Erbb2 Promotes Src Synthesis and Stability: Novel Mechanisms of Src Activation That Confer Breast Cancer Metastasis

Research Article

ErbB2 Promotes Src Synthesis and Stability: Novel Mechanisms of Src Activation That Confer Breast Cancer Metastasis

Ming Tan,1 Ping Li,1 Kristine S. Klos,1 Jing Lu,1 Keng-Hsueh Lan,1 Yoichi Nagata,1 Dexing Fang,1 Tong Jing,1 and Dihua Yu1,2

Departments of 1Surgical Oncology and 2Molecular and Cellular Oncology, University of Texas M.D. Anderson Cancer Center, Houston, Texas

Abstract The ErbB2 RTK has intrinsic tyrosine kinase activity (1, 6). Activation of the ErbB2 tyrosine kinase leads to autophosphor- Activation of Src kinase plays important roles in the development of many neoplasias. Most of the previous Src ylation of tyrosine residues in its carboxyl-terminal domain. The studies focused on the deregulation of Src kinase activity. phosphorylated tyrosine residues then function as docking sites The deregulated Src protein synthesis and stability in for ErbB2 downstream signaling molecules (6, 8). Previous mediating malignant phenotypes of cancer cells, however, studies showed that kinase-activated ErbB2 physically associates have been neglected. While investigating the signal trans- with Shc (9), Grb-7 (10), Ras-GTPase-activating protein (11), phospholipase Cg (11, 12), the 85-kDa regulatory subunit (p85) duction pathways contributing to ErbB2-mediated metasta- h sis, we found that ErbB2-activated breast cancer cells that of phosphatidylinositol 3-kinase (12, 13), Src (14, 15), -catenin (16), Csk-homologous kinase (17), Erbin (18), Ralt (19), c-Cbl had higher metastatic potentials also had increased Src Cdc2 activity compared with ErbB2 low-expressing cells. The (20), and p34 (21). However, the overall picture of how ErbB2 exerts its function through these signaling molecules is increased Src activity in ErbB2-activated cells paralleled higher Src protein levels, whereas Src RNA levels were not unclear, and the signaling events responsible for ErbB2- significantly altered. Our studies revealed two novel mech- mediated cancer metastasis remain elusive. Because most anisms that are involved in Src protein up-regulation and patients with ErbB2-overexpressing breast cancer die from activation by ErbB2: (a) ErbB2 increased Src translation metastasis, unraveling the molecular mechanisms and signaling through activation of the Akt/mammalian target of rapamy- events that underlie ErbB2-mediated metastasis is extremely cin/4E-BP1 pathway and (b) ErbB2 increased Src stability important and may facilitate the development of therapies or most likely through the inhibition of the calpain protease. prevention strategies for patients with ErbB2-overexpressing Furthermore, inhibition of Src activity by a Src-specific tumors. The human c-Src proto-oncogene encodes an intracellular inhibitor, PP2, or a Src dominant-negative mutant dramat- tyrosine kinase, pp60c-Src (22, 23). c-Src was originally identified as ically reduced ErbB2-mediated cancer cell invasion in vitro a homologue of the transforming gene of the Rous sarcoma virus and metastasis in an experimental metastasis animal model. (24) and is the prototype of a family of highly conserved genes Together, activation of ErbB2 and downstream signaling that includes yes, fyn, fgr, lyn, lck, hck, blk, and yrk. Src family pathways can lead to increased Src protein synthesis and proteins have been implicated in many signal transduction decreased Src protein degradation resulting in Src up- pathways and a wide variety of cellular functions (25). Activation regulation and activation, which play critical roles in of Src kinase has been linked to the development of many human ErbB2-mediated breast cancer invasion and metastasis. neoplasias, especially those of the colon, breast, lung, and (Cancer Res 2005; 65(5): 1858-67) pancreas (26). Most of the previous Src studies focused on the deregulation of Src kinase activity but not the altered expression Introduction levels of Src protein. The deregulated Src protein synthesis and ErbB2 is a member of the epidermal growth factor receptor stability in mediating malignant phenotypes of cancer cells have family called class I receptor tyrosine kinases (RTK; refs. 1, 2). The been largely neglected. erbB2 gene has been found amplified or overexpressed in f30% Tumors and cell lines that overexpress ErbB2 have been of human breast cancers (3) and in many other cancer types (4). reported to contain activated Src family kinases, especially Patients with ErbB2-overexpressing breast cancer have substan- pp60c-Src (14, 15, 27). Although these previous studies suggest tially lower survival rates and shorter times to relapse than that Src activation might play a role in the malignant patients without the overexpression. Moreover, many clinical and phenotypes induced by ErbB2, whether Src is important in laboratory investigations have shown that overexpression of ErbB2-mediated breast cancer metastasis has not been investi- ErbB2 leads to increased breast cancer metastasis (5, 6). Results gated, and the mechanism by which ErbB2 activates Src has not from our laboratory also showed that ErbB2 overexpression been well established. enhances the intrinsic metastatic potential of human breast To elucidate the signaling pathways downstream of ErbB2 that cancer cells (7). contribute to ErbB2-mediated breast cancer metastasis, we expressed wild-type (wt) or mutant ErbB2 proteins with altered signaling capacity in the MDA-MB-435 human breast cancer cell line that expresses low levels of ErbB2. We then examined the Requests for reprints: Dihua Yu, Department of Surgical Oncology, University of metastatic potentials of these cells and their relationship with the Texas M.D. Anderson Cancer Center, Unit 107, Houston, TX 77030. Phone: 713-792-3636; Fax: 713-794-4830; E-mail: [email protected]. activation of ErbB2 downstream signals that might be involved in I2005 American Association for Cancer Research. metastasis. We found that ErbB2 activated Src kinase through the

