized by the SH2 domain of Grb2 (pY-x-Asparagine) (Gay et al., 1997).

Our immunoprecipitation data indicate that Met-phosphorylated plexin-B1 can interact with Grb2. Interestingly, we previously identified Grb2 as an interaction partner of ErbB-2-phosphorylated tyrosine 1708 (Giubellino et al., 2008; Swiercz et al., 2009). However, in contrast to our previous findings, we observed that PLCγ1/2 are not required for plexin-B1–Grb2 interaction in Met-expressing cells. This fact indicates that Grb2 directly interacts with Met-phosphorylated plexin-B1. Indeed, we were able to observe a strong interaction between phosphorylated peptide containing tyrosine 2094 of plexin-B1 and the purified SH2 domain of Grb2.

The classic paradigm of signaling through Grb2 involves binding of the SH2 domain to activated receptor tyrosine kinases, and subsequential binding of the downstream signaling partners through the SH3 domains (Giubellino et al., 2008), the process including both SH3 domains of Grb2, which might interact simultaneously with different sites on a single target molecule, thereby increasing both the apparent affinity and selectivity of Grb2-target interaction (Mayer, 2001). By contrast, we observed that the N-terminal SH3 domain of Grb2 is dispensable for its interaction with p190 RhoGAP. Thus, our data identify a novel mode of interaction between Grb2 and their downstream signaling partners, leaving at the same time the possibility that the N-terminal SH3 domain might play an as yet unknown function in plexin-B signaling.

It has been shown that p190 RhoGAP might be activated by a variety of processes including tyrosine phosphorylation via Src or direct interaction with small GTPases of the Rnd family (Fincham et al., 1999; Wennerberg et al., 2003). By contrast, we did not observe tyrosine phosphorylation of p190 RhoGAP in response to Sema4D, suggesting that tyrosine phosphorylation is not required for p190-RhoGAP-mediated RhoA inactivation. Another interesting possibility is activation of p190 RhoGAP through direct binding with Rnd proteins, indeed Rnd1–Rnd3 were shown to directly bind members of plexin-B family and are required for plexin-mediated intrinsic GAP activity toward M-Ras and R-Ras (Oinuma et al., 2003; Ridley et al., 2003; Oinuma et al., 2004b; Puschel, 2007; Saito et al., 2009). Our data, however, clearly show, that in contrast to the intrinsic GAP activity of plexin-B1, the expression of Rnd proteins is not necessary for p190-RhoGAP-mediated Rho inactivation. Additionally, we observed that knockdown of Grb2 or p190 RhoGAP had no effect on plexin-mediated RasGAP activity. Thus indicating that plexin-B1 GAP activity towards M-Ras/R-Ras and RhoA are regulated differentially.

Cell migration is a complex process that can be regulated at various levels. It involves the detachment of adhesions, the formation of polarized cellular protrusions, the formation of new adhesive structures, as well as actomyosin-based cell body contraction (Ridley et al., 2003). RhoA has been shown to play key role in the regulation of cellular migration (Burridge and Wennerberg, 2004; Raftopoulou and Hall, 2004). It has been shown recently that RhoA activity is particularly high in the leading edge as well as in the retracting tail of migrating cells (Pertz et al., 2006). Sema4D–plexin-B1-mediated signaling might influence cell migration in a dual fashion. In endothelial cells, activation of plexin-B1 has promigratory effects, which require the activation of RhoA and the PI3K–Akt pathway (Basile et al., 2004; Basile et al., 2005). Under certain conditions, Sema4D inhibits cell migration through the inhibition of integrin function, a process that involves R-RasGAP activity of plexin-B1 (Oinuma et al., 2004b). In breast carcinoma cells expressing either ErbB-2 or Met, we observed that Sema4D had promigratory or antimigratory effects, respectively (Swiercz et al., 2008). This is consistent with the observation that the promigratory effects of Sema4D on endothelial cells were independent of Met (Basile et al., 2004). Our data show that Sema4D exerts different biological activities as a result of the differential association of its receptor, plexin-B1, with the receptor tyrosine kinases Met and ErbB-2. We observe that Sema4D-mediated, Met-dependent inhibition of a migration of breast carcinoma cells is dependent on the presence of Grb2 and p190 RhoGAP, further supporting the regulatory role of RhoA inactivation in cellular processes involving the plexin-B1–Met receptor complex.

