SEARCH

SEARCH BY CITATION

Keywords:

  • ErbB2;
  • endocytosis;
  • Herceptin;
  • Akt;
  • PTEN

Abstract

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The anti-proliferative effect of the ErbB2 specific antibody Herceptin in cells overexpressing ErbB2 has previously been explained by endocytic downregulation of ErbB2. However, in the following, we demonstrate that Herceptin inhibited proliferation of ErbB2 overexpressing cells without downregulating ErbB2. Herceptin did also not induce endocytosis of ErbB2. Herceptin was found to blunt proliferation of SKBr3 cells overexpressing EGFR, ErbB2, and ErbB3 and expressing functional PTEN, probably by recruiting PTEN to the plasma membrane. Akt was found to be constitutively phosphorylated both in SKBr3 cells overexpressing EGFR, ErbB2 and ErbB3, and in SKOv3 cells, overexpressing EGFR and ErbB2. However, phosphorylation of Akt was inhibited by Herceptin only in SKBr3 cells. SKOv3 cells, which lack the tumour suppressor protein Ras homolog member I, was found to have constitutively phosphorylated mitogen activated protein kinase and functionally increased Ras activity. SKOv3 cells further had low expression levels of PTEN. We thus confirm that the anti-proliferative effect of Herceptin in SKBr3 cells is due to recruitment of PTEN to the plasma membrane and conclude that Herceptin does not blunt phosphatidyl inositol 3 kinase-induced growth in cells with constitutive Ras activity. We further conclude that endocytic downregulation of ErbB2 does not contribute to Herceptin's antiproliferative effect. © 2005 Wiley-Liss, Inc.

ErbB2 (HER2/Neu) is a member of the epidermal growth factor (EGF) receptor (EGFR) family of receptor tyrosine kinases, also called the ErbB family. The ErbB family includes 4 members, namely, the EGFR (ErbB1), ErbB2, ErbB3 and ErbB4, which all dimerize with each other. Dimerized receptors activate signaling pathways involved in the regulation of cellular proliferation and apoptosis (recently reviewed in 1,2). Altered expression of ErbB proteins has been associated with early transformation of mammary epithelial cells.3, 4 ErbB2/ErbB3 heterodimers have been demonstrated to be potent in mitogenic signaling5, 6, 7 and have been suggested to function as oncogenic units responsible for driving breast tumor cell proliferation by activating the phosphatidyl inositol 3 kinase (PI3 K)/Akt pathway.8 ErbB3 has docking sites for the p85 subunit of PI3 K,9, 10 and targeting ErbB2 has been demonstrated to decrease activation of the PI3 K/Akt pathway.11, 12, 13, 14

ErbB2 is frequently overexpressed in epithelial cancers, and the overexpression is associated with poor clinical outcome (reviewed in 15). ErbB2 is therefore an attractive target for immunotherapy. Accordingly, Herceptin, a recombinant humanized version of the anti-ErbB2 monoclonal antibody 4D5, is currently used in treatment of patients with breast cancer overexpressing ErbB2. Herceptin and 4D5 inhibit proliferation in cells overexpressing ErbB2 in vitro12, 14, 16, 17, 18, 19, 20 and have been demonstrated to accumulate cells in the G1-phase of the cell cycle,12, 21 possibly as a result of reduced Akt phosphorylation. Inhibited phosphorylation of Akt increases the level of p27 and cyclin E/Cdk2 complexes in the nucleus and thereby reduces Cdk2 activity.11, 12, 14, 21 Herceptin has additionally been demonstrated to downregulate ErbB2 expression at the plasma membrane22, 23 and to increase the turnover of ErbB2.17, 18 This has been suggested to be the result of Herceptin-induced endocytosis of ErbB2.17, 19, 23 As reducing the level of ErbB2 at the plasma membrane will reduce the signaling from ErbB2, Herceptin-induced endocytosis and downregulation of ErbB2 has been considered an important mechanism whereby Herceptin downregulates ErbB2-induced proliferation.20, 21 It has also been suggested that Herceptin can reduce the signaling from ErbB2 by directly preventing dimerization of ErbB2 with other members of the ErbB family,24 and that Herceptin inhibits metalloprotease cleavage of ErbB2.25

Our current findings demonstrate that even though Herceptin inhibits proliferation of SKBr3 cells, neither ErbB2 nor Herceptin is endocytosed in SKBr3, SKOv3 or Hep2 cells. Our data further indicate that the anti-proliferative effect of Herceptin is not caused by downregulation of ErbB2. PTEN (phosphatase, tensin homologue, deleted on chromosome TEN), a major lipid phosphatase inhibiting PI3 K/Akt signaling pathways, was recently demonstrated to be recruited to the plasma membrane when cells overexpressing ErbB2 were incubated with Herceptin.26 We confirm that Herceptin inhibits constitutive phosphorylation of Akt and recruits PTEN to the plasma membrane in SKBr3 cells. Our results further demonstrate that in SKOv3 cells, where Herceptin does not induce dephosphorylation of Akt, the level of PTEN is reduced. Moreover, the tumor suppressor protein Ras homolog member I (ARHI) has been found to be lacking in these cells.27 Consistently, we found that Ras activity was functionally high in SKOv3 cells. We therefore conclude that the anti-proliferative effect of Herceptin depends on the ability to interfere with increased PI3 K signaling, and we argue that Herceptin-induced inhibition of cell proliferation is not caused by downregulation of ErbB2 upon endocytosis.