Cancer Res 2005; 65: (5). March 1, 2005 1858 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. ErB2 Promotes Src Protein Synthesis and Stability up-regulation of the Src protein level via increasing Src protein Diego, CA). The survival curves were compared by the Mantel-Haenszel synthesis and stability. Moreover, we showed that increased Src log-rank test. All animal experiments in this study were done under the activities were required for ErbB2-mediated breast cancer animal use guidelines of the NIH (Bethesda, MD) and the Animal Care and metastasis. Use Committee of the University of Texas M.D. Anderson Cancer Center (Houston, TX). Src Kinase Assay. Cells were lysed and immunoprecipitated with anti- Materials and Methods Src antibodies. The immunoprecipitates were washed thrice with Antibodies and Reagents. Antibodies and reagents were purchased immunoprecipitation buffer and twice with a kinase buffer containing from commercial sources: antibodies against Src and ErbB2 (Ab3) were 10 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 10 mmol/L MgCl2, from Oncogene Science (Cambridge, MA); antibodies against Akt, 1 mmol/L DTT, and 0.5 mmol/L orthovanadate. Kinase activity was phospho-Akt (Ser473), 4E-BP1, phospho-4E-BP1 (Ser65), phosphospecific assayed by incubating the immunoprecipitates with 40 AL kinase buffer ErbB2 (Tyr1248), and rapamycin were purchased from Cell Signaling containing 250 Ag/mL enolase, 2 ACi [g-32P]ATP, and 3 AL of 100 mmol/L Technology (Beverly, MA); antibodies against phosphotyrosine (PY20) ATP. After 30 minutes at 37jC, the reaction was terminated by adding were from BD Transduction Laboratories (San Diego, CA); polyclonal 15 ALof6Â SDS sample buffer and boiling for 5 minutes. The samples antibodies against [35S]methionine were from Santa Cruz Biotechnology were subjected to SDS-PAGE. The [g-32P]ATP-labeled enolase was (Santa Cruz, CA); anti-hemagglutinin (HA) monoclonal antibodies (HA.11) visualized by autoradiography. were from Babco (Richmond, CA); and the Src-specific inhibitor PP2 and RNA Interference. BT474 breast cancer cells were cultured to f30% its control chemical PP3, calpain-specific inhibitor N-acetyl-leucyl-leucyl- confluence and then transfected with various concentrations of ErbB2 norleucinal, calpain-specific inhibitor N-acetyl-leucyl-leucyl-methional, siRNA. After 3 hours of transfection, cells were incubated with serum- and MG-132 were purchased from Calbiochem (San Diego, CA). The Src containing medium and cell lysates were collected for Western blot analysis kinase substrate elonase, purified calpain, and cycloheximide were from 48 hours after transfection. A siRNA concentration of 66.7 nmol/L was able Sigma (St. Louis, MO). to significantly decrease ErbB2 protein levels compared with that in the Design and Construction of Small Interfering RNA. Small interfering control GFP siRNA- or mock-transfected cells. RNAs (siRNA) were designed to target ErbB2-specific sequences. Assay of Protein Half-life by Pulse-Chase Experiment. The cells We synthesized two oligonucleotides: 5V-AAGTACACGATGCGGA- were washed thrice with [35S]methionine-free DMEM (Life Technologies) GACTGCCTGTCTC-3V and 5V-AACAGTCTCCGCATCGTGTACCCTGTCTC- supplemented with 3% fetal bovine serum. The cells were incubated for 3V. The Silencer siRNA Construction kit (Ambion, Austin, TX) was used to 1 hour with the wash medium containing 500 ACi/mL [35S]methionine obtain the double-stranded siRNA according to the manufacturer’s (Amersham Pharmacia, Piscataway, NJ). The cells were washed thrice and instruction. incubated in normal growth medium. At the different chase time points, Cell Lines and Cell Cultures. The MDA-MB-435 human breast cancer the cells were lysed in immunoprecipitation buffer, and Src protein was cell line, 435.eB cells (which are MDA-MB-435 cells stably expressing wt immunoprecipitated by the Src antibody. The immunoprecipitates were ErbB2), and 435.neo cells were obtained, established, and cultured as subjected to SDS-PAGE, and radiolabeled Src was detected by reported previously (28). The ErbB2 mutants were subcloned from the pSV2 autoradiography. vectors into pcDNA3 expression vectors (Invitrogen, Carlsbad, CA) in which Protein Synthesis Assay. The cells were washed thrice with transcription is driven by a cytomegalovirus promoter. The resulting [35S]methionine-free DMEM (Life Technologies) supplemented with 3% plasmids were transfected into MDA-MB-435 cells as described (29), and the fetal bovine serum and then incubated for 1 or 4 hours with the wash stable transfectants were selected by G418 (geneticin, Life Technologies, medium containing 500 ACi/mL [35S]methionine. Cell lysates were Rockville, MD). Western blot was done to identify transfectants that extracted and equal amounts of protein lysates were subjected to expressed the ErbB2 protein. Stable wt ErbB2 transfectants of the human immunoprecipitation with the antibody against Src or h-actin, and the breast cancer cell line MDA-MB-231 were established using the same immunoprecipitates were analyzed by SDS-PAGE and radiolabeled method. The dominant-negative mutant sequence of Src (DNSrc) was proteins were detected by autoradiography. subcloned from the expression vector pcR295 into the pIRESpuro2 vector Experimental Metastasis Assays. We injected 1 Â 106 cancer cells into (Clontech, Palo Alto, CA). The resulting plasmid (DNSrc) was transfected the tail veins of female ICR-SCID mice. On the same day of the injections, into 435.neo or 435.eB cells, and the transfectants were selected using the mice were injected i.p. with 1 mg/kg body weight of Src kinase inhibitor puromycin (Clontech). HA-tagged ubiquitin expression vector (pCGN-Ub) PP2 or its solvent, 10% DMSO-PBS. Three days later, the mice were injected was constructed and expressed as described previously (30). All breast with a second dose of PP2 or 10% DMSO-PBS. The injection schedules were cancer cells used in this study were grown in DMEM and Ham’s F-12 then changed to once a week, and the injections continued for 12 weeks. (DMEM/F-12, Life Technologies). Osteosacoma cell line Saos-2 was cultured The mice were euthanized 90 days later, and their lungs were instilled with in RPMI 1640 (Life Technologies). India ink. Metastatic lung nodules >0.5 mm in diameter were counted. Western Blot and Immunoprecipitation Analyses. Western blot and Statistical analyses were done by Student’s t test. immunoprecipitation analyses were done as described previously (31). The The experimental metastasis assay by i.v. injection of the 435.eB or protein inputs in the Western blot and immunoprecipitation analyses were eB.DNSrc was done and analyzed similar to the above described, except normalized by loading the equal amount of total protein lysates onto the that 2 Â 106 cancer cells were injected, mice were not treated with PP2, and SDS-PAGE gels. The nonradioactive signals were quantified by Alpha Imager mice were euthanized 60 days after injection. 2000 (Alpha Innotech, San Leandro, CA). The radioactive signals were quantified by PhosphoImager (Molecular Dynamics, Sunnyvale, CA). In vitro Cell Invasion Assay. Cell invasion was assayed as described Results previously (7). Breast Cancer Cells with Increased ErbB2 Kinase Activity In vivo Metastatic Survival Rates. Single-cell suspensions of each cell Have Increased Metastatic Potential. To investigate the ErbB2 line were injected into the lateral tail veins of 8-week-old female severe downstream signals that contribute to ErbB2-mediated breast combined immunodeficient (ICR-SCID) mice (1 Â 106 cells per mouse, cancer metastasis, we chose to determine the metastatic potential Harlan, Madison, WI). The injected mice were maintained under identical conditions and were monitored regularly. Once a mouse became moribund, of isogenic breast cancer cells that overexpress wt or mutant forms euthanasia and necropsy were done, and a veterinary pathologist was of ErbB2 and exhibit impaired or enhanced ErbB2 RTK activities 659 consulted to determine the cause of the animal’s death. The times of death (32). The V659E mutant has a Val Glu mutation at the for the mice that died of metastasis were plotted. Survival curves were transmembrane domain that leads to constitutive activation of drawn and analyzed using Prism software (GraphPad Software, San the ErbB2 tyrosine kinase. The K753M mutant, in addition to www.aacrjournals.org 1859 Cancer Res 2005; 65: (5). March 1, 2005

