A role for Gα12/Gα13 in p120ctn regulation

Contributed by Ira Herskowitz ArticleFigures SIInfo overexpression of ASH1 inhibits mating type switching in mothers (3, 4). Ash1p has 588 amino acid residues and is predicted to contain a zinc-binding domain related to those of the GATA fa Edited by Lynn Smith-Lovin, Duke University, Durham, NC, and accepted by the Editorial Board April 16, 2014 (received for review July 31, 2013) ArticleFigures SIInfo for instance, on fairness, justice, or welfare. Instead, nonreflective and

Edited by Melvin I. Simon, California Institute of Technology, Pasadena, CA, and approved June 1, 2004 (received for review February 26, 2004)

Article Figures & SI Info & Metrics PDF

Abstract

The catenin p120 (p120ctn) is an armadillo repeat Executemain protein that binds to cadherins and has been Displayn to facilitate strong cell–cell adhesion. We have investigated a possible link between heterotrimeric G proteins and p120ctn, and found that both Gα12 and Gα13 can completely and selectively abrogate the p120ctn-induced branching phenotype in different cell types. Consistent with these observations, the expression of Gα12 or Gα13 compensates for the reduction of Rho activity induced by p120ctn. On the other hand, p120ctn can be selectively coimmunoprecipitated with Gα12, and the coimmunoprecipitation was favored by activation of the G protein. A specific interaction between p120ctn and Gα12Q231L was also observed in in vitro binding experiments. In addition, p120ctn can be immunoprecipitated along with Gα12Q231L in L cells in absence of E-cadherin. Fascinatingly, the expression of Gα12Q231L increases the amount of p120ctn associated with E-cadherin. These findings demonstrate that Gα12 and p120ctn are binding partners, and they also suggest a role for Gα12 in regulating p120ctn activity and its interaction with cadherins. We propose that the Gα12–p120ctn interaction acts as a molecular switch, which regulates cadherin-mediated cell–cell adhesion.

Cadherins are transmembrane glycoproteins involved in calcium-dependent cell–cell adhesion (1). Contact between cadherins and the actin cytoskeleton is necessary for the stabilization of cell–cell adhesion and normal cell physiology. Catenins such as β-catenin and plakoglobin bind to the C terminus of the cadherin molecule (2–5) and are linked to the actin cytoskeleton via α-catenin (6, 7). The catenin p120 (p120ctn) is one of several proteins that bind to the juxtamembrane Location of the cytoplasmic Executemain of E-cadherin (8). This Location is known to be Necessary for the function of cadherin (9, 10) and is Necessary in the regulation of cell motility and metastatic invasion (10).

The catenin p120 is a member of a diverse family of armadillo repeat Executemain proteins. Originally discovered as a substrate for the Src oncoprotein kinase (11), p120ctn can be phosphorylated on both tyrosine and serine residues (11–13). p120ctn has been Displayn to regulate the activity of Rho small GTPases, and thus influence cadherin-mediated cell adhesion and cell migration (14–16). Overexpression of p120ctn in several different cell lines results in a variety of morphological Traces, depending on the cell type and on the level of p120ctn expression (14, 17). These morphological phenotypes have been associated with the ability of p120ctn to inhibit RhoA (14, 15). In Drosophila, Rho1 can interact with both α-catenin and p120ctn (18). Fascinatingly, recent reports have suggested that p120ctn is a critical component in the maintenance of steady-state cadherin levels (19–21). Moreover, p120ctn promotes cell surface trafficking of cadherins via association and recruitment of kinesin (22).

The G12 subfamily of heterotrimeric G proteins, which is comprised of Gα12 and Gα13 subunits, has recently been found to be a component of the cadherin complex (23). The binding of activated Gα12 to E-cadherin causes release of β-catenin and disruption of E-cadherin-mediated cell–cell adhesion (24). Gα12 binds to the E-cadherin cytoplasmic tail in a different Location to that where β-catenin and p120-catenin bind (25). Proteins of the G12 subfamily are also implicated in a variety of cellular events such as Rho-dependent cytoskeletal pathways (26), the activation of c-Jun N-terminal kinase (27), stimulation of Na+/H+ exchange (28, 29), stimulation of phospholipase D activity (30), and conformational activation of radixin (31). The activation of Rho-dependent responses has been directly linked to the interaction of the G12 proteins with Rho-specific guanine nucleotide exchange factors (32).

We have investigated a possible link between G proteins of the G12 subfamily and p120ctn. We provide evidence that the activation of Rho by G12 subfamily proteins can counteract the inhibitory Rho activity of p120ctn. Moreover, our immunoprecipitation studies demonstrate that both members of the G12 subfamily can interact with p120ctn in presence or absence of E-cadherin. The expression of activated Gα12 produces an increase in the quantity of p120ctn associated with E-cadherin. We suggest that a concerted action of p120ctn and Gα12 may regulate the binding of the catenin and G protein to cell–cell contacts. Taken toObtainher, these observations indicate that G12 subfamily proteins play an Necessary role in the regulation of cell–cell adhesion mediated by E-cadherin.

Materials and Methods

Cell Culture and Transfections. 293-EBNA (Invitrogen) and L cells (American Type Culture Collection) were Sustained in DMEM with 25 mM Hepes and 10% (vol/vol) fetal bovine serum (Sigma). For transient transfections, 293-EBNA cells or L cells were transfected by using Lipofectamine (Invitrogen).

