Introduction

The ERBB2 gene encodes an M_r 185 kDa member of the ErbB family of transmembrane receptor tyrosine kinases, whose expression in certain normal tissues plays a role in regulating cell growth and differentiation (Sibilia and Wagner, 1995; Threadgill et al., 1995). In addition, aberrant activation of ErbB2 in breast and other adenocarcinomas predicts for a poor clinical outcome (Slamon et al., 1987, 1989; Gullick et al., 1991; Bacus et al., 1994). Consequently, therapeutic monoclonal antibodies or small molecules, tyrosine kinase inhibitors targeting ErbB2 and/or the epidermal growth factor receptor (EGFR or ErbB1), have been developed. For example, trastuzumab a humanized anti-ErbB2 monoclonal antibody is approved for treating the 25–30% of breast cancers that overexpress ErbB2 or demonstrate ErbB2 gene amplification (Cobleigh et al., 1999).

The regulation of ErbB receptor signaling is in large part mediated by a family of peptide ligands that bind to and activate ErbB receptors, for example, epidermal growth factor (EGF) and transforming growth factor (TGF) bind EGFR, while heregulin (HRG) binds ErbB3 and ErbB4 (Salomon et al., 1995; Reise and Stern, 1998). Ligand binding induces ErbB receptor phosphorylation with subsequent formation of either receptor homo- or heterodimers. Phosphorylation of specific tyrosine residues contained within the cytoplasmic domain of ErbB receptors establishes docking sites for the phosphotyrosine-binding domain and/or Src-homology 2 (SH2) containing proteins. The activation of these adapter proteins initiates a signaling cascade involved in regulating cell growth and survival (Luttrell et al., 1994; Levkowitz et al., 1996; Tzahar et al., 1996; Olayioye et al., 1998; Hackel et al., 1999; Klapper et al., 2000). ErbB2 is the preferred heterodimeric partner, potentiating ErbB receptor signaling (Stern and Kamps, 1988; Ullrich and Schlessinger, 1990; Wada et al., 1990; Karunagaran et al., 1996; Graus-Porta et al., 1997). Although it lacks a peptide ligand, ErbB2 is transactivated through heterodimerization.

The biological effects elicited by ErbB receptor complexes vary depending upon the peptide ligands expressed in the tumor microenvironment as well as the cellular repertoire of ErbB receptors. Certain SH2 and phosphotyrosine-binding-domain proteins preferentially associate with specific ErbB receptors. For example, ErbB3 contains multiple cytoplasmic docking sites for the p85 subunit of phosphatidylinositol 3-kinase (PI3K), a prosurvival factor involved in resistance to hormonal/chemotherapy (Cheng et al., 1992; Prigent and Gullick, 1994; Soltoff et al., 1994; Bacus et al., 2002; Vivanco and Sawyers, 2002; Yakes et al., 2002). Accordingly, ErbB3-containing heterodimers, specifically ErbB2–ErbB3 complexes, elicit potent mitogenic and prosurvival signals (Alimandi et al., 1995; Siegel et al., 1999).

Full-length ErbB2 (p185^ErbB2) undergoes proteolytic cleavage, shedding its extracellular domain (ECD), which is detectable in cell culture medium or patients' sera (Lin and Clinton, 1991; Zabrecky et al., 1991; Pupa et al., 1993). Elevated levels of ErbB2 ECD in sera of breast cancer patients correlate with a poorer response to therapy (Kandl et al., 1994; Brandt-Rauf, 1995; Yamauchi et al., 1997; Colomer et al., 2000). Truncated ErbB2 receptor (p95^ErbB2) exhibits increased autokinase activity and enhanced transforming efficiency compared with p185^ErbB2, implicating the ECD as a negative regulator of ErbB2 kinase and oncogenic activity (Di Fiore et al., 1987; Bargmann and Weinberg, 1988; Segatto et al., 1988).

p95^ErbB2 expression correlates with positive lymph node metastasis in ErbB2-overexpressing breast cancers (Christianson et al., 1998; Molina et al., 2002). It is therefore tempting to speculate that enhanced signaling through p95^ErbB2-containing receptor complexes plays a role in metastatic progression. We now show that GW572016, a reversible inhibitor of ErbB2 and EGFR tyrosine kinases (Cockerill et al., 2001), blocks p95^ErbB2 phosphorylation in breast cancer cell lines and tumor xenografts, whereas trastuzumab has no effect. In contrast to the full-length receptor, p95^ErbB2 almost exclusively associates with ErbB3 compared with EGFR, providing an explanation for the activation of p95^ErbB2 and ErbB3 by HRG. ErbB3-p95^ErbB2 heterodimer formation and subsequent activation of the PI3K-AKT prosurvival pathway provide a possible explanation for the link between p95^ErbB2 expression and metastatic progression in breast cancer. The inhi-bition of activated p95^ErbB2 by GW572016 has clinical implications for treating metastatic breast cancer and early-stage tumors that express p95^ErbB2.

