C-Cbl-Dependent Epha2 Protein Degradation Is Induced by Ligand Binding

Vol. 1, 79–87, November 2002 Molecular Cancer Research 79 c-Cbl-Dependent EphA2 Protein Degradation Is Induced by Ligand Binding

Jennifer Walker-Daniels,1 David J. Riese II,2 and Michael S. Kinch1,3

Departments of 1Basic Medical Sciences and 2Medicinal Chemistry and Molecular Pharmacology, Purdue University Cancer Center, West Lafayette, IN; and 3MedImmune, Inc., Gaithersburg, MD

Abstract EphA2 is frequently overexpressed in a large number of The EphA2 receptor protein tyrosine kinase is overex- cancer cells (20, 23–27). Elevated EphA2 protein levels are pressed and functionally altered in a large number of observed in many different types of cancers, and the highest human carcinomas. Despite its elevated levels in cancer, levels of EphA2 protein are consistently found on the most the EphA2 on the surface of malignant cells demon- aggressive tumor cells (24, 27). For example, EphA2 protein strates lower levels of ligand binding and tyrosine levels are elevated in breast (25), prostate (24), and colon (20) phosphorylation than the EphA2 on non-transformed carcinomas, as well as in aggressive melanomas (28). epithelial cells. In our present study, we demonstrate Furthermore, ectopic overexpression of EphA2 is sufficient to that ligand-mediated stimulation causes EphA2 to be confer tumorigenic and metastatic potential upon non-trans- internalized and degraded. The mechanism of this formed epithelial cells (27). Together, these findings implicate response involves ligand-mediated autophosphorylation EphA2 as a powerful oncogene in metastatic cancer. of EphA2, which promotes an association between Despite its overexpression, the EphA2 on malignant cells EphA2 and the c-Cbl adaptor protein. We also show that exhibits lower levels of ligand binding than the EphA2 on non- c-Cbl promotes stimulation-dependent EphA2 degrada- transformed epithelial cells (25). This deficiency arises because tion. These findings are important for understanding the malignant cells generally have unstable cell-cell contacts (29– causes of EphA2 overexpression in malignant cells and 31), which prevent EphA2 from productively interacting with provide a foundation for investigating EphA2 as a its membrane-anchored ligands on neighboring cells (25, 27). potential target for therapeutic intervention. In particular, EphA2-ligand binding is highly sensitive to the proper expression and function of the E-cadherin adhesion system (25). Due to decreased EphA2-ligand binding, the Introduction EphA2 on malignant cells has lower levels of tyrosine Malignant cells often demonstrate elevated levels of intra- phosphorylation than the EphA2 in non-transformed epithelial cellular signaling by tyrosine kinases (1–3). Much recent cells that does bind ligand (25, 27). However, unlike other interest has focused on receptor tyrosine kinases, which control receptor tyrosine kinases, ligand binding is not necessary for many aspects of tumorigenesis (4–7) because receptor tyrosine EphA2 enzymatic activity (25). Instead, EphA2 ligands kinases provide much-needed targets for therapeutic interven- regulate EphA2 subcellular localization and its interactions tion (8–10). For example, antibody targeting of the HER2 with downstream adaptor and signaling proteins (25, 32, 33). receptor tyrosine kinase can treat certain breast cancers (11– Decreased levels of EphA2-ligand binding in cancer cells can 13). In part, HER2 antibodies function by promoting the be overcome in vitro by using either monoclonal antibodies or binding of adaptor molecules such as c-Cbl to HER2, and soluble forms of ligand, neither of which require stable cell-cell facilitating clearance of HER2 protein by degradation (14). contacts to stimulate EphA2 (25, 32, 33). These strategies have However, HER2 is overexpressed on a narrow range of human been used to demonstrate that EphA2 ligands negatively cancers (15) and there remains a need to identify other regulate tumor cell growth and invasiveness (25, 32). Of molecules that might be similarly targeted. particular note, ligand binding reverses the oncogenic effects of EphA2 is a transmembrane receptor tyrosine kinase that is EphA2 overexpression (27). found at low levels on adult epithelial cells (16). EphA2 is one Recent studies in our laboratory have sought to determine member of the Eph family of receptor tyrosine kinases, which how EphA2 ligands reverse the oncogenic effects of EphA2 are unique in that they recognize ligands, known as ephrins, overexpression. In the present study, we demonstrate that ligand which are anchored to the membrane of adjacent cells (17–19). binding causes EphA2 to be internalized and degraded. We also However, the cellular functions of EphA2 in epithelial cells study the mechanisms responsible for ligand-mediated EphA2 remain largely unknown (20–22). degradation and find that EphA2 interacts with the c-Cbl adaptor protein, which promotes EphA2 degradation.

Received 4/4/02; revised 7/22/02; accepted 7/26/02. 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 Results accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Grant support: American Cancer Society (CSM) Grant RPG-97-105-01 and the Stimulation Induces EphA2 Protein Degradation NIH Grant CA91318-01. EphrinA1-Fc is a soluble fusion protein in which the Requests for reprints: Michael S. Kinch, MedImmune, Inc., 35 West Watkins extracellular domain of the EphA2 ligand, Ephrin-A1, has been Mill Road, Gaithersburg, MD 20878. Phone: (240) 632-4639; Fax: (301) 527- 4200. E-mail: [email protected] linked to the human immunoglobulin Fc region (33). To study Copyright D 2002 American Association for Cancer Research. the biochemical consequences of ligand stimulation of EphA2,

we treated MDA-MB-231 cultured breast carcinoma cells with h-catenin to confirm equal sample loading. Identical results 0–2 Ag/ml EphrinA1-Fc. The samples were extracted and cell were obtained in response to antibody-mediated aggregation of lysates were resolved by SDS-PAGE before Western blot EphA2 on other cells systems, including PC-3 and MCFEphA2 analyses with EphA2-specific antibodies (D7). EphA2 immu- cells (Fig. 2B and data not shown). Because EphrinA1-Fc and noreactivity decreased by at least 80% in response to EphrinA1- B2D6 antibody aggregation provided consistent results, these Fc stimulation, and in a dose-dependent manner (Fig. 1A). When two stimuli were used interchangeably for the experiments MDA-MB-231 cells were subsequently treated with 1 Ag/ml described below. EphrinA1-Fc for 0–24 h, EphA2 immunoreactivity began to We then sought to determine the mechanism for decreased decrease within 1 h. This decrease in EphA2 was durable, as low EphA2 immunoreactivity in response to stimulation. One levels of EphA2 persisted for at least 24 h (data not shown). possibility was that loss of EphA2 immunoreactivity could Ligand binding similarly decreased EphA2 immunoreactiv- have resulted from loss of a particular antibody epitope. ity in other cell models. Multiple cell systems in which EphA2 However, lower levels of EphA2 were detected using multiple has low levels of endogenous ligand binding, as indicated by EphA2 antibodies, each of which recognizes a different epitope low levels of EphA2 tyrosine phosphorylation, were utilized. (data not shown). An alternative was that EphA2 stimulation These models included PC-3, MDA-MB-435, MCFEphA2 (27), induces EphA2 protein degradation. To test this idea, EphA2 and EphA2-transfected LNCaP and NIH-3T3 cells.1 In each protein stability was measured by pulse-chase analyses with or case, EphrinA1-Fc decreased EphA2 immunoreactivity in a without stimulation by B2D6 cross-linking. EphA2 was dose- and time-dependent manner that was analogous to our metabolically labeled by incubating MDA-MB-231 cells in results with MDA-MB-231 cells (Fig. 1B and data not shown). The decreased EphA2 immunoreactivity did not reflect a generalized decrease in cellular proteins, because the levels of A