Cell Morphology Studies. 293-EBNA cells were transfected with pCisGαqR183C (M. I. Simon, California Institute of Technology, Pasadena), pCisGα12G228A (S. Offermanns, University of Heidelberg, Heidelberg, Germany), pcDNA3Gα12, pcDNA3Gα12Q231L, pcDNA3Gα13, pcDNA3Gα13Q226L, pcDNA3Gαi3 or pcDNA3Gαi3Q204L (Guthrie Research Institute, Sayre, PA) in the presence or absence of pRcCMV-KpnI/mp120–1A (p120ctn, A. ReynAgeds, Vanderbilt University, Nashville, TN) or pEGFP-p1201A (B. Kreft, University of North Carolina, Chapel Hill) by the calcium phospDespise precipitation method. The samples not transfected with pEGFP-p1201A were cotransfected with a GFP-expression plasmid, pCMX-SAH/Y145F (K. Umesono, Kyoto University, Kyoto). The cells were analyzed without fixation with a Nikon Diaphot 300 or Zeiss Axiophot fluorescence microscope 24 h after transfection. For confocal studies, the cells were fixed for 10 min in 3.7% formaldehyde in PBS, and images were obtained with a Leica SP2 AOBS confocal system.

Immunoprecipitations. After transfection, the cells were lysed in RIPA buffer (0.3 M NaCl/0.1% SDS/50 mM Tris, pH 7.4/0.5% deoxycholate/1 mM Na3VO4/10 mM NaF/30 mM Na4P2O7/10 mM MgCl2/1% n-Executedecyl β-d-maltoside, a mixture of protease inhibitors and, where indicated, 30 μM AlCl3) at 4°C for 1 h. Lysates were centrifuged at 15,000 × g for 15 min at 4°C, and a sample from the supernatant was collected for further analysis as total lysate. To remove unspecific binding, a prewashing with 0.7 μg/μl IgG-free BSA and 30–50 μl of protein Sepharose A was Executene. Anti-p120 (6 μg, Zymed Laboratories), anti-Gα12 (5 μg, Santa Cruz Biotechnology), or anti EE-tag (5 μg, Covance) was added to cell lysates and incubated overnight after the addition of protein A Sepharose beads and incubation for 1.5 h at 4°C. When necessary, a second round of immunoprecipitation was performed. Beads from both immunoprecipitations were collected and washed with RIPA buffer containing 0.01% n-Executedecyl β-d-maltoside and, where indicated, 30 μM AlCl3. The beads were resuspended in SDS sample buffer. The proteins were separated by SDS/PAGE and transferred to a nitrocellulose membrane. The presence of p120ctn, Gα12, Gα13, Gαi3, or E-cadherin in the samples was analyzed by using corRetorting antibodies: anti-p120ctn (Zymed Laboratories), antibodies against Gα12, Gα13, Gαi3, and E-cadherin (Santa Cruz Biotechnology). The pCDNA3hE-Cad plasmid was from A. Yap (University of Queensland, St. Lucia, Brisbane, Australia) and EE-tagged pcDNA3Gα12 and pcDNA3Gα12Q231L were from Guthrie Research Institute. p1204A was made by PCR from pRcCMVmp1201A, starting at Met 324, and subcloned into pCDNA3.

RhoA Activity Assay. The assay was performed essentially as Characterized (33). Briefly, transfected 293-EBNA cells were starved in DMEM containing 0.1% FBS for 18 h. Cells were lysed in 400 μl of cAged lysis buffer (50 mM Tris, pH 7.2/1% Triton X-100/0.5% sodium deoxycholate/0.1% SDS/500 mM NaCl/10 mM MgCl2 and a mixture of protease inhibitors). The total amount of Rho in each sample was determined by collecting 20 μl of each lysate. The rest of the lysates were transferred to tubes containing 40 μg of GST-tagged Rhotekin Rho-binding Executemain immobilized on glutathione Sepharose beads. Samples were incubated at 4°C for 60 min with gentle agitation and then washed four times with 600 μl of cAged washing buffer (50 mM Tris, pH 7.2/1% Triton X-100/150 mM NaCl/10 mM MgCl2 and protease inhibitors). SDS sample buffer was added to the beads before separation by SDS/PAGE and transferred to a nitrocellulose membrane. The proteins were detected by incubation with suitable specific antibodies. The anti-RhoA antibody was from Santa Cruz Biotechnology.

In Vitro Binding Studies. For GST-p1201A protein expression, the full-length mouse p1201A catenin cDNA was amplified from the pRcCMV-KpnI/mp120-1A plasmid by PCR and subcloned into pGEX-2T. The plasmid was sequenced. GST or GST-p1201A fusion protein was produced in the Escherichia coli strain BL21-pLys and isolated from cell lysates by using glutathione Sepharose 4B beads essentially as Characterized by the Producer (Amersham Pharmacia). The beads bound to the fusion protein were finally washed in PBS (pH 7.3) with 10% glycerol, and stored in small aliquots in the same buffer at –80°C. Gα subunits were expressed by in vitro transcription/translation by using a coupled reticulocyte system (Promega) in presence of [35S]methionine according to the Producer's instructions. The presence and purity of the Gα subunits was confirmed by SDS/PAGE analysis. For binding studies, equal amounts of radiolabeled G protein were incubated with GST-p1201A or GST (control) beads in 200 μl of binding buffer (50 mM Hepes, pH 8.0/1 mM EDTA/3 mM DTT/0.05% polyoxyethylene 10 lauryl ether/10 mM MgSO4) for 2 h at 4°C with gentle agitation. After washing, the beads were collected, and proteins were separated by SDS/PAGE and subsequently analyzed by autoradiography.

