Cancer Research The Future of Cancer Research: Science and Patient Impact Translational Medicine Conference in Israel
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, S. E.
Right arrow Articles by Arteaga, C. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, S. E.
Right arrow Articles by Arteaga, C. L.
[Cancer Research 66, 9591-9600, October 1, 2006]
© 2006 American Association for Cancer Research

Cell, Tumor, and Stem Cell Biology

HER2/Neu (ErbB2) Signaling to Rac1-Pak1 Is Temporally and Spatially Modulated by Transforming Growth Factor ß

Shizhen Emily Wang1,5, Incheol Shin6, Frederick Y. Wu2, David B. Friedman3,4 and Carlos L. Arteaga1,2,5

Departments of 1 Cancer Biology, 2 Medicine, and 3 Biochemistry; 4 Mass Spectrometry Research Center; 5 Breast Cancer Research Program, Vanderbilt-Ingram Comprehensive Cancer Center, Vanderbilt University School of Medicine, Nashville, Tennessee; and 6 Department of Life Sciences, Hanyang University, Seoul, Korea

Requests for reprints: Carlos L. Arteaga, Division of Oncology, Vanderbilt University Medical Center, 2220 Pierce Avenue, 777 PRB, Nashville, TN 37232-6307. Phone: 615-936-3524; Fax: 615-936-1790; E-mail: carlos.arteaga{at}vanderbilt.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
In HER2 (ErbB2)-overexpressing cells, transforming growth factor ß (TGF-ß), via activation of phosphoinositide-3 kinase (PI3K), recruits actin and actinin to HER2, which then colocalizes with Vav2, activated Rac1, and Pak1 at cell protrusions. This results in prolonged Rac1 activation, enhanced motility and invasiveness, Bad phosphorylation, uncoupling of Bad/Bcl-2, and enhanced cell survival. The recruitment of the HER2/Vav2/Rac1/Pak1/actin/actinin complex to lamellipodia was abrogated by actinin siRNAs, dominant-negative (dn) p85, gefitinib, and dn-Rac1 or dn-Pak1, suggesting that the reciprocal interplay of PI3K, HER2 kinase, and Rac GTPases with the actin cytoskeleton is necessary for TGF-ß action in oncogene-overexpressing cells. Thus, by recruiting the actin skeleton, TGF-ß "cross-links" this signaling complex at cell lamellipodia; this prolongs Rac1 activation and increases metastatic properties and survival of HER2-overexpressing cells. (Cancer Res 2006; 66(19): 9591-600)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Transforming growth factor ß (TGF-ß) has been shown to synergize with transforming oncogenes in cancer progression (1). An oncogenic signaling network for which this has been shown is the ErbB family of receptor tyrosine kinases. Overexpression of active TGF-ß1 or active mutants of the type I TGF-ß receptor (Alk5) in the mammary gland of bitransgenic mice also expressing mouse mammary tumor virus-Neu (ErbB2) accelerates metastases from Neu-induced mammary cancers (2–4). In addition, a genetic modifier screen in nontumorigenic mammary epithelial cells identified TGF-ß1 and TGF-ß3 as molecules that cooperate with HER2 in inducing cell motility and invasion (5). Inhibition of HER2 with the HER2-neutralizing antibody trastuzumab blocked the promigratory effect of TGF-ß on HER2-overexpressing mammary epithelial cells (6), suggesting that oncogene function is required for the transforming effect of TGF-ß. Furthermore, TGF-ß can activate signaling pathways downstream ErbB receptor tyrosine kinases such as Ras/mitogen-activated protein kinase (7) and phosphoinositide-3 kinase (PI3K)/Akt (8, 9). Inhibition of endogenous TGF-ß function with dominant-negative (dn) type I TGF-ß receptor has been shown to block growth of Ras-transformed tumors in vivo (10), suggesting that oncogenes can engage and, in turn, become dependent on TGF-ß function for tumor progression. However, the molecular mechanisms of cross-talk between TGF-ß and ErbB receptor signaling remain largely unknown.

The Rho family of small GTPases (RhoA, Rac1, and Cdc42) regulate the actin cytoskeleton and focal adhesion complexes, which are essential for morphogenesis, cell motility, and invasiveness (11). They also signal for cell proliferation and apoptosis through targeting of downstream effectors such as Pak, c-jun NH2-terminal kinase (JNK), p38, Rock, signal transducers and activators of transcription 3, and nuclear factor {kappa}B (NF-{kappa}B; ref. 12). Their activation is regulated by guanine nucleotide exchange factors, which catalyze the exchange of GDP for GTP and hence "switch on" the GTPase to its active GTP-bound form (13). GTPases like Rac1 have been shown to contribute to TGF-ß–mediated cellular and transcriptional responses (14, 15). In addition, TGF-ß can rapidly activate RhoA and Rac1, contributing to an epithelial to mesenchymal transition and enhanced cell motility (16, 17). Rac1 activity in situ is higher in mouse mammary cancers expressing Neu (ErbB2) and active TGF-ß1 transgenes compared with transgenic tumors expressing the Neu oncogene alone (3). MCF10A human mammary epithelial cells that overexpress HER2/ErbB2 exhibit higher Rac1 activity than cells with low HER2 levels and this activity is blocked by treatment with a HER2-neutralizing antibody (6). Because HER2, as well as coreceptors in the ErbB family, can also activate Rac1 (18–20), these data suggest that, like for PI3K/Akt, HER2/ErbB2 and TGF-ß can converge at Rac1 and synergize for tumor progression.

We have examined mechanisms of TGF-ß/HER2 interaction in MCF10A cells stably transfected with a HER2 vector and breast cancer lines that naturally overexpress the proto-oncogene. In these cells, treatment with TGF-ß activated PI3K, leading to the recruitment of actin and actinin to HER2, which then colocalized with the Vav2 guanine nucleotide exchange factor, activated Rac1, and its effector Pak1 at cell protrusions. The stabilization of this complex at cell lamellipodia resulted in prolonged activation of Rac1, enhanced cell invasiveness, Bad phosphorylation at Ser136, uncoupling of Bad from Bcl-2, and enhanced cell survival.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cells, plasmids, and viruses. Primary mammary tumor cells from Neu transgenic tumors were generated and maintained as described (2–4). MCF10A/VEC and MCF10A/HER2 cells were generated and maintained as previously described (6, 21). Phoenix-Ampho cells were grown in DMEM (Cambrex, Rockland, ME) containing 10% fetal bovine serum (FBS; Hyclone, Logan, UT) in a humidified 5% CO2 incubator at 37°C. Human breast tumor cell lines MCF7, MDA453, MDA361, and BT474 were purchased from the American Type Tissue Culture Collection (Manassas, VA) and maintained in IMEM/10% FBS. The following reagents were used: recombinant human TGF-ß1 (R&D Systems, Minneapolis, MN), wortmannin (Sigma, St. Louis, MO), gefitinib (AstraZeneca Pharmaceuticals, Wilmington, DE), and puromycin (Calbiochem, La Jolla, CA). The dn-Pak1-205 (22) cDNA, encoding kinase-dead truncated 1-205 Pak1, was a gift from Gary Bokoch (Scripps Research Institute, La Jolla, CA). Retroviruses expressing dn-Rac1 (17), dn-Pak1, or pBabe vector were produced by transfecting Phoenix-Ampho cells with corresponding plasmid DNA as described (21). Stably transfected cells were selected in 1 µg/mL puromycin.

Peptide mass mapping and mass spectrometry. Neu immune complexes were separated by SDS-PAGE and the gel was stained with Coomassie blue. Protein bands corresponding to apparent molecular masses of 102 and 47 kDa were excised and subjected to in-gel digestion (details provided in Supplemental data).