Results

Inhibition of p95 by GW572016 in ErbB2 overexpressing breast cell lines and its identification as the truncated ErbB2 receptor (p95^ErbB2)

Exposure of ErbB2 overexpressing BT474 breast cancer cells to GW572016 (1 M) for 16 h not only inhibited phosphorylation of p185^ErbB2 (Figure 1, upper arrow) but also that of a 95 kDa phosphotyrosine protein (p95) (arrowhead). In addition, GW572016 inhibited p185^ErbB2 and p95 phosphorylation in S1 cells, a cell line established by single cell cloning of Hb4ac5.2 cells, a nonmalignant mammary epithelial line stably transfected with ErbB2 (Xia et al., 2002). In contrast, p95 was not identified in the EGFR-overexpressing head and neck squamous cell carcinoma line HN5 or in parental Hb4a cells, although exposure to 5 M GW572016 inhibited phosphorylation of p170^EGFR (lower arrow) and p185^ErbB2, respectively (upper arrow).

Figure 1.

Effects of GW572016 on the expression of phospho-ErbB2 and EGFR as well as a 95-kDa phosphotyrosine protein (p95) in BT474, HN5, S1, and Hb4a cell lines. Western blot analysis was performed on equal amounts of protein from whole-cell extracts using pTyr mAb. Steady-state protein levels of phosphorylated p185^ErbB2, p170^EGFR (arrows), and p95 (arrowhead) are shown. Cells were treated with vehicle alone (-) (DMSO at a final concentration of 0.1%) or GW572106 (1 or 5 M as indicated for 24 h)

Full figure and legend (48K)

Proteolytic cleavage of the ECD of p185^ErbB2 results in a phosphorylated, truncated 95 kDa ErbB2 receptor (p95^ErbB2) (Christianson et al., 1998). To determine if p95 and p95^ErbB2 are the same molecules, equal amounts of protein from the total cell extracts were analysed by Western blot using ErbB2 mAb recognizing an epitope within the ErbB2 ECD (Figure 2, lanes 1–3). Steady-state p185^ErbB2 protein levels (arrow) were unchanged in BT474 cells treated with vehicle alone (control) (lane 1), 0.5 M GW572016 (lane 2), or 10 g/ml trastuzumab (lane 3). In contrast, p95 was not identified using this particular mAb (Figure 2, lanes 1–3).

Figure 2.

Identification of p95 as the truncated ErbB2 receptor (p95^ErbB2). Exponentially growing BT474 cells were cocultured with 0.5 M GW572016 (lanes 2, 5, 8) or 10 g/ml trastuzumab (lanes 3, 6, 9) for 24 h. Equal amounts of protein were separated by SDS–PAGE and then ErbB2, p95^ErbB2, pTyr/p95^ErbB2, and pTyr/ErbB2 steady-state protein levels assessed by Western blot. Blots were probed with the following Abs: (a) ErbB2 ECD (lanes 1–3), (b) intracytoplasmic ErbB2 peptide (aa 1243–1255) (lanes 4–6), and (c) pTyr (lanes 7–9). Controls were treated with vehicle (0.1% DMSO) alone (lanes 1, 4, 7)

Full figure and legend (50K)

However, p185^ErbB2 (arrow) and p95 (arrowhead) were identified in Western blots from BT474 whole-cell extracts using ErbB2 Ab recognizing peptide 1243–1255, an intracytoplasmic ErbB2 sequence distinct from EGFR or ErbB3 (Figure 2, lanes 4–6). Thus, p95 is the truncated ErbB2 receptor, p95^ErbB2. Exposure to GW572016 (0.5 M) for 24 h inhibited tyrosine phosphorylation of both p185^ErbB2 and p95^ErbB2 in BT474 cells (lane 8). Although p95^ErbB2 steady-state protein levels were somewhat reduced in trastuzumab (10 g/ml) treated BT474 cells (lane 6), it had no effect on p95^ErbB2 phosphorylation (lane 9).

GW572016 inhibits both p95^ErbB2 and p185^ErbB2 in breast cancer xenografts

We next examined the in vivo effects of GW572016 on p95^ErbB2 in mice bearing established BT474 tumor xenografts. As shown in Figure 3a, steady-state protein levels of total p185^ErbB2 and p95^ErbB2 were unchanged in tumor xenografts from vehicle (lanes 1–3), GW572016 (lanes 4–6), or trastuzumab (lanes 7–9) treated mice.

Figure 3.

GW572016 inhibits steady-state activated p95^ErbB2 and p185^ErbB2 protein levels without affecting the total p95^ErbB2 and p185^ErbB2 protein levels. BT474 tumor xenografts were established in CD-1 nude mice. When tumors were palpable, animals were administered GW572016 (lanes 4–6), trastuzumab (lanes 7–9), or vehicle alone (lanes 1–3) (see Materials and methods). (a) The total p185^ErbB2 and p95^ErbB2 steady-state protein levels were assessed by Western blot using an Ab recognizing an intracytoplasmic peptide (aa 1243–1255) of ErbB2. (b) Activated phospho-p185^ErbB2 (pTyr/p185^ErbB2) and p95^ErbB2 (pTyr/p95^ErbB2) were assessed using pTyr mAb. (c) Phosphorylated p95^ErbB2 and p185^ErbB2 were assessed using phospho-tyrosine-specific mAb recognizing Y1248

Full figure and legend (138K)

Using an antiphosphotyrosine antibody for Western blot analysis, we show that phospho-p185^ErbB2 steady-state protein levels were inhibited in BT474 tumors isolated from GW572016-treated mice (100 mg/kg/dose) (Figure 3b, lanes 4–6). Similarly, GW572016 completely inhibited phospho-p95^ErbB2 steady-state protein levels (lanes 4–6). We also examined the effect of GW572016 on the phosphorylation state of tyrosine 1248 (Y1248), a key ErbB2 autophosphorylation site linked to downstream MAPK-Erk1/2 signaling. Using a phosphopeptide-specific antibody, we show that GW572016 inhibited Y1248 phosphorylation in p95^ErbB2 (Figure 3c, lanes 4–6), whereas trastuzumab had little effect (lanes 7–9).