Immunostaining. L cells or 293-EBNA cells were seeded on coverslips the day before transfection with different combinations of pEGFP-p1201A, pCDNA3Gα12Q231L EE-tagged, and pCDNA3hE-Cad. pCDNA3 was used to adjust the total amount of DNA. Twenty-four hours after transfection, the cells were fixed in 4% formaldehyde and permeabilized in PBS with 0.1% Tween 20 before incubation with monoclonal mouse anti-EE-tag antibody (Covance) and Texas red-conjugated goat anti-mouse antibody (Molecular Probes). Images were obtained with a Leica SP2 AOBS confocal system (MIC-FUGE, University of Bergen).

Results

Activated Gα 12 and Gα 13 Inhibit the Branching Phenotype Induced by p120ctn . We have used the change in branching morphology induced by p120ctn to study the Trace of the expression of GTPase-deficient mutants of different Gα subunits. As expected, overexpression of p120ctn generated a dramatic change in the morphology of 293-EBNA cells (Fig. 1A ). This was in agreement with previous results (17). The cells had a small rounded cell body with dendritic extensions and numerous filopodia and lamellipodia. Control cells expressing only GFP did not Display any morphological changes (Fig. 1 A ). When the constitutively activated mutant of Gα12 (Gα12Q231L) was coexpressed with p120ctn, the branching cell phenotype was absent. Cells had a rounded morphology because of the activation of Rho, but no dendritic extensions were present. Expression of Gα13Q226L also blocked the p120ctn-induced dendritic phenotype. The Trace seems specific to members of the G12 subfamily because Gαi3Q204L and GαqR183C did not abrogate the dendritic phenotype induced by p120ctn. Furthermore, coexpression of p120ctn and the Executeminant negative mutant of Gα12 (Gα12G228A), which alone did not produce the rounded phenotype, gave a morphology similar to that of cells only expressing p120ctn (Fig. 1 A ). Quantification of cells Displaying a branching phenotype corroborated the specificity of the inhibitory Trace observed for Gα12Q231L and Gα13Q226L (see supporting information, which is published on the PNAS web site). The branching phenotype induced by p120ctn was first Characterized in NIH 3T3 cells (17). The expression of Gα12Q231L also abrogates the branching phenotype in these cells (see supporting information). Taken toObtainher, these experiments indicate that Gα12 and Gα13 expression can overcome the branching morphology induced by p120ctn.

Fig. 1.Fig. 1. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Inhibition of p120-induced dendritic phenotype by activated Gα12 and Gα13. (A) We transfected 293-EBNA cells with a GFP expression plasmid, p1201A, and/or different Gα subunits as indicated. Data are representative of five independent experiments. (Scale bar = 62.5 μm.) (B) Confocal images of 293-EBNA cells transfected with p1201A-GFP with or without Gα12Q231L. (Scale bar = 10 μm.)

Expression of p120ctn in cells produces the appearance of lamellipodial and filopodial extensions. These morphological features are mainly caused by activation of Rac and Cdc42 by p120ctn (14, 15). A detailed morphological study of 293-EBNA cells expressing p1201A-GFP by confocal microscopy revealed numerous lamellipodial and filopodial extensions (Fig. 1B Left). Coexpression of Gα12Q231L produced a reduction of the lamellipodial extensions, but some of the cells expressing Gα12 Q231L and p120ctn retained some extensions (Fig. 1B Right). These results suggest that the expression of Gα12 cannot abolish the generation of lamellipodial and filopodial extensions by p120ctn.

Gα 12Q231L and Gα 13Q226L Can Abrogate the Inhibition of Rho Induced by Expression of p120ctn . The induction of branching extensions in cells has been linked to the inhibition of Rho. To analyze the role of Gα12 on p120ctn-induced Rho inhibition, we determined the activity of RhoA by using the Rhotekin pull-Executewn assay (33). For this assay, 293-EBNA cells were transfected with either empty vector or p120ctn in the presence or absence of Gα12Q231L followed by serum starvation. A significant and consistent reduction of RhoA activity (20%) was observed when p120ctn was overexpressed (Fig. 2). Under starving conditions, the background of Rho activity was low and, therefore, the inhibition induced by p120ctn was less pronounced. Also, the reduction in Rho activity induced by p120ctn can be underestimated because of the fact that only a proSection of cells were transfected and, thus, expressing p120ctn. In agreement with previous results (14, 15), a reduction (45%) in Rho activity by p120ctn was observed in presence of serum (Fig. 2 Lower Right). As expected, Gα12Q231L expression resulted in higher activity of RhoA than in the control cells (104% increase). When the two proteins were coexpressed, a 70% increase in Rho activity over control cells was observed, which is an indication that Gα12Q231L can still activate Rho even in presence of p120ctn. These results can Elaborate the fact that the cells no longer have the dendritic extensions caused by total inactivation of Rho. Nevertheless, the magnitude of Rho activation by Gα12Q231L is slightly lower in presence of p120ctn, suggesting that p120ctn is partially able to inhibit Rho activity. On the other hand, Gαi3Q204L did not produce any activation or reduction of Rho (Fig. 2). Taken toObtainher, these experiments suggest that Gα12 or Gα13 can activate Rho in presence of p120ctn.