Immunoprecipitation, Rac activity assay, and immunoblotting. Cells were washed twice with ice-cold PBS and lysed in NP40 lysis buffer (20 mmol/L Tris, 150 mmol/L NaCl, 1% NP40, 0.1 mmol/L EDTA, and protease and phosphatase inhibitors). After sonication for 10 seconds and centrifugation (14,000 rpm), protein concentration in the supernatants was measured using the bicinchoninic acid protein assay reagent (Pierce, Rockford, IL). Immunoprecipitation and immunoblotting were done as described (21) with horseradish peroxidase–conjugated secondary antibodies (Promega, Madison, WI). Primary antibodies include actinin, Pak1, Vav2, Smad2/3, phospho-tyrosine, Bcl-2 (Santa Cruz Biotechnology, Santa Cruz, CA), phospho-Smad2(Ser465/467), Rac1 (Transduction Laboratories, Lexington, KY), phospho-Pak1(Thr423), phospho-Akt(Ser473), Akt, phospho-Bad(Ser136), Bad (Cell Signaling, Beverly, MA), HER2/ErbB2 (NeoMarkers, Fremont, CA), and actin (Sigma). To pull down GTP-bound (active) Rac, a glutathione S-transferase (GST) fusion protein of Pak binding domain (PBD) precoupled to agarose-glutathione beads (Cytoskeleton Inc., Denver, CO) was used as previously described (6). Eluted Rac1 was detected by immunoblot.

PI3K assay. Cell lysates were prepared as previously described (8) and (500 µg) precipitated with an antibody against p85 (Upstate Biotechnology, Lake Placid, NY). Immune complexes were subjected to a PI3K enzymatic assay (8) in a reaction mixture containing 0.2 mg/mL crude brain phosphoinositides (2 mg/mL; Sigma), 50 µmol/L ATP, 10 µCi of [{gamma}-32P]ATP (specific activity, 6,000 Ci/mmol; Amersham, Piscataway, NJ), 5 mmol/L MgCl2, and 20 mmol/L HEPES buffer (pH 7.5). The reactions were carried out for 15 minutes at room temperature and stopped by the addition of 100 µL of 1 N HCl followed by 200 µL of chloroform/methanol (1:1, v/v). Lipids were extracted and resolved on potassium oxalate–coated TLC plates (silica gel 60) with a solvent system of n-propyl alcohol/2 mol/L acetic acid (65:35, v/v). TLC plates were air-dried and the incorporation of 32P into phosphoinositide was detected by autoradiography.

Apoptosis assay and flow cytometric analysis of cell cycle distribution. Approximately 5 x 105 cells were suspended in 8 mL of serum-free medium with or without 2 ng/mL TGF-ß for 16 hours. Both suspended cells and adherent cells were harvested and washed with PBS, before being subjected to Apo-5-bromo-2'-deoxyuridine (BrdUrd) analysis using an APO-BRDU assay kit (Phoenix Flow Systems, San Diego, CA) according to the protocol of the manufacturer in a FACS/Calibur flow cytometer (BD Biosciences, Mansfield, MA). Cell cycle analysis was done as described elsewhere (21). To detect caspase-3 cleavage, cells suspended for 0 to 16 hours were lysed and subjected to SDS-PAGE and immunoblot with caspase-3 antibody (Cell Signaling). Cytosolic DNA was extracted as described (23) and analyzed on 2% agarose gels.

Cell adhesion and motility. Cell adhesion and wound closure motility assays were done as previously described (21). In wound closure assays, cell monolayers were incubated in serum-free medium overnight before being wounded and replenished with serum-free medium ± 2 ng/mL TGF-ß1 and different inhibitors. Phase-contrast images were photographed at different time points after wounding.

Three-dimensional morphogenesis and invasion assay. Cells were seeded on Growth Factor Reduced Matrigel (BD Biosciences) or 1:1 mixture of Matrigel/collagen I (2 mg/mL collagen; Upstate) in eight-well chamber slides following the protocol described by Debnath et al. (24). TGF-ß1 (2 ng/mL) was added into the medium 12 hours after cell seeding. Invasion assay was done in BD BioCoat Growth Factor Reduced Matrigel invasion chambers (BD Biosciences) according to the protocol of the manufacturer. Immunofluorescence staining of three-dimensional acini was done as previously described (24). Confocal analyses were done with Zeiss inverted LSM510 confocal microscopy system.

Indirect immunofluorescence assay. Immunofluorescence assay was done as previously described (21). For detection of GTP-bound Rac1, cells were fixed with methanol and incubated with 10 µg/mL GST-PBD protein (Cytoskeleton, Inc.) in 1% milk for 15 minutes and next with a GST antibody (Santa Cruz Biotechnology) for 1 hour at room temperature and with Oregon Green-{alpha}-mouse secondary immunoglobulin G (IgG). Fluorescent images were captured using a Princeton Instruments (Trenton, NJ) cooled charge-coupled device digital camera from a Zeiss Axiophot upright microscope.

Transfection of siRNA. siRNA oligonucleotides designed to target the nucleotide sequence 5'-AAGTGCCAGCTGGAGATCAAC (for actinin) or 5'-AACTCGGAGCTGGAGATCAAC (as a control sequence; mismatched nucleotides are underlined) were obtained from Qiagen (Valencia, CA). The siRNA oligonucleotides were transfected into MCF10A/HER2 cells grown on six-well plates or coverslips using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) following the procedure of the manufacturer. Expression of actinin was examined by immunoblot.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
TGF-ß induces association of HER2 with the actin cytoskeleton. To determine whether TGF-ß induced recruitment of proteins to the Neu transmembrane tyrosine kinase, we treated Neu-expressing primary mammary tumor cells (3) with TGF-ß1 for 16 hours. Neu antibody or control IgG precipitations from these cells were separated by SDS-PAGE. Commassie blue staining of the gel revealed 102-kDa and 47-kDa bands in the precipitates from treated but not from untreated cells (Fig. 1A ). Mass spectrometry followed by database interrogation of the excised protein bands identified them as actinin-4, an actin-bundling protein that connects the actin cytoskeleton to the cell membrane (25), and actin, respectively. We next confirmed this association in MCF-10A human mammary epithelial cells stably expressing HER2 (ErbB2), the human homologue of Neu. By immunoprecipitation followed by immunoblot analysis, actin and actinin coprecipitated with HER2 in TGF-ß–treated cells (Fig. 1B). Because TGF-ß reorganizes the actin cytoskeleton and induces the formation of filamentous actin stress fibers, the association of HER2 with actin/actinin suggested ligand-induced changes in the cellular distribution of HER2. Immunofluorescence staining of untreated cells revealed low levels of F-actin with HER2 distribution throughout the cell membrane. Treatment with TGF-ß for 16 hours induced F-actin formation and colocalization of F-actin and HER2 predominantly at cell lamellipodia (Fig. 1C).


Figure 1
View larger version (35K):
[in this window]
[in a new window]
  		Figure 1. TGF-ß induces association of HER2 and the actin cytoskeleton at membrane protrusions. A, primary mammary tumor cells from Neu transgenic tumors were serum starved for 24 hours before treatment with TGF-ß (2 ng/mL) or not treated for 16 hours, and lysed for immunoprecipitation (IP) with a HER2/Neu (ErbB2) antibody or control IgG. The precipitated samples were separated by SDS-PAGE and the gel was stained with Commassie blue. The identities of the indicated protein bands (arrows) were determined by peptide mass mapping and single-stage mass spectrometry as indicated in Materials and Methods. Positions and molecular weights of the protein ladder bands are indicated. B, MCF10A/HER2 cells were serum starved for 24 hours and treated with TGF-ß, or not treated, for 16 hours in serum-free medium. Cells were then lysed for immunoprecipitation with a HER2 antibody or control IgG. The pull-downs were subjected to immunoblotting with the indicated antibodies. C, serum-starved MCF10A/HER2 cells were treated with TGF-ß for 16 hours or not treated before fixation for immunofluorescence assay with a HER2 antibody (red) and Alexa Fluor 488 phalloidin (green; for F-actin). Colocalization (yellow) was observed in TGF-ß–treated cells. HER2 localization enriched at cell protrusions (arrows). Hoechst staining (blue) indicates cell nuclei. D, MCF10A/HER2 cells grown on six-well plates were transiently transfected with siRNA oligonucleotides targeting actinin or a control sequence. Seventy-two hours after transfection, cells were treated with TGF-ß for 16 hours or not treated in serum-free medium and then lysed before precipitation with a HER2 antibody followed by immunoblot analysis (left). Cells grown on glass coverslips were also transfected with actinin siRNAs for 72 hours and then treated with TGF-ß for 16 hours in serum-free medium before staining with a HER2 (green) or Alexa Fluor 488 phalloidin (green). Hoechst, nuclear staining (blue). Bar, 50 µm.