0 0.25 0.5 2.0 µ other proteins, including h-catenin, vinculin, or paxillin, were EphrinA1-Fc ( g/mL) not changed in response to EphrinA1-Fc (Fig. 1, A and B and data not shown). Thus, these proteins were used throughout this Blot: EphA2 study as loading controls. These findings suggested that 116- EphA2-ligand binding decreased EphA2 immunoreactivity in 97- β multiple cell model systems. In our subsequent studies that Blot: -catenin focused on the molecular mechanism of ligand-induced EphA2 MDA-MB-231 protein degradation, we therefore primarily used MDA-MB- 231 cells, which are a model of invasive human breast carcinomas, and PC-3 cells, which are a model for invasive -+EphrinA1-Fc human prostate carcinomas. B We next asked whether the decreased EphA2 immunor- Blot: EphA2 eactivity detected in cell lysates of ligand-stimulated MDA- 116- MB-231 cells reflected a decrease in cell-surface associated EphA2. Cell-surface proteins were biotinylated and the cells 97- Blot: β -catenin were left untreated or stimulated with EphrinA1-Fc. Biotiny- lated proteins were immunoprecipitated with immobilized NeutrAvidin and were analyzed by Western blot analyses with -+ EphrinA1-Fc EphA2-specific antibodies. Stimulation with EphrinA1-Fc led C to a marked decrease in cell-surface associated EphA2 protein Blot: EphA2 levels (Fig. 1C). 116- IP: Avidin The EphA ligand used in these studies, EphrinA1, can stimulate EphA receptor tyrosine kinases other than EphA2 (19, PC-3 34). Therefore, we next asked whether decreased EphA2 FIGURE 1. Ligand activation decreases EphA2 immunoreactivity in a protein levels following EphrinA1-Fc stimulation of cells was dose- and time-dependent manner. MDA-MB-231 breast cancer cells A j caused by EphA2 stimulation exclusively or indirectly through were incubated with 0 – 2 g/ml EphrinA1-Fc at 37 Cfor1h.A. Equal amounts of cell lysates were resolved by SDS-PAGE and subjected to the stimulation of other EphA kinases. MDA-MB-231 cells anti-EphA2 Western blot analysis (D7). The membrane was stripped and were treated with B2D6, a monoclonal antibody that specifi- reprobed with h-catenin antibodies as a loading control. Note that EphA2 protein levels decrease in a dose-dependent manner following ligand cally recognizes an epitope on the extracellular domain of activation, but h-catenin levels remain unchanged. The most substantial EphA2, and a secondary antibody (rabbit anti-mouse IgG). decrease in EphA2 levels was with 2 Ag/ml EphrinA1-Fc treatment. B. PC- Western blot analyses of cell lysates demonstrated that EphA2 3 prostate cancer cells were left untreated (À) or were stimulated (+) with 1 Ag/ml EphrinA1-Fc for 2 h. Cell lysates probed for EphA2 levels showed cross-linking by B2D6 reduced EphA2 immunoreactivity (Fig. that ligand stimulation decreases EphA2 protein in PC-3 cells but does not 2A). In contrast, neither the primary nor secondary antibodies alter the levels of the loading control h-catenin. C. MDA-MB-231 cells alone changed EphA2 immunoreactivity. The membranes were were incubated with 0.5 mg/ml EZ-Link Sulfo-NHS-LC-Biotin and were left unstimulated (À) or were stimulated with 1 Ag/ml EphrinA1-Fc (+) for 3 h. stripped and reprobed with antibodies that recognize vinculin or Biotinylated proteins were immunoprecipitated (IP) with NeutrAvidin, and the presence of EphA2 protein was detected by Western blot analyses with EphA2-specific antibodies. Note that levels of cell-surface associated EphA2 decrease following EphrinA1-Fc treatment. Molecular mass stand- 1M. S. Kinch, unpublished observations. ards are indicated to the left.

Downloaded from mcr.aacrjournals.org on October 1, 2021. © 2002 American Association for Cancer Research. Molecular Cancer Research 81 o Multiple analyses confirmed that EphA2 was internalized in +2 o o o NT 1 2 A 1 Antibodies response to stimulation. First, we aggregated EphA2 with specific antibodies as described above and stripped the labeled antibodies from the cell surface using a low-pH medium. The Blot: EphA2 116- EphA2 localization in acid-stripped samples was identical to that of samples in Fig. 4A (data not shown), which suggested that this EphA2 had been internalized. Immunoelectron 116- Blot: Vinculin microcopy confirmed that stimulated EphA2 had been internalized (Fig. 4B). For these studies, EphA2 was labeled MDA-MB-231 at the surface of MDA-MB-231 cells with B2D6 antibodies and gold-conjugated secondary antibodies. The antibody