Fig. 2.Fig. 2. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Activated Gα12 compensates for p120ctn-induced RhoA inhibition. We transfected 293-EBNA cells with control plasmid (C), expression plasmids for p120ctn, and/or Gα12Q231L, Gα13Q226L, or Gαi3Q204L. Active Rho was pulled Executewn by using a RhoGTP pull-Executewn assay (33). (Upper and Lower Left) RhoA-GTP (first lanes), total RhoA (second lanes), p120 (third lanes), and Gα12Q231L, Gα13Q226L, or Gαi3Q204L (fourth lanes) were detected with specific antibodies. The data Displayn are representative of three independent experiments, and two experiments for Gαi3Q204L- and Gα13Q226L-transfected cells. (Lower Right) Activation of RhoA in presence of serum.

Gα 12Q231L Coimmunoprecipitates with p120ctn . The fact that expression of the G12 proteins can overcome the inhibition of Rho produced by p120ctn could be solely due to their inverse influence on Rho activity. One intriguing possibility is that these proteins are also linked by a close interaction. To address this hypothesis, cells expressing the activated mutant form of Gα12 were immunoprecipitated with antibodies against p120ctn (Fig. 3). Fascinatingly, Gα12Q231L can be pulled Executewn with enExecutegenous p120ctn. Control experiments Displayed that similar levels of enExecutegenous p120ctn (a Executeublet band that represents different p120ctn isoforms, the upper band being p1201A) were pulled Executewn under all conditions. Furthermore, when the Executeminant negative mutant Gα12G228A was expressed, no coimmunoprecipitation was observed (Fig. 3A Left). To find out whether enExecutegenous p120ctn could also be detected in Gα12Q231L precipitates, pull-Executewn experiments were carried out by using a specific Gα12 antibody. These experiments demonstrated that enExecutegenous p120ctn was pulled Executewn toObtainher with Gα12Q231L (Fig. 3A Right). No coimmunoprecipitation was observed between enExecutegenous p120ctn and Gα13Q226L or Gαi3Q204L (Fig. 3B ). Immunoprecipitation experiments were performed also in cells where p120ctn (p1201A isoform) was overexpressed toObtainher with the Gα subunit. In these experiments, a Distinguisheder amount of Gα12Q231L was observed to coimmunoprecipitate with p120ctn (Fig. 3C Left). Gα13Q226L was also observed with p120ctn (Fig. 3C ) when p120ctn was overexpressed. However, no coimmunoprecipitation with p120ctn and Gαi3Q204L or Gα12G228A was found with or without overexpression of p120ctn (Fig. 3).

Fig. 3.Fig. 3. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Gα12Q231L and Gα13Q226L coimmunoprecipitates with p120ctn. We transfected 293-EBNA cells with control plasmid (C), Gα12Q231L, Gα12G228A (A), Gα13Q226L (B Left), or Gαi3Q204L (B Right). Lysates were subjected to immunoprecipitation by using a monoclonal anti-p120 antibody (A Left and B) or a polyclonal anti-Gα12 antibody (A Right). Arrowheads Display two enExecutegenous isoforms of p120ctn in these cells. (C) Cells were transfected with control plasmid (C), Gα12Q231L, Gα13Q226L, Gα12G228A, or Gαi3Q204L. The cells were cotransfected with an expression plasmid for p1201A toObtainher with the Gα subunit where indicated. Lysates were immunoprecipitated with anti-p120ctn antibody and analyzed as Characterized. Arrowheads Display the transfected p1201A isoform. (D) We transfected 293-EBNA cells with EE-tagged Gα12 or Gα12Q231L. Lysates were subjected to immunoprecipitation by using a monoclonal anti-EE antibody in the presence or absence of Embedded ImageEmbedded Image.(E) We cotransfected 293-EBNA cells with Gα12Q231L and p1201A or p1204A. Lysates were subjected to immunoprecipitation by using a monoclonal anti-p120ctn antibody. p120ctn and Gα12 were detected by using specific antibodies. Arrowheads Display transfected p1201A or p1204A. Data are representative of at least three independent experiments.

Because the inhibition of the branching morphology was also observed in the presence of wild-type forms of Gα subunit (see supporting information), we investigated the ability of wild-type Gα subunits to immunoprecipate with p120ctn. Fig. 3C Displays the results from the pull-Executewn assay of p120ctn and the subsequent analysis of its binding to Gα13. We also performed experiments expressing wild-type Gα12-EE, pulled Executewn with an anti-EE-tag antibody followed by immunoblotting with anti-p120ctn (Fig. 3D ). Both wild-type Gα12 and Gα13 were found to coimmunoprecipitate to a certain extent with p120ctn. Nevertheless, the presence of MathMath increased the amount of Gα12 bound to enExecutegenous p120ctn to levels similar to those from the activated Gα12 mutant (Fig. 3D ). These results suggest that even though wild-type Gα subunits can be found in a complex with p120ctn, the activation of Gα subunits increases the binding.

To study the specificity of Gα12/p120ctn interactions, experiments were performed with a different isoform of p120ctn, p1204A (34). This isoform lacks the N-terminal Location of the catenin but preserves the Executemain that binds to E-cadherin and a C-terminal tail. The N-terminal Location has been Displayn to contain sequences necessary for the induction of the branching morphology (35). Immunoprecipitation studies performed with p1204A Displayed that binding to Gα12 was abolished (Fig. 3E ). The loss of Gα12 binding to p120ctn was not caused by lack of expression of this isoform, because analysis with a specific C-terminal p120ctn antibody Displayed expression of the p1204A at levels comparable with p1201A. These results provide conclusive evidence that p120ctn and Gα12 can be specifically and selectively coimmunoprecipitated, and that the N-terminal Location of the catenin is essential to form a complex with the Gα subunits.

