• paramagnetic microspheres;
  • ErbB tyrosine kinases;
  • epidermal growth factor receptor;
  • ErbB2;
  • receptor phosphorylation;
  • localized signal spreading;
  • confocal laser scanning microscopy


  1. Top of page
  2. Abstract
  6. Acknowledgements


ErbB2 (HER-2), a member of the epidermal growth factor (EGF) receptor family, is a class I transmembrane receptor tyrosine kinase. Although erbB2 has no known physiologic ligand, it can form complexes with other members of the family and undergo transactivation of its very potent kinase activity, thereby initiating downstream signaling and cell proliferation. ErbB2 is a frequent pathologic marker in ductal invasive breast carcinomas and is targeted by using a specific humanized monoclonal antibody, trastuzumab (Herceptin). The antibody is effective in only 20% to 50% of erbB2-positive tumors, and this resistance, as yet poorly understood, constitutes a major therapeutic challenge.


Magnetic microspheres coated with ligands or antibodies are widely used for separation of proteins and cells and allow localized, high intensity, and precisely timed stimulation of cells. We used EGF- and trastuzumab-covered paramagnetic microspheres, quantitative confocal laser scanning microscopy, and digital image processing to investigate the (trans)activation of and local signal propagation from erbB1 and erbB2 on trastuzumab sensitive and resistant carcinoma cell lines expressing these receptors at high levels.


On A431 cells expressing high levels of endogenous erbB1 and transfected erbB2-mYFP (A4-erbB2-mYFP F4 cell line), EGF-coupled-microspheres activated erbB1 and transactivated erbB2-mYFP. In two other cell lines with comparable erbB2 expression but lower levels of erbB1, EGF microspheres transactivated erbB2 less efficiently. Trastuzumab in solution activated erbB2 on A4-erbB2-mYFP and the trastuzumab sensitive SKBR-3 cells, but only negligibly on the resistant JIMT-1 cells that showed a 10 times higher Kd for the antibody. Nevertheless, pronounced erbB2 activation and tyrosine phosphorylation could be detected after stimulation with trastuzumab-coupled microspheres in all cell lines, although transactivation of erbB1 was negligible. Receptor phosphorylation was restricted to the immediate proximity of the microspheres, i.e., receptor clusters external to these locations remained inactive.


ErbB1 ligand and erbB2 specific antibody attached to magnetic microspheres are efficient tools in assessing erbB activation, localized signal propagation, and erbB heterodimer formation. Trastuzumab coupled to microspheres is more efficient at accessing erbB2 and activating it than trastuzumab in solution. © 2005 International Society for Analytical Cytology

ErbB2, a member of the erbB family of receptor tyrosine kinases, plays an important role in the initiation and development of ductal mammary carcinomas. Almost 25% to 30% of these tumors overexpress erbB2, which is associated with very poor prognosis (1). Only about 20% to 50% of the erbB2 overexpressing tumors are sensitive to trastuzumab (Herceptin), a humanized monoclonal antibody that binds to an extracellular domain of erbB2 adjacent to the plasma membrane (2–5). The mechanism of action of trastuzumab is complex (6–10); thus, the possible origins of trastuzumab resistance also remain obscure. In view of the large number of suspected mechanisms and molecular partners possibly interacting with erbB2 (11, 12), many alterations of the erbB2 receptor itself and its signaling pathways may be the source for trastuzumab resistance. Although trastuzumab was thought to act, at least in part, by inducing downregulation of erbB2 (13), Austin et al. (10) recently demonstrated that trastuzumab does not cause efficient degradation of internalized erbB2 but rather recycles with the receptor to the cell surface after endocytosis. Another recent study has proposed that the mechanism of action of trastuzumab involves the rapid activation of the lipid phosphatase PTEN (14), and others have suggested that the in vivo response is due to induction of an immune response (15). In either case, a simple cause of resistance could be the decreased binding of trastuzumab to its target, caused by mutations in erbB2, or blocking by other molecules such as the MUC4 sialomucin complex (16, 17).

JIMT-1, a newly established trastuzumab resistant cell line (18), provides the means for investigating the mechanism of trastuzumab resistance using an in vitro model. JIMT-1 expresses erbB1 to approximately the same level as SKBR-3, a widely used mammary carcinoma line that is sensitive to trastuzumab, and overexpresses erbB2 as a result of gene amplification to about half the level found on SKBR-3. Recently, it was reported that the epitope of erbB2 binding trastuzumab is masked on JIMT-1 cells, and that the local density of MUC4 sialomucins is negatively correlated with trastuzumab binding (17). However, decreased binding of trastuzumab might not be the single cause of diminished response to therapy. Because trastuzumab does not disrupt association of erbB2 with other members of the erbB family, we assessed heterodimer formation and erbB activation in the sensitive and resistant cell lines via transactivation by erbB1 and erbB2 specific ligands attached to magnetic microspheres so as to achieve high local concentrations.

Microspheres covered with ligands or antibodies are