The inhibition of p185^ErbB2 and p95^ErbB2 phosphorylation by GW572016 in turn abrogated downstream signaling pathways involved in regulating tumor cell growth and survival. GW572016 inhibited p-Erk1/2 and p-AKT steady-state protein levels without affecting the total steady-state protein levels of either molecule (Figure 4, lanes 4–6). In addition, cyclin D steady-state protein levels were inhibited by GW572016. Conversely, trastuzumab (100 mg/kg) had little effect on p-Erk1/2, p-AKT, or cyclin D protein (lanes 7–9). Actin steady-state protein levels demonstrate equal loading of protein.

Figure 4.

GW572016 blocks the activation of MAPK-Erk1/2 and PI3K-AKT pathways and reduces cyclin D1/2 total steady-state protein levels in BT474 xenografts. Equal amounts of protein from tumor xenograft whole-cell extracts from vehicle (lanes 1–3), GW572016 (lanes 4–6), and trastuzumab-treated animals (lanes 7–9) were separated by SDS–PAGE and steady-state protein levels of total and phosphorylated forms of Erk1/2, and AKT assessed by Western blot. Cyclin D1/2 steady-state protein levels were also determined. Steady-state protein levels of actin confirm equal loading of protein

Full figure and legend (96K)

p95^ErbB2 preferentially associates with ErbB3

BT474 cells also express EGFR and ErbB3. To determine whether truncated p95^ErbB2 forms heterodimers, whole-cell extracts were subjected to immunoprecipitation (IP) with either EGFR, ErbB2, or ErbB3 Abs followed by Western blot analysis using pTyr mAb. As shown, EGF (50 ng/ml) increased EGFR phosphorylation in BT474 cells (Figure 5, upper panel). GW572016 (1 M) markedly inhibited baseline EGFR phosphorylation and partially blocked that induced by EGF. In contrast, HRG had little effect on phosphorylation of EGFR (upper panel). Moreover, phospho-p95^ErbB2 did not coprecipitate with EGFR (upper panel). Using an Ab directed against the intracytoplasmic ErbB2 peptide (1243–1255) to immunoprecipitate both full-length and truncated receptors, we show that GW572016 inhibited phosphorylation of p185^ErbB2 and p95^ErbB2 in the presence or absence of HRG and EGF (Figure 5, middle panel).

Figure 5.

Appearance of p95^ErbB2-ErbB3 heterodimers in BT474 cells and the effect of GW572016 on EGF- or HRG-induced phosphorylation of p95^ErbB2, p185^ErbB2, and ErbB3. Cells were cultured in serum-free medium containing 1.5% BSA with or without 1 M GW572016 for 16 h and then exposed to EGFR (50 ng/ml) or HRG (5 mM) for 15 min prior to harvesting. Phospho-p95^ErbB2, p185^ErbB2, and ErbB3 were assessed by IP (as indicated in the upper, middle, and lower panels) Western blot using pTyr mAb. Quantification was performed using an Odyssey Infrared Imaging System. These results are representative of three independent experiments

Full figure and legend (101K)

Importantly, p95^ErbB2 coimmunoprecipitated with ErbB3 (Figure 5, lower panel). HRG is an ErbB3 ligand. ErbB3 phosphorylation increased 6.3-fold following stimulation with HRG, while phosphorylation of p95^ErbB2 that coprecipitated with ErbB3 increased 4.5-fold (Figure 5, lower panel). Treatment with GW572016 blocked HRG-induced phosphorylation of coimmunoprecipitated p95^ErbB2 by 50%. In contrast, EGF had no effect on ErbB3 phosphorylation or the phosphorylation of coprecipitated p95^ErbB2. HRG is also a ligand for ErbB4; however, we were unable to demonstrate ErbB4 expression in these cells (data not shown).

In vivo, treatment with GW572016 inhibited phosphorylation of p185^ErbB2 and p95^ErbB2 in BT474 tumor xenografts when compared with vehicle treatment alone (Figure 6a, upper and lower panels). We next examined BT474 tumor xenografts for the presence of ErbB3-p95^ErbB2 heterodimers in mice treated with GW572016 (100 mg/kg/dose) or vehicle alone. ErbB3 IPs were analysed by Western blot using either pTyr (Figure 6b, upper panel) or ErbB2 (peptides 1243–1253) (Figure 6b, lower panel) Abs. ErbB3-p95^ErbB2 heterodimers were identified in tumor xenografts (Figure 6b, upper and lower panels). Furthermore, GW572016 treatment inhibited ErbB3 and p95^ErbB2 tyrosine phosphorylation compared with vehicle treatment alone (Figure 6b, upper panel).