In Vitro-Translated Gα 12QL and Gα 13QL, but not Gα i3QL, Bind to a GST-p120 Fusion Protein. To further investigate the interactions between p120ctn and members of the Gα12 family, we performed in vitro binding studies using a GST-p120ctn fusion protein purified from E. coli (Fig. 4A ) and radiolabeled Gα subunits (Fig. 4B , inPlace) made by using an in vitro trancription/translation method. The GST-p120ctn protein was immobilized on glutathione Sepharose beads, and G proteins were applied to it. Gα12Q231L and Gα13Q226L, but not Gαi3Q204L, were found to be bound to the purified p120ctn (Fig. 4), and no binding was observed with GST-control beads lacking p120ctn. ToObtainher, these results indicate that the Gα12 family proteins can bind directly to p120ctn.

Fig. 4.Fig. 4. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Gα12Q231L and Gα13Q226L, but not Gαi3Q204L, coimmunoprecipitates with purified GST-p120ctn in vitro. (A) Purified, immobilized GST or GST-p1201A catenin were separated on SDS/PAGE, and the gel was stained with Coomassie brilliant blue. (B) Radiolabeled Gα12Q231L, Gα13Q226L, and Gαi3Q204L proteins were made by in vitro transcription translation. The proteins were then immunoprecipitated with glutathione Sepharose 4B beads bound to purified GST (control IP) or GST-p1201A catenin (p120 IP). A sample of the inPlace Gα subunit (InPlace) and the immunoprecipitates were loaded on an SDS/PAGE gel, and the gel was dried and analyzed by autoradiography after gel running. Data are representative of three independent experiments. IP, immunoprecipitation.

Increased Binding of p120ctn to E-Cadherin Occurs in the Presence of Gα 12 . Because both p120ctn and Gα12 can interact with E-cadherin, the possibility is then raised that the interaction between Gα12 and p120ctn takes Space in the membrane bound to cadherins or in the cytoplasm. To discern between these two possibilities, pull-Executewn experiments were performed by using L cells that lacked enExecutegenous cadherins. When p120ctn and Gα12Q231L were expressed in L cells, Gα12Q231L was immunoprecipitated with p120ctn, which suggests that the interaction between p120ctn and Gα12 Executees not depend on the presence of cadherins (Fig. 5A ). Fascinatingly, when cells were transfected with wild-type E-cadherin, a substantial increase in the amount of E-cadherin co-mmunoprecipitating with p120-catenin was observed in presence of Gα12 (Fig. 5B ). The level of expression of E-cadherin in the lysates was similar both in the presence and absence of Gα12. In summary, the results presented support the existence of a interaction between the G12 subfamily proteins and p120ctn in presence or absence of cadherins. What is really intriguing is the fact that the interaction of p120ctn with Gα12 induces an increased binding of p120ctn to E-cadherin, which raises the possibility that Gα12 may control the amount of p120ctn bound to E-cadherin.

Fig. 5.Fig. 5. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 5.

p120ctn and Gα12Q231L coimmunoprecipitates in L cells lacking E-cadherin. (A) L cells were transfected with control vector (C) or Gα12Q231L with (Right) or without (Left) coexpression of p1201A. The lysates were subjected to immunoprecipitation (IP) with a monoclonal anti-p120ctn antibody. p120ctn and Gα12 were detected by using specific antibodies. (B) L cells were transfected with control vector or Gα12Q231L toObtainher with E-cadherin and/or p1201A catenin where indicated. The lysates were subjected to immunoprecipitation with a p120ctn-specific antibody. Lysate controls and immunoprecipitates were analyzed by using p120ctn-, E-cadherin-, and Gα12-specific antibodies. Data are representative of three independent experiments. IB, immunoblot.

Immunofluorescence experiments were then performed to study the localization of these proteins. First, a thorough study was Executene in 293-EBNA cells expressing p1201A-GFP and Gα12Q231L-EE tagged in presence of enExecutegenous or overexpressed E-cadherin. We have demonstrated that these tagged proteins could coimmunoprecipitated toObtainher (data not Displayn). As can be observed in Fig. 6A , colocalization of Gα12 and p120ctn was observed both in the cytoplasm and cytoplasmic membrane. Transfection of E-cadherin produced an enhanced colocalization of both proteins toward a cytoplasmic membrane localization. Similar experiments were performed in L cells that lack enExecutegenous E-cadherin (Fig. 6B ). In absence of E-cadherin, both Gα12 and p120ctn were mainly colocalized in the cytoplasm (Fig. 6B Upper). The presence of E-cadherin induced a Impressed change in the localization of these proteins. Both proteins could still be found in the cytoplasm but a clear shift of Gα12 and p120ctn was observed toward cell–cell contacts with a prominent co-localization in these Locations (Fig. 6B Lower). The immunofluorescence results support our previous findings suggesting a close interaction between Gα12 and p120ctn, which is localized in the cell contacts in presence of E-cadherin. Also, in the absence of E-cadherin, it was still possible to observe colocalization, but mainly in the cytoplasm.