 
To determine if that the association with actinin played a role in the recruitment of both HER2 and F-actin to cell protrusions, we knocked down the expression of actinin with siRNA oligonucleotides. In cells transfected with siRNA targeting actinin but not control siRNA, TGF-ß–induced HER2 association with actinin and actin was abrogated (Fig. 1D, left). Consistent with this result, in cells transfected with actinin siRNA, TGF-ß did not induce polymerization of F-actin or translocation of HER2 to lamellipodia (Fig. 1D, right).

Rac1-Pak1 function is required for TGF-ß–induced association of HER2 and the actin cytoskeleton. We have previously reported that TGF-ß rapidly increases the interaction of HER2 with Rac1 and Pak1 as well as Rac1 activity (6, 17). Pak1, a main effector of Rac1, interacts with cytoskeletal components and induces cytoskeletal reorganization (26). Therefore, we speculated that the association of HER2 and the actin cytoskeleton in TGF-ß–treated cells also involves an interaction with Rac1 and Pak1. To test this, we examined if Pak1 associated with the HER2/actin/actinin complex in TGF-ß–treated cells. TGF-ß induced a time-dependent association of Pak1 with HER2, actin, and actinin, which was maximal at 8 hours (Fig. 2A ). We next tested Rac1 and Cdc42 activities with GST-PBD fusion protein to capture the GTP-bound active form of Rac1 and Cdc42, followed by immunoblot with Rac1 and Cdc42 antibodies, respectively. Rac1 activity exhibited a biphasic response to TGF-ß in MCF10A/HER2 cells but not in cells expressing vector alone with two peaks of activation at 4 and 24 hours (Fig. 2B). Cdc42 activity was not affected (data not shown). Increasing levels of HER2, actinin, and actin associated with GST-PBD as a function of time of treatment with TGF-ß, suggesting an association of these with GTP-Rac1. In MCF10A/VEC cells, the kinetics of TGF-ß–induced Rac1 activation were markedly slower with maximal stimulation at 8 hours and a more delayed association of actin and actinin with GST-PBD (Fig. 2B).


Figure 2
View larger version (77K):
[in this window]
[in a new window]
  		Figure 2. TGF-ß–induced HER2/cytoskeleton association, cell migration, and adhesion require Rac1-Pak1. A, MCF10A/HER2 cells were serum starved overnight, treated with TGF-ß, lysed, and precipitated with a Pak1 antibody. Pak1 pull-downs and whole-cell lysates were subjected to immunoblotting with the indicated antibodies. B, MCF10A/HER2 and MCF10A/VEC cells were serum starved overnight, treated with TGF-ß at the indicated times, and lysed before incubation with GST-PBD. GST-PBD–bound proteins were eluted, separated by SDS-PAGE, and subjected to the indicated immunoblot analyses. Ponceau staining of the GST-PBD fusion was used as loading control. C, MCF10A/HER2 cells stably expressing dn-Rac1, dn-Pak1, or pBabe vector were serum starved overnight and treated with TGF-ß for 12 hours. Cell lysates were prepared for immunoprecipitation with Pak1, Rac1, or HER2 antibodies. The presence of actin, actinin, HER2, Pak1, and Rac1 in the precipitated samples was detected by immunoblot analysis as indicated. Immunoblots of total cell lysates showed increased p-Smad2 in all TGF-ß–treated samples. D, MCF10A/HER2 cells expressing dn-Rac1, dn-Pak1, or pBabe vector were grown on six-well plates to 100% confluence and serum starved overnight before wounding and the addition of 2 ng/mL TGF-ß1. Wound closure was captured at the indicated times. E, the above cell lines growing on cell culture dishes were trypsinized and resuspended in serum-free medium. Equal numbers of cells were seeded to six-well plates and allowed to attach in the presence or absence of 2 ng/mL TGF-ß. After 2 hours, nonattached cells were removed and adherent cells were washed with PBS, trypsinized, and counted. Columns, mean of four wells; bars, SD.

 
To determine if Rac1 and Pak1 are required for HER2 interactions with the actin cytoskeleton, we established MCF10A/HER2 cells stably expressing dn-Rac1 (Rac1-T17N; ref. 17), dn-Pak1 (Pak1-205; ref. 22), or pBabe vector. Precipitation from TGF-ß–treated pBabe control cell lysates with Pak1, Rac1, or HER2 antibodies followed by immunoblot analyses indicated ligand-induced association of Pak1, Rac1, HER2, actin, and actinin. In cells expressing dn-Rac1 or dn-Pak1, this association was abrogated but ligand-induced phosphorylation of Smad2 was unaffected (Fig. 2C). Although at lower levels than in the presence of TGF-ß, Rac1 and Pak1 were constitutively associated with HER2 (Fig. 2C, lane 1) and, interestingly, the basal phosphorylation of HER2 was markedly reduced in cells expressing dn-Rac1 and dn-Pak1 (Fig. 2C, lanes 7-12, first row). Finally, we examined if Rac1-Pak1 function is required for TGF-ß–induced cell motility and adhesion. In cells transfected with vector alone, TGF-ß accelerated wound closure and cell adhesion. These responses were impaired in MCF10A/HER2 cells expressing dn-Rac1 or dn-Pak1 (Fig. 2D and E), suggesting that Rac1-Pak1 function is required for TGF-ß–induced cell motility and adhesion.

HER2 phosphorylation is required for TGF-ß–induced activation of Rac1 and HER2 translocation. We then asked if the phosphorylation of HER2 is required to form complexes with Rac1, Pak1, and cytoskeletal components. We used the small-molecule ErbB tyrosine kinase inhibitor gefitinib, which in epidermal growth factor receptor (EGFR)–positive cells has been shown to inhibit HER2 phosphorylation (27). Treatment with gefinib markedly reduced HER2 phosphorylation (Fig. 3A, third row ). In addition, it reduced TGF-ß–mediated activation of Rac1 and abolished ligand-induced association of HER2 with Pak1, Rac1, actin, and actinin (Fig. 3A), suggesting that activation of HER2 is necessary for its association with the actin cytoskeleton and the engagement of Rac1. Pretreatment with gefitinib did not prevent ligand-induced phosphorylation of Smad2, suggesting that the effect was ErbB specific and that TGF-ß receptors were unaffected by the small-molecule inhibitor. Consistent with the coprecipitation experiments, preincubation with gefitinib prevented TGF-ß–induced translocation of HER2 to lamellipodia as well as MCF10A/HER2 cell motility in a wound closure assay (Fig. 3B and C). These results imply that TGF-ß requires a threshold of HER2 activation to engage Rac1-dependent cell motility.


Figure 3
View larger version (68K):
[in this window]
[in a new window]
  		Figure 3. Phosphorylation of HER2 is required for TGF-ß–induced Rac1 activity and HER2 translocation. A, MCF10A/HER2 cells were serum starved overnight and pretreated with gefitinib (1 µmol/L) for 1 hour before adding TGF-ß for 0 to 24 hours. Cell lysates were harvested for immunoprecipitation with Pak1 or HER2 antibodies or GST-PBD. The antibody or fusion protein pull-downs were subjected to immunoblot analysis with the antibodies shown on the right. Immunoblots of total cell lysates showed equal loading in each lane. p-Smad2 was induced in a time-dependent fashion by added TGF-ß. B, MCF10A/HER2 cells grown on coverslips were serum starved overnight and pretreated with gefitinib for 1 hour before adding TGF-ß for 16 hours. Cells were then fixed and subjected to immunofluorescence assay with a HER2 antibody. Arrows, HER2 translocation to cell protrusions. Hoechst, nuclear staining. C, MCF10A/HER2 cell monolayers were serum starved and wounded as described in Fig. 2D. TGF-ß ± gefitinib was added and wound closure monitored 16 hours later.