In summary, we show that the activation of the plexin-B1–Met receptor complex results in the phosphorylation of specific tyrosine residues of plexin-B1, thereby providing docking sites for the SH2 domain of Grb2. After recruitment of Grb2 into the plexin-B1 receptor complex, Grb2 via its SH3 domain, mediates the interaction with p190 RhoGAP and its subsequent activation. This Grb2-dependent signaling mechanism is required for Sema4D-induced, plexin-B1-mediated regulation of cell migration. These data identify an important new component of plexin-B-mediated signaling and fill the gap between Met-mediated phosphorylation of plexin-B1 and its ability to deactivate RhoA via p190 RhoGAP.

Materials and Methods

Antibodies and reagents

The following antibodies were used: goat polyclonal anti-plexin-B2 (Santa Cruz Biotechnology, Heidelberg); rabbit polyclonal anti-GST and rabbit polyclonal anti-Myc (Sigma-Aldrich, Munich); rabbit polyclonal anti-PLCγ1, anti-Grb2, anti-RhoA, anti-R-Ras and mouse monoclonal Met antibodies (Cell Signaling Technology, Frankfurt); mouse monoclonal anti-phosphotyrosine (4G10) (Millipore, Schwalbach); mouse monoclonal anti-Rac (23A8) (Invitrogen, Karlsruhe); goat polyclonal anti-plexin-B1 and mouse monoclonal anti-plexin-B1 (R&D Systems, Wiesbaden-Nordenstadt); goat polyclonal anti-VSV (Thermo, Bonn); mouse monoclonal anti-p190 RhoGAP (BD, Heidelberg).

Plasmids

Eukaryotic expression plasmids carrying the cDNAs of PDZ-RhoGEF, Sema4D, (Myc)-RhoA, (GST)-RhoA, Met, Rnd1 and plexin-B1 Y/F mutants were described previously (Swiercz et al., 2002; Swiercz et al., 2004; Swiercz et al., 2008; Swiercz et al., 2009); (VSV)-plexin-B1 was kindly provided by Luca Tamagnone (University of Torino, Torino, Italy); Met (Y1356F) was kindly provided by Silvia Giordano (University of Torino, Torino, Italy). R-Ras was provided by Monique Dail (The Burnham Institute, La Jolla, CA). GST-RacQ61L, Grb2 and p190 RhoGAP were provided by Dominique Brandt (University of Marburg, Marburg, Germany). Human (GST)-Grb2 and (Myc)-Grb2 mutants lacking amino acids 1–58 (ΔSH3.N), 60–152 (ΔSH2), 156–215 (ΔSH3.C ), and mouse (FLAG)-Grb2 or its mutant lacking amino acids 156–217 (ΔSH3) were generated using standard molecular biology methods. Bacterial expression constructs containing SH2 domains of PLCγ1 and Grb2 (amino acids 532–635, 646–735 and 156–215 respectively) and constructs containing a full length mouse Grb2 (amino acids 1–217), its mutant (amino acids 1–155) or C-terminal portion of p190 RhoGAP (amino acids 1240–1499) were generated using standard molecular biology methods.