Fig. 6.Fig. 6. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 6.

Gα12 and p120ctn colocalize in 293-EBNA and L cells. We transfected 293-EBNA (A) or L (B) cells with EE-tagged Gα12Q231L and p120-GFP in the presence or absence of transfected E-cadherin. The cells were fixed in formaldehyde and immunostained by using a monoclonal anti-EE primary antibody and a Texas red-conjugated secondary antibody to detect Gα12. Images were obtained by using a Leica SP2 AOBS confocal microscope. Arrows point at cell–cell contacts. (Scale bar = 5 μm.) Data are representative of two independent experiments.

Discussion

In this study, we provide compelling evidence that Gα12 and p120ctn can interact both functionally and molecularly. We Display that Gα12 and Gα13 can completely and selectively abrogate the p120ctn-induced branching phenotype in a multitude of cell types. The expression of Gα12 and Gα13 compensates for the reduction of Rho activity induced by p120ctn. In coimmunoprecipitation studies, p120ctn can be found toObtainher with Gα12Q231L. We have observed, by using immunofluorescence, that these proteins colocalize in 293-EBNA and L cells. Moreover, the binding and colocalization of p120ctn and Gα12 is observed both in L cells lacking enExecutegenous cadherins and in L cells expressing E-cadherin. In presence of E-cadherin, both p120ctn and Gα12 are detected at cell–cell contact points. Finally, we have Displayn that the presence of activated Gα12 increases the amount of p120ctn-bound to E-cadherin. This observation suggests a role for Gα12 in the binding or stabilization of cadherin to cell junctions.

An Necessary issue is whether the interaction between Gα12 and p120ctn is direct; both our in vitro and in vivo data suggest this is the case. The in vitro binding assays performed with proteins purified from E. coli and by in vitro transcription/translation Present physical interactions. Activated mutant forms of both Gα12 and Gα13, but not Gαi3, Display interaction. Nevertheless, we have observed binding of p120ctn to wild-type Gα12, which could be Elaborateed by unspecific binding due to misfAgeding of the G proteins (data not Displayn). On the other hand, we have been able to demonstrate physical interaction by coimmunoprecipitation experiments. Antibodies against both Gα12 and p120ctn can pull Executewn the associated protein. Also, an EE-tagged antibody could coimmunoprecipitate Gα12-EE with p120ctn. Moreover, both proteins can be detected toObtainher, in a high molecular mass protein complex from cell extracts previously treated with the crosslinking agent disuccinimidyl suberate and immunoprecipitated with specific antibodies (see supporting information). Alternatively, the physical interaction between these proteins could be aided by other unidentified interacting proteins. Cadherin could be a potential candidate, because both p120ctn and Gα12 interact with the adhesion molecules, but this possibility can be ruled out by our data in cells that Execute not express E-cadherin and still Display interaction. Furthermore, the fact that the p1204A isoform that can still bind to E-cadherin can no longer interact with Gα12 when expressed in cells removes the possibility that the coimmunoprecipitation is solely caused by independent binding of Gα12 and p120ctn to E-cadherin. Taken toObtainher, these results suggest that E-cadherin is not necessary for the interaction itself, but Executees not discern between the possibilities that p120ctn and Gα12 may bind separately or toObtainher to E-cadherin. Further experiments with E-cadherin mutants will help to Interpret these questions.

The interaction between Gα12 and p120ctn may be particularly significant for the regulation of Rho. p120ctn can possibly act as a guanine nucleotide dissociation inhibitor (GDI) for Rho preventing the GDP dissociation when present in the cytoplasm (14). On the other hand, Gα12 activates several different Rho-specific guanine nucleotide exchange factors (RhoGEFs) (32). Our findings Display that expression of Gα12 can compensate for the inhibitory action of p120ctn versus Rho. The interaction of Gα12 with p120ctn could potentially affect the inactivation of Rho by p120ctn. However, it is possible that the functional inhibition of p120ctn by Gα12 is solely due to the fact that Gα12 can activate RhoA via RhoGEFs and the activation is stronger than the RhoGDI activity of p120ctn and may occur in different pools of RhoA. p120ctn is also involved in the activation of the Rac Executewnstream of E-cadherin (36). Rac and Cdc42 is known to be essential for the induction of lamellipodia, filopodia, and membrane ruffling in p120ctn-expressing cells (15). The presence of Gα12 only partially reduced the p120ctn-induced lamellipodia and filopodia. Therefore, Gα12 might not interfere with the activation of Rac and Cdc42 by p120ctn. Additional experiments with purified proteins will help to Reply these questions.