having the Val659Glu mutation, has a Lys753Met mutation at the ATP binding site of the ErbB2 kinase domain, resulting in the loss of its tyrosine kinase activity. We stably transfected these two mutants, the wt ErbB2 or the pcDNA3 vector, into the ErbB2 low- expressing MDA-MB-435 human breast cancer cell line. ErbB2 protein expression levels were detected by immunoblotting on protein lysates from the parental MDA-MB-435 cells, vector- transfected control 435.neo cells, and wt or mutant ErbB2 transfectants. ErbB2 transfectants that expressed the ErbB2 protein at comparable levels were selected for further studies (Fig. 1A, top). ErbB2 tyrosine autophosphorylation corresponds to the kinase activity of ErbB2 (33–35). Therefore, we determined the changes in ErbB2 tyrosine autophosphorylation in wt and mutant ErbB2- expressing MDA-MB-435 cells by immunoblotting using an ErbB2 phosphotyrosine-specific antibody (Fig. 1A, top). As expected, little ErbB2 tyrosine phosphorylation was detected in parental MDA- MB-435, 435.neo, or kinase-defective K753M cells, indicating very low levels of ErbB2 kinase activation in these cells. However, 435.eB and V659E cells overexpressing wt or activated ErbB2 had high levels of ErbB2 phosphorylation, indicating that the ErbB2 kinase was activated in these cells. Transphosphorylation of other cellular proteins was detected on whole cell lysate by immunoblotting with antibody against phosphotyrosine (Fig. 1A, bottom). The 435.neo and K753M cells showed low levels of tyrosine transphosphorylation (Fig. 1A, bottom). In contrast, 435.eB and V659E cells, which had higher levels of ErbB2 autophosphorylation (Fig. 1A, top), had higher levels of transphosphorylation (Fig. 1A, bottom). These results confirm that ErbB2 kinase activities are increased in 435.eB and V659E cells but not in K753M cells. To determine the impact of ErbB2 activation on breast cancer metastasis, we examined MDA-MB-435 cells, 435.neo cells, and wt or mutant ErbB2 transfectants for their metastatic potential in vitro and in vivo. An invasion assay showed that 435.eB and V659E cells were highly invasive compared with MDA-MB-435, 435.neo, and K753M cells (Fig. 1B), which correlated with increased ErbB2 kinase activities (Fig. 1A). To test the metastatic potential of these cells in vivo, we injected the cells into the lateral tail veins of immunodeficient ICR-SCID mice. The mice later died of metastases, and the metastatic survival rates of the mice were recorded. The mice injected with 435.eB or V659E cells had significantly lower metastatic survival rates than those injected with MDA-MB-435 and 435.neo cells (435.eB versus 435.neo, P = 0.0420; V659E versus 435.neo, P = 0.0252; Fig. 1C). The metastatic Figure 1. ErbB2 enhances human breast cancer cell invasion and metastasis. survival rates of mice injected with K753M cells were similar to A, ErbB2 protein expression and tyrosine phosphorylation in MDA-MB-435 cells, those of mice injected with MDA-MB-435 and 435.neo cells (Fig. 435.neo cells, and ErbB2 transfectants were detected by using an ErbB2 1C). In summary, 435.eB and V659E cells with high ErbB2 kinase antibody and a phosphospecific ErbB2 antibody (top). Levels of tyrosine phosphorylation on ErbB2 and other proteins were detected by whole cell lysate activities were highly metastatic, whereas MDA-MB-435, 435.neo, Western blotting with PY20 antibody (bottom). B, ErbB2 increased the and K753M cells with low kinase activities were not. These results invasiveness of ErbB2-activated transfectants. Transwells with an 8 Am pore polycarbonate filter were coated with Matrigel. The lower compartment of the show that ErbB2 kinase activity plays a critical role in ErbB2- transwells contained 0.6 mL of 40 Ag/mL laminin in DMEM/F-12 as a mediated metastasis in these breast cancer cells. chemoattractant. Cells (1 Â 105 in 0.1 mL DMEM/F-12 containing 0.1% bovine Up-regulation of Src Protein by ErbB2. We next investigated serum albumin) were placed in the upper compartment and incubated for 24 hours at 37jC. After incubation, the filters were fixed with 3% glutaraldehyde multiple ErbB2 downstream signaling pathways that may in PBS and stained with Giemsa. Cells on the upper surface of the filter were contribute to ErbB2-mediated metastasis, and we focused on removed with a cotton swab, and the invasive ability of the cells was determined Src because it was tightly regulated by altered ErbB2 RTK by counting the number of cells per high-power field (200Â) that had migrated to the lower side of the filter. Matrigel-invaded cells were counted by at least three activities and closely correlated with the metastatic potential of high-power fields per filter. Each sample was assayed in triplicate, and the our panel of breast cancer cells (see below). When we compared assays were repeated at least twice. *, P < 0.05, statistically significant when compared with 435.neo cells (Student’s t test). C, ErbB2 increased the Src activity in MDA-MB-435 cells, 435.neo cells, and wt and metastatic potential of activated ErbB2 transfectants in vivo. We injected 1 Â 106 mutant ErbB2 transfectants by immunocomplex kinase assay cells of each cell line into the lateral tail veins of 8-week-old female ICR-SCID (Fig. 2A), we found that the Src kinase was activated at mice. Survival curves were drawn and analyzed using Prism software. Metastatic survival curves were compared using the Mantel-Haenszel log-rank dramatically higher levels in 435.eB and V659E cells than in the test. *, P < 0.05, statistically significant when compared with 435.neo cells. control MDA-MB-435, 435.neo, and K753M cells. Thus, activation