Cell culture, immunoprecipitation studies and transfection

HEK293, MCF-7 and MDA-MB-468 cells were cultured and immunoprecipitations were performed as described previously (Swiercz et al., 2008), HEK293 cells were transfected using the calcium phosphate method, and cells were transfected with siRNAs using HiPerFect reagent according to the manufacturer's instructions (Qiagen, Hilden). Cell migration was assayed using polystyrene Transwell inserts (Greiner, Frickenhausen) with pore sizes of 0.8 µm as described (Swiercz et al., 2008), alternatively cell migration shown in Fig. 7C was analysed using ImageJ software (Version 1.46e). The area of Toluidine Blue stain was counted after applying a fixed threshold on scanned migration filters. The total area of toluidine blue stain was estimated using the ‘analyze particles’ function. Wound-healing assays and scratch assays were performed as described previously (Swiercz et al., 2008).

Determination of activated RhoA and R-Ras

The amount of activated cellular RhoA and R-Ras were determined by precipitation with a fusion protein, consisting of GST and the Rho-binding domain of Rhotekin (GST-RBD) or the Ras-binding domain of Raf1 (GST-Raf1) as described previously (Ren and Schwartz, 2000; van Triest and Bos, 2004).

Small interfering RNAs (siRNAs)

The target sequence for siRNAs specific to ErbB-2, Met, PLCγ1/2 and Grb2 used in this study were published previously (Faltus et al., 2004; Swiercz et al., 2008). Target sequence of siRNAs specific for p190 RhoGAP were: 5′-CAGGATGTTCTGGGAGAGGAA-3′ and 5′-AAGGTGTTGAGCGGTACATTA-3′. AllStars Negative Control siRNA was purchased from Qiagen (Hilden).

Production and purification of recombinant Sema4D

Generation of Lec3.2.8.1 CHO cells expressing the extracellular part of human Sema4D (residues 1–657 followed by a lysine residue and a C-terminal histidine tag for purification) was described previously (Love et al., 2003).

Peptide synthesis and isolation of binding partners

Peptides C-VLRENQDYVPGERT and C-ALHELYKYINKYYDQ, corresponding to residues 1857–1871 and 2087–3001 of plexin-B1, respectively, were synthesized in the Analytical Laboratory of the University Hospital in Düsseldorf (Germany). A phosphotyrosine was inserted in positions corresponding to amino acid residues 1864 and 2094 of plexin-B1 in order to obtain phosphorylated peptides. Peptides were coupled to the SulfoLink Coupling Resin (Thermo, Bonn) and the non-specific binding was reduced by incubation with 50 mM cysteine. HEK293 cells were lysed in RIPA buffer, and cleared lysates were incubated with the peptide-coupled SulfoLink for 2 hours at 4°C. The resin was then extensively washed, boiled in Laemmli buffer, and proteins bound to peptides were separated using SDS-PAGE. Proteins were visualized using Coomassie Blue staining.

Mass spectrometry

Mass spectrometry analysis was performed in the Core Facility for Mass Spectrometry and Proteomics, University of Heidelberg, Germany. Spots were excised from gels and digested with trypsin. Matrix-assisted laser desorption ionization of quadruple time-of-flight mass spectrometry (MALDI-TOF MS) was performed using ultraflex TOF (Bruker Daltonik, Bremen). For protein identification, the peptide mass fingerprint was run against the NCBI-nr database using Mascot (Matrix Science, London, UK).

Protein purification

Proteins were produced and purified in Escherichia coli strain DE3. GST fusion proteins were eluted from GSSH-beads (GE Healthcare, Munich) followed by gel filtration chromatography and were then concentrated by Amicon Ultracentrifugal filters (Millipore, Schwalbach).

Statistical analysis

Statistical significance was evaluated by Student's t-test.

Acknowledgments

The authors thank Evelyne Jones (London) for kindly providing Lec3.2.8.1 CHO cells expressing part of human Sema4D. We also thank H.-P. Gensheimer and D. Magalei for technical assistance and S. Huemmer for help with the preparation of the manuscript.

Footnotes

  • Funding

    J.M.S. and T.S. are supported by the German Research Foundation [grant number SW 148/1-1]. R.K. received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

  • Accepted March 21, 2012.

References

View Abstract