o incubation was performed on ice to permit binding but to

+2 prevent antibody-dependent receptor internalization before o B NT 1 Antibodies sample analysis. The samples were incubated at 37jCfor0– 20 min to permit EphA2 internalization, and were then Blot: EphA2 immediately processed for observation by electron micro- 116- scopy using standard techniques. Before incubation at 37jC, gold-labeled EphA2 was diffusely distributed over the cell 97- Blot: β-catenin Fig. 4B, 0 min). Upon incubation at 37jC, EphA2 became PC-3 clustered at the cell surface and was internalized within endosomal vesicles (Fig. 4B, 20 min). Thus, these studies FIGURE 2. EphA2-specific antibody aggregation decreases EphA2 demonstrate that antibody aggregation of EphA2 causes immunoreactivity. A. MDA-MB-231 breast cancer cells were incubated with EphA2-specific antibodies (B2D6, 1j) alone, secondary antibodies EphA2 internalization. alone (2j), or aggregated at the cell surface with both (1j+2j) for 1 h. Cell lysates were resolved by SDS-PAGE and EphA2 levels were assessed by Western blot analyses with anti-EphA2 antibodies. Note decreased EphA2 Stimulated EphA2 Interacts with c-Cbl levels following aggregation of EphA2 with both primary and secondary We next sought to determine the biochemical mechanism by antibodies relative to control samples. The blot was stripped and reprobed for the loading control vinculin. B. PC-3 cells were left untreated or EphA2 which EphA2 is internalized and degraded in response to stimu- was aggregated with both EphA2-specific primary antibodies and lation. In response to ligand binding, many receptor tyrosine secondary antibodies (1j+2j) for 1 h. Again, EphA2 protein levels decrease following antibody aggregation of EphA2, but the levels of the loading control h-catenin remain constant. A 0 0.5 1 2 4 8 Chase (hours) medium containing 35S-labeled methionine and cysteine. In NT untreated cells, little EphA2 degradation was detected over the 116- 8-h time period (Fig. 3A). Upon EphA2 stimulation, however, EphA2 was rapidly degraded and had an estimated half-life of o o 3.5 h (Fig. 3B). These results indicated that the decreased 1 +2 116- EphA2 immunoreactivity detected following EphA2 stimulation represented EphA2 protein degradation. B 1.5 EphA2 is Internalized in Response to Stimulation Ligand-mediated aggregation of transmembrane receptors often precedes their internalization and degradation (35). Thus, 1.0 we asked whether cell surface aggregation of EphA2 would similarly induce its internalization. To test this, the EphA2 on 0.5 the surface of MDA-MB-231 cells was labeled with B2D6 antibodies and this EphA2 was aggregated using fluorescein- Labeled EphA2 conjugated secondary antibodies for 0–60 min at 37jC. 0 Fluorescence microscopy revealed that EphA2 at the cell 02468 surface of control samples (the 0 min time point) was mostly Chase (hours) diffuse. Some aggregation of EphA2 within membrane ruffles FIGURE 3. Stimulation induces EphA2 protein degradation. A. MDA- was observed, which is consistent with our previous studies of MB-231 cells were metabolically labeled with Tran35S-Label and EphA2 EphA2 subcellular localization in metastatic cells (25). After 5 was left untreated [not treated (NT)] or was stimulated by antibody j j min of aggregation with secondary antibodies, EphA2 had aggregation (1 +2 ). The radiolabel was chased for the indicated times and EphA2 was immunoprecipitated from cell lysates and resolved by clustered into distinct patches and after 20 min, much of this SDS-PAGE. Note that whereas EphA2 in control cells (top) is stable over material appeared to have been internalized (Fig. 4A). Cells 8 h, stimulated EphA2 (bottom) is degraded. B. Quantitation of band intensities in A estimates that the half-life of antibody-aggregated EphA2 treated with secondary antibodies only had low levels of (&) is 3.5 h, whereas non-treated EphA2 in control cells (6) has a half-life nonspecific staining at all time points (data not shown). greater than 8 h.

kinases undergo autophosphorylation, which creates docking A-Sepharose (PAS)] to prevent unwanted binding of sites for signaling or adaptor proteins (36, 37). Indeed, EphA2 EphrinA1-Fc and associated EphA2. MDA-MB-231 cells was tyrosine phosphorylated within 10 min following either were then left untreated or stimulated with EphrinA1-Fc for ligand binding or antibody aggregation (Fig. 5A). The phos- 10 min at 37jC. This short duration of EphA2 stimulation photyrosine (p-Tyr) content of EphA2 decreased over time, was chosen to evaluate the protein interactions of EphA2 which coincided with decreased levels of EphA2 protein (data before its degradation. Cell lysates were precipitated with not shown). Cbl-PAS or with an irrelevant RPS. Relative to unstimulated The c-Cbl adaptor protein has been shown to preferentially controls, EphrinA1-Fc-stimulated cells had more EphA2 in associate with, and induce the degradation of, tyrosine- association with c-Cbl. RPS-PAS precipitated very little phosphorylated receptor tyrosine kinases (38–40). Thus, we EphA2, regardless of ligand stimulation (Fig. 6B). These postulated that an interaction with c-Cbl might facilitate EphA2 experiments provide additional support for the hypothesis protein degradation. To test this hypothesis, we first examined that ligand stimulation promotes the association of EphA2 the subcellular localization of EphA2 and c-Cbl before and after with c-Cbl. EphA2 stimulation. EphA2 was aggregated using B2D6 and fluorescein-conjugated secondary antibodies to track its local- c-Cbl Regulates EphA2 Degradation ization. The samples were then immunostained for c-Cbl and The activation-dependent association of EphA2 with c-Cbl examined by epifluorescence microscopy. In mock-stimulated led us to ask whether binding to c-Cbl contributed to activation- cells, EphA2 exhibited a characteristic membrane-staining dependent EphA2 protein degradation. c-Cbl functions in part pattern that overlapped with the distribution of a population by targeting proteins for degradation via the proteasomal of c-Cbl (Fig. 5B). EphA2 and c-Cbl coclustered proximal to pathway (39, 40). Thus, our first experiments utilized the the cell membrane following EphA2 aggregation, and chemical proteasome inhibitor MG-132, which potently but subsequently colocalized into intracellular vesicles (Fig. 5B). reversibly decreases the degradation of ubiquitin-conjugated Vertical sectioning using confocal microscopy confirmed the proteins by the 26S proteasome complex. PC-3 cell monolayers colocalization of EphA2 and c-Cbl following ligand binding were treated with vehicle alone or 10 AM MG-132 and then left (Fig. 6A). untreated or stimulated by EphrinA1-Fc for 2 h. MG-132 Based on the colocalization of EphA2 and c-Cbl in prevented 67% of activation-dependent EphA2 degradation response to EphA2 activation, we asked if the two molecules (Fig. 6A). We were able to rule out that MG-132 had prevented could interact in vivo. To test this, we stimulated MDA-MB- EphA2 from productively responding to ligand because the 231 cells with EphrinA1-Fc and asked if EphA2 could EphA2 in cells that had been treated with both MG-132 and coimmunoprecipitate with c-Cbl. In designing this study, we EphrinA1-Fc was highly tyrosine phosphorylated (Fig 7A, considered that the EphrinA1-Fc itself might unintentionally bottom panel). Similar results were obtained using MDA-MB- precipitate EphA2 and thereby lead to an erroneous inter- 231 cells (data not shown). pretation of the results. To overcome this, Sepharose beads To determine whether c-Cbl played a role in activation- were directly conjugated to c-Cbl antibodies [Cbl-protein induced degradation of EphA2, we blocked c-Cbl function

A 0 min 5 min 20 min 25 µm

FIGURE 4. EphA2 aggregation causes its internalization. A. The EphA2 on MDA-MB- 231 breast cancer cells was stimulated using the anti-EphA2 antibody B2D6 and rhodamine- conjugated rabbit anti-mouse IgG secondary antibodies at 37jC for the times shown. Over time, EphA2 clustered at the cell surface (5min) and was internalized (20 min). B. Immunoelec- tron microscopy was performed by labeling B 0 min 20 min MDA-MB-231 cells with the anti-EphA2 antibody 200 nm 200 nm B2D6. EphA2 was then aggregated using gold- conjugated rabbit anti-mouse IgG secondary antibodies at 37jC. Gold particles are the small uniform electron dense structures indicated by the arrows. 0min, the localization of gold particles was assessed immediately after adding secondary antibodies. Note the non-clustered EphA2 at the cell surface (arrows). 20 min,the localization of gold particles was assessed after adding secondary antibodies and returning cells to 37jC for 20 min. Note that EphA2 clusters at the cell surface (small arrows) and is internalized into early endosomal compartments (big arrow) within 20 min of aggregation.