Cancer Res 2005; 65: (5). March 1, 2005 1860 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. ErB2 Promotes Src Protein Synthesis and Stability of ErbB2 RTK by overexpression or activating mutation can lead to Src activation in these breast cancer cells. To explore the mechanism by which ErbB2 activates Src, we examined the Src protein levels in these cells by Western blot (Fig. 2A) and quantified and correlated the Src protein level and kinase activity in each cell line (Fig. 2A and B). Surprisingly, Src protein levels dramatically increased in ErbB2-overexpressing 435.eB cells and ErbB2-activated V659E cells, which correlated very well with their higher Src kinase activities. This indicates that the higher Src kinase activities in 435.eB and V659E cells were most likely due to the increased Src protein level rather than the increased kinase activity per Src molecule. We next did Northern blotting to determine whether the increased Src protein level resulted from an increased Src RNA level (Fig. 2C). Src RNA levels were similar among all of the cell lines, indicating that the up-regulation of Src expression by ErbB2 did not occur at the RNA level. To further confirm that ErbB2 regulates Src protein expression levels, the ErbB2 high-expressing breast cancer cell line BT474 was transfected with siRNA to ErbB2 (Fig. 2D, left). Immunoblotting results showed that ErbB2 siRNA down-regulated ErbB2 expression in BT474 cells and subsequently reduced Src protein levels. In addition, compared with 435.eB, the Src protein level was reduced in the 435eB.R spontaneous revertant, a 435.eB derivative that partially lost the transfected erbB2 gene and expresses lower levels of ErbB2 (Fig. 2D, middle). Moreover, elevated Src protein expression could also be observed in the wt ErbB2-transfected breast cancer cell line MDA-MB-231 (231.eB) compared with their parental cells (MDA-MB-231; Fig. 2D, right). Furthermore, by immunohistochemistry analysis, we studied Src protein expression levels in breast cancer patient samples, which were strongly correlated with ErbB2 expression levels (data not shown) as reported previously (36). These data show that Src protein expression is regulated by ErbB2 in breast cancer cells. Taken together, we conclude that ErbB2 activation can up-regulate Src at the post-transcriptional level and thereby increase Src protein and kinase activity in ErbB2-activated human breast cancer cells. ErbB2 Increases Src Protein Synthesis by Activating the Akt/Mammalian Target of Rapamycin/4E-BP1 Protein Figure 2. Up-regulation of Src protein by ErbB2. A, ErbB2 overexpression Translation Pathway. To investigate the mechanism of Src or activation led to increased Src protein levels and Src activity in MDA-MB-435 protein up-regulation by ErbB2, we first tested whether the rate human breast cancer cells. Cell lysates were immunoprecipitated with Src of Src protein synthesis was greater in ErbB2-overexpressing cells antibodies. Immunocomplexes were subjected to Src kinase assays using enolase as the substrate (top). Src (middle) and h-actin (bottom) protein levels than in ErbB2 low-expressing cells. MDA-MB-435 and 435.eB cells were determined by Western blot. B, quantification of the Src protein level and were cultured with medium containing [35S]methionine to kinase activity in MDA-MB-435 cells and ErbB2 transfectants. Results from A were quantified by Alpha Imager 2000. Relative Src protein levels and kinase metabolically label the newly synthesized proteins for 1 or 4 hours. activities were calculated using the protein level and kinase activity of 435.neo as Src proteins from the cell lysates were immunoprecipitated with a 1.0. C, similar RNA levels among MDA-MB-435 cells and ErbB2 transfectants. Src-specific antibody, analyzed by SDS-PAGE, and visualized by Src RNA level was detected by Northern blot using a Src-specific sequence as the probe. D, ErbB2 regulates Src protein levels in human breast cancer cell autoradiography. The results showed that ErbB2-overexpressing lines. Cell lysates from BT474 cells treated with ErbB2 siRNA or control siRNA 435.eB cells had higher levels of [35S]methionine-labeled Src protein (left), 435.eB and its revertant 435eB.R (middle), and MDA-MB-231 and its wt than MDA-MB-435 cells at both 1- and 4-hour 35S labeling (Fig. 3A, ErbB2-transfectant 231.eB cells (right) were immunoblotted with anti-ErbB2 (top), anti-Src (middle), and anti-h-actin antibodies (bottom). left and middle), although the overall rate of incorporation of [35S]methionine was similar between the two cell lines (data not shown). Meanwhile, immunoprecipitation on the same [35S]methi- An important regulator of protein synthesis and translation is onine-labeled cell extracts with the h-actin antibody and the mammalian target of rapamycin (mTOR), which is a subsequent analysis showed that the h-actin synthesis rates in phosphatidylinositol 3-kinase family protein that functions MDA-MB-435 and 435.eB cells were similar (Fig. 3A, right). These downstream of Akt (37, 38). We therefore examined whether the results, in conjunction with the observation that the Src protein Akt/mTOR pathway was involved in ErbB2-mediated increase of degradation during the 4-hour [35S]methionine labeling was Src protein synthesis. We detected higher levels of Akt minimal in these cells (see Fig. 4A), suggest that ErbB2 phosphorylation in ErbB2-activated 435.eB and V659E cells but preferentially increased the Src protein synthesis rate in ErbB2- not in MDA-MB-435, 435.neo, or K753M cells (Fig. 3B), whereas overexpressing 435.eB cells. the total amount of Akt remained similar. Because activated Akt www.aacrjournals.org 1861 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Research is known to activate mTOR, and mTOR in turn can phosphorylate To test whether activation of the mTOR pathway was its downstream substrate 4E-BP1, a translation repressor, on the responsible for increasing 4E-BP1-Ser65 phosphorylation and inhibitory phosphorylation site at Ser65 (37), we tested whether ErbB2-mediated Src protein synthesis, we treated MDA-MB-435 ErbB2, which activates Akt, may further increase 4E-BP1 and 435.eB cells with rapamycin, which is a mTOR-specific phosphorylation and lead to its inhibition. We examined the inhibitor (39). Inhibition of mTOR by rapamycin effectively phosphorylation of 4E-BP1 by Western blot with a 4E-BP1-Ser65 reduced 4E-BP1-Ser65 phosphorylation (Fig. 3C). Notably, MDA- phosphorylation-specific antibody. Compared with MDA-MB-435 MB-435 cells had a minimal reduction in Src protein levels after cells, 435.eB cells had increased 4E-BP1 phosphorylation (Fig. 3C), rapamycin treatment, whereas ErbB2-overexpressing 435.eB cells indicating that there were higher levels of mTOR kinase activity had a evident reduction in Src protein levels (Fig. 3D), indicating in 435.eB cells than in MDA-MB-435 cells. The increased that activation of the Akt/mTOR/4E-BP1 pathway by ErbB2 in phosphorylation of 4E-BP1 on Ser65 could also be observed in 435.eB cells is involved in the enhanced Src protein synthesis. To wt ErbB2-transfected MDA-MB-231 human breast cancer cells but further test this hypothesis, we treated the cells with or without not in their parental cells (data not shown). rapamycin and labeled the cells with [35S]methionine for 4 hours. Immunoprecipitation of Src protein was done, and the immunoprecipitates were analyzed by SDS-PAGE. The results showed that without rapamycin more [35S]methionine-labeled Src protein was detected in 435.eB cells than that of in MDA-MB-435 cells (Fig. 3A, left and middle). However, in rapamycin-treated cells, the [35S]methionine-labeled Src protein in 435.eB cells was reduced to a level similar to that in MDA-MB-435 cells (Fig. 3A, left), indicating that ErbB2-dependent mTOR activation is required for ErbB2-promoted Src protein translation. ErbB2 Stabilizes Src by Inhibiting Calpain-Mediated Src Degradation. To test whether the elevated Src protein level detected in ErbB2-activated cells may also result from reduced Src protein degradation, pulse-chase analyses were done to test whether ErbB2 increases the stability of Src protein. MDA-MB-435 and 435.eB cells were incubated with [35S]methionine for 1 hour. Free [35S]methionine was washed out, and Src protein was chased at different time points by immunoprecipitation with the Src- specific antibody and visualized by autoradiography. The results showed that the Src degradation was slower in 435.eB cells than in MDA-MB-435 cells and that the Src protein half-life was longer in 435.eB cells (52 hours) than in MDA-MB-435 cells (20 hours; Fig. 4A). Thus, activation of ErbB2 leads to Src protein stabilization, which can also contribute to the increased Src protein levels in the ErbB2-activated cells. It has been shown that Src could be degraded by the proteasome-ubiquitin-dependent pathway (40). To investigate the mechanism of Src stabilization by ErbB2, we investigated whether ErbB2 activation inhibits ubiquitination-mediated Src protein degradation. A HA-tagged ubiquitin-expressing construct was transfected into MDA-MB-435 and V659E cells. The expression of the HA-tagged ubiquitin was confirmed by Western blot with antibody against HA (Fig. 4B). To detect the ubiquitination of Src, Figure 3. ErbB2 increases Src protein synthesis by activating a cells were treated or untreated with proteasome inhibitor MG-132, mTOR-mediated translational mechanism. A, Src protein synthesis increased in ErbB2-overexpressing 435.eB cells. MDA-MB-435 and 435.eB cells were and immunoprecipitation with the Src-specific antibody and treated without or with 25 nmol/L rapamycin and labeled with [35S]methionine for subsequent Western blot with the anti-HA antibody were done 1 or 4 hours. Immunoprecipitations with Src-specific antibodies (left and middle) and h-actin specific antibodies (right). Immunoprecipitates were washed, (Fig. 4B, left). Src ubiquitination could not be detected under these analyzed by SDS-PAGE, and visualized by autoradiography. IgG was used as a experimental conditions. Contrarily, control experiment showed control for immunoprecipitation. Rel. intensity, relative protein band intensity. that ubiquitinated Met, a control protein, was readily detected in Experiments were repeated thrice. SD V 0.21. B, Akt activation in MDA-MB-435 cells and ErbB2 transfectants. Western blot was done with phosphorylated the HA-ubiquitin-expressing osteosarcoma cell line Saos-2 with Akt-specific antibodies (top). Total Akt proteins were immunoblotted with MG-132 treatment (Fig. 4B, right), indicating that the assay is antibodies specifically against Akt (bottom). C, rapamycin inhibited a higher sufficiently sensitive for detecting the ubiquitination of proteins. 4E-BP1-Ser65 phosphorylation level in ErbB2-overexpressing 435.eB cells. MDA-MB-435 and 435.eB cells were treated with or without 25 nmol/L Next, we examined whether blocking the proteasome pathway with rapamycin for 4 hours, protein lysates were extracted, and Western blot was MG-132 will stabilize Src in MDA-MB-435 and V659E cells. Cell done with 4E-BP1-Ser65 phosphorylation-specific antibodies (top) and 4E-BP1 specific antibodies (bottom). D, Src protein levels were more obviously reduced lysates were collected from MG-132-treated or untreated cells and by rapamycin treatment in ErbB2-overexpressing 435.eB cells than in the immunoblotted with Src and cyclin B1 antibodies. The results parental MDA-MB-435 cells. Cells were treated with or without rapamycin for the showed that MG-132 treatment led to an obvious accumulation of indicated concentrations for 24 hours, protein lysates were extracted, and Western blot was done with Src-specific antibodies (top) and antibodies against the control protein cyclin B1 that was well known to be degraded h-actin (bottom). Experiments were repeated thrice. SD V 0.19. through proteasome pathway (41). In contrast, MG-132 treatment