Figure 6.

p95^ErbB2 preferentially heterodimerizes with ErbB3 in BT474 xenografts; inhibition by GW572016. The activation state of EGFR, p185^ErbB2, p95^ErbB2, and ErbB3 was assessed by IP Western blot. (a) Western blot analysis of pTyr IP probed with ErbB2 mAb recognizing the cytoplasmic peptide 1243–1255 (upper panel); ErbB2 (aa 1243–1255) immunoprecipitated proteins probed with pTyr mAb (lower panel). (b) Western blot analysis of an ErbB3 IP probed with pTyr mAb (upper panel); ErbB3 IP probed with ErbB2 (aa 1243–1255) Ab (lower panel). (c) Western blot analysis of an EGFR IP probed with pTyr mAb (upper panel); EGFR IP probed with ErbB2 (aa 1243–1255) Ab (lower panel). Tumor-bearing mice were administered vehicle alone or GW572016 (see Material and methods). Quantification was performed using an Odyssey Infrared Imaging System

Full figure and legend (204K)

p95^ErbB2 did not associate with EGFR (Figure 6c, upper and lower panels). In contrast, full-length p185^ErbB2 formed heterodimers with EGFR (lower panel). Moreover, GW572016 inhibited EGFR tyrosine phosphorylation (upper panel).

The cytoplasmic domain of ErbB3 contains at least seven tyrosine residues that serve as docking sites for the p85 subunit of PI3K (Prigent and Gullick, 1994; Soltoff et al., 1994). Phosphorylation of these tyrosine residues leads to AKT phosphorylation and activation. Since p95^ErbB2 preferentially associates with ErbB3 in BT474 breast cancer cells, we next examined the effects of HRG, an ErbB3 ligand, on the MAPK-Erk1/2 and PI3K-AKT pathways. Exposing BT474 cells to EGF (50 ng/ml) for 15 min increased p-Erk1/2 steady-state protein levels, an effect blocked by GW572016 (1 M) (Figure 7). AKT phosphorylation increased approximately 40% in response to HRG (5 mM). Although GW572016 (1 M) blocked HRG induction of AKT phosphorylation above the baseline, it did not reduce p-AKT expression below constitutively phosphorylated levels (Figure 7).

Figure 7.

GW572016 inhibits activation of Erk1/2 and AKT by EGF and HRG in BT474 cells. Exponentially growing cells were treated as indicated in Material and methods. GW572016 was added for 24 h. Equal amounts of protein were analysed by Western blot for steady-state protein levels of total and phosphorylated forms of Erk1/2 and AKT. Proteins were visualized using fluorescent-labeled secondary Abs and quantified by Odyssey Infrared Imaging System. These results are representative of three independent experiments

Full figure and legend (67K)

Discussion

Despite recent advances in breast cancer therapy, nearly 40 000 people died from metastatic breast cancer in 2001. Elucidating the pathways regulating breast cancer cell growth and survival might provide additional targets for therapeutic intervention. Proteolytic cleavage of the ErbB2 ECD results in the expression of a truncated ErbB2 receptor (p95^ErbB2), which exerts potent oncogenic signals in preclinical models (Di Fiore et al., 1987; Bargmann and Weinberg, 1988; Segatto et al., 1988). Moreover, tumor expression of p95^ErbB2 has been linked to metastatic progression in ErbB2-overexpressing breast cancers (Christianson et al., 1998; Molina et al., 2002). Whereas the role of p185^ErbB2 in regulating breast cancer cell growth and survival has been extensively studied, relatively little is known about p95^ErbB2. Here, we show that p95^ErbB2 is constitutively activated in breast cancer cell lines and tumor xenografts. Since ErbB2 tyrosine phosphorylation is a biochemical marker of increased cell transformation activity, these data implicate activated, phospho-p95^ErbB2 as playing a role in metastatic progression of breast cancer. We now show that p95^ErbB2 preferentially heterodimerizes with ErbB3 in BT474 breast cancer cells and tumor xenografts. Consequently, HRG, but not EGF, stimulates p95^ErbB2 phosphorylation, which has consequences on downstream mediators of tumor growth and survival (e.g., AKT). These data identify p95^ErbB2 as a target for therapeutic intervention.

Engineered truncated ErbB2 receptors possess enhanced cell transformation activity compared with p185^ErbB2 (Di Fiore et al., 1987; Bargmann and Weinberg, 1988; Segatto et al., 1988). Deletions within the ECD of ErbB2 increase ErbB2 autokinase and transformation activities, implicating sites within the ECD as exerting repressive effects on ErbB2 activity. These findings are reminiscent of a deletion mutation within the ECD of EGFR (vIII mutation) present in a variety of malignancies and associated with constitutive, ligand-independent activation of EGFR (Wong et al., 1992). The increased oncogenic properties of p95^ErbB2 are consistent with the link between p95^ErbB2 expression and metastatic progression in ErbB2-overexpressing breast cancers.

To elucidate the link between p95^ErbB2 expression in breast cancer and metastatic progression, we investigated the regulation of p95^ErbB2 in human breast cancer cells and tumor xenografts. Since ErbB receptors signal primarily through heterodimer complexes, it was necessary to address whether p95^ErbB2 forms heterodimers, and if so, with what ErbB receptors. ErbB3 and p185^ErbB2 are frequently coexpressed in breast cancers (Lemoine et al., 1992; Rajkumar and Gullick, 1994; Alimandi et al., 1995; Siegel et al., 1999). Moreover, ErbB3-p185^ErbB2 heterodimers represent one of the most potent mitogenic, transforming receptor complexes. Coexpression of ErbB3 and p185^ErbB2, but not EGFR and ErbB3, synergizes to transform NIH3T3 cells, whereby ErbB2 transphosphorylates and activates ErbB3 (Alimandi et al., 1995). Since ErbB3 contains multiple docking sites for the SH2 domain of the p85 subunit of PI3K, ErbB3-containing heterodimers are potent activators of the PI3K-AKT growth and survival pathway (Prigent and Gullick, 1994; Soltoff et al., 1994). Consequently, cotransfection of ErbB3 and p185^ErbB2 activates the PI3K-AKT pathway. In addition, heterodimerization with p185^ErbB2 enhances the binding affinity of HRG for ErbB3. In this context, aberrant activation of the PI3K-AKT pathway in breast and other carcinomas predicts for a poorer clinical outcome (Cheng et al., 1992; Vivanco and Sawyers, 2002; Yakes et al., 2002).