A o +2 o NT EphrinA1-F NT 1 Blot: p-Tyr IP: EphA2 116- 116- Blot: EphA2 116- 116- IP: EphA2

B c-Cbl EphA2

FIGURE 5. EphA2 stimulation increases its association with c-Cbl. A. EphA2 on MDA-MB- 231 breast cancer cells was stimulated with either EphrinA1-Fc or the EphA2-specific antibody B2D6 plus secondary antibodies (1j+2j) for 10 min. Cells were lysed and EphA2 was immunoprecipi- Control tated (IP) and probed for its p-Tyr content with the antibody 4G10 (top) or for EphA2 levels (bottom). 100 µm EphA2 stimulation did not affect EphA2 protein levels at this short time point, but either form of EphA2 stimulation dramatically increased the tyrosine phosphorylation of EphA2. B. The subcellular localization of EphA2 and c-Cbl was assessed by immunofluorescence microscopy in o

MDA-MB-231 cells before and after stimulation of +2 o

EphA2 by antibody aggregation (1j+2j). Note 1 that whereas EphA2 and c-Cbl are diffusely distributed in control cultures (top), the two molecules colocalize in the cytosol in response to EphA2 stimulation within 20 min (bottom).

using dominant-negative Cbl mutants. MDA-MB-231 cells mimic the actions of endogenous ligand. We also demonstrate were infected with a recombinant retrovirus that encodes v- that EphA2 interacts with the c-Cbl adaptor protein in a cbl (41), which inhibits c-Cbl function by competing with it stimulation-dependent manner and that proteasome and c-Cbl for binding to phosphorylated tyrosine residues (42). In the functions promote stimulation-dependent EphA2 protein absence of EphA2 stimulation, vector-infected and v-cbl- degradation. infected cells had similar levels of EphA2 protein (Fig. 7B). Recent studies by our laboratory and others have demon- Stimulation of EphA2 by antibody aggregation caused strated that ligand stimulation of EphA2 on tumor cells has efficient degradation of EphA2 in control samples (68% several biological consequences, including decreased tumor cell degradation) but higher levels of EphA2 persisted in growth, migration, and invasiveness (25, 27, 32). At first stimulated v-cbl-infected cells (25% degradation, Fig. 7B). glance, these biological outcomes were presumed to have arisen Thus, v-Cbl prevented 63% of stimulation-induced EphA2 from direct induction of downstream signaling. Tyrosine- protein degradation (P < 0.03). Similar results were obtained phosphorylated EphA2 interacts with a variety of adaptor and using 70Z-Cbl (43) (P < 0.0004, data not shown). The inverse signaling proteins including PI 3-kinase (44), SHC,1 SHP2 experimentation was also performed. Overexpression of c- (32), and SLAP (44). EphA2 also contains a consensus binding Cbl in MDA-MB-231 cells was sufficient to decrease the motif for c-Cbl (YxxxP) at tyrosine 813, although it is presently levels of EphA2 protein (Fig. 7C). Thus, consistent data unclear whether this site is necessary or sufficient for the confirmed the link between c-Cbl and the regulation of EphA2 interactions described herein. Our present findings suggest that protein levels. the biological consequences of ligand-mediated EphA2 stimulation may arise in part from the removal of EphA2 from the cell surface and its subsequent degradation. High levels of Discussion EphA2 are sufficient to confer oncogenic and metastatic The major finding of this study is that the EphA2 receptor potential upon non-transformed cells (27). It logically follows tyrosine kinase is rapidly and efficiently degraded in response that stimulation-dependent degradation of EphA2 could to ligand- or antibody-mediated stimulation. EphA2 levels in negatively regulate malignant behavior by decreasing the carcinoma cells decreased following treatment with stimuli that expression of a powerful oncoprotein. In fact, recent studies

A c-Cbl EphA2

Cbl

EphA2

Control Merge Vertical Section FIGURE 6. Activation-dependent interaction of EphA2 with c-Cbl. A. Confocal microscopic

c Cbl analyses confirm that c-Cbl (green) and EphA2 (red) colocalize in response to antibody aggregation in MDA-MB-231 cells. Note that vertical EphA2 sectioning (right) reveals that EphA2 redistributes from the cell surface to intracellular sites, where it colocalizes with c-Cbl (yellow). B. MDA-MB-231

EphrinA1-F Merge cells were left untreated (À) or stimulated (+) for 10 min at 37jC with EphrinA1-Fc. Total protein levels of EphA2 and c-Cbl remained constant in cell lysates (left) because the l0-min stimulation period was insufficient to cause EphA2 degrada- B RPS c-Cbl IP Antibody tion. Immunoprecipitates with a negative control -+ -+ -+ EphrinA1-F rabbit polyclonal sera with PAS [rabbit polyclonal c serum (RPS)] or c-Cbl PAS (c-Cbl) were probed for EphA2 (top) or c-Cbl (bottom). No detectable Blot: EphA2 amounts of c-Cbl and little EphA2 precipitated 116- with the negative control RPS-PAS regardless of EphA2 stimulation. However, EphA2 coimmuno- 116- Blot: c-Cbl precipitated with c-Cbl PAS in a stimulation- dependent manner.