Cancer Res 2005; 65: (5). March 1, 2005 1862 www.aacrjournals.org

Figure 4. ErbB2 stabilizes Src by inhibiting calpain-mediated Src degradation. A, pulse-chase analysis of Src proteins in MDA-MB-435 and 435.eB cells. Cells were washed and pulse labeled with 500 ACi/mL [35S]methionine for 1 hour. Cells were washed thrice and incubated in DMEM/F-12 containing 10% fetal bovine serum and harvested at the indicated time points. Cell lysates were immunoprecipitated by Src antibodies and separated by SDS-PAGE. [35S]Methionine-labeled Src protein was visualized by autoradiography. IgG was used as a control for the immunoprecipitation. Results from the pulse-chase analysis were quantified by Alpha Imager 2000. Average percentage of the remaining [35S]methionine-labeled Src protein from the three repeated assays was plotted. B, Src ubiquitination could not be detected in MDA-MB-435 and 435.eB cells. A HA-tagged ubiquitin-expressing construct and its control vector were transfected into the MDA-MB-435 and 435.eB cells. Cells were treated or untreated with 10 Amol/L proteasome inhibitor MG-132 for 3 hours, and immunoprecipitation with a Src-specific antibody and Western blot with an anti-HA antibody were done (left). As a control, HA-tagged ubiquitin-expressing constructs were transfected into the osteosarcoma cell line Saos-2. Cells were treated with or without 10 Amol/L MG-132. Met was immunoprecipitated and the ubiquitination of Met was detected by Western blot (right). Expression of the HA-tagged ubiquitin was detected by immunoblotting with the anti-HA antibody. C, MG-132 treatment did not lead to Src accumulation. Cell lysates were collected from 10 Amol/L MG-132 24-hour-treated or untreated cells and immunoblotted with Src, cyclin B1, or h-actin antibodies. D, top, in vitro degradation of Src by purified calpain. Protein lysates from 435.eB cells were incubated for 2 hours at 30jC with 0, 5, or 50 AU purified calpain in varying conditions as indicated, and Src protein degradation was determined by Western blot analysis (top). Control 14-3-3 protein levels were detected by Western blot (bottom). Bottom, extended incubation of calpain with cell lysates led to faster Src protein degradation. Protein lysates from 435.eB cells were treated with 50 AU calpain in the presence of 5 mmol/L CaCl2 for different times. Immunoblotting with Src antibody was done and results were quantified by Alpha Imager 2000. Relative Src protein levels were calculated using the protein level at 0-hour time point as 1.00. Experiments were repeated thrice. SD V 0.17. E, calpain inhibitors repressed Src protein degradation more obviously in ErbB2 low-expressing MDA-MB-435 cells than in ErbB2 high-expressing 435.eB cells. Cells were treated with 100 Ag/mL cycloheximide (CHX) alone, cycloheximide plus 10 Amol/L calpain inhibitor N-acetyl-leucyl-leucyl-norleucinal (ALLN), or cycloheximide plus 40 Amol/L calpain inhibitor N-acetyl-leucyl-leucyl-methional (ALLM) for 0, 12, or 24 hours; protein lysates were extracted; and Western blot was done with Src-specific antibodies and antibodies against h-actin. Relative Src protein levels were calculated using the protein level at 0-hour time point as 1.00. Experiments were repeated thrice. SD V 0.19. did not lead to Src protein accumulation in either of the cell lines including apoptosis, proliferation, and migration (42). Previous (Fig. 4C). This, in conjunction with the failure of detecting Src reports suggested that calpain could degrade Src and lead to protein ubiquitination, suggests that the ubiquitin-proteasome decreased Src kinase activity in human platelets (43, 44). Therefore, pathway may not be a significant pathway for Src degradation in we hypothesized that in ErbB2-overexpressing breast cancer cells these cells. ErbB2 may inhibit calpain-mediated Src protein degradation, Calpains belong to a family of Ca2+-dependent proteases and which could be involved in stabilizing Src. We first examined have been implicated in a wide range of cellular functions, whether Src was a substrate of calpain in our cell system. Protein www.aacrjournals.org 1863 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Research lysates from 435.eB cells were incubated with or without 5 or 50 AU veins of female ICR-SCID mice. The mice were sacrificed 90 days purified calpain under various conditions, and the Src protein level after the injection. Metastatic lung nodules were counted and was determined by Western blot analysis (Fig. 4D, top). In the statistically analyzed by Student’s t test (Table 1). Consistent with presence of 5 mmol/L EGTA, a chelator of Ca2+, there was no the results from the in vitro invasion assay (Fig. 5A and B), PP2 obvious calpain activity toward Src. In the absence of added Ca2+, dramatically inhibited the metastatic potential of ErbB2-activated calpain activity toward Src was minimal. However, in the presence V659E cells (P < 0.05) but did not significantly inhibit metastasis of of 5 mmol/L CaCl2, calpain caused a dramatic Src protein cleavage 435.neo control cells (P > 0.05). To further investigate the role of Src into smaller fragments in a concentration-dependent manner. In in ErbB2-mediated cancer cell metastasis, another experimental contrast, immunoblotting with antibody against 14-3-3, a control metastasis assay was done to compare the metastatic potentials of protein, did not detect 14-3-3 degradation by calpain (Fig. 4D, top), 435.eB cells and 435.eB cells expressing a DNSrc (eB.DNSrc). The indicating that the proteolytic cleavage is Src specific. To examine result showed that compared with the mice injected with 435.eB the time course of calpain-mediated Src degradation, protein cells mice injected with eB.DNSrc cells formed fewer pulmonary lysates from 435.eB cells were treated with 50 AU calpain in the nodules (P < 0.05; Fig. 5C), and their lungs weighed less (P < 0.01; presence of 5 mmol/L CaCl2 for different times (Fig. 4D, bottom). Fig. 5D). These in vitro and in vivo results firmly show that Src Immunoblotting with Src antibody showed that the total Src plays a critical role in ErbB2-mediated human breast cancer cell protein detected by the antibody was reduced with prolonged invasion and metastasis. incubation, suggesting that calpain can trigger complete degradation of Src. These data indicate that Src could be a substrate of calpain in these cells. Discussion To test whether an inhibition of calpain-mediated Src degrada- In this study, we found that ErbB2 activation could up-regulate tion was involved in ErbB2-mediated Src stabilization, we Src protein levels and thereby increase Src kinase activity in ErbB2- incubated MDA-MB-435 and 435.eB cells in the presence or activated human breast cancer cells. The increased Src protein absence of cycloheximide to block protein synthesis or cyclohex- levels in ErbB2-activated cells resulted from at least two imide plus calpain-specific inhibitor N-acetyl-leucyl-leucyl-norleu- mechanisms: (a) ErbB2 promoted Src protein synthesis by cinal. At different time points, protein lysates were collected and activating the Akt/mTOR/4E-BP1 protein translation pathway the Src protein level was determined by Western blot analysis and (b) ErbB2 stabilized the Src protein most likely by inhibiting (Fig. 4E). In the presence of the calpain inhibitor N-acetyl-leucyl- calpain-mediated Src protein degradation. Moreover, we showed leucyl-norleucinal, degradation of Src protein was delayed in both that Src played a critical role in ErbB2-mediated breast cancer MDA-MB-435 and 435.eB cells compared with cells treated with invasion and metastasis (Fig. 5E). cycloheximide alone (Fig. 4E, left). Notably, the delay of Src protein Understanding the molecular mechanisms of ErbB2-mediated degradation was more evident in MDA-MB-435 cells than in 435.eB metastasis is important for the development of new therapies to cells, suggesting that Src degradation by the calpain pathway was treat metastases derived from ErbB2-overexpressing breast cancer. less active in 435.eB cells than in MDA-MB-435 cells most likely as a ErbB2 overexpression can be detected in 50% to 60% of high-grade result of ErbB2 activation. These findings were supported by using ductal carcinoma in situ breast cancers (46). Interestingly, another calpain-specific inhibitor N-acetyl-leucyl-leucyl-methional, activation of ErbB2 receptor in preformed mammary acini grown which has been shown to have little inhibition of proteasome- in three-dimensional cell cultures has led to altered mammary mediated protein degradation (ref. 41; Fig. 4E, right). Thus, in acini structures that share properties with ductal carcinoma in situ. addition to up-regulating the Src protein level by increasing Src No invasive phenotypes were found in these altered acini structures protein synthesis via activation of the Akt/mTOR/4E-BP1 pathway, (47). The findings suggest that activation of ErbB2 can induce early ErbB2 may also stabilize Src protein via inhibition of calpain- stages of mammary carcinogenesis but is not sufficient to induce mediated Src degradation. invasive phenotypes (47). It has been suggested that additional Src Plays a Crucial Role in ErbB2-Mediated Breast Cancer genetic or epigenetic events are needed for mammary epithelial Metastasis. To assess the role of Src activation in ErbB2-induced cells to acquire invasive properties (47). The MDA-MB-435 is a breast cancer metastasis, we tested whether PP2, a Src-specific highly aggressive metastatic breast cancer cell line and it harbors inhibitor (45), can inhibit the invasiveness of the ErbB2 trans- multiple genetic or epigenetic alterations that may facilitate the fectants that contain activated Src. The cells were treated with PP2 development of metastasis (48). or its control chemical PP3 and transferred into transwell units for Little was known on the role of Src in ErbB2-induced breast chemoinvasion assays. PP2 dramatically inhibited the basement cancer metastasis and on the mechanisms of Src regulation by membrane invasion abilities of ErbB2-overexpressing, Src-activated ErbB2. In this study, we showed that Src plays a critical role in 435.eB and V659E cells and to a much lesser extent the invasion ErbB2-mediated breast cancer invasion and metastasis. To our abilities of 435.neo control cells (Fig. 5A). Moreover, expression of a knowledge, this study reports the first experimental evidence that DNSrc in 435.neo and 435.eB cells reduced their invasive abilities in the higher metastatic potential of ErbB2-overexpressing cancer the chemoinvasion assay (Fig. 5B). Similar to the treatment with cells is mediated at least in part through the activation of Src. PP2, the inhibition of invasiveness by the DNSrc was more Therefore, this study provides a rationale for targeting Src as a dramatic in 435.eB cells compared with 435.neo cells. These results therapeutic strategy for the treatment or prevention of ErbB2- show that Src kinase plays a critical role in ErbB2-induced cancer mediated breast cancer metastasis. It may also have important cell invasion. We then did experimental metastasis assays to implications in treating other RTK-mediated malignant cancer determine the role of Src in ErbB2-mediated metastasis in vivo. The behaviors. 435.neo cells, which had lower ErbB2 expression and Src kinase Tumor samples from a substantial portion of colon cancers activity, and the V659E cells, which overexpressed activated ErbB2 (49, 50) and other cancer types (50), including f48% of breast and had a higher Src kinase activity, were injected into the tail cancers (51), were shown to have elevated Src protein levels in