Here, we show that p95^ErbB2 preferentially associates with ErbB3, whereas p185^ErbB2 heterodimerizes with both EGFR and ErbB3. This provides an explanation for the regulation of p95^ErbB2 phosphorylation by HRG, but not EGF with subsequent downstream activation of AKT. However, HRG-induced ErbB3 phosphorylation was not effectively blocked by GW572016 despite inhibition of p185^ErbB2 and p95^ErbB2 tyrosine phosphorylation (Figure 5, lower panel). Although HRG also activates ErbB4, we were unable to detect ErbB4 in these cells (data not shown). These data are consistent with a recent observation, suggesting that some of the biological effects of HRG may be mediated via a mechanism independent of ErbB3 or ErbB4 (Beerli et al., 1995).

GW572016 blocks phosphorylation of EGFR, p185^ErbB2, Erk1/2, and AKT, as well as inhibiting cyclin D protein levels in human tumor cell lines and xenografts (Rusnak et al., 2001; Xia et al., 2002). Although preliminary, GW572016 has shown antitumor activity in early-phase clinical trials in heavily pretreated subjects with metastatic cancers, notably breast cancer (manuscript in preparation). In this context, we now show that GW572016 inhibits p95^ErbB2 phosphorylation, blocking the activation of downstream Erk1/2, AKT, and cyclin D protein levels in BT474 breast cancer cells and tumor xenografts. In contrast, trastuzumab, which binds to the ECD of ErbB2, did not block p95^ErbB2 activation. In our hands, trastuzumab had little effect on steady-state protein levels of p-Erk1/2, p-AKT, and cyclin D in tumor cell lines and tumor xenografts. The mechanism(s) by which trastuzumab exerts its antitumor activity remains unknown and is likely multifactorial, mediated through both immunological effects, which are independent of its modulation of ErbB2 phosphorylation, as well as effects on signal transduction pathways. Nonetheless, it is tempting to speculate that signaling through p95^ErbB2-containing heterodimers, which would not be affected by an antibody like trastuzumab directed against the ErbB2 ECD, might contribute to trastuzumab resistance, which is most likely a multifactorial process.

p95^ErbB2 has been implicated in the progression of ErbB2-overexpressing breast cancers. Cleavage of the ECD of ErbB2 appears to be mediated by an MMP family member (Codony-Servat et al., 1999). Increased MMP activity stimulates tumor angiogenesis and degradation of the extracellular matrix, both key events for metastases (Stamenkovic, 2000). During metastatic progression, increased MMP activity may explain the increased cleavage of p185^ErbB2. In addition, ErbB3-p95^ErbB2 heterodimers potently activate the PI3K-AKT pathway, the latter being implicated in hormonal/chemotherapy resistance. This explanation is particularly relevant to breast cancer where coexpression of ErbB3 and ErbB2 occurs frequently. However, the nature of the signal generated by ErbB receptor complexes is not only dependent upon the repertoire of ErbB receptors expressed by tumors but also the ErbB receptor ligands expressed in the tumor microenvironment. HRG, an ErbB3/ErbB4 ligand, is expressed in the microenvironment of many tumors, notably breast cancer (Tsai et al., 2003). Our data suggest that HRG, but not EGF, activates p95^ErbB2-ErbB3 heterodimers, which has downstream effects on AKT activation and clinical implications related to disease progression and resistance to therapy.

In the future, cancer therapy will be based on molecular tumor profiles rather than histology. Elucidating the biological effects of targeted therapies will help identify those factors regulating tumor cell growth and survival, thereby facilitating identification of likely responders and providing a rationale for therapeutic strategies to overcome resistance. Although trastuzumab may reduce ErbB2 cleavage by binding to the ECD (Molina et al., 2001), in our hands it did not block p95^ErbB2 activation or downstream signaling. Resistance to trastuzumab therapy may therefore be partly mediated through increased expression of p95^ErbB2 during disease progression. One potential clinical advantage of GW572016 over trastuzumab is that the former potently inhibits not only the activation of EGFR and p185^ErbB2 but also that of p95^ErbB2. Identification of p95^ErbB2-ErbB3 heterodimers and their activation by HRG implicate this receptor complex as a mediator of prosurvival signals and a target for therapeutic intervention. Hence, the expression of p95^ErbB2 may identify patients more likely to benefit from GW572016 therapy compared with trastuzumab. Combining trastuzumab with GW572016 is an appealing therapeutic strategy in breast cancer since it may abrogate p95^ErbB2 signaling by simultaneously inhibiting ErbB2 cleavage and blocking p95^ErbB2 autophosphorylation.