reveal that EphA2 can serve as a target for monoclonal become ubiquitinated (58) and such changes could also alter antibodies and that growth-inhibitory EphA2 antibodies induce intracellular signaling. Although we cannot exclude that some efficient degradation of EphA2 (45). However, we cannot of the biological consequences of EphA2 stimulation result exclude the possibility that active induction of downstream from c-Cbl-mediated signaling, EphA2 stimulation did not alter signals following EphA2 stimulation also contributes to the the p-Tyr content or protein levels of c-Cbl (data not shown). negative regulation of tumor cell growth, survival, and Thus, our results suggest that c-Cbl primarily serves to facilitate invasiveness. Thus, future studies should assess the relative EphA2 degradation. contributions of direct EphA2 signaling versus receptor Stable cell-cell contacts, particularly those mediated by E- degradation as mechanisms of ligand-mediated inhibition of cadherin, favor ligand-mediated EphA2 autophosphorylation malignant cell behavior. (25), which could promote interactions between EphA2 and c- The interaction of EphA2 with c-Cbl has interesting Cbl, and thereby favor EphA2 degradation. Malignant cells implications for understanding the biological actions of both often have unstable cell-cell contacts (31, 59) and one pre- proteins. c-Cbl is an adaptor protein that interacts with certain diction of our model is that the resulting decrease in EphA2- tyrosine-phosphorylated proteins via an internal SH2 domain ligand binding (25, 26) allows EphA2 to accumulate at the cell (38). c-Cbl has been shown to negatively regulate several surface. Our present findings suggest that decreased ligand receptor tyrosine kinases [e.g., epidermal growth factor receptor binding might favor EphA2 accumulation, triggering a cycle (46, 47), platelet-derived growth factor receptor (48), HER2 that would increase the amount of EphA2 in tumor cells. (14)] by promoting their ubiquitination and degradation Consistent with this, the highest levels of EphA2 protein are following ligand stimulation (48, 49). This function involves consistently found in aggressive tumor cells where EphA2 has a a zinc finger domain in c-Cbl that functions as an ubiquitin low p-Tyr content (27). Indeed, the levels of EphA2 protein can ligase (49–51). Based on this knowledge, we sought to be disproportionately higher than EphA2 mRNA levels in determine whether EphA2 becomes ubiquitinated in response malignant epithelial cells (26). Such a model may explain why to ligand binding, but technical limitations have precluded this high levels of EphA2 are found in a large number of human line of investigation. cancers. c-Cbl itself can be tyrosine phosphorylated (52) and serve as The aforementioned model of EphA2 degradation was an adaptor for downstream signaling (53, 54). c-Cbl-mediated based upon artificial stimulation of EphA2 on tumor cells. It signal transduction is not well understood, but c-Cbl has been is presently unclear whether soluble ligands or antibodies implicated in both the positive (55, 56) and negative (57) faithfully reproduce the consequences of endogenous transmission of intracellular signals. Moreover, c-Cbl can ligands, which are anchored to the surface of adjacent cells

(60). For example, differences in the timing or duration Cells and Cell Culture of EphA2 stimulation, or changes in EphA2 subcellular MDA-MB-231 cultured human breast tumor cells, and localization in response to a soluble versus a membrane an- LNCaP and PC-3 cultured human prostate tumor cells were chored, could have profound biochemical or biological con- propagated in RPMI 1640 supplemented with 10% fetal bovine sequences. serum and antibiotic solution as described (24, 25). MDA-MB- In summary, our present findings indicate that the EphA2 435 and MCFEphA2 cells were cultured as described previously receptor tyrosine kinase is rapidly and efficiently degraded in (27, 62). response to antibody or ligand stimulation. Stimulated EphA2 becomes tyrosine phosphorylated and interacts with the c-Cbl Transfection adaptor protein, which promotes EphA2 degradation. Our The HA-tagged human c-cbl in pGEM4Z was a kind gift of findings have implications for understanding the prevalence of W. Langdon (The University of Western Australia). HA-c-cbl EphA2 overexpression in cancer cells and for the development was cut from this vector and cloned into the mammalian of targeted therapeutics designed to selectively remove a pow- expression vector pCDNA3. It was then transfected into MDA- erful oncoprotein. MB-231 cells using the FuGENE system according to the manufacturer’s recommendations (Roche, Indianapolis, IN) and A Materials and Methods cells were selected with 400 g/ml G418. Transfectants were serially diluted and subcloned. Data reported are representative Antibodies and Reagents of at least three clones. The proteasome inhibitor, MG-132, was purchased from Calbiochem (San Diego, CA). Monoclonal antibodies specific for EphA2 (D7 and B2D6) were generated in the laboratory Immunoprecipitations and Western Blot Analyses (61) or purchased from Upstate Biotechnology, Inc. (Lake Cells were lysed and protein expression was analyzed by Placid, NY). Rabbit polyclonal antibodies specific for EphA2 immunoprecipitation and Western blotting as described (24). and c-Cbl were purchased from Santa Cruz Biotechnology For the Cbl immunoprecipitations, PAS beads were incubated (Santa Cruz, CA). Antibodies specific for vinculin and overnight with c-Cbl polyclonal antibodies (Santa Cruz paxillin were a generous gift of Dr. K. Burridge (University Biotechnology) and washed three times before using this of North Carolina). Monoclonal antibodies specific for h- material to precipitate cell lysates as described above. Immuno- catenin and c-Cbl were purchased from Transduction precipitations with NeutrAvidin (Pierce, Rockford, IL) were Laboratories (Lexington, KY) and 4G10 p-Tyr antibodies performed by incubating 40 Al cross-linked immobilized were purchased from Upstate Biotechnology. EphrinA1-Fc NeutrAvidin in a 50% slurry with lysates of cell-surface was the generous gift of Dr. B. Wang (Case-Western Reserve biotinylated cells for 1.5 h at 4jC. To confirm equal sample University). loading, membranes were stripped and reprobed with antibodies

A C FIGURE 7. c-Cbl contributes to EphA2 protein degradation. A. PC-3 cells were pretreated with EphrinA1-Fc - + - + the proteasome inhibitor MG-132 or a vehicle MG-132

- - + + Vector c-Cbl control as indicated. Cells were left untreated (À) or stimulated (+) with EphrinA1-Fc for 2 h at 37jC. Western blot analyses of cell lysates revealed that Blot: EphA2 MG-132 prevented at least 60% of the stimulation- Blot: HA 116- 116- induced EphA2 degradation (top). The blot was IP: HA stripped and reprobed for h-catenin as a loading control (middle). EphA2 was immunoprecipitated Blot: β-catenin 97- from the same cell lysates and probed for tyrosine phosphorylation (bottom). Note that MG-132 treatment did not prevent stimulation-induced tyrosine Blot: p-Tyr phosphorylation of EphA2. B. MDA-MB-231 cells IP: EphA2 116- Blot: c-Cbl were infected with vector control retroviral particles 116- or with retroviral particles that encode the v-cbl PC-3 gene. EphA2 on these cells was stimulated by antibody aggregation (+) for 2 h at 37jC or was left B untreated (À). Cells infected with the retrovirus Vector v-Cbl Blot: EphA2 that encodes v-cbl prevented 63% of activation- dependent EphA2 protein degradation compared 1o+2o Antibodies - + - + 116- to vector control cells. Levels of the loading control paxillin remained constant. C. The protein levels of EphA2 were evaluated following overexpression of Blot: EphA2 97- Blot: β-catenin HA-tagged c-Cbl in MDA-MB-231 cells. The overexpression of c-Cbl was confirmed using both HA 116- and c-Cbl-specific antibodies and equal sample loading was confirmed using h-catenin-specific Blot: Paxillin antibodies. Note that EphA2 protein levels are 66- lower in the c-Cbl overexpressing cells than in control cells. MDA-MB-231