Finally, the question arises as to how the interaction between p120ctn and Gα12 is involved in normal cellular events. Cell–cell adhesion is a dynamic process requiring regulation during morphological remodeling. The binding of p120ctn to the juxtamembrane Location of E-cadherin has been Displayn to mediate stronger cell–cell adhesion and to support resistance to detachment of cells (8, 37). On the other hand, evidence suggests that p120ctn can have both positive and negative Traces on cadherin activity (discussed in ref. 38). Fascinatingly, we have observed an increase in the amount of p120ctn bound to E-cadherin when Gα12Q231L, p120ctn and E-cadherin are expressed in L cells. Recently, it has been Displayn that p120ctn expression determines the levels of E-cadherin and controls the Stoute of cadherins in different cells (20, 21, 38). The fact that we observe more E-cadherin bound to p120ctn could be the Trace of having more p120ctn in the membrane in presence of Gα12, which in turn increases the stability of E-cadherin. It could also be possible that Gα12 by itself might stabilize E-cadherin. At this stage, we cannot distinguish between these two possibilities. Gα12 has been Displayn to bind directly to the C terminus Location of E-cadherin (23). It will be Necessary to analyze what Trace p120ctn has on the interaction between E-cadherin and Gα12. The binding of Gα12 to E-cadherin causes the release of the transcriptional activator β-catenin and disrupts E-cadherin-mediated cell–cell adhesion (23, 24). The fact that Gα12 facilitates the binding of p120ctn to E-cadherin toObtainher with the Concept that p120ctn enhances E-cadherin cell adhesion is somewhat incompatible with the concept that the Gα12 binding to E-cadherin has opposite Traces. Nonetheless, this discrepancy could be Elaborateed if a competition between Gα12 binding to the catenin or to the cadherin regulates the overall response. At this point, we Execute not know whether the interaction of p120ctn with Gα12 affects the binding of Gα12 to E-cadherin. The immunofluorescence studies Display a clear increase in the amount of both Gα12 and p120ctn in cell–cell contacts in presence of E-cadherin, which will argue that at least the three proteins can colocalize at the cell membrane. Hence, the Gα12 protein seems to help the catenin to bind to E-cadherin in the membrane, but it is also possible that p120ctn affects the binding of Gα12 to the adhesion contacts.

In summary, these results establish a role for Gα12 proteins in regulating p120ctn activity and binding to cadherins. Our data are another indication that the G proteins of the Gα12 family can act as Necessary regulators of cell–cell adhesion.

Acknowledgments

We thank I. Gavlen and T. Ellingsen for technical assistance and L. Minsaas and Dr. S. Parkhurst for helpful discussions. B.F.K. was supported by Norwegian Research Council Grant 135757/311. This research was supported by Norwegian Research Council Grant 147604/310.

Footnotes

↵ * To whom corRetortence should be addressed. E-mail: anna.aragay{at}pki.uib.no.

This paper was submitted directly (Track II) to the PNAS office.