Cancer Res 2005; 65: (5). March 1, 2005 1864 www.aacrjournals.org

Figure 5. Src is critical in ErbB2-mediated breast cancer metastasis. A, Src inhibitor, PP2, inhibited ErbB2-mediated invasiveness of ErbB2-activated transfectants. Cells were treated with 10 Amol/L PP2 or its control chemical PP3 and transferred to transwell units for chemoinvasion assays as described in Fig. 1B. *, P < 0.05, statistically significant compared with PP3-treated 435.eB or V659E cells. B, transfection of a DNSrc decreased the invasiveness of ErbB2-activated 435.eB cells. Chemoinvasion assay was done as in Fig. 1B.*,P < 0.05, statistically significant compared with vector alone–transfected 435.eB cells. C and D, DNSrc inhibited the metastatic potentials of ErbB2-overexpressing cells in vivo. Experimental metastasis assay by tail vein injection of the 435.eB or eB.DNSrc was done and analyzed as described in Materials and Methods. Numbers of the metastatic lung nodules found in each mouse (C; P < 0.05) and weights of the lungs of the mice (D; P < 0.01). E, schematic model of Src activation in ErbB2-mediated human breast cancer metastasis. In addition to the known pathway that direct association between Src and ErbB2 leads to Src activation, in this study, we found that the increased Src kinase activity in ErbB2-activated cells was due to increased Src protein levels that resulted from at least two mechanisms: (a) ErbB2 promoted Src protein synthesis by activating the mTOR-mediated protein translation pathway and (b) ErbB2 stabilized Src protein most likely by inhibiting calpain-mediated Src protein degradation. Moreover, Src activation played a critical role in ErbB2-mediated breast cancer invasion and metastasis.