specific for vinculin, h-catenin, or paxillin. To measure and ice with B2D6 anti-EphA2 antibodies for 20 min, rinsed, and compare protein levels, the autoradiograms were analyzed with incubated on ice with 10 nM colloidal gold-labeled goat anti- KODAK 1D image analysis software. Band intensity was mouse IgG antibodies (Sigma). Cells were kept on ice or determined after correcting for nonspecific background. returned to 37jC for various times and then fixed in glu- taraldehyde and prepared for electron microscopic evaluation EphA2 Stimulation Studies according to standard methods. The EphA2 on tumor cells was stimulated with a soluble EphrinA1 fusion protein (EphrinA1-Fc) or EphA2-specific Retroviral Infections antibodies (B2D6) as described (21, 25). Briefly, cells were The v-cbl cDNA constructs were a generous gift from Dr. W. incubated with ligand at 37jC for the times indicated, Langdon (University of Western Australia) and were subcloned or incubated with antibodies on ice and then returned to into the recombinant retroviral vector pRV-HygR (64). The 37jC for the given times. Mock-stimulated cells were incubated resulting constructs were packaged into amphotropic retrovirus with ligand or antibodies on ice, but never returned to 37jC. particles that were used to infect MDA-MB-231 cells as described (64). Infected cells were selected using hygromycin B Biotinylation of Cell Surface Proteins (Alexis Corp., San Diego, CA) and drug-resistant colonies were Cell-surface associated proteins were biotinylated by pooled to generate bulk cultures that were utilized for the incubating adherent cells with 0.5 mg/ml EZ-Link Sulfo- studies described herein. NHS-LC-Biotin (Pierce) for 30 min at room temperature on an orbital shaker. Excess biotin was removed, and the cells were Acknowledgments The authors gratefully thank Drs. J. Turek, D. Zelinski, N. D. Zantek, D. Van then left unstimulated or stimulated with 1 Ag/ml EphrinA1-Fc Horn, and R. Pratt for helpful advice and technical expertise and Drs. W. for 3 h. Biotinylated proteins were immunoprecipitated with Langdon and B. Wang for reagents. immobilized NeutrAvidin (Pierce) and subjected to Western blot analyses with EphA2-specific antibodies. References 1. Kolibaba, K. S. and Druker, B. J. Protein tyrosine kinases and cancer. Biochim. Biophys. Acta., 1333: F217 – F248, 1997. Metabolic Labeling 2. Dickson, R. B. and Lippman, M. E. New approaches in the therapy of breast MDA-MB-231 cells were cultured in 6-cm tissue culture cancer—introduction. Breast Cancer Res. Treat., 38: 1 – 2, 1996. dishes (Costar, Acton, MA) overnight at 37jC. The samples 3. Patarca, R. Protein phosphorylation and dephosphorylation in physiologic and were then incubated in media lacking cysteine and methionine oncologic processes. Crit. Rev. Oncog., 7: 343 – 432, 1996. for 1 h before labeling with 200 ACi Tran35S-Label (ICN 4. van der Geer, P., Hunter, A. J., and Lindberg, R. A. Receptor protein-tyrosine kinases and their signal transduction pathways. Annu. Rev. Cell Biol., 10: 251 – Radiochemicals, Irvine, CA) for 1 h. After removing the labeling 337, 1994. medium, the samples were incubated for 0–8 h in standard cell 5. Rubin, R. and Baserga, R. Insulin-like growth factor-I receptor. Its role in cell culture medium supplemented with 1 mg/ml of L-cysteine and L- proliferation, apoptosis, and tumorigenicity. Lab. Invest., 73: 311 – 331, 1995. methionine (Sigma Chemical Co., St. Louis, MO). EphA2 6. Kim, H. G., Kassis, J., Souto, J. C., Turner, T., and Wells, A. EGF receptor antibody aggregation was performed coincident with the res- signaling in prostate morphogenesis and tumorigenesis. Histol. Histopathol., 14: 1175 – 1182, 1999. toration of normal cell culture medium. The samples were 7. Malitschek, B., Fornzler, D., and Schartl, M. Melanoma formation in resolved by SDS-PAGE, the gel was dried, and the resulting Xiphophorus: a model system for the role of receptor tyrosine kinases in autoradiogram was analyzed using a Cyclone Storage Phosphor tumorigenesis. BioEssays, 17: 1017 – 1023, 1995. System (Packard, Meriden, CT). OptiQuant image analysis 8. Shak, S. Overview of the trastuzumab (Herceptin) anti-HER2 monoclonal software (Packard) was used to measure band intensities. antibody clinical program in HER2-overexpressing metastatic breast cancer. Herceptin Multinational Investigator Study Group. Semin. Oncol., 26: 71 – 77, 1999. Immunofluorescence 9. Levitzki, A. and Gazit, A. Tyrosine kinase inhibition: an approach to drug Cells were grown on glass coverslips overnight before development. Science, 267: 1782 – 1788, 1995. fixation and extraction as described (63). The samples were 10. Rusch, V., Mendelsohn, J., and Dmitrovsky, E. The epidermal growth factor receptor and its ligands as therapeutic targets in human tumors. Cytokine Growth stained with specific primary antibodies and rhodamine- Factor Rev., 7: 133 – 141, 1996. conjugated donkey anti-mouse or FITC-conjugated donkey 11. Weiner, L. M. Monoclonal antibody therapy of cancer. Semin. Oncol., 26: anti-rabbit antibodies (Chemicon International, Inc., Temecula, 43 – 51, 1999. CA) and antibody binding was visualized using epifluorescence 12. Press, M. F., Jones, L. A., Godolphin, W., Edwards, C. L., and Slamon, D. J. HER-2/neu oncogene amplification and expression in breast and ovarian cancers. microscopy (BX60 Olympus, Lake Success, NY). Images were Prog. Clin. Biol. Res., 354A: 209 – 221, 1990. recorded by photomicroscopy using a 35-mm camera and Tmax 13. Baselga, J., Norton, L., Albanell, J., Kim, Y-M., and Mendelsohn, J. 400 film (Eastman Kodak, Rochester, NY). To observe inter- Recombinant humanized anti-HER2 antibody (Herceptin) enhances the antitumor nalized antibodies, antibodies at the cell surface were stripped activity of paclitaxel and doxorubicin against HER2/neu overexpressing breast cancer xenografts. Cancer Res., 58: 2825 – 2831, 1998. off of cell monolayers by incubation in a low-pH medium (pH 2) 14. Klapper, L. N., Waterman, H., Sela, M., and Yarden, Y. Tumor-inhibitory for 5 min on ice. Confocal images were taken with a Bio-Rad antibodies to HER-2/ErbB-2 may act by recruiting c-Cbl and enhancing MRC1024-UV on a Nikon Diaphot 300 microscope. ubiquitination of HER-2. Cancer Res., 60: 3384 – 3388, 2000. 15. Reese, D. M. and Slamon, D. J. HER-2/neu signal transduction in human Electron Microscopy breast and ovarian cancer. Stem Cells, 15: 1 – 8, 1997. 16. Lindberg, R. A. and Hunter, T. cDNA cloning and characterization of eck, an Cell monolayers were cultured in six-well tissue culture epithelial cell receptor protein-tyrosine kinase in the eph/elk family of protein dishes at 37jC overnight. The samples were then incubated on kinases. Mol. Cell. Biol., 10: 6316 – 6324, 1990.