 
TGF-ß and HER2 induce a PI3K-dependent interaction of Vav2 with Rac1. Both activated TGF-ß receptors and HER2 (28) have been shown to activate the PI3K/Akt pathway. Inhibition of PI3K has been shown to reverse a TGF-ß–induced fibroblastoid phenotype and cell motility (29–31). In addition, PI3K has been shown to activate Rac1 and Pak1 (32–35) as well as the Rac guanine nucleotide exchange factor Vav2 (36). Because Vav2 has been shown to associate with ErbB receptors (19, 37, 38), we examined the role of PI3K and Vav2 in TGF-ß–induced Rac1 activity in MCF10A/HER2 cells. Pretreatment with the PI3K inhibitor wortmannin prevented TGF-ß–induced cell motility (not shown) and translocation of HER2 to lamellipodia (Fig. 4A ). Subsequent immunoprecipitation experiments showed basal association of Vav2 and Rac1, which increased 1 hour after TGF-ß treatment, corresponding temporally to an increase in Rac1 activity (Fig. 4B). These changes were still detectable in the presence of 100 µg/mL cycloheximide (data not shown). Like Rac1, HER2 associated constitutively with Vav2, but this association was not increased by treatment with TGF-ß. Both basal and ligand-enhanced association of HER2 with Vav2 were prevented by gefitinib. On the other hand, neither gefitinib nor wortmannin reduced the constitutive basal Rac1/Vav2 but they both eliminated any enhanced association induced by TGF-ß (Fig. 4B). This suggests that the HER2/Vav2 complex requires active HER2 but not PI3K. On the other hand, both HER2 and PI3K functions are permissive for Rac1/Vav2 complex formation.


Figure 4
View larger version (67K):
[in this window]
[in a new window]
  		Figure 4. TGF-ß and HER2 induce a PI3K-dependent interaction of Vav2 with Rac1. A, MCF10A/HER2 cells grown on coverslips were serum starved overnight, preincubated with wortmannin (50 nmol/L) for 1 hour, treated or not with TGF-ß for 16 hours, and then fixed and stained with a HER2 antibody. Arrows, HER2 translocation to the cell protrusions. Hoechst, nuclear staining. B, cells were serum starved overnight and pretreated with gefitinib or wortmannin for 1 hour before the addition of TGF-ß for 0.5 to 4 hours. Cell lysates were prepared for immunoprecipitation with Rac1 or Vav2 antibodies or GST-PBD as indicated in Materials and Methods. Precipitates and eluates were tested in immunoblot procedures with the indicated antibodies. C, cells were serum starved overnight and pretreated with gefitinib for 1 hour or not before adding TGF-ß for 0.5 to 4 hours. Cell lysates were prepared and precipitated with a p85 antibody. The precipitates were divided in half and tested in a PI3K in vitro assay as described in Materials and Methods or subjected to a p85 immunoblot procedure. D, cells were infected with an adenovirus encoding dn-p85 or ß-gal (multiplicity of infection, 1.5). Sixteen hours postinfection, the cells were treated or not with TGF-ß for 16 hours in serum-free medium, lysed, and precipitated with a HER2 antibody, followed by the HER2, actin, and actinin immunoblot analyses. Dn-p85 transduction was confirmed by p85 immunoblot of total cell lysates. E, cells were transduced with a dn-p85 or ß-gal adenovirus. Cell monolayers were serum starved overnight and wounded. Wounded cells were incubated in serum-free medium with or without TGF-ß for 16 hours.

 
The similar effects of gefitinib and wortmannin on the ligand-stimulated Rac1/Vav2 complex and on Rac1 activity suggested that the effects of the ErbB inhibitor were a result of PI3K inhibition. Thus, we examined the effect of gefitinib on PI3K catalytic activity as measured by the ability of immune complexes precipitated with p85 antibodies to stimulate formation of 3'-phosphorylated inositol lipids in vitro. In MCF10A/HER2 cells, TGF-ß activated PI3K within 1 hour after treatment. In the presence of gefitinib, both basal and ligand-induced PI3K activities were markedly reduced (Fig. 4C), implying that HER2 phosphorylation is permissive and/or required for TGF-ß–stimulated PI3K activity. Consistent with this result, gefitinib and wortmannin each prevented ligand-mediated phosphorylation of Akt and Pak1 (Fig. 4B, bottom four rows). Finally, expression of adenoviruses encoding dn-p85 (Ax{Delta}p85; ref. 39) completely abolished the TGF-ß–induced association of HER2 with actin and actinin, as well as Akt activation (Fig. 4D) and cell motility, in a wound closure assay (Fig. 4E), further suggesting a central role for PI3K in the crosstalk between TGF-ß and HER2 signaling.

TGF-ß–induced HER2 translocation to lamellipodia recruits Vav2 and locally activates Rac1. Inhibition of HER2 with gefitinib blocked TGF-ß–induced recruitment of HER2 to cell lamellipodia, activation of Rac1 and PI3K, and the association of HER2 with Vav2. These data suggested that treatment with TGF-ß results in HER2- and PI3K-dependent redistribution of HER2-associated Vav2 to lamellipodia, which in turn regulates the local activation of Rac1. To test this possibility, we double-stained HER2 and Vav2 in TGF-ß–treated MCF10A/HER2 cells. To localize Rac1 activation, we incubated fixed cells with GST-PBD and then with a GST antibody to detect GTP-bound Rac1. In ligand-stimulated cells, HER2 colocalized with Vav2 and with activated Rac1 at cell lamellipodia (Supplementary Fig. S1A and B). Because GST-PBD also recognizes Cdc42, we cannot completely rule out the involvement of this GTPase although it is known to trigger filopodial extensions. The colocalization of HER2 and activated Rac1 was abolished in cells treated with gefitinib or wortmannin as well in cells expressing dn-Rac1, dn-Pak1, or dn-p85 (Supplementary Fig. S1C), implying that both HER2 phosphorylation and PI3K are required for Vav2 recruitment and local Rac1 activation.

TGF-ß and HER2 cooperate in tumor cell invasiveness. TGF-ß and HER2 have been shown to synergize in tumor metastasis. Thus, we examined whether they also cooperated in tumor cell invasiveness via Rac1-Pak1 ex vivo. MCF10A cells form polarized, quiescent acini-like spheroids in three-dimensional basement membrane gels. Activation of HER2 in these cells has been shown to reinitiate proliferation, disrupt tight junctions and apical polarity, and induce acinar expansion without invasion of the surrounding matrix (40). TGF-ß treatment of MCF10A/HER2 cells induced three-dimensional acinar expansion and invasive protrusions into the surrounding Matrigel (Fig. 5A, left ). To investigate branching morphogenesis, we used a mixture of Matrigel/collagen I, which has intermediate stiffness that is optimal for cell invasion compared with a softer matrix such as Matrigel (41). In a matrix containing 2 mg/mL collagen I (a 1:1 mixture of Matrigel/collagen I), the MCF10A/HER2 cells exhibited an elongated shape not observed in the vector controls. Addition of TGF-ß to the HER2-overexpressing cells enhanced their elongated fibroblastoid shape and arranged them as invasive cords into the surrounding matrix; these effects were abrogated in cells expressing dn-Rac1 or dn-Pak1 (Fig. 5A, right). In MCF10A/HER2 cells within the TGF-ß–stimulated invasive cords, F-actin was induced and it colocalized with HER2 as measured by immunofluorescence (Fig. 5B). Finally, we tested the ability of the cells to invade through Matrigel-coated chambers. In MCF10A/HER2 cells, but not in control cells, TGF-ß increased invasion into Matrigel 1.6-fold. Ligand-induced invasion was completely blocked in cells expressing dn-Rac1 or dn-Pak1 (Fig. 5C).