Materials and methods

Materials

HN5, an EGFR-overexpressing LICR-LON-HN5 head and neck carcinoma cell line, was kindly provided by Helmout Modjtahedi at the Institute of Cancer Research, Surrey, UK. The ErbB2 overexpressing human breast adenocarcinoma cell line, BT474, was obtained from the American Type Culture Collection (Manassas, VA, USA). HB4a cells are derived from human mammary luminal tissue; ErbB2 transfection of parental HB4a cells yielded the HB4a C5.2 cell line (Harris et al., 1999). S1 cells, which express elevated levels of p-ErbB2, were established by subcloning HB4a C5.2 (Xia et al., 2002). EGF was purchased from Sigma Chemical (St Louis, MO, USA). Recombinant human NRG-1-B1/HRGB1 EGFR domain (HRG) was obtained from RD system (Minneapolis, MN, USA). Antiphosphotyrosine antibody was purchased from Sigma and Upstate (Lake Placid, NY, USA). Anti-EGFR (Ab-12) and anti-c-ErbB2 (Ab-11) mAbs were obtained from Neo Markers (Union City, CA, USA). Anti-ErbB2 (AA1243-1255), antiphospho-ErbB2 (Y1248), and anticyclin D1/2 were obtained from Upstate. Antiphospho-AKT (Ser 437) mAb was obtained from Cell Signaling Technology, Inc. (Beverly, MA, USA). Anti-ErbB3 (C17), anti-AKT1/2, antiphospho-Erk1/2, anti-Erk1, and anti-Erk2 antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Trastuzumab was purchased from Genentech, Inc. (South San Francisco, CA, USA). SuperSignal West Femto Maximum Sensitivity Substrate was obtained from Pierce (Rockford, IL, USA). Protein G agarose was purchased from Boehringer Mannheim (Germany). IRDye800 Conjugated Affinity Purified Anti-Rabbit IgG and anti-Mouse IgG were obtained from Rockland (Gilbertsville, PA, USA). Alexa Fluor680 goat anti-rabbit IgG was obtained from Molecular Probes (Eugene, OR, USA). GW572016, N-{3-Chloro-4-[(3-fluorobenzyl)oxy]phenyl}-6-[5-({[2-(methylsulfonyl)ethyl]amino}methyl)-2-furyl]-4-quinazolinamine, was synthesized as previously described (Cockerill et al., 2001). GW572016 for cell culture work was dissolved in DMSO.

Cell cultures

BT474 cells were cultured in RPMI 1640 supplemented with L-glutamine, 10% FBS (Hyclone), and 5 /ml insulin. HB4a cells were cultured under identical conditions to BT474 cells, in addition, with 10 g/ml hydrocortisone. S1 cells were cultured in RPMI 1640 supplemented with L-glutamine, 10% FBS, and 50 g/ml hygromycin. HN5 cells were cultured in DMEM supplemented with high glucose and 10% fetal bovine serum (FBS). All cell cultures were maintained in a humidified atmosphere of 5% CO₂ at 37°C.

EGF and HRG stimulation experiments

Cells were seeded at low density in serum-free medium supplemented with 1.5%. BSA, and then exposed for 6–24 h to GW572016 at various concentrations indicated in the figure legends, or 10 g/ml trastuzumab. Cells were stimulated with 50 ng/ml EGF or 5 nM HRG for 15 min, harvested on ice, and then lysed in RIPA buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 0.25% (w/v) deoxycholate, 1% NP-40, 5 mM sodium orthovanadate, 2 mM sodium fluoride, and a protease inhibitor cocktail).

IP and Western blots

IPs and Western blots were performed as previously described (Xia et al., 2002). Briefly, protein concentrations were determined using a modification of the Bradford method (Bio-Rad Laboratory), and equal amounts of protein subjected to IP and Western blot. Efficiency and equal loading of proteins were evaluated by Ponceau S staining. Membranes were blocked for 1 h in TBS (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 2.7 mM KCl) containing 4% (w/v) low-fat milk or 3% BSA (w/v). Membranes were then probed with specific antibodies recognizing target proteins, which were visualized with the SuperSignal West Femto Maximum sensitivity substrate kit (Pierce). Other blots were visualized and quantification performed using the Odyssey Infrared Imaging System (LI-COR, Inc., Lincoln, NE, USA). For the Odyssey, membranes were incubated with fluorescent-labeled secondary antibody at 1 : 10 000 dilution with 3% BSA in PBS for 60-min protected from light. After washing in PBS +0.1% Tween-20, the membranes were scanned using an Odyssey Infrared Imaging System.

Tumor xenografts

BT474 tumors were maintained by serial passage of fragments into female C.B-17 SCID mice, for up to 10 passages. Once tumor implants were palpable, mice were administered either vehicle (0.5% hydroxypropylmethylcellulose/0.1% Tween 80) given orally (p.o.), five doses of GW572016 at 100 mg/kg (p.o.) twice daily at 8 h intervals, or trastuzumab at 100 mg/kg given intraperitoneally (i.p.) daily for 3 days. Tumors were removed 4 h after the last dose, frozen in liquid nitrogen, and stored at -80°C until analysis. For the terminal biopsy, mice were euthanized with CO₂ inhalation. Cell extracts were prepared by homogenization in RIPA buffer at 4°C.