17. Bartley, T. D., Hunt, R. W., Welcher, A. A., Boyle, W. J., Parker, V. P., 42. Thien, C. B. and Langdon, W. Y. EGF receptor binding and transformation Lindberg, R. A., Lu, H. S., Colombero, A. M., Elliott, R. L., and Guthrie, B. A. by v-cbl is ablated by the introduction of a loss-of-function mutation from the B61 is a ligand for the ECK receptor protein-tyrosine kinase. Nature, 368: 558 – Caenorhabditis elegans sli-1 gene. Oncogene, 14: 2239 – 2249, 1997. 560, 1994. 43. Andoniou, C. E., Thien, C. B., and Langdon, W. Y. Tumour induction by 18. Gale, N. W. and Yancopoulos, G. D. Ephrins and their receptors: a repulsive activated abl involves tyrosine phosphorylation of the product of the cbl topic? Cell Tissue Res., 290: 227 – 241, 1997. oncogene. EMBO J., 13: 4515 – 4523, 1994. 19. Pasquale, E. B. The Eph family of receptors. Curr. Opin. Cell Biol., 9: 608 – 44. Pandey, A., Duan, H., and Dixit, V. M. Characterization of a novel src-like 615, 1997. adapter protein that associates with the Eck receptor tyrosine kinase. J. Biol. 20. Rosenberg, I. M., Goke, M., Kanai, M., Reinecker, H. C., and Podolsky, Chem., 270: 19201 – 19204, 1995. D. K. Epithelial cell kinase-B61: an autocrine loop modulating intestinal epithelial 45. Carles-Kinch, K., Kilpatrick, K. E., Stewart, J. C., and Kinch, M. S. Anti- migration and barrier function. Am. J. Physiol., 273: G824 – G832, 1997. body targeting of the EphA2 tyrosine kinase inhibits malignant cell behavior. 21. Pandey, A., Shao, H., Marks, R. M., Polverini, P. J., and Dixit, V. M. Role of Cancer Res., 62: 2840 – 2847, 2002. B61, the ligand for the Eck receptor tyrosine kinase, in TNF-a-induced 46. Waterman, H., Levkowitz, G., Alroy, I., and Yarden, Y. The RING finger of angiogenesis. Science, 268: 567 – 569, 1995. c-Cbl mediates desensitization of the epidermal growth factor receptor. J. Biol. 22. Magal, E., Holash, J. A., Toso, R. J., Chang, D., Lindberg, R. A., and Chem., 274: 22151 – 22154, 1999. Pasquale, E. B. B61, a ligand for the Eck receptor protein-tyrosine kinase, 47. Ueno, H., Sasaki, K., Miyagawa, K., Honda, H., Mitani, K., Yazaki, Y., and exhibits neurotrophic activity in cultures of rat spinal cord neurons. J. Neurosci. Hirai, H. Antisense repression of proto-oncogene c-Cbl enhances activation of the Res., 43: 735 – 744, 1996. JAK-STAT pathway but not the ras pathway in epidermal growth factor receptor 23. Easty, D. J., Herlyn, M., and Bennett, D. C. Abnormal protein tyrosine kinase signaling. J. Biol. Chem., 272: 8739 – 8743, 1997. gene expression during melanoma progression and metastasis. Int. J. Cancer, 60: 48. Miyake, S., Lupher, M. L., Jr., Druker, B., and Band, H. The tyrosine kinase 129 – 136, 1995. regulator Cbl enhances the ubiquitination and degradation of the platelet-derived 24. Walker-Daniels, J., Coffman, K., Azimi, M., Rhim, J. S., Bostwick, D. G., growth factor receptor a. Proc. Natl. Acad. Sci. USA, 95: 7927 – 7932, 1998. Snyder, P., Kerns, B. J., Waters, D. J., and Kinch, M. S. Overexpression of the 49. Levkowitz, G., Waterman, H., Ettenberg, S. A., Katz, M., Tsygankov, A. Y., EphA2 tyrosine kinase in prostate cancer. Prostate, 41: 275 – 280, 1999. Alroy, I., Lavi, S., Iwai, K., Reiss, Y., Ciechanover, A., Lipkowitz, S., and Yarden, 25. Zantek, N. D., Azimi, M., Fedor-Chaiken, M., Wang, B., Brackenbury, R., Y. Ubiquitin ligase activity and tyrosine phosphorylation underlie suppression of and Kinch, M. S. E-cadherin regulates the function of the EphA2 receptor growth factor signaling by c-Cbl/Sli-1. Mol. Cell, 4: 1029 – 1040, 1999. tyrosine kinase. Cell Growth & Differ., 10: 629 – 638, 1999. 50. Joazeiro, C. A., Wing, S. S., Huang, H., Leverson, J. D., Hunter, T., and Liu, 26. Zantek, N. D., Walker-Daniels, J., Stewart, J. C., Hansen, R. K., Robinson, Y. C. The tyrosine kinase negative regulator c-Cbl as a RING-type, E2-dependent D., Miao, H., Wang, B., Kung, H. J., Bissell, M. J., and Kinch, M. S. MCF-10A- ubiquitin-protein ligase. Science, 286: 309 – 312, 1999. NeoST: a new cell system for studying cell-ECM and cell-cell interactions in 51. Yokouchi, M., Kondo, T., Houghton, A., Bartkiewicz, M., Horne, W. C., breast cancer. Clin. Cancer Res., 7 (11): 3640 – 3648, 2001. Zhang, H., Yoshimura, A., and Baron, R. Ligand-induced ubiquitination of the 27. Zelinski, D. P., Zantek, N. D., Stewart, J., Irizarry, A., and Kinch, M. S. epidermal growth factor receptor involves the interaction of the c-Cbl RING EphA2 overexpression causes tumorigenesis of mammary epithelial cells. Cancer finger and UbcH7. J. Biol. Chem., 274: 31707 – 31712, 1999. Res., 61: 2301 – 2306, 2001. 52. Feshchenko, E. A., Langdon, W. Y., and Tsygankov, A. Y. Fyn, Yes, and Syk 28. Easty, D. J., Guthrie, B. A., Maung, K., Farr, C. J., Lindberg, R. A., Toso, phosphorylation sites in c-Cbl map to the same tyrosine residues that become R. J., Herlyn, M., and Bennett, D. C. Protein B61 as a new growth factor: phosphorylated in activated T cells. J. Biol. Chem., 273: 8323 – 8331, 1998. expression of B61 and up-regulation of its receptor epithelial cell kinase during 53. de Jong, R., ten Hoeve, J., Heisterkamp, N., and Groffen, J. Crkl is melanoma progression. Cancer Res., 55: 2528 – 2532, 1995. complexed with tyrosine-phophorylated cbl in PH-positive leukemia. J. Biol. 29. Birchmeier, W. E-cadherin as a tumor (invasion) suppressor gene. BioEssays, Chem., 270: 21468 – 21471, 1995. 