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Materials

Human recombinant EGF was from Bachem Feinchemikalien AG (Budendorf, Switzerland), Na125I and 3H-thymidine were from Amersham Biosciences (Buckinghamshire, UK). Geldanamycin (GA) was from Merck Biosciences Ltd.(Nottingham, UK). Ras activation assay kit was from Upstate (Lake Placid, NY). Other chemicals were from Sigma Chemical Co. (St. Louis, MO), unless otherwise noted.

Plasmids

pcDNA3.1-ErbB2 was generated by PCR amplification of full length ErbB2 from the pRK5-Her2-GFP plasmid (a gift from A. Chantry, University of East Anglia, Norwich, UK) using gene-specific primers 5′-AGA AGC TTC ACA CTG GCA CGT CCA GAC CCA G-3′ and 5′-AGG CTA GCC GCA GTG AGC ACC ATG G-3′ (Invitrogen, Paisley, UK) with restriction sites for Nhe I and Hind III included. The PCR product was directly cloned into the pCR-Blunt II-TOPO (Invitrogen). A positive clone was digested with Nhe I and Hind III and ligated into the respective sites of pcDNA3.1/Zeo (Invitrogen).

Stable transfection of cells

Porcine aortic endothelial (PAE) cells and PAE.B2 cells were stably transfected with the plasmid pcDNA3.1-ErbB2. Stably transfected cells were established using FuGENE 6 transfection reagent, standard single-cell cloning procedures and Zeocin selection (30 μg/ml). The resulting clones of PAE.ErbB2 and of PAE. B2.ErbB2 will be described in more detail elsewhere.

Cell culture and treatment

The laryngeal carcinoma cell line Hep2 was provided by K. Sandvig, The Norwegian Radium Hospital, Oslo, Norway, and the mammary carcinoma cell line SKBr3, as well as the ovarian carcinoma cell line SKOv3, were from ATCC (Borås, Sweden). Hep2, SKBr3 and SKOv3 cells were grown in Dublecco's modified Eagle's medium (DMEM) (BioWhittaker, Walkersville, MD) containing penicillin-streptomycin mixture (0.5×) (BioWhittaker), L-glutamine (2 mM) (BioWhittaker), and fetal bovine serum (FBS) (5% vol/vol Hep2 cells, 15% vol/vol SKBr3 and SKOv3 cells) (PAA Innovations, Linz, Austria). PAE cells were obtained from C.-H. Heldin, Ludwig Institute for Cancer Research, Biomedical Centre, Uppsala, Sweden, and stably transfected PAE cells expressing wt EGFR (PAE.B2) were obtained from A. Sorkin, University of Colorado Health Sciences Center, Denver, CO. The cells were grown in Ham's F12 (BioWhittaker) supplemented with 10% (v/v) FBS (PAA Innovations) and 0.5× penicillin-streptomycin mixture (BioWhittaker), and PAE.B2 cells were additionally incubated with 400 μg/ml G418 sulfate (Invitrogen). All cells were grown in dishes 2 or 3 days prior to experiments to reach 80% confluency. During the experiments, cells were incubated with the indicated compounds either in minimal essential medium (MEM) (Gibco, Life Technologies, Paisley, UK) with 0.1% bovine serum albumine (BSA) at 37°C or in DMEM if the experiment lasted more than 5 hr.

Antibodies

Herceptin® was from Genentech, Inc. (San Francisco, CA). Rabbit anti-ErbB2 (intracellular domain), mouse anti-ErbB2 (extracellular domain) and mouse anti-transferrin receptor (TfR) antibodies were from Zymed Laboratories, Inc. (San Francisco, CA). Mouse anti-ErbB2 (intracellular domain, Ab-8) and mouse anti-EGFR (Ab-3) antibodies were from Neomarkers (Fremont, CA). Rabbit anti-phospho ErbB2 (pTyr1248), mouse anti-phospho EGFR (pTyr1173) and rabbit anti-p85 antibodies were from Upstate Biotechnology (Lake Placid, NY). Rabbit anti-ErbB3, rabbit anti-mitogen activated protein kinase (MAPK) and mouse anti-PTEN antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-phospho Akt (Ser 473) and rabbit anti-phospho-p44/p42 MAPK (Thr202/Tyr204) antibodies were from Cell Signaling Technology (Beverly, MA). Sheep anti-EGFR antibody was from Fitzgerald Industries International, Inc. (Concord, MA) and mouse anti-Grb2 and mouse anti-EEA1 antibodies were from Transduction Laboratories (Lexington, UK). Peroxidase-conjugated donkey anti-mouse, peroxidase-conjugated donkey anti-sheep, peroxidase-conjugated donkey anti-rabbit, Rhodamine Red-X-conjugated donkey anti-rabbit, Cy5-conjugated donkey anti-mouse and Phycoerythrin-conjugated goat anti-mouse antibodies were from Jackson Immuno Research Laboratories (West Grove, PA). Peroxidase-conjugated goat anti-rabbit antibody was from Sigma Chemical Co. and Alexa 488-conjugated goat anti-mouse antibody was from Molecular Probes, Inc. (Eugene, OR).