cancer cells compared with normal cells. However, few studies were nism (38, 52–55). Our study provides an excellent example of how devoted to investigating the role and mechanisms of deregulated signaling molecules can be regulated at the protein level by Src protein synthesis and stability in mediating malignant modulating their protein translational mechanism. To our phenotypes. Our study revealed that the up-regulation of Src knowledge, this is the first report that ErbB2 activates the protein is an important mechanism for Src activation in ErbB2- overexpressing breast cancer cells. This mechanism may also be exploited by other growth factor receptors or other types of cancer Table 1. PP2 inhibited metastatic potential of ErbB2 transfectants cells in addition to ErbB2 and breast cancer cells. Based on our findings, the overexpressed Src protein in the cancer patient samples may conceivably be due at least in part to growth factor Cell lines Pulmonary metastasis receptor–induced protein up-regulation. The new mechanisms revealed in this study may also have important clinical implications Mice with Nodules per P for designing Src-targeting therapies. For example, drug designers metastasis/mice mouse* may enhance Src blocking efficiency by inhibiting Src protein injected (%) [mean (range)] synthesis and/or inducing Src protein degradation in addition to inhibiting Src kinase activity. 435.neo + solvent 4/8 (50) 5 (0-20) — In cancer research, studies on how signaling pathways regulate 435.neo + PP2 4/8 (50) 4 (0-20) >0.05 V659E + solvent 8/9 (89) 17 (0-41) — the transcriptional control of gene expression have been pursued V659E + PP2 4/7 (57) 5 (0-13) <0.05 extensively. In contrast, fewer studies have focused on how these pathways control other fundamental processes (52, 53). Recently, *Lung metastasis nodules >0.5 mm in diameter were counted in the modulating protein translational control by signaling pathway assay. activation has emerged as another important regulation mecha- www.aacrjournals.org 1865 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. Cancer Research mTOR/4E-BP1 translation pathway and thereby increases Src mechanism of this modulation is, and what other biological protein synthesis. Notably, ErbB2 preferentially promotes the functions this modulation may exert. Given the fact that both translation of Src, which is a typical growth-regulated protein, ErbB2 and Src play important roles in cancer development and but not the translation of h-actin, which is a typical constitutive progression and the ubiquitous expression and important func- protein in most cells (Fig. 3A). This is consistent with previous tions of calpain protease in the human body, future investigations reports that the activation of mTOR preferentially up-regulates the of these questions will lead to a better understanding of how translation of mRNAs of growth-regulated proteins (56). cancer metastasis signaling is regulated as well as to better Previous reports suggested that Src might be a substrate of the strategies for metastasis intervention. Ca2+-dependent protease calpain in human platelets (43, 44). However, calpain inhibitors or activators were used to modulate Src protein in these previous studies, and it was not clear whether Acknowledgments Src was a direct substrate of the calpain protease. By using purified calpain, we provided another evidence to support that Src might be Received 7/6/2004; revised 11/9/2004; accepted 12/23/2004. Grant support: NIH grants P30-CA16672 (M.D. Anderson Cancer Center), 2RO1- a substrate of calpain (Fig. 4D). Furthermore, in our study, we CA60488, RO1-CA109570, and 1PO1-CA099031-1 (D. Yu) and M.D. Anderson Cancer discovered a biological function for calpain-mediated Src degra- Center institutional research grant (M. Tan). dation as a way of regulating Src signaling by RTK, which might 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 play an important role in ErbB2 (or other RTKs)–enhanced cancer with 18 U.S.C. Section 1734 solely to indicate this fact. metastasis. These were not addressed by previous studies. However, We thank Dr. Tadashi Yamamoto for providing the ErbB2 mutant expression vectors; Dr. David Shalloway for providing the Src mutant expression vectors; Drs. the following questions remain to be answered: whether ErbB2 Mien-Chie Hung, Jun Yao, and Chris Neal for valuable discussions and comments; and directly or indirectly modulates calpain protease activity, what the Lore Feldman and Katie Matias for editing the article.

References 15. Muthuswamy SK, Siegel PM, Dankort DL, Webster MA, p21Cip1, which inhibits p34Cdc2 kinase. Mol Cell Muller WJ. Mammary tumors expressing the neu proto- 1998;2:581–91. 1. Yarden Y, Sliwkowski MX. Untangling the ErbB oncogene possess elevated c-Src tyrosine kinase activity. 30. Nagata Y, Anan T, Yoshida T, et al. The stabilization signalling network. Nat Rev Mol Cell Biol 2001;2:127–37. Mol Cell Biol 1994;14:735–43. mechanism of mutant-type p53 by impaired ubiquiti- 2. Holbro T, Hynes NE. ErbB receptors: directing key 16. Ochiai A, Akimoto S, Kanai Y, Shibata T, Oyama T, nation: the loss of wild-type p53 function and the hsp90 signaling networks throughout life. Annu Rev Pharma- Hirohashi S. C-erbB-2 gene product associates with association. Oncogene 1999;18:6037–49. col Toxicol 2004;44:195–217. catenins in human cancer cells. Biochem Biophys Res 31. Tan M, Grijalva R, Yu D. Heregulin h1-activated 3. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the Commun 1994;205:73–8. phosphatidylinositol 3-kinase enhances aggregation of HER-2/neu proto-oncogene in human breast and 17. Zrihan-Licht S, Lim J, Keydar I, Sliwkowski MX, MCF-7 breast cancer cells independent of extracellular ovarian cancer. Science 1989;244:707–12. Groopman JE, Avraham H. Association of csk-homolo- signal-regulated kinase. Cancer Res 1999;59:1620–5. 4. Yu D, Hung MC. Role of erbB2 in breast cancer gous kinase (CHK) (formerly MATK) with HER-2/ErbB-2 32. Akiyama T, Matsuda S, Namba Y, Saito T, chemosensitivity. BioEssays 2000;22:673–80. in breast cancer cells. J Biol Chem 1997;272:1856–63. Toyoshima K, Yamamoto T. The transforming poten- 5. Moody SE, Sarkisian CJ, Hahn KT, et al. Conditional 18. Borg JP, Marchetto S, Le Bivic A, et al. ERBIN: a tial of the c-erbB-2 protein is regulated by its activation of Neu in the mammary epithelium of basolateral PDZ protein that interacts with the autophosphorylation at the carboxyl-terminal domain. transgenic mice results in reversible pulmonary metas- mammalian ERBB2/HER2 receptor. Nat Cell Biol Mol Cell Biol 1991;11:833–42. tasis. Cancer Cell 2002;2:451–61. 2000;2:407–14. 33. Bargmann CI, Weinberg RA. Increased tyrosine 6. Holbro T, Civenni G, Hynes NE. The ErbB receptors 19. Fiorentino L, Pertica C, Fiorini M, et al. Inhibition of kinase activity associated with the protein encoded by and their role in cancer progression. Exp Cell Res ErbB-2 mitogenic and transforming activity by RALT, a the activated neu oncogene. Proc Natl Acad Sci U S A 2003;284:99–110. mitogen-induced signal transducer which binds to the 1988;85:5394–8. 7. Tan M, Yao J, Yu D. Overexpression of the c-erbB-2 ErbB-2 kinase domain. Mol Cell Biol 2000;20:7735–50. 34. Segatto O, Lonardo F, Pierce JH, Bottaro DP, Di Fiore gene enhanced intrinsic metastatic potential in human 20. Levkowitz G, Oved S, Klapper LN, et al. c-Cbl is a PP. The role of autophosphorylation in modulation of breast cancer cells without increasing their transfor- suppressor of the neu oncogene. J Biol Chem erbB-2 transforming function. New Biol 1990;2:187–95. mation abilities. Cancer Res 1997;57:1199–205. 2000;275:35532–9. 35. Weiner DB, Kokai Y, Wada T, Cohen JA, Williams WV, 8. Van der Geer P, Hunter T, Lindberg RA. Receptor 21. Tan M, Jing T, Lan KH, et al. Phosphorylation on Greene MI. Linkage of tyrosine kinase activity with protein-tyrosine kinases and their signal transduction tyrosine-15 of p34(Cdc2) by ErbB2 inhibits p34(Cdc2) transforming ability of the p185neu oncoprotein. Onco- pathways. Annu Rev Cell Biol 1994;10:251–337. activation and is involved in resistance to taxol-induced gene 1989;4:1175–83. 9. Segatto O, Pelicci G, Giuli S, et al. Shc products are apoptosis. Mol Cell 2002;9:993–1004. 36. Bieche I, Onody P, Tozlu S, Driouch K, Vidaud M, substrates of erbB-2 kinase. Oncogene 1993;8:2105–12. 22. Martin GS. The hunting of the Src. Nat Rev Mol Cell Lidereau R. Prognostic value of ERBB family mRNA 10. Stein D, Wu J, Fuqua SA, et al. The SH2 domain Biol 2001;2:467–75. expression in breast carcinomas. Int J Cancer protein GRB-7 is co-amplified, overexpressed and in a 23. Yeatman TJ. A renaissance for SRC. Nat Rev Cancer 2003;106:758–65. tight complex with HER2 in breast cancer. EMBO J 2004;4:470–80. 37. Gingras AC, Raught B, Sonenberg N. Regulation of 1994;13:1331–40. 24. Stehelin D, Varmus HE, Bishop JM, Vogt PK. DNA translation initiation by FRAP/mTOR. Genes Dev 11. Fazioli F, Kim UH, Rhee SG, Molloy CJ, Segatto O, related to the transforming gene(s) of avian sarcoma 2001;15:807–26. Di Fiore PP. The erbB-2 mitogenic signaling pathway: viruses is present in normal avian DNA. Nature 38. Holland EC, Sonenberg N, Pandolfi PP, Thomas G. tyrosine phosphorylation of phospholipase C-g and 1976;260:170–3. Signaling control of mRNA translation in cancer GTPase-activating protein does not correlate with 25. Biscardi JS, Tice DA, Parsons SJ. c-Src, receptor pathogenesis. Oncogene 2004;23:3138–44. erbB-2 mitogenic potency. Mol Cell Biol 1991;11:2040–8. tyrosine kinases, and human cancer. Adv Cancer Res 39. Sekulic A, Hudson CC, Homme JL, et al. A direct 12. Peles E, Levy RB, Or E, Ullrich A, Yarden Y. Oncogenic 1999;76:61–119. linkage between the phosphoinositide 3-kinase-AKT forms of the neu/HER-2 tyrosine kinase are permanently 26. Summy JM, Gallick GE. Src family kinases in tumor signaling pathway and the mammalian target of coupled to phospholipase Cg. EMBO J 1991;10:2077–86. progression and metastasis. Cancer Metastasis Rev rapamycin in mitogen-stimulated and transformed 13. Ram TG, Ethier SP. Phosphatidylinositol 3-kinase 2003;22:337–58. cells. Cancer Res 2000;60:3504–13. recruitment by p185erbB-2 and erbB-3 is potently 27. Belsches-Jablonski AP, Biscardi JS, Peavy DR, Tice DA, 40. Hakak Y, Martin GS. Ubiquitin-dependent degrada- induced by neu differentiation factor/heregulin during Romney DA, Parsons SJ. Src family kinases and HER2 tion of active Src. Curr Biol 1999;9:1039–42. mitogenesis and is constitutively elevated in growth interactions in human breast cancer cell growth and 41. Choi YH, Lee SJ, Nguyen P, et al. Regulation of cyclin factor-independent breast carcinoma cells with c-erbB-2 survival. Oncogene 2001;20:1465–75. D1 by calpain protease. J Biol Chem 1997;272:28479–84. gene amplification. Cell Growth Differ 1996;7:551–61. 28. Yu D, Liu B, Tan M, Li J, Wang SS, Hung MC. 42. Glading A, Lauffenburger DA, Wells A. Cutting to the 14. Muthuswamy SK, Muller WJ. Activation of Src family Overexpression of c-erbB-2/neu in breast cancer cells chase: calpain proteases in cell motility. Trends Cell kinases in Neu-induced mammary tumors correlates confers increased resistance to Taxol via mdr-1- Biol 2002;12:46–54. with their association with distinct sets of tyrosine independent mechanisms. Oncogene 1996;13:1359–65. 43. Oda A, Druker BJ, Ariyoshi H, Smith M, Salzman EW. phosphorylated proteins in vivo. Oncogene 1995;11: 29. Yu D, Jing T, Liu B, et al. Overexpression of ErbB2 pp60src is an endogenous substrate for calpain in 1801–10. blocks Taxol-induced apoptosis by upregulation of human blood platelets. J Biol Chem 1993;268:12603–8.