Figure 5
View larger version (49K):
[in this window]
[in a new window]
  		Figure 5. TGF-ß and HER2 cooperate in tumor cell invasiveness. A, left, phase-contrast images of MCF10A/VEC or MCF10A/HER2 acini cultured for 6 days on Matrigel in the absence (top) or presence (bottom) of TGF-ß (2 ng/mL). Right, phase-contrast images of MCF10A/VEC or MCF10A/HER2 acini expressing dn-Rac1, dn-Pak1, or pBabe vector cultured for 5 days on a 1:1 mixture of Matrigel/collagen I ± TGF-ß (2 ng/mL). Bar, 50 µm. B, MCF10A/HER2 cells from experiment shown in (A, right) were stained with a HER2 antibody (red) or Alexa Fluor 488 phalloidin (green). Colocalization (yellow) was observed in TGF-ß–treated acini. HER2 localization enriched at cell protrusions (arrows). Dark pink, TOPRO-3 nuclear staining. Bar, 50 µm. C, MCF10A/VEC cells or MCF10A/HER2 cells expressing dn-Rac1, dn-Pak1, or pBabe vector were seeded at 2.5 x 104 per well on Matrigel-coated transwells and allowed to invade toward growth medium ± TGF-ß. At 24 hours, the cells invading into the Matrigel were counted; columns, mean of three wells; bars, SD. D, the indicated human breast cancer cell lines were seeded at 5 x 104 per well on Matrigel-coated transwells and allowed to invade toward growth medium ± TGF-ß1. At 48 hours, cells invading into the Matrigel were counted; columns, mean of three wells; bars, SD. *, P < 0.001. E, cells were treated with TGF-ß1 (2 ng/mL) for 24 hours or not treated and lysed. Protein extracts were precipitated with a HER2 antibody. The pull-downs or whole-cell lysates were subjected to immunoblotting with the indicated antibodies. F, BT474 cells were treated in serum-free medium with TGF-ß for 30 hours before fixation for immunofluorescence assay with a HER2 antibody (red) and Alexa Fluor 488 phalloidin (green). Colocalization (yellow) was observed in TGF-ß–treated cells. HER2 localization enriched at cell protrusions (arrows). Bar, 50 µm.

 
We next studied if the effects of TGF-ß on cell invasiveness correlated with the levels of HER2 in a panel of human breast cancer cell lines. MCF7 cells underexpress HER2. MDA453 and MDA361 overexpress HER2 protein but do not exhibit gene amplification, whereas in BT474, the HER2 gene is amplified (42). In all lines, except MCF7, treatment with TGF-ß enhanced cancer cell invasion through Matrigel; this increase was 40% to 60% for MDA453 and MDA361 and 180% for BT474 cells (Fig. 5D). A similar trend was observed in terms of the ability of TGF-ß to induce association of HER2 with cytoskeletal elements. Both basal and ligand-stimulated association of HER2 with actin and actinin were absent in MCF7, detectable in MDA453 and MDA361, and highest in BT474 cells (Fig. 5E). Finally, TGF-ß induced colocalization of HER2 and F-actin at cell protrusions only in BT474 cells (Fig. 5F), implying that a threshold of HER2 is required for TGF-ß–stimulated cell invasiveness.

Rac1-Pak1 function is required for TGF-ß–induced cell survival in HER2-overexpressing cells. The Rac1-Pak1 pathway has been shown to regulate cell survival by down-regulating proapoptotic pathways (43, 44). TGF-ß protects tumor cells from serum starvation– or detachment-induced apoptosis (anoikis; refs. 23, 29, 45, 46). Thus, we next examined if Rac1-Pak1 are involved in TGF-ß–mediated protection from anoikis using an Apo-BrdUrd assay in suspended cells expressing dn-Rac1, dn-Pak1, or vector alone. Compared with adherent cells, suspended MCF10A/HER2 control cells showed 38% apoptosis, which could be partially protected by adding TGF-ß. In cells stably expressing dn-Rac1 or dn-Pak1, TGF-ß failed to protect cells from anoikis (Fig. 6A ). Consistent with this result, cells suspended in medium containing TGF-ß exhibited lower levels of cleaved caspase-3 and DNA fragmentation in agarose gels, but this was not observed when dn-Rac1 or dn-Pak1 was expressed (Fig. 6B).


Figure 6
View larger version (61K):
[in this window]
[in a new window]
  		Figure 6. Rac1-Pak1 pathway is required for TGF-ß–induced cell survival. A, MCF10A/HER2 cells expressing dn-Rac1, dn-Pak1, or pBabe vector were suspended and incubated in serum-free medium with or without TGF-ß for 16 hours, followed by Apo-BrdUrd assay. FL3-H images show propidium iodide staining for DNA content. The mean percentage ± SD of apoptotic (gated) cells from three experiments is indicated. Top row, similar number of cells that were allowed to attach on a 100-mm culture dish for 16 hours in serum-free medium. B, the same cell lines were trypsinized and resuspended in serum-free medium with or without TGF-ß. Cells were collected at 0 to 16 hours after resuspension and prepared for caspase-3 immunoblot analysis (top) or DNA laddering in agarose gels (bottom). C, MCF10A/HER2 cells were trypsinized and resuspended in serum-free medium with or without TGF-ß. Cells were harvested at 0 to 8 hours after resuspension and lysates prepared for GST-PBD pull-down and immunoblot with a Rac1 antibody. D, MCF10A/HER2 cells expressing dn-Rac1, dn-Pak1, or pBabe vector were trypsinized and resuspended in serum-free medium with or without TGF-ß. Cells were harvested 8 hours after resuspension and cell lysates prepared for immunoblot analyses with the indicated antibodies. The level of Bad-associated Bcl-2 was determined by immunoprecipitation with a Bad antibody followed by Bcl-2 immunoblot.

 
We next examined GTP-bound Rac1 levels over time in cells suspended in medium with or without TGF-ß. Cells in medium containing TGF-ß exhibited an increase in GTP-bound Rac1 levels as early as 1 hour on suspension (Fig. 6C), suggesting that ligand-induced Rac1 activation does not require cell adhesion. Rac1 and Pak1 have been reported to induce Bad phosphorylation at Ser112 and Ser136, leading to reduced interaction with Bcl-2 and thus enhanced survival (43, 44). Phosphorylation of Bad was induced by TGF-ß in suspended MCF10A/HER2 cells expressing pBabe vector alone but not in cells expressing dn-Rac1 or dn-Pak1. This increase of Bad phosphorylation in TGF-ß–treated cells correlated temporally with a decrease of Bad-associated Bcl-2 (Fig. 6D). As expected, both dn-Rac1 and dn-Pak1 inhibited ligand-stimulated Pak1 phosphorylation in Thr423, which corresponds to an autophosphorylation site in the activation loop of its kinase domain (26). However, they did not affect basal or TGF-ß–stimulated Ser473 Akt phosphorylation (Fig. 6D, seventh row) or PI3K activity (data not shown), suggesting that Rac1-Pak1 are downstream of activated PI3K.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
There is increasing evidence that in advanced cancers, TGF-ß predominantly contributes to tumor progression. Although this outcome may involve TGF-ß–mediated paracrine effects that modulate the tumor microenvironment and the host immune system, other data indicate that the effects of TGF-ß on the tumor cells themselves have an essential role in cancer cell viability and progression (47). In addition to activating the Smad tumor suppressor pathway, TGF-ß can engage signaling programs associated with transformation such as Ras/MAPK, PI3K/Akt, Rho GTPases, NF-{kappa}B, and JNK (8, 48, 49). In some cases, blockade of these pathways abrogates the transforming effects of TGF-ß. For example, MAPK and PI3K inhibitors block TGF-ß–induced motility in oncogene-overexpressing mammary epithelial cells (5, 6). TGF-ß–induced epithelial to mesenchymal transition, protection from apoptosis, and/or tumor cell migration is blocked by the PI3K inhibitor LY294002 or with expression of dn-Akt (23, 29–31, 46). The ErbB receptor network, which includes the HER2/Neu (ErbB2) tyrosine kinase, has been shown to synergize with TGF-ß in transformation (see Introduction). Molecular mechanisms of ErbB/TGF-ß cooperation that have been proposed include (i) differential transcriptional modulation in cooperation with Smads; (ii) activation of Smad-independent signal transducers, such as those listed above; (iii) attenuation of TGF-ß–mediated, Smad-dependent antimitogenic effects via up-regulation of Smad7; and (iv) autocrine induction of ligands that activate ErbB tyrosine kinases (50).