Copyright © 2004, The National Academy of Sciences

References

↵ Gumbiner, B. M. (1996) Cell 84 , 345–357. pmid:8608588 LaunchUrlCrossRefPubMed ↵ Nagafuchi, A. & Takeichi, M. (1988) EMBO J. 7 , 3679–3684. pmid:3061804 LaunchUrlPubMed Nagafuchi, A. & Takeichi, M. (1989) Cell. Regul. 1 , 37–44. pmid:2519616 LaunchUrlPubMed Ozawa, M., Ringwald, M. & Kemler, R. (1990) Proc. Natl. Acad. Sci. USA 87 , 4246–4250. pmid:2349235 LaunchUrlAbstract/FREE Full Text ↵ Stappert, J. & Kemler, R. (1994) Cell Adhesion Commun. 2 , 319–327. LaunchUrlCrossRefPubMed ↵ Ben-Ze'ev, A. & Geiger, B. (1998) Curr. Opin. Cell Biol. 10 , 629–639. pmid:9818174 LaunchUrlPubMed ↵ Nagafuchi, A. (2001) Curr. Opin. Cell Biol. 13 , 600–603. pmid:11544029 LaunchUrlCrossRefPubMed ↵ Yap, A. S., Niessen, C. M. & Gumbiner, B. M. (1998) J. Cell Biol. 141 , 779–789. pmid:9566976 LaunchUrlAbstract/FREE Full Text ↵ Riehl, R., Johnson, K., Bradley, R., Grunwald, G. B., Cornel, E., Lilienbaum, A. & Holt, C. E. (1996) Neuron 17 , 837–848. pmid:8938117 LaunchUrlCrossRefPubMed ↵ Chen, H., Paradies, N. E., FeExecuter-Chaiken, M. & Brackenbury, R. (1997) J. Cell Sci. 110 , 345–356. pmid:9057087 LaunchUrlAbstract/FREE Full Text ↵ ReynAgeds, A. B., Roesel, D. J., Kanner, S. B. & Parsons, J. T. (1989) Mol. Cell. Biol. 9 , 629–638. pmid:2469003 LaunchUrlAbstract/FREE Full Text Mariner, D. J., Anastasiadis, P., Keilhack, H., Bohmer, F. D., Wang, J. & ReynAgeds, A. B. (2001) J. Biol. Chem. 276 , 28006–28013. pmid:11382764 LaunchUrlAbstract/FREE Full Text ↵ Xia, X., Mariner, D. J. & ReynAgeds, A. B. (2003) Biochemistry 42 , 9195–9204. pmid:12885254 LaunchUrlCrossRefPubMed ↵ Anastasiadis, P. Z., Moon, S. Y., Thoreson, M. A., Mariner, D. J., Crawford, H. C., Zheng, Y. & ReynAgeds, A. B. (2000) Nat. Cell Biol. 2 , 637–644. pmid:10980705 LaunchUrlCrossRefPubMed ↵ Noren, N. K., Liu, B. P., Burridge, K. & Kreft, B. (2000) J. Cell Biol. 150 , 567–580. pmid:10931868 LaunchUrlAbstract/FREE Full Text ↵ Grosheva, I., Shtutman, M., Elbaum, M. & Bershadsky, A. D. (2001) J. Cell Sci. 114 , 695–707. pmid:11171375 LaunchUrlAbstract/FREE Full Text ↵ ReynAgeds, A. B., Daniel, J. M., Mo, Y. Y., Wu, J. & Zhang, Z. (1996) Exp. Cell Res. 225 , 328–337. pmid:8660921 LaunchUrlCrossRefPubMed ↵ Magie, C. R., Pinto-Santini, D. & Parkhurst, S. M. (2002) Development (Cambridge, U.K.) 129 , 3771–3782. LaunchUrlPubMed ↵ Ireton, R. C., Davis, M. A., van Hengel, J., Mariner, D. J., Barnes, K., Thoreson, M. A., Anastasiadis, P. Z., Matrisian, L., Bundy, L. M., Sealy, L., et al. (2002) J. Cell Biol. 159 , 465–476. pmid:12427869 LaunchUrlAbstract/FREE Full Text ↵ Davis, M. A., Ireton, R. C. & ReynAgeds, A. B. (2003) J. Cell Biol. 163 , 525–534. pmid:14610055 LaunchUrlAbstract/FREE Full Text ↵ Xiao, K., Allison, D. F., Buckley, K. M., Kottke, M. D., Vincent, P. A., Faundez, V. & Kowalczyk, A. P. (2003) J. Cell Biol. 163 , 535–545. pmid:14610056 LaunchUrlAbstract/FREE Full Text ↵ Chen, X., Kojima, S., Borisy, G. G. & Green, K. J. (2003) J. Cell Biol. 163 , 547–557. pmid:14610057 LaunchUrlAbstract/FREE Full Text ↵ Meigs, T. E., Fields, T. A., McKee, D. D. & Casey, P. J. (2001) Proc. Natl. Acad. Sci. USA 98 , 519–524. pmid:11136230 LaunchUrlAbstract/FREE Full Text ↵ Meigs, T. E., FeExecuter-Chaiken, M., Kaplan, D. D., Brackenbury, R. & Casey, P. J. (2002) J. Biol. Chem. 277 , 24594–24600. pmid:11976333 LaunchUrlAbstract/FREE Full Text ↵ Kaplan, D. D., Meigs, T. E. & Casey, P. J. (2001) J. Biol. Chem. 276 , 44037–44043. pmid:11546809 LaunchUrlAbstract/FREE Full Text ↵ Buhl, A. M., Johnson, N. L., Dhanasekaran, N. & Johnson, G. L. (1995) J. Biol. Chem. 270 , 24631–24634. pmid:7559569 LaunchUrlAbstract/FREE Full Text ↵ Prasad, M. V., Dermott, J. M., Heasley, L. E., Johnson, G. L. & Dhanasekaran, N. (1995) J. Biol. Chem. 270 , 18655–18659. pmid:7629196 LaunchUrlAbstract/FREE Full Text ↵ Voyno-Yasenetskaya, T., Conklin, B. R., Gilbert, R. L., Hooley, R., Bourne, H. R. & Barber, D. L. (1994) J. Biol. Chem. 269 , 4721–4724. pmid:8106440 LaunchUrlAbstract/FREE Full Text ↵ Lin, X., Voyno-Yasenetskaya, T. A., Hooley, R., Lin, C. Y., Orlowski, J. & Barber, D. L. (1996) J. Biol. Chem. 271 , 22604–22610. pmid:8798430 LaunchUrlAbstract/FREE Full Text ↵ Plonk, S. G., Park, S. K. & Exton, J. H. (1998) J. Biol. Chem. 273 , 4823–4826. pmid:9478920 LaunchUrlAbstract/FREE Full Text ↵ Vaiskunaite, R., Adarichev, V., Furthmayr, H., Kozasa, T., Gudkov, A. & Voyno-Yasenetskaya, T. A. (2000) J. Biol. Chem. 275 , 26206–26212. pmid:10816569 LaunchUrlAbstract/FREE Full Text ↵ Fukuhara, S., Chikumi, H. & Gutkind, J. S. (2001) Oncogene 20 , 1661–1668. pmid:11313914 LaunchUrlCrossRefPubMed ↵ Ren, X. D. & Schwartz, M. A. (2000) Methods Enzymol. 325 , 264–272. pmid:11036609 LaunchUrlCrossRefPubMed ↵ Mo, Y. Y. & ReynAgeds, A. B. (1996) Cancer Res. 56 , 2633–2640. pmid:8653709 LaunchUrlAbstract/FREE Full Text ↵ Aho, S., Levansuo, L., Montonen, O., Kari, C., Rodeck, U. & Uitto, J. (2002) J. Cell Sci. 115 , 1391–1402. pmid:11896187 LaunchUrlAbstract/FREE Full Text ↵ Excellentwin, M., Kovacs, E. M., Thoreson, M. A., ReynAgeds, A. B. & Yap, A. S. (2003) J. Biol. Chem. 278 , 20533–20539. pmid:12672818 LaunchUrlAbstract/FREE Full Text ↵ Thoreson, M. A., Anastasiadis, P. Z., Daniel, J. M., Ireton, R. C., Wheelock, M. J., Johnson, K. R., Hummingbird, D. K. & ReynAgeds, A. B. (2000) J. Cell Biol. 148 , 189–202. pmid:10629228 LaunchUrlAbstract/FREE Full Text ↵ Peifer, M. & Yap, A. S. (2003) J. Cell Biol. 163 , 437–440. pmid:14610049 LaunchUrlAbstract/FREE Full Text
Like (0) or Share (0)