References

1. Alimandi M, Romano A, Curia MC, Muraro R, Fedi P, Aaronson SA, Di Fiore PP and Kraus MH. (1995). Oncogene, 10, 1813–1821. | PubMed | ISI | ChemPort |
2. Bacus SS, Altomare DA, Lyass L, Chin DM, Farrell MP, Gurova K, Gudkov A and Testa JR. (2002). Oncogene, 21, 3532–3540. | Article | PubMed | ISI | ChemPort |
3. Bacus SS, Zelnick CR, Plowman G and Yarden Y. (1994). Am. J. Clin. Pathol., 102, S13–S24. | PubMed | ISI | ChemPort |
4. Bargmann CI and Weinberg RA. (1988). EMBO J., 7, 2043–2052. | PubMed | ISI | ChemPort |
5. Beerli RR, Graus-Porta D, Woods-Cook K, Chen X, Yarden Y and Hynes NE. (1995). Mol. Cell. Biol., 15, 6496–6505. | PubMed | ISI | ChemPort |
6. Brandt-Rauf PW. (1995). Mutat. Res., 333, 203–208. | PubMed |
7. Cheng J, Godwin AK, Bellacosa A, Taguchi T, Franke TF, Hamilton TC, Tsichlis PN and Testa JR. (1992). Proc. Natl. Acad. Sci. USA, 89, 9267–9271. | Article | PubMed | ChemPort |
8. Christianson TA, Doherty JK, Lin YJ, Ramsey EE, Holmes R, Keenan EJ and Clinton GM. (1998). Cancer Res., 58, 5123–5129. | PubMed | ISI | ChemPort |
9. Cobleigh MA, Vogel CL, Tripathy D, Robert NJ, Scholl S, Fehrenbacher L, Wolter J, Paton V, Shak S, Lieberman G and Slamon DJ. (1999). J. Clin. Oncol., 17, 2639–2648. | PubMed | ISI | ChemPort |
10. Cockerill S, Stubberfield C, Stables J, Carter M, Guntrip S, Smith K, McKeown S, Shaw R, Topley P, Thomsen L, Affleck K, Jowett A, Hayes D, Willson M, Woollard P and Spalding D. (2001). Bioorg. Med. Chem. Lett., 11, 1401–1405. | Article | PubMed | ChemPort |
11. Codony J, Albanell J, Lopez-Talavera JC, Arribas J and Baselga J. (1999). Cancer Res., 59, 1196–1201. | PubMed | ISI | ChemPort |
12. Colomer R, Montero S, Lluch A, Ojeda B, Barnadas A, Casado A, Massuti A, Cortes-Funes J and Lloveras B. (2000). Clin. Cancer Res., 6, 2356–2362. | PubMed | ChemPort |
13. Di Fiore PP, Pierce JH, Kraus MH, Segatto O, King CR and Aaronson SA. (1987). Science, 237, 178–182. | Article | PubMed | ChemPort |
14. Graus-Porta D, Beerli RR, Daly JM and Hynes NE. (1997). EMBO J., 16, 1647–1655. | Article | PubMed | ChemPort |
15. Gullick WJ, Love SB, Wright C, Barnes DM, Gusterson B, Harris AJ and Altman DG. (1991). Br. J. Cancer, 63, 434–438. | PubMed | ISI | ChemPort |
16. Hackel PO, Zwick E, Prenzel N and Ullrich A. (1999). Curr. Opin. Cell Biol., 11, 184–189. | Article | PubMed | ISI | ChemPort |
17. Harris RA, Eichholtz TJ, Hiles JD, Page MT and O'Hare MJ. (1999). Int. J. Cancer, 80, 477–484. | Article | PubMed | ISI | ChemPort |
18. Kandl H, Seymour L and Bezwoda WR. (1994). Br. J. Cancer, 70, 739–742. | PubMed | ISI | ChemPort |
19. Karunagaran D, Tzahar E, Beerli RR, Chen X, Graus-Porta D, Ratzkin BJ, Seger R, Hynes NE and Yarden Y. (1996). EMBO J., 15, 254–264. | PubMed | ISI | ChemPort |
20. Klapper LN, Kirschbaum MH, Sela M and Yarden Y. (2000). Adv. Cancer Res., 77, 25–79. | PubMed | ISI | ChemPort |
21. Lemoine NR, Barnes DM, Hollywood DP, Hughes CM, Smith P, Dublin E, Prigent SA, Gullick WJ and Hurst HC. (1992). Br. J. Cancer, 66, 1116–1121. | PubMed | ISI | ChemPort |
22. Levkowitz G, Klapper LN, Tzahar E, Freywald A, Sela M and Yarden Y. (1996). Oncogene, 12, 1117–1125. | PubMed | ISI | ChemPort |
23. Lin YJ and Clinton GM. (1991). Oncogene, 6, 639–643. | PubMed | ChemPort |
24. Luttrell DK, Lee TJ, Lansing TJ, Crosby RM, Jung KD, Willard D, Luther M, Rodriguez J, Berman J and Gilmer TM. (1994). Proc. Natl. Acad. Sci. USA, 91, 83–87. | Article | PubMed | ChemPort |
25. Molina MA, Codony-Servat J, Albanell J, Rojo F, Arribas J and Baselga J. (2001). Cancer Res., 61, 4744–4749. | PubMed | ISI | ChemPort |
26. Molina MA, Saez R, Ramsey EE, Garcia-Barchino M-J, Rojo F, Evans AJ, Albanell J, Keenan EJ, Lluch A, Garcia-Conde J, Baselga J and Clinton GM. (2002). Clin. Cancer Res., 8, 347–353. | PubMed | ISI | ChemPort |