We have shown that in HER2-overexpressing mammary epithelial cells, TGF-ß, via activation of PI3K, recruits the actin cytoskeleton to HER2, which then colocalizes with Vav2, activated Rac1, and its effector Pak1 at cell lamellipodia, leading to prolonged Rac1 activation and enhanced cell motility and survival. These effects occurred in the presence of cycloheximide as well as in suspended cells, suggesting that they did not require gene expression and/or new protein synthesis or cell adhesion. As such, they represent a novel mechanism of ErbB/TGF-ß/cytoskeleton signaling cross-talk that may not invoke transcriptional effects.

It is generally accepted that in cells with low HER2 levels, the receptor is unable to generate a transforming signal. Studies with natural and transfected cell lines and primary human tumors that overexpress HER2 have shown ligand-independent phosphorylation of HER2, as shown in this study in MCF10A/HER2 cells. The biochemical basis of this constitutive phosphorylation is not clear but concurs with the reported ability of wild-type Neu to spontaneously multimerize and become activated when present in cells at high density (51). Another possible mechanism of HER2 phosphorylation is via ligand-induced activation of EGFR (ErbB1), ErbB3, or ErbB4 (28). Indeed, in MCF10A/HER2 cells, HER2 was constitutively phosphorylated and associated with Rac1, Pak1, Vav2, actin, and actinin. This resulted in activation of PI3K/Akt in the absence of added ligands. Interestingly, inhibition of Rac1 and Pak1 with dominant-negative constructs markedly reduced basal HER2 phosphorylation without affecting HER2 protein levels (Fig. 2C), suggesting that Rac1-Pak1 are permissive for ligand-independent HER2 activity. This is consistent with the role of Rac1-Pak1 in transformation induced by other oncogenes (33, 52). However, the magnitude and stability of these protein-protein associations and signaling output in the absence of TGF-ß were not above a threshold necessary to recruit the HER2/Vav2/Rac1/Pak1/actin/actinin complex to the leading edge of the cell to initiate and maintain cell motility. By recruiting the actin skeleton, TGF-ß stabilizes and/or "cross-links" this complex at cell lamellipodia, thus prolonging Rac1 activation (Fig. 7 ). In turn, this localization of HER2 at membrane protrusions may further enhance its resistance for internalization and prolong signaling (53).


Figure 7
View larger version (30K):
[in this window]
[in a new window]
  		Figure 7. Temporal and spatial modulation of HER2 signaling to Rac1-Pak1 by TGF-ß. In cells with low HER2 levels, the receptor is not activated or associated with signal transducers or the actin cytoskeleton (A). In cells with HER2 gene amplification, HER2 is constitutively phosphorylated and associated with PI3K, Rac1, Pak1, Vav2, actin, and actinin. Although this complex may be enough to generate survival or proliferative signals, because it is not enriched at cell protrusions, it cannot induce cell motility (B). In cells with overexpressed and activated HER2, TGF-ß induces PI3K above a necessary threshold required for the recruitment and stabilization of F-actin and the HER2 complex at cell protrusions, thus leading to prolonged Rac1 activation, cell movement, and increased survival on cell detachment (C).

 
The recruitment of the HER2/Vav2/Rac1/Pak1/actin/actinin complex to cell protrusions and cell motility were abrogated by the inhibition of PI3K with either dn-p85 or wortmannin, by inhibition of HER2 phosphorylation with gefitinib, by stable expression of either dn-Rac1 or dn-Pak1, and by RNAi against actinin, suggesting that the reciprocal interplay of PI3K, HER2 kinase, and Rac GTPase activities with the actin cytoskeleton is necessary for TGF-ß action in oncogene-overexpressing cells. Treatment with wortmannin blocked TGF-ß–induced activation of Rac1 and its association with Vav2, but expression of dn-Rac1 and dn-Pak1 did not affect ligand-induced p-Akt or PI3K catalytic activity, suggesting that Rac1-Pak1 are downstream of PI3K. Similar data have been reported by Wilkes et al. (9, 54) in fibroblasts where TGF-ß can induce PI3K-dependent and Akt-independent Pak2 activity. We speculate, however, that PI3K-dependent, locally activated Rac1 may mediate the accumulation of phosphatidilinositol-3,4,5 trisphosphate and actin polymers at membrane protrusions as it has been shown in nontransformed cell systems (55, 56). This speculation requires further investigation.

Both Rac1 and Pak1 have been shown to suppress apoptosis (57–59). Rac1 stimulates Bad phosphorylation on Ser75, thereby suppressing drug-induced caspase activation and apoptosis in human lymphoma cells (44). Pak1 can also phosphorylate Bad in vitro and in vivo on Ser112 and Ser136 (43). Pak1 also contributes to cell survival via phosphorylation of the forkhead transcription factor FKHR, leading to the exclusion of FKHR from the nucleus (60). Akt can stimulate Pak1 through a GTPase-independent mechanism and, in turn, induce Bad phosphorylation and enhance survival, which can be abolished by dn-Pak1 (33). Levels of p-Akt were similar in MCF10A/HER2 cells expressing dn-Rac1, dn-Pak1, and vector alone either in the presence or absence of TGF-ß. TGF-ß prevented anoikis, induced phosphorylation of Bad at Ser136, and uncoupled Bcl-2 from Bad in control cells but not in those expressing dn-Rac1 or dn-Pak1 (Fig. 6D). These data suggest that TGF-ß and HER2 synergize for enhanced tumor cell survival via PI3K and Rac1-Pak1 and that Akt is not sufficient to protect these cells from anoikis.

In summary, TGF-ß modulates HER2 signaling to Rac1-Pak1 by compartmentalizing a HER2/Vav2/Rac1/Pak1/actin/actinin complex to the cell lamellipodia. This results in enhanced cell motility and survival. If operative in cancer cells, this mechanism would enhance tumor cell dissemination from primary tumor sites and their subsequent metastatic progression. In addition, this cross-talk suggests a rationale for the combined use of HER2 and TGF-ß inhibitors in cancer patients with HER2 overexpression and evidence of active TGF-ß signaling.