Measurement of DNA synthesis

Cells were incubated with or without Herceptin, as described in legend to Figure 1, prior to incubation with 3H-thymidine (1 μCi/ml) for 6 hr at 37°C. The medium was removed, and the cells were incubated twice with trichloroacetic acid (5% v/v) for 10 min at room temperature. The precipitated DNA was solubilized with 0.1 M KOH. Ultima gold scintillation fluid (Packard Instrument B.V., Groningen, The Netherlands) was added before liquid scintillation counting in a Packard (Tri-Carb 2300 TR) beta counter. The mean of 3 independent measurements of counts per min (cpm) in cells not incubated with Herceptin was defined as 100% (control). Other cpm values were converted to percent compared to the control.

thumbnail image

Figure 1. Herceptin inhibits proliferation of SKBr3 cells but not of SKOv3 and Hep2 cells. (a) SKBr3, SKOv3 and Hep2 cells were incubated with Herceptin (15 μg/ml) at 37°C for the indicated time periods before incubation for 6 hr with 3H-thymidine (1 μCi/ml). The TCA-precipitable radioactivity was measured as described in Material and Methods. The data represent the mean of 3 independent experiments with 3 parallels ± SEM. (b) Cell lysates from SKBr3, SKOv3 and Hep2 cells were subjected to SDS-PAGE and immunoblotting with sheep anti-EGFR antibody, mouse anti-ErbB2 antibody and rabbit anti-ErbB3 antibody.

Download figure to PowerPoint

Immunoblotting

Cells were lysed in lysis buffer (10 mM Tris-HCl (pH 6.8), 5 mM EDTA (Merck KGaA, Darmstadt, Germany), 50 mM NaF, 30 mM sodium pyrophosphate, sodium dodecyl sulphate (SDS) (2% wt/vol) (AppliChem GmbH, Darmstadt, Germany), 1 mM Na3VO4 (Strem Chemicals, Inc., Newburyport, MA) and 1 mM phenyl methyl sulphonyl fluoride (PMSF) (AppliChem) on ice for 10 min. Then a buffer containing β-mercaptoethanol (4% vol/vol), glycerol (4% vol/vol) and bromophenolblue (0.005% wt/vol) was added. The cell lysates were incubated at 95°C for 5 min, centrifuged at 14,000g for 2 min and the supernatant fractions were subsequently subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE). After being electrotransferred to nitrocellulose membranes, the reactive proteins were detected using an enhanced chemiluminescence method (ECL; Amersham Biosciences).

Immunoprecipitation

Cells were lysed in immunoprecipitation buffer (10 mM EDTA, 1% Triton X-100, 10 mM NaF, 1 mM PMSF,1 mM Na3VO4 (Strem Chemicals, Inc.), 20 μg/ml Leupeptin, 10 μg/ml Aprotinin and 1 mM N-Ethylmaleimide) for 10 min on ice. Protein A- or Protein G-coupled sepharose beads (Amersham Biosciences) were washed twice with Tris-HCl buffer (50 mM, pH 7.0), incubated with antibody for 1 hr at room temperature and subsequently washed 4 times with immunoprecipitation buffer, before the cell lysates were added. Immunoprecipitation was performed at 4°C for 1 hr, and the precipitates were washed 3 times with immunoprecipitation buffer and once with phosphate-buffered saline (PBS, 1:10). Sample buffer [20 mM Tris-HCl (pH 6,8), 10 mM EDTA (Merck), 100 mM NaF, 60 mM sodium pyrophosphate, 4% SDS (Applichem), 2% β-Mercaptoethanol, 20% Glycerol and 0.006% bromophenolblue] was added to the precipitates and the samples were boiled for 5 min, subjected to SDS-PAGE and immunoblotting, as described. To investigate noncomplexed receptors, cells were lysed in SDS (1%), incubated at 100°C for 5 min and chilled on ice before homogenization using a QIA-shredder column (Qiagen, Inc., Valencia, CA). The lysates were added to Protein G- or Protein A- coupled magnetic beads (Dynal Biotech ASA, Oslo, Norway) precoupled to EGFR or ErbB2 (essentially as above). After washing, the beads were dissolved in 2× immunoprecipitation (IP) buffer (2% (v/v) Triton X-100, 0.5% (w/v) sodium deoxycholate, 1% (w/v) BSA, 2 mM EDTA, 40 mM NaF, 2 mM PMSF, 4 mM Na3VO4, 40 μg/ml leupeptin, 20 μg/ml aprotinin and 2 mM NEM). Antibody-coupled magnetic beads and cell lysates were gently mixed for 1 hr at 4°C. The beads were then washed in 1× immunoprecipitation buffer (50% 2× IP buffer + 50% 1% SDS in PBS), eluted in sample buffer as above and eventually subjected to SDS-PAGE and immunoblotting as described.

Immunocytochemistry and confocal microscopy

The cells were seeded on 12 mm coverslips (Menzel-Gläser®, Braunschweig, Germany) in 24-well plates. After incubation with indicated compounds, cells were washed once in PBS and subsequently fixed with paraformaldehyde (4% wt/vol) (Riedel-de Haën AG, Seelze-Hannover, Germany) in Soerensen's phosphate buffer28 for 20 min on ice. Cells were washed 3 times with PBS prior to permeabilization with Triton X-100 (0.1% vol/vol in PBS) for 10 min. The fixed and permeabilized cells were preincubated with BSA (1% wt/vol in PBS) for 30 min prior to incubation with primary antibodies for 1 hr, washed with PBS and incubated with secondary antibodies for 30 min. The coverslips were mounted using Dako fluorescent mounting medium (Dako, Copenhagen, Denmark), and the cells were examined using a Leica TSC XP confocal microscope (Leica Microsystems AG, Wetzlar, Germany).