Cancer Res 2005; 65: (5). March 1, 2005 1866 www.aacrjournals.org

44. Schoenwaelder SM, Yuan Y, Cooray P, Salem HH, 48. Sellappan S, Grijalva R, Zhou X, et al. Lineage 53. Lasko P. Ribosomes rule: translation, not transcrip- Jackson SP. Calpain cleavage of focal adhesion proteins infidelity of MDA-MB-435 cells: expression of melano- tion, is the primary target of two major intercellular regulates the cytoskeletal attachment of integrin aIIbh3 cyte proteins in a breast cancer cell line. Cancer Res signaling pathways. Dev Cell 2003;5:671–2. (platelet glycoprotein IIb/IIIa) and the cellular retrac- 2004;64:3479–85. 54. Wilson KF, Cerione RA. Signal transduction and post- tion of fibrin clots. J Biol Chem 1997;272:1694–702. 49. Iravani S, Mao W, Fu L, et al. Elevated c-Src protein transcriptional gene expression. Biol Chem 2000; 45. Hanke JH, Gardner JP, Dow RL, et al. Discovery of a expression is an early event in colonic neoplasia. Lab 381:357–65. novel, potent, and Src family-selective tyrosine kinase Invest 1998;78:365–71. 55. Rajasekhar VK, Viale A, Socci ND, Wiedmann M, inhibitor. Study of Lck- and FynT-dependent T cell 50. Irby RB, Yeatman TJ. Role of Src expression and Hu X, Holland EC. Oncogenic Ras and Akt signaling activation. J Biol Chem 1996;271:695–701. activation in human cancer. Oncogene 2000;19:5636–42. contribute to glioblastoma formation by differential 46. Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, 51. Reissig D, Clement J, Sanger J, Berndt A, Kosmehl H, recruitment of existing mRNAs to polysomes. Mol Cell McGuire WL. Human breast cancer: correlation of Bohmer FD. Elevated activity and expression of Src- 2003;12:889–901. relapse and survival with amplification of the HER-2/ family kinases in human breast carcinoma tissue 56. Mendez R, Myers MG, White MF, Rhoads RE. neu oncogene. Science 1987;235:177–82. versus matched non-tumor tissue. J Cancer Res Clin Stimulation of protein synthesis, eukaryotic translation 47. Muthuswamy SK, Li D, Lelievre S, Bissell MJ, Oncol 2001;127:226–30. initiation factor 4E phosphorylation, and PHAS-I Brugge JS. ErbB2, but not ErbB1, reinitiates proliferation 52. Prendergast GC. Signal transduction: putting phosphorylation by insulin requires insulin receptor and induces luminal repopulation in epithelial acini. translation before transcription. Cancer Cell 2003; substrate 1 and phosphatidylinositol 3-kinase. Mol Cell Nat Cell Biol 2001;3:785–92. 4:244–5. Biol 1996;16:2857–64.

www.aacrjournals.org 1867 Cancer Res 2005; 65: (5). March 1, 2005

Downloaded from cancerres.aacrjournals.org on September 26, 2021. © 2005 American Association for Cancer Research. ErbB2 Promotes Src Synthesis and Stability: Novel Mechanisms of Src Activation That Confer Breast Cancer Metastasis

Ming Tan, Ping Li, Kristine S. Klos, et al.

Cancer Res 2005;65:1858-1867.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/65/5/1858

Cited articles This article cites 56 articles, 20 of which you can access for free at: http://cancerres.aacrjournals.org/content/65/5/1858.full#ref-list-1

Citing articles This article has been cited by 20 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/65/5/1858.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/65/5/1858. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.