27. Olayioye MA, Graus-Porta D, Beerli RR, Rohrer J, Gay B and Hynes NE. (1998). Mol. Cell. Biol., 18, 5042–5051. | PubMed | ISI | ChemPort |
28. Prigent SA and Gullick WJ. (1994). EMBO J., 13, 2831–2841. | PubMed | ISI | ChemPort |
29. Pupa SM, Crepaldi L, Gandino S, Longati P and Comoglio P. (1993). Oncogene, 8, 2917–2923. | PubMed | ISI | ChemPort |
30. Rajikumar T and Gullick WJ. (1994). Breast Cancer Res. Treat., 29, 3–9. | PubMed |
31. Reise DJ and Stern DF. (1998). Bioessays, 20, 41–48. | Article | PubMed | ISI |
32. Rusnak DW, Lackey K, Affleck K, Wood ER, Alligood KJ, Rhodes N, Keith BR, Murray DM, Glennon K, Knight WB, Mullin RJ and Gilmer TM. (2001). Mol. Cancer Ther., 1, 85–94. | PubMed | ISI | ChemPort |
33. Salomon DD, Brandt R, Ciardiello F and Normanno N. (1995). Crit. Rev. Oncol. Hematol., 19, 183–232. | Article | PubMed | ISI | ChemPort |
34. Segatto O, King CR, Pierce JH, DiFiore PP and Aaronson SA. (1988). Mol. Cell. Biol., 8, 5570–5574. | PubMed | ISI | ChemPort |
35. Sibilia M and Wagner EF. (1995). Science, 269, 234–238. | Article | PubMed | ISI | ChemPort |
36. Siegel PM, Ryan ED, Cardiff RD and Muller WJ. (1999). EMBO J., 18, 2149–2164. | Article | PubMed | ISI | ChemPort |
37. Slamon DJ, Clark SG, Wong WJ, Levin A, Ullrich A and McGuire WL. (1987). Science, 235, 177–182. | Article | PubMed | ISI | ChemPort |
38. Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DE, Levin WJ, Stuart SG, Udove J, Ullrich A and Press MF. (1989). Science, 244, 707–712. | Article | PubMed | ISI | ChemPort |
39. Soltoff SP, Carraway III KL, Prigent SA, Gullick WJ and Cantley LC. (1994). Mol. Cell. Biol., 14, 3550–3558. | PubMed | ISI | ChemPort |
40. Stamenkovic I. (2000). Semin. Cancer Biol., 10, 415–433. | Article | PubMed | ISI | ChemPort |
41. Stern DF and Kamps MP. (1988). EMBO J., 7, 995–1001. | PubMed | ISI | ChemPort |
42. Threadgill DW, Dlugosz AA, Hansen LA, Tennenbaum T, Lichti U, Yee D, Lamantia C, Mourton T, Herrup K, Harris RC, Barnard JA, Yuspa SH, Coffey RJ and Magnuson T. (1995). Science, 269, 230–234. | Article | PubMed | ISI | ChemPort |
43. Tsai MS, Shamon-Taylor LA, Mehmi I, Tang CK and Lupu R. (2003). Oncogene, 22, 761–768. | Article | PubMed | ISI | ChemPort |
44. Tzahar E, Waterman H, Chen X, Levkowitz G, Karunagaran D, Lavi S, Ratzkin BJ and Yarden Y. (1996). Mol. Cell. Biol., 16, 5276–5287. | PubMed | ISI | ChemPort |
45. Ullrich A and Schlessinger J. (1990). Cell, 61, 203–212. | Article | PubMed | ISI | ChemPort |
46. Vivanco I and Sawyers CL. (2002). Nat Rev/Cancer, 2, 489–501.
47. Wada T, Qian XL and Greene MI. (1990). Cell, 61, 1339–1347. | Article | PubMed | ISI | ChemPort |
48. Wong AJ, Ruppert JM, Bigner SH, Grzeschik CH, Humphrey PA, Bigner DS and Vogelstein B. (1992). Proc. Natl. Acad. Sci. USA, 89, 2965–2969. | Article | PubMed | ChemPort |
49. Xia W, Mullin RJ, Keith BR, Liu L-H, Ma H, Rusnak DW, Owens G, Alligood KJ and Spector NL. (2002). Oncogene, 21, 6255–6263. | Article | PubMed | ISI | ChemPort |
50. Yakes FM, Chinratanalab W, Ritter CA, King W, Seelig S and Arteaga CL. (2002). Cancer Res., 62, 4132–4141. | PubMed | ISI | ChemPort |
51. Yamauchi H, O'Neill A, Gelman R, Carney W, Tenney DY, Hosch S and Hayes DF. (1997). J. Clin. Oncol., 15, 2518–2525. | PubMed | ISI | ChemPort |
52. Zabrecky JR, Lam T, McKenzie SJ and Carney W. (1991). J. Biol. Chem., 266, 1716–1720. | PubMed | ISI | ChemPort |

Acknowledgements

We thank Barry Keith, Robert Mullin, and James Onori for providing assistance with tumor xenografts. We also thank Arthur Moseley, Wanda Bodnar, Mary Moyer, Bob Hollingsworth, and Roderick Davis for their assistance in helping to identify p95 as well as Tona Gilmer and Allen Oliff for helpful discussions.