Flow cytometry

Cells were trypsinized and washed twice with PBS containing FBS (2%) and EDTA (2 mM, pH 7.4) prior to fixation with paraformaldehyde (4% wt/vol) for 15 min on ice. The fixed cells were incubated with primary antibodies for 30 min, washed twice with PBS with FBS and EDTA and incubated with secondary antibodies for 30 min. The cells were finally resuspended in PBS with FBS and EDTA and analyzed using a FACS calibur flow cytometer (BD Biosciences IS, San Jose, CA). To compare cell growth in cells expressing different levels of ErbB2 and EGFR, flow cytometry was performed essentially as described.29 Data were analyzed using MultiPlus AV (Phoenix Flow System, San Diego, CA).

125I-Herceptin internalization experiments

Herceptin was dialyzed (Spectra/Phore®1,2,3,4 and 5 molecular porous Dialysis membrane, Spectrum Medical Industries, Inc., Laguna Hills, CA) and labeled with Na125I (0.5 mCi) by incubation with Iodogen (Pierce, Rockford, IL).19125I-Herceptin was separated from the free iodide using a Shepadex G-25 column (PD-10 G25M from Amersham Biosciences). After incubation with 125I-Herceptin at 37°C, the cells were placed on ice and washed 3 times with ice-cold PBS and then twice with ice-cold PBS-azide buffer (PBS containing 0.1% sodium azide (Fluka Chemie AG, Buchs, Switzerland) and 1% BSA). Surface-bound Herceptin was removed by treating the cells with acetate stripping buffer [0.5 M NaCl, 0.5 M acetic acid, (pH 2.5) (Merck)] for 30 min on ice. PBS-azide buffer was added, and the released radioactivity was subsequently measured in a gamma counter (Wallac, 1470 Wizard, Turku, Finland). The cells were lysed by incubation with SDS (1% wt/vol) (Applichem) for 10 min at room temperature, and the acid resistant cell bound 125I-Herceptin was counted in a gamma counter.

Separation of membrane bound and cytosolic proteins by ultracentrifugation

SKBr3 and SKOv3 cells were grown in T75 flasks until 70% confluency. Trypsinized cells were pelleted by centrifugation and resuspended in MEM (Gibco BRL) and 0.1% BSA. Then the cells were aliquoted and stimulated as described in the legend to Figure 7e. Upon stimulation, the cells were washed twice with PBS and resuspended in Buffer A (150 mM NaCl, 1 mM EDTA, 50 mM Tris-HCl, pH 7.4, 1 mM PMSF, 10 μg/ml aprotinin, 20 μg/ml leupeptin and 1 mM Na3VO4). The cells were bursted 4 times by liquid nitrogen, and the lysates were centrifuged at 800g for 10 min. The postnuclear supernatant was then transferred to an ultracentrifuge tube and centrifuged (Beckman Optima LE 80K centrifuge, Sw55Ti rotor) at 100,000g for 60 min at 4°C. The pellets were then resuspended in Buffer A, and the protein concentration was measured using the DC Protein Assay from BioRad (Hercules, UK) before the fractions were analyzed by SDS-PAGE and immunoblotting.

thumbnail image

Figure 7. Constitutive phosphorylation of Akt was inhibited and PTEN was recruited to the plasma membrane upon incubation of SKBr3 cells with Herceptin. (a,b) SKBr3, SKOv3 and Hep2 cells were incubated with Herceptin (15 μg/ml) for 1 hr at 37°C or with EGF (10 nM) for 15 min at 37°C. The cells were lysed, and cell lysates were subjected to SDS-PAGE and immunoblotting with (a) rabbit anti-phospho ErbB2 (pTyr1248) antibody or (b) mouse anti-phospho EGFR (pTyr1173) antibody. (c,d) SKBr3 and SKOv3 cells were incubated with Herceptin (15 μg/ml) for the indicated time periods at 37°C. The cells were lysed, and the cell lysates were subjected to SDS-PAGE and immunoblotting with rabbit anti-phospho Akt (pSer 473) antibody and sheep anti-EGFR antibody (loading control). (e) SKBr3 cells were incubated with Herceptin (15 μg/ml) for 15 min at 37°C, and the soluble and membrane fractions were separated by ultracentrifugation followed by SDS-PAGE and immunoblotting with mouse anti-PTEN antibody and mouse anti-ErbB2 antibody. (f) SKBr3 and SKOv3 cells were lysed, and the cell lysates were subjected to SDS-PAGE and immunoblotting with anti-PTEN antibody.

Download figure to PowerPoint

Results

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Herceptin inhibits growth of SKBr3 cells overexpressing EGFR, ErbB2 and ErbB3

Upon incubation of SKBr3, SKOv3 and Hep2 cells with Herceptin (15 μg/ml) for up to 72 hr, we found that Herceptin inhibited proliferation of SKBr3 cells only (Fig. 1a). Growth was measured as incorporation of 3H-thymidine into cellular DNA. As demonstrated in Figure 1a, the anti-proliferative effect of Herceptin in SKBr3 cells increased with time, and upon incubation for 72 hr, thymidine incorporation was inhibited by approximately 70%. Herceptin did not inhibit DNA synthesis in Hep2 and SKOv3 cells (Figure 1a). We did not detect ErbB3 by immunoblotting of Hep2 and SKOv3 cell lysates, while increasing amounts of ErbB2 were found in Hep2, SKOv3, and SKBr3 cells, respectively. EGFR was expressed in all cells with lowest amount in SKBr3 cells, more in SKOv3 cells, and most in Hep2 cells (Fig. 1b).

ErbB2 is neither internalized, nor down-regulated, in SKBr3 cells upon incubation with Herceptin.