   Acknowledgments
 
Grant support: NIH R01 grants CA62212 and CA80195 (C.L. Arteaga), Breast Cancer Specialized Program of Research Excellence grant P50 CA98131, and Vanderbilt-Ingram Comprehensive Cancer Center Support grant P30 CA68485.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Drs. Shimian Qu, Jae Youn Yi, and Yasuhiro Koh for helpful comments and Teresa Dugger for administrative and technical support.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Received 6/ 6/06. Revised 7/21/06. Accepted 8/16/06.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Siegel PM, Massague J. Cytostatic and apoptotic actions of TGF-ß in homeostasis and cancer. Nat Rev Cancer 2003;3:807–20.[CrossRef][Medline]
  2. Siegel PM, Shu W, Cardiff RD, Muller WJ, Massague J. Transforming growth factor ß signaling impairs Neu-induced mammary tumorigenesis while promoting pulmonary metastasis. Proc Natl Acad Sci U S A 2003;100:8430–5.[Abstract/Free Full Text]
  3. Muraoka RS, Koh Y, Roebuck LR, et al. Increased malignancy of Neu-induced mammary tumors overexpressing active transforming growth factor ß1. Mol Cell Biol 2003;23:8691–703.[Abstract/Free Full Text]
  4. Muraoka-Cook RS, Shin I, Yi JY, et al. Activated type I TGFß receptor kinase enhances the survival of mammary epithelial cells and accelerates tumor progression. Oncogene 2006;25:3408–23.[CrossRef][Medline]
  5. Seton-Rogers SE, Lu Y, Hines LM, et al. Cooperation of the ErbB2 receptor and transforming growth factor ß in induction of migration and invasion in mammary epithelial cells. Proc Natl Acad Sci U S A 2004;101:1257–62.[Abstract/Free Full Text]
  6. Ueda Y, Wang S, Dumont N, Yi JY, Koh Y, Arteaga CL. Overexpression of HER2 (erbB2) in human breast epithelial cells unmasks transforming growth factor ß-induced cell motility. J Biol Chem 2004;279:24505–13.[Abstract/Free Full Text]
  7. Mulder KM. Role of Ras and Mapks in TGFß signaling. Cytokine Growth Factor Rev 2000;11:23–35.[CrossRef][Medline]
  8. Yi JY, Shin I, Arteaga CL. Type I transforming growth factor ß receptor binds to and activates phosphatidylinositol 3-kinase. J Biol Chem 2005;280:10870–6.[Abstract/Free Full Text]
  9. Wilkes MC, Mitchell H, Penheiter SG, et al. Transforming growth factor-ß activation of phosphatidylinositol 3-kinase is independent of Smad2 and Smad3 and regulates fibroblast responses via p21-activated kinase-2. Cancer Res 2005;65:10431–40.[Abstract/Free Full Text]
  10. Ventura JJ, Kennedy NJ, Flavell RA, Davis RJ. JNK regulates autocrine expression of TGF-ß1. Mol Cell 2004;15:269–78.[CrossRef][Medline]
  11. Jaffe AB, Hall A. Rho GTPases: biochemistry and biology. Annu Rev Cell Dev Biol 2005;21:247–69.[CrossRef][Medline]
  12. Etienne-Manneville S, Hall A. Rho GTPases in cell biology. Nature 2002;420:629–35.[CrossRef][Medline]
  13. Schmidt A, Hall A. Guanine nucleotide exchange factors for Rho GTPases: turning on the switch. Genes Dev 2002;16:1587–609.[Free Full Text]
  14. Mucsi I, Skorecki KL, Goldberg HJ. Extracellular signal-regulated kinase and the small GTP-binding protein, Rac, contribute to the effects of transforming growth factor-ß1 on gene expression. J Biol Chem 1996;271:16567–72.[Abstract/Free Full Text]
  15. Atfi A, Djelloul S, Chastre E, Davis R, Gespach C. Evidence for a role of Rho-like GTPases and stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK) in transforming growth factor ß-mediated signaling. J Biol Chem 1997;272:1429–32.[Abstract/Free Full Text]
  16. Bhowmick NA, Ghiassi M, Bakin A, et al. Transforming growth factor-ß1 mediates epithelial to mesenchymal transdifferentiation through a RhoA-dependent mechanism. Mol Biol Cell 2001;12:27–36.[Abstract/Free Full Text]
  17. Bakin AV, Rinehart C, Tomlinson AK, Arteaga CL. p38 mitogen-activated protein kinase is required for TGFß-mediated fibroblastic transdifferentiation and cell migration. J Cell Sci 2002;115:3193–206.[Abstract/Free Full Text]
  18. Adam L, Vadlamudi R, Kondapaka SB, Chernoff J, Mendelsohn J, Kumar R. Heregulin regulates cytoskeletal reorganization and cell migration through the p21-activated kinase-1 via phosphatidylinositol-3 kinase. J Biol Chem 1998;273:28238–46.[Abstract/Free Full Text]
  19. Bourguignon LY, Zhu H, Zhou B, Diedrich F, Singleton PA, Hung MC. Hyaluronan promotes CD44v3-2 interaction with Grb2-185(HER2) and induces Rac1 and Ras signaling during ovarian tumor cell migration and growth. J Biol Chem 2001;276:48679–92.[Abstract/Free Full Text]
  20. Yang C, Liu Y, Lemmon MA, Kazanietz MG. Essential role for Rac in heregulin ß1 mitogenic signaling: a mechanism that involves epidermal growth factor receptor and is independent of ErbB4. Mol Cell Biol 2006;26:831–42.[Abstract/Free Full Text]
  21. Wang SE, Wu FY, Shin I, Qu S, Arteaga CL. Transforming growth factor ß (TGF-ß)-Smad target gene protein tyrosine phosphatase receptor type {kappa} is required for TGF-ß function. Mol Cell Biol 2005;25:4703–15.[Abstract/Free Full Text]
  22. Kiosses WB, Daniels RH, Otey C, Bokoch GM, Schwartz MA. A role for p21-activated kinase in endothelial cell migration. J Cell Biol 1999;147:831–44.[Abstract/Free Full Text]
  23. Shin I, Bakin AV, Rodeck U, Brunet A, Arteaga CL. Transforming growth factor ß enhances epithelial cell survival via Akt-dependent regulation of FKHRL1. Mol Biol Cell 2001;12:3328–39.[Abstract/Free Full Text]
  24. Debnath J, Muthuswamy SK, Brugge JS. Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods 2003;30:256–68.[CrossRef][Medline]
  25. Honda K, Yamada T, Endo R, et al. Actinin-4, a novel actin-bundling protein associated with cell motility and cancer invasion. J Cell Biol 1998;140:1383–93.[Abstract/Free Full Text]
  26. Daniels RH, Bokoch GM. p21-activated protein kinase: a crucial component of morphological signaling? Trends Biochem Sci 1999;24:350–5.[CrossRef][Medline]
  27. Moulder SL, Yakes FM, Muthuswamy SK, Bianco R, Simpson JF, Arteaga CL. Epidermal growth factor receptor (HER1) tyrosine kinase inhibitor ZD1839 (Iressa) inhibits HER2/neu (erbB2)-overexpressing breast cancer cells in vitro and in vivo. Cancer Res 2001;61:8887–95.[Abstract/Free Full Text]
  28. Yarden Y, Sliwkowski MX. Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2001;2:127–37.[CrossRef][Medline]
  29. Muraoka-Cook RS, Kurokawa H, Koh Y, et al. Conditional overexpression of active transforming growth factor ß1 in vivo accelerates metastases of transgenic mammary tumors. Cancer Res 2004;64:9002–11.[Abstract/Free Full Text]
  30. Bakin AV, Tomlinson AK, Bhowmick NA, Moses HL, Arteaga CL. Phosphatidylinositol 3-kinase function is required for transforming growth factor ß-mediated epithelial to mesenchymal transition and cell migration. J Biol Chem 2000;275:36803–10.[Abstract/Free Full Text]
  31. Gotzmann J, Huber H, Thallinger C, et al. Hepatocytes convert to a fibroblastoid phenotype through the cooperation of TGF-ß1 and Ha-Ras: steps towards invasiveness. J Cell Sci 2002;115:1189–202.[Abstract/Free Full Text]
  32. Benard V, Bohl BP, Bokoch GM. Characterization of rac and cdc42 activation in chemoattractant-stimulated human neutrophils using a novel assay for active GTPases. J Biol Chem 1999;274:13198–204.[Abstract/Free Full Text]
  33. Tang Y, Zhou H, Chen A, Pittman RN, Field J. The Akt proto-oncogene links Ras to Pak and cell survival signals. J Biol Chem 2000;275:9106–9.[Abstract/Free Full Text]