17: 97 – 99, 1995. 54. Jain, S. K., Langdon, W. Y., and Varticovski, L. Tyrosine phosphorylation of 30. Volberg, T., Zick, Y., Dror, R., Sabanay, I., Gilon, C., Levitzki, A., and p120cbl in BCR/abl transformed hematopoietic cells mediates enhanced Geiger, B. The effect of tyrosine-specific protein phosphorylation on the assembly association with phosphatidylinositol 3-kinase. Oncogene, 14: 2217 – 2228, 1997. of adherens-type junctions. EMBO J., 11: 1733 – 1742, 1992. 55. Tanaka, S., Amling, M., Neff, L., Peyman, A., Uhlmann, E., Levy, J. B., and 31. Zetter, B. R. Adhesion molecules in tumor metastasis. Semin. Cancer Biol., Baron, R. c-Cbl is downstream of c-Src in a signalling pathway necessary for 4: 219 – 229, 1993. bone resorption. Nature, 383: 528 – 531, 1996. 32. Miao, H., Burnett, E., Kinch, M. S., Simon, E., and Wang, B. EphA2 kinase 56. Garcia-Guzman, M., Larsen, E., and Vuori, K. The proto-oncogene c-Cbl is a associates with focal adhesion kinase and upon activation, inhibits integrin- positive regulator of Met-induced MAP kinase activation: a role for the adaptor mediated cell adhesion and migration. Nat. Cell Biol., 2: 62 – 69, 2000. protein Crk. Oncogene, 19: 4058 – 4065, 2000. 33. Pandey, A., Lazar, D. F., Saltiel, A. R., and Dixit, V. M. Activation of the Eck 57. Yoon, C. H., Lee, J., Jongeward, G. D., and Sternberg, P. W. Similarity of receptor protein tyrosine kinase stimulates phosphatidylinositol 3-kinase activity. sli-1, a regulator of vulval development in C. elegans, to the mammalian proto- J. Biol. Chem., 269: 30154 – 30157, 1994. oncogene c-cbl. Science, 269: 1102 – 1105, 1995. 34. Pandey, A., Lindberg, R. A., and Dixit, V. M. Cell signalling. Receptor 58. Wang, Y., Yeung, Y. G., Langdon, W. Y., and Stanley, E. R. c-Cbl is orphans find a family. Curr. Biol., 5: 986 – 989, 1995. transiently tyrosine-phosphorylated, ubiquitinated, and membrane-targeted following CSF-1 stimulation of macrophages. J. Biol. Chem., 271: 17 – 20, 1996. 35. Carpenter, G. and Cohen, S. Epidermal growth factor. J. Biol. Chem., 265: 7709 – 7712, 1990. 59. Kinch, M. S. and Burridge, K. Altered adhesions in ras-transformed breast epithelial cells. Biochem. Soc. Trans., 23: 446 – 450, 1995. 36. Carpenter, G. The EGF receptor: a nexus for trafficking and signaling. BioEssays, 22: 697 – 707, 2000. 60. Davis, S., Gale, N. W., Aldrich, T. H., Maisonpierre, P. C., Lhotak, V., Pawson, T., Goldfarb, M., and Yancopoulos, G. D. Ligands for EPH-related 37. Sorkin, A., Waters, C., Overholser, K. A., and Carpenter, G. Multiple receptor tyrosine kinases that require membrane attachment or clustering for autophosphorylationsitemutationsoftheepidermalgrowthfactor receptor.Analysis activity. Science, 266: 816 – 819, 1994. of kinase activity and endocytosis. J. Biol. Chem., 266: 8355 – 8362, 1991. 61. Kinch, M. S., Kilpatrick, K., and Zhong, C. Identification of tyrosine 38. Meng, W., Sawasdikosol, S., Burakoff, S. J., and Eck, M. J. Structure of the phosphorylated adhesion proteins in human cancer cells. Hybridoma, 17: 227 – amino-terminal domain of Cbl complexed to its binding site on ZAP-70 kinase. 235, 1998. Nature, 398: 84 – 90, 1999. 62. Bae, S., Arand, G., Azzam, H., Pavasant, P., Torri, J., Frandsen, T., and 39. Miyake, S., Lupher, M. L., Andoniou, C. E., Lill, N. L., Ota, S., Douillard, P., Thompson, E. Molecular and cellular analysis of basement membrane invasion by Rao, N., and Band, H. The Cbl protooncogene product: from an enigmatic human breast cancer cells in Matrigel-based in vitro assays. Breast Cancer Res. oncogene to center stage of signal transduction. Crit. Rev. Oncog., 8: 189 – 218, Treat., 24: 241 – 255, 1993. 1997. 63. Kinch, M. S., Clark, G. J., Der, C. J., and Burridge, K. Tyrosine 40. Langdon, W. Y. The cbl oncogene: a novel substrate of protein tyrosine phosphorylation regulates the adhesions of ras-transformed breast epithelia. kinases. Aust. N. Z. J. Med., 25: 859 – 864, 1995. J. Cell Biol., 130: 461 – 471, 1995. 41. Langdon, W. Y., Hartley, J. W., Klinken, S. P., Ruscetti, S. K., and Morse, H. 64. Riese, D. J. and DiMaio, D. An intact PDGF signaling pathway is required C., III. v-cbl, an oncogene from a dual-recombinant murine retrovirus that induces for efficient growth transformation of mouse C127 cells by the bovine early B-lineage lymphomas. Proc. Natl. Acad. Sci. USA, 86: 1168 – 1172, 1989. papillomavirus E5 protein. Oncogene, 10: 1431 – 1439, 1995.

Downloaded from mcr.aacrjournals.org on October 1, 2021. © 2002 American Association for Cancer Research. c-Cbl-Dependent EphA2 Protein Degradation Is Induced by Ligand Binding 11 American Cancer Society (CSM) Grant RPG-97-105-01 and the NIH Grant CA91318-01.

Jennifer Walker-Daniels, David J. Riese II and Michael S. Kinch

Mol Cancer Res 2002;1:79-87.

Updated version Access the most recent version of this article at: http://mcr.aacrjournals.org/content/1/1/79

Cited articles This article cites 63 articles, 27 of which you can access for free at: http://mcr.aacrjournals.org/content/1/1/79.full#ref-list-1

Citing articles This article has been cited by 35 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/1/1/79.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://mcr.aacrjournals.org/content/1/1/79. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.