By confocal microscopy and flow cytometry, we studied the amount of ErbB2 at the plasma membrane and the subcellular localization of ErbB2 in SKBr3 and SKOv3 cells upon incubation with Herceptin. Such studies demonstrated similar results for SKBr3 and SKOv3 cells, and only results obtained in SKBr3 cells are demonstrated in Figures 2 and 3. Confocal microscopy studies, using antibodies either to the extracellular domain of ErbB2 alone (data not shown) or the intracellular domain of ErbB2 and to early endosome antigen 1 (EEA1) (Fig. 2a), demonstrated that ErbB2 localized to the plasma membrane and that ErbB2 did not reach early endosomes or other intracellular vesicles upon incubation with Herceptin for 1 hr or upon incubation for 15 or 30 min, 2, 4, 6, 8, 12, 18, 24, 48 or 72 hr (data not shown). Thus, Herceptin does not induce endocytosis of ErbB2. In order to make sure that we would be able to detect ErbB2 in endosomes in case endocytosis took place, we incubated SKBr3 cells with geldanamycin (GA), which has been demonstrated to cleave ErbB2 and to direct the N-terminal part of ErbB2 to endosomes.30 Upon incubation with GA and immunostaining with an antibody to ErbB2, we indeed detected ErbB2 in endosomes (Fig. 2b). This fact strengthens the conclusion that Herceptin does not induce endocytosis of ErbB2. Flow cytometry analyses demonstrated that in SKBr3 cells, the amount of ErbB2 at the plasma membrane was not downregulated upon incubation with Herceptin for 24 hr (Fig. 2c) nor for 72 hr (data not shown). Importantly, however, incubation of SKBr3 cells with Herceptin for 24 hr resulted in 50% inhibition of DNA synthesis (see Fig. 1a). Incubation of SKOv3 cells with Herceptin did, however, not inhibit DNA synthesis (see Fig. 1a).

thumbnail image

Figure 2. ErbB2 at the plasma membrane is not internalized and not downregulated upon incubation of SKBr3 cells with Herceptin. (A) SKBr3 cells were incubated at 37°C with or without Herceptin (15 μg/ml) for 1 hr. The cells were fixed and immunostained using rabbit anti-ErbB2 antibody followed by Rhodamine Red-X conjugated donkey anti-rabbit antibody. To detect early endosomes, the cells were also immunostained using mouse anti-EEA1 followed by Cy5 conjugated donkey anti-mouse antibodies and studied by confocal microscopy. One representative experiment (out of 3) is demonstrated. (b) SKBr3 cells were incubated with or without 3 μM GA for 1 hr at 37°C. The cells were fixed and immunostained using mouse anti-ErbB2 antibody (to the extracellular domain) followed by Rhodamine Red-X conjugated donkey anti-mouse antibody. (c) SKBr3 cells were incubated at 37°C with or without Herceptin (15 μg/ml) for 24 hr. The cells were fixed and immunostained using mouse anti-ErbB2 antibody (to the extracellular domain) followed by Phycoerythrin conjugated goat anti-mouse antibody and analyzed by flow cytometry. One representative experiment (out of 3) is demonstrated.

Download figure to PowerPoint

thumbnail image

Figure 3. EGFR is endocytosed in SKBr3 cells. SKBr3 cells were incubated with or without EGF (10 nM) for 15 min at 37°C. (a) The cells were fixed and immunostained using mouse anti-EGFR antibody followed by Alexa Fluor 488 conjugated goat anti-mouse antibody and rabbit anti-ErbB2 antibody followed by Rhodamine Red-X conjugated donkey anti-rabbit antibody. Bar, 10 μm. (b) SKBr3 cells were fixed and immunostained using mouse anti-EGFR antibody followed by Phycoerythrin conjugated goat anti-mouse antibody and analyzed by flow cytometry. One representative experiment (out of 3) is demonstrated.

Download figure to PowerPoint

To investigate whether the activated EGFR was endocytosed in SKBr3 cells, EGF (10 nM) was added for 15 min at 37°C. The cells were fixed and labeled with antibody to the extracellular domain of EGFR followed by confocal microscopy and flow cytometry. As demonstrated, the EGFR was endocytosed and localized to intracellular vesicles upon incubation with EGF (Fig. 3a). EGF-induced internalization of the EGFR was further found to downregulate the amount of the EGFR at the plasma membrane (Fig. 3b). However, upon incubation of SKBr3 cells with EGF, ErbB2 was not detected in intracellular vesicles (Fig. 3a) nor downregulated from the plasma membrane (data not shown). The findings that ErbB2 could not be observed in endosomes and was not down-regulated from the plasma membrane are consistent with a lack of endocytosis of ErbB2. Alternatively, ErbB2 could be endocytosed and subsequently be rapidly recycled to the plasma membrane. To investigate whether ErbB2 was rapidly recycled, we incubated SKBr3 cells with monensin, which is known to inhibit endocytic recycling. Transferrin (Tf) is trafficked both through a rapid rab4-dependent recycling pathway and a slower, rab11-dependent route.31 Consistently, we found that incubation of SKBr3 cells with 10 μM monensin efficiently inhibited the constitutive recycling of the Tf receptor (TfR). As demonstrated, upon incubating cells with monensin, the TfR accumulated intracellularly (Fig. 4d). However, upon incubation of cells with monensin, no accumulation of ErbB2 was observed in intracellular vesicles even when Herceptin and monensin had been added. The small amount of ErbB2 observed intracellularly was the same with (Fig. 4c) and without (Fig. 4a) monensin and probably represents ErbB2 in the biosynthetic pathway. This finding strongly argues that the reason ErbB2 is not downregulated from the plasma membrane is lack of endocytosis and not rapid recycling upon endocytosis.

thumbnail image

Figure 4. ErbB2 does not accumulate intracellularly upon incubation of SKBr3 cells with monensin. SKBr3 cells were incubated with Herceptin and with (c,d) or without (a,b) 10 μM monensin for 1 hr at 37°C and subsequently prepared for immunofluorescence by labeling with antibodies to ErbB2 (a,c) and TfR (b,d).

Download figure to PowerPoint

125I-Herceptin is not endocytosed in SKBr3 cells

We further incubated SKBr3 cells with 125I-Herceptin to investigate whether Herceptin could be internalized by endocytosis. Cells were incubated at 37°C with 125I-Herceptin for various time periods from 5 min up to 24 hr. To separate surface bound Herceptin from internalized Herceptin, the cells were stripped using acetic acid. 125I-Herceptin released in the stripping buffer represents the fraction of 125I-Herceptin bound to the cell surface, while 125I-Herceptin in the cell lysates represents the internalized Herceptin. As endocytosis does not take place when cells are incubated on ice, control cells (0 hr) were incubated with 125I-Herceptin for 15 min on ice and not chased at 37°C. As shown in Figure 5, as much as 18% of 125I-Herceptin bound to cells on ice could not be released by the low pH buffer and is therefore acid resistant surface-bound 125I-Herceptin. Upon incubating cells with 125I-Herceptin at 37°C, the maximal increase in acid resistant radioactivity was only 4%. This demonstrates that there was no significant endocytosis of 125I-Herceptin in SKBr3 cells.

thumbnail image

Figure 5. 125I-Herceptin is not internalized in SKBr3 cells. SKBr3 cells were incubated with 125I-Herceptin diluted with unlabeled Herceptin to a total Herceptin concentration of 1.5 μg/ml (specific activity 25,000 cpm/mg) for 5, 15, 30 min, 1 hr and 24 hr at 37°C. Control cells (0 hr) were incubated with 125I-Herceptin for 15 min on ice. Subsequently, the cells were washed and treated with acetic acid (0.5 M, pH 2.5) to release surface bound 125I-Herceptin. The data represent the mean of 3 independent experiments with 6 parallels ± SEM.

Download figure to PowerPoint

Herceptin induces reduced expression of ErbB2 at the plasma membrane, but not growth arrest, in Hep2 cells

By flow cytometry, a 3.5-fold reduction of ErbB2 at the plasma membrane was demonstrated in Hep2 cells (Fig. 6a). However, by immunocytochemistry, ErbB2 was found localized to the plasma membrane in Hep2 cells. Incubation with Herceptin for 1 hr at 37°C did not cause relocalization of ErbB2 to endocytic vesicles (data not shown). It should be noted that even though incubation of Hep2 cells with Herceptin resulted in decreased expression of ErbB2, the growth of Hep2 cells was not affected upon incubation with Herceptin (see Fig. 1a). To investigate whether expression of ErbB2 was reduced due to fragmentation, we incubated Hep2 cells with Herceptin for up to 48 hr and immunoblotted cell lysates using a mixture of antibodies to the intracellular part of ErbB2. In Figure 6b, a significant downregulation of full-length ErbB2 upon incubation with Herceptin for 4, 24 and 48 hr at 37°C is demonstrated. The mixture of anti-ErbB2 antibodies detected full-length ErbB2 and additionally fragments of approximately 105 kDa MW (Fig. 6b).

thumbnail image

Figure 6. Herceptin induces fragmentation of ErbB2 in Hep2 cells. Hep2 cells were incubated with or without Herceptin (15 μg/ml) at 37°C for the indicated time periods. (a) The cells were fixed and immunostained, using a mouse anti-ErbB2 antibody followed by Phycoerythrin conjugated goat anti-mouse antibody and analyzed using flow cytometry. (b) The cells were lysed, and the cell lysates were subjected to SDS-PAGE and immunoblotting with a mixture of antibodies to various intracellular domains of ErbB2.

Download figure to PowerPoint

Herceptin inhibits the constitutive phosphorylation of Akt in SKBr3 cells but not in SKOv3 cells

As demonstrated in Figure 7a,b, ErbB2, but not the EGFR, was constitutively phosphorylated in SKBr3 and SKOv3 cells. Incubation of SKBr3 and SKOv3 cells with Herceptin did not alter the tyrosine phosphorylation of ErbB2, as demonstrated with antibody to pTyr1248 (Figure 7a). When ErbB2 was immunoprecipitated from Herceptin-treated SKBr3 and SKOv3 cells, and the immunoprecipitated ErbB2 was investigated by immunoblotting using antibody to pTyr, there was no detectable alteration (data not shown). To study whether Herceptin had an effect on signaling downstream of ErbB2, the phosphorylation of Akt was investigated. Akt was found to be constitutively phosphorylated in SKBr3 cells and in SKOv3 cells (Figure 7c,d) but not in Hep2 cells (data not shown). In SKBr3 cells, incubation with Herceptin caused reduced phosphorylation of Akt already upon 15 min of incubation, and the reduced phosphorylation of Akt was found to be sustained (Figure 7c). In SKOv3 cells, incubation with Herceptin did however not alter the phosphorylation of Akt (Figure 7d). The inhibited phosphorylation of Akt in SKBr3 cells was found to correlate with a slightly reduced interaction of ErbB3 with p85 observed upon incubation of SKBr3 cells with Herceptin (data not shown). However, consistent with recently published findings,26 we found that PTEN was recruited to the plasma membrane in SKBr3 cells (Fig. 7e), while in SKOv3 cells, we could not detect recruitment of PTEN to the plasma membrane (data not shown). This is consistent with our finding that the expression of PTEN was very low in SKOv3 cells compared to SKBr3 (Fig. 7f).

Ras is constitutively activated in SKOv3 cells

While MAPK was phosphorylated upon incubating SKBr3 cells with EGF, MAPK was found to be constitutively phosphorylated in SKOv3 cells (Fig. 8a). This fact suggested a high Ras activity and could potentially be explained by the reported lack of the Ras-like tumor suppressor protein ARHI in SKOv3 cells.27 Consistently, a Ras activity assay demonstrated constitutively active Ras in SKOv3 cells (Fig. 8b).

thumbnail image

Figure 8. MAPK and Ras are constitutively activated in SKOv3 cells. (a) SKBr3 and SKOv3 cells were incubated with EGF (10 nM) at 37°C for the times indicated. Cells were lysed and subjected to SDS-PAGE and immunoblotting with rabbit anti-phospho p44/42 MAPK (Thr 202/Tyr 204) and rabbit anti-MAPK antibodies (loading control). (b) Ras activity in SKOv3 cells was analyzed using a Ras activity kit, where GTP-bound Ras was precipitated using Raf-1's Ras binding domain (RBD) coupled to glutathione agarose beads and visualized by Western blotting with anti-Ras antibody.

Download figure to PowerPoint

ErbB2 homodimers do not induce hyperproliferation

It has been unclear whether ErbB2 homodimers can signal independently of other members of the ErbB family. In order to investigate this closer, we took advantage of an endothelial cell line lacking ErbB proteins (PAE cells). These cells were stably transfected with ErbB2 only, and selected clones were characterized with respect to ErbB2 overexpression by flow cytometry (unpublished) and Western blotting (Fig. 9a). One of the ErbB2 overexpressing clones (PAE.ErbB2) was compared to a cell clone derived from the same parental cells and stably transfected with both EGFR and ErbB2 (PAE.B2.ErbB2) (Fig. 9a). Flow cytometry experiments demonstrated that PAE cells with overexpression of ErbB2 only had no increased growth compared to nontransfected cells, while in PAE cells expressing both EGFR and ErbB2, there was an increased fraction of cells in S-phase even in the absence of EGF (Fig. 9b). In cells expressing ErbB2 only, there was no readily detectable phosphorylation of Akt compared to in nontransfected cells, while in isogenic cells expressing both ErbB2 and EGFR, phosphorylation of Akt was clearly observed both in the absence and presence of EGF (Fig. 9c). However, ErbB2 was hyperphosphorylated regardless of whether the ErbB2-overexpressing cells additionally expressed EGFR (Fig. 9c). This demonstrates that functional ErbB2 homodimers are formed, but that such homodimers cannot activate Akt, in the absence of other members of the ErbB family.

thumbnail image

Figure 9. ErbB2 homodimers do not induce hyperproliferation. (a) Cell lysates of untransfected PAE cells, PAE cells transfected with ErbB2 (PAE.ErbB2) or PAE cells transfected with ErbB2 and EGFR (PAE.B2.ErbB2) were subjected to SDS-PAGE and immunoblotting with sheep anti-EGFR and mouse anti-ErbB2 antibodies. (b) Cell cycle analysis was performed by flow cytometry as described in Material and Methods, and the distribution of cells in S-phase is demonstrated for each of the different cell lines. (c) PAE, PAE.ErbB2, and PAE.B2.ErbB2 cells were incubated with or without EGF (10 nM) for 15 min at 37°C. The cells were then lysed, and cell lysates were subjected to SDS-PAGE and immunoblotting with rabbit anti-phospho Akt (pSer 473) antibody (upper panel), rabbit anti-phospho ErbB2 (pTyr1248) antibody (middle panel) and mouse anti-Grb2 antibody (loading control) (lower panel).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Proliferation is inhibited by the humanized ErbB2-binding monoclonal antibody Herceptin in SKBr3 cells, which have high expression levels of EGFR, ErbB2 and ErbB3. We have demonstrated that in SKBr3 cells and in SKOv3 cells, ErbB2 is not downregulated from the plasma membrane upon incubation with Herceptin. However, in Herceptin-treated Hep2 cells, ErbB2 is downregulated due to extensive fragmentation. Herceptin has been reported to downregulate ErbB2 in some cells, but not in others.12, 17, 18, 22, 23 This difference in downregulation could potentially be explained by Herceptin's ability to activate proteases in some cell lines. Since Herceptin inhibited proliferation of SKBr3 cells, where EbB2 was not downregulated, but not of Hep2 cells, where ErbB2 was downregulated, the growth-inhibiting effect of Herceptin does not correlate with ErbB2 down-regulation. We thus conclude that Herceptin inhibits DNA synthesis independently of ErbB2 downregulation.

Internalization of ErbB2 upon binding antibodies has been discussed in a number of articles.17, 18, 22, 23 In none of the studies described has endocytosis of ErbB2 as such been studied, but rather increased turnover and downregulation of the receptor has been reported. Sarup et al.19 incubated cells with 125I-4D5 prior to treating the cells with a high salt/low pH buffer to separate internalized 125I-4D5 from surface-bound 125I-4D5 and reported internalization of 125I-4D5. We have repeated such experiments in which cells were incubated with radiolabelled Herceptin before the radioactivity was fractionated using acetic acid buffer. In our experiments, we found that Herceptin bound to the cell surface was not released by the buffer used by Sarup et al. and that the fraction reported as internalized was in fact antibody firmly attached to the cell surface (unpublished). Upon stripping with acetic acid, more 125I-Herceptin was released, and we found that 125I-Herceptin was not internalized in ErbB2 overexpressing cells even upon incubation with Herceptin at 37°C for as long as 24 hr. Our results are consistent with the published reports by Baulida et al.32 and Hommelgaard et al.,33 demonstrating that ErbB2 is in fact endocytosis-impaired. Hommelgaard et al.33 suggested that ErbB2 is endocytosis-impaired due to retention on membrane protrusions. ErbB2 has also recently been demonstrated to bind PDZ domain-containing proteins like Erbin, PICK1 and Lin-7,34, 35, 36 and such interactions have been suggested to inhibit endocytosis and to increase recycling of ErbB2.36 Like Hommelgaard et al., we were unable to detect ErbB2 in endosomes, and in SKBr3 cells, ErbB2 was not downregulated from the plasma membrane. Our demonstration that ErbB2 is not downregulated by endocytosis is partly consistent with the view recently presented by Austin et al.37 However, these authors explained the high surface distribution of ErbB2 by rapid recycling of ErbB2 upon endocytosis and not by lack of endocytosis. It should be noted that our inability to find endocytosed ErbB2 in Herceptin-treated cells is not based on the inability to observe ErbB2 had it been endocytosed. When SKBr3 cells were incubated with GA, ErbB2 could clearly be observed in endocytic vesicles consistent with the fact that GA results in cleavage and endocytosis of ErbB2.30 Our current data demonstrate that while the TfR accumulates in intracellular compartments upon blocking recycling with monensin, there is no concentration of ErbB2 in similar recycling organelles. Our data are thus not consistent with the data recently published by Austin et al.37 but in agreement with the recent findings of Hommelgaard et al.33 It should be noted that while Austin et al.37 only investigated the subcellular localization of Herceptin, we investigated the subcellular localization of ErbB2, as did Hommelgaard et al.33 This difference in approach could potentially explain the different conclusions reached.

Herceptin's anti-proliferative effect has been linked to inhibited Akt activity.12, 14, 22, 38 In SKBr3 cells, we observed constitutively phosphorylated Akt. Consistent with the growth promoting activity of pAkt, we found that incubation of SKBr3 cells with Herceptin resulted in reduced phosphorylation of Akt. By immunoprecipitation experiments, we further observed that the recruitment of p85 to ErbB3 was slightly inhibited upon incubation with Herceptin. However, in agreement with the recent publication by Nagata et al.26, we could demonstrate that PTEN was recruited to the plasma membrane in Herceptin-treated SKBr3 cells. As membrane recruitment of functional PTEN has been demonstrated to efficiently decrease the amount of phosphatidylinositol-3,4,5- trisphosphate (PIP3), we consider recruitment of PTEN to be the prime growth-inhibitory mechanism in SKBr3 cells.

The constitutive phosphorylation of Akt was not inhibited by Herceptin in SKOv3 cells. We found that both MAPK and Akt were constitutively phosphorylated in SKOv3 cells, pointing to an increased Ras activity. Consistently, we also demonstrated constitutive Ras activity in SKOv3 cells. SKOv3 cells, like a number of other ovarial cancer cells, lack the functional tumor suppressor protein ARHI27, and lack of this Ras-like protein has been reported to result in increased Ras-dependent activation of MAPK.39 Our findings therefore strongly suggest that Akt is constitutively phosphorylated due to Ras-induced PI3 K activity. Furthermore, SKOv3 cells were found to have low expression of PTEN, suggesting inefficient PIP3 phosphatase activity. This argues that for Herceptin to be growth inhibitory, the activity of PTEN must compete out the activity of PI3 K and that this will be impossible in cells with constitutive Ras activity and reduced levels of PTEN. It should be noted that germline inactivation of 1 copy of PTEN in mice was sufficient to predispose animals to tumors in multiple organs. Most of these tumors were associated with inactivation of the wild type allele. However, some were not, suggesting that haploinsufficiency of PTEN was able to promote tumor development in some instances.40

It has been proposed that inhibition of ErbB2 phosphorylation is by itself insufficient in determining the response of ErbB2 overexpressing cells to Herceptin.12 This is consistent with our current results demonstrating no Herceptin-induced change in phosphorylation of ErbB2, while both DNA synthesis and phosphorylation of Akt were inhibited in SKBr3 cells. In fact, our present results demonstrated that overexpression of ErbB2 in the absence of EGFR, ErbB3 and ErbB4 did not result in increased growth compared to cells lacking ErbB2. This is consistent with the notion that overexpression of ErbB2 homodimers only do not transform cells.41 Our current findings further demonstrate that ErbB2 homodimers are incapable of inducing phosphorylation of Akt. This is consistent with the notion that ErbB2 homodimers are unable to recruit the p85 regulatory subunit of PI3 K.42

In conclusion, we have demonstrated that Herceptin does not induce endocytic downregulation of ErbB2. We have further confirmed that Herceptin effects reduced phosphorylation of Akt by recruitment of PTEN to the plasma membrane. Furthermore, our data indicate that prerequisites for Herceptin to be growth inhibitory are lack of constitutive Ras activity and lack of reduced expression of PTEN.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

M.S. Rødland is acknowledged for excellent technical assistance. K.E. Longva was supported by a fellowship from The Norwegian Women's Public Health Association, N.M. Pedersen by a fellowship from The Norwegian Cancer Society, C. Haslekås and E. Stang by fellowships from the functional genomics programme (FUGE) in The Research Council of Norway.

References

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References