  34. Papakonstanti EA, Stournaras C. Association of PI-3 kinase with PAK1 leads to actin phosphorylation and cytoskeletal reorganization. Mol Biol Cell 2002;13:2946–62.[Abstract/Free Full Text]
  35. King CC, Gardiner EM, Zenke FT, et al. p21-activated kinase (PAK1) is phosphorylated and activated by 3-phosphoinositide-dependent kinase-1 (PDK1). J Biol Chem 2000;275:41201–9.[Abstract/Free Full Text]
  36. Han J, Luby-Phelps K, Das B, et al. Role of substrates and products of PI 3-kinase in regulating activation of Rac-related guanosine triphosphatases by Vav. Science 1998;279:558–60.[Abstract/Free Full Text]
  37. Marcoux N, Vuori K. EGF receptor mediates adhesion-dependent activation of the Rac GTPase: a role for phosphatidylinositol 3-kinase and Vav2. Oncogene 2003;22:6100–6.[CrossRef][Medline]
  38. Tamas P, Solti Z, Bauer P, et al. Mechanism of epidermal growth factor regulation of Vav2, a guanine nucleotide exchange factor for Rac. J Biol Chem 2003;278:5163–71.[Abstract/Free Full Text]
  39. Kotani K, Ogawa W, Hino Y, et al. Dominant negative forms of Akt (protein kinase B) and atypical protein kinase C{lambda} do not prevent insulin inhibition of phosphoenolpyruvate carboxykinase gene transcription. J Biol Chem 1999;274:21305–12.[Abstract/Free Full Text]
  40. Debnath J, Brugge JS. Modelling glandular epithelial cancers in three-dimensional cultures. Nat Rev Cancer 2005;5:675–88.[CrossRef][Medline]
  41. Irie HY, Pearline RV, Grueneberg D, et al. Distinct roles of Akt1 and Akt2 in regulating cell migration and epithelial-mesenchymal transition. J Cell Biol 2005;171:1023–34.[Abstract/Free Full Text]
  42. Yakes FM, Chinratanalab W, Ritter CA, King W, Seelig S, Arteaga CL. Herceptin-induced inhibition of phosphatidylinositol-3 kinase and Akt is required for antibody-mediated effects on p27, cyclin D1, and antitumor action. Cancer Res 2002;62:4132–41.[Abstract/Free Full Text]
  43. Schurmann A, Mooney AF, Sanders LC, et al. p21-activated kinase 1 phosphorylates the death agonist bad and protects cells from apoptosis. Mol Cell Biol 2000;20:453–61.[Abstract/Free Full Text]
  44. Zhang B, Zhang Y, Shacter E. Rac1 inhibits apoptosis in human lymphoma cells by stimulating Bad phosphorylation on Ser-75. Mol Cell Biol 2004;24:6205–14.[Abstract/Free Full Text]
  45. Sachsenmeier KF, Sheibani N, Schlosser SJ, Allen-Hoffmann BL. Transforming growth factor-ß1 inhibits nucleosomal fragmentation in human keratinocytes following loss of adhesion. J Biol Chem 1996;271:5–8.[Abstract/Free Full Text]
  46. Horowitz JC, Lee DY, Waghray M, et al. Activation of the pro-survival phosphatidylinositol 3-kinase/AKT pathway by transforming growth factor-ß1 in mesenchymal cells is mediated by p38 MAPK-dependent induction of an autocrine growth factor. J Biol Chem 2004;279:1359–67.[Abstract/Free Full Text]
  47. Dumont N, Arteaga CL. Targeting the TGFß signaling network in human neoplasia. Cancer Cell 2003;3:531–6.[CrossRef][Medline]
  48. Derynck R, Zhang YE. Smad-dependent and Smad-independent pathways in TGF-ß family signalling. Nature 2003;425:577–84.[CrossRef][Medline]
  49. Lu T, Burdelya LG, Swiatkowski SM, et al. Secreted transforming growth factor ß2 activates NF-{kappa}B, blocks apoptosis, and is essential for the survival of some tumor cells. Proc Natl Acad Sci U S A 2004;101:7112–7.[Abstract/Free Full Text]
  50. Grunert S, Jechlinger M, Beug H. Diverse cellular and molecular mechanisms contribute to epithelial plasticity and metastasis. Nat Rev Mol Cell Biol 2003;4:657–65.[CrossRef][Medline]
  51. Samanta A, LeVea CM, Dougall WC, Qian X, Greene MI. Ligand and p185c-neu density govern receptor interactions and tyrosine kinase activation. Proc Natl Acad Sci U S A 1994;91:1711–5.[Abstract/Free Full Text]
  52. Tang Y, Chen Z, Ambrose D, et al. Kinase-deficient Pak1 mutants inhibit Ras transformation of Rat-1 fibroblasts. Mol Cell Biol 1997;17:4454–64.[Abstract]
  53. Hommelgaard AM, Lerdrup M, van Deurs B. Association with membrane protrusions makes ErbB2 an internalization-resistant receptor. Mol Biol Cell 2004;15:1557–67.[Abstract/Free Full Text]
  54. Wilkes MC, Murphy SJ, Garamszegi N, Leof EB. Cell-type-specific activation of PAK2 by transforming growth factor ß independent of Smad2 and Smad3. Mol Cell Biol 2003;23:8878–89.[Abstract/Free Full Text]
  55. Wang F, Herzmark P, Weiner OD, Srinivasan S, Servant G, Bourne HR. Lipid products of PI(3)Ks maintain persistent cell polarity and directed motility in neutrophils. Nat Cell Biol 2002;4:513–8.[CrossRef][Medline]
  56. Cote JF, Motoyama AB, Bush JA, Vuori K. A novel and evolutionarily conserved PtdIns(3,4,5)P3-binding domain is necessary for DOCK180 signalling. Nat Cell Biol 2005;7:797–807.[CrossRef][Medline]
  57. Senger DL, Tudan C, Guiot MC, et al. Suppression of Rac activity induces apoptosis of human glioma cells but not normal human astrocytes. Cancer Res 2002;62:2131–40.[Abstract/Free Full Text]
  58. Pervaiz S, Cao J, Chao OS, Chin YY, Clement MV. Activation of the RacGTPase inhibits apoptosis in human tumor cells. Oncogene 2001;20:6263–8.[CrossRef][Medline]
  59. Frost JA, Swantek JL, Stippec S, Yin MJ, Gaynor R, Cobb MH. Stimulation of NF{kappa}B activity by multiple signaling pathways requires PAK1. J Biol Chem 2000;275:19693–9.[Abstract/Free Full Text]
  60. Mazumdar A, Kumar R. Estrogen regulation of Pak1 and FKHR pathways in breast cancer cells. FEBS Lett 2003;535:6–10.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Mol. Cell. Biol.Home page
S. E. Wang, B. Xiang, M. Guix, M. G. Olivares, J. Parker, C. H. Chung, A. Pandiella, and C. L. Arteaga
Transforming Growth Factor {beta} Engages TACE and ErbB3 To Activate Phosphatidylinositol-3 Kinase/Akt in ErbB2-Overexpressing Breast Cancer and Desensitizes Cells to Trastuzumab
Mol. Cell. Biol., September 15, 2008; 28(18): 5605 - 5620.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
S. Kikuchi, K. Honda, H. Tsuda, N. Hiraoka, I. Imoto, T. Kosuge, T. Umaki, K. Onozato, M. Shitashige, U. Yamaguchi, et al.
Expression and Gene Amplification of Actinin-4 in Invasive Ductal Carcinoma of the Pancreas
Clin. Cancer Res., September 1, 2008; 14(17): 5348 - 5356.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
Y. Hu, X. Hu, L. Boumsell, and L. B. Ivashkiv
IFN-{gamma} and STAT1 Arrest Monocyte Migration and Modulate RAC/CDC42 Pathways
J. Immunol., June 15, 2008; 180(12): 8057 - 8065.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
J. J. Northey, J. Chmielecki, E. Ngan, C. Russo, M. G. Annis, W. J. Muller, and P. M. Siegel
Signaling through ShcA Is Required for Transforming Growth Factor {beta}- and Neu/ErbB-2-Induced Breast Cancer Cell Motility and Invasion
Mol. Cell. Biol., May 15, 2008; 28(10): 3162 - 3176.
[Abstract] [Full Text] [PDF]

Home page
 JCOHome page
M. Guix, N. de Matos Granja, I. Meszoely, T. B. Adkins, B. M. Wieman, K. E. Frierson, V. Sanchez, M. E. Sanders, A. M. Grau, I. A. Mayer, et al.
Short Preoperative Treatment With Erlotinib Inhibits Tumor Cell Proliferation in Hormone Receptor-Positive Breast Cancers
J. Clin. Oncol., February 20, 2008; 26(6): 897 - 906.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
C. A. Ritter, M. Perez-Torres, C. Rinehart, M. Guix, T. Dugger, J. A. Engelman, and C. L. Arteaga
Human Breast Cancer Cells Selected for Resistance to Trastuzumab In vivo Overexpress Epidermal Growth Factor Receptor and ErbB Ligands and Remain Dependent on the ErbB Receptor Network
Clin. Cancer Res., August 15, 2007; 13(16): 4909 - 4919.
[Abstract] [Full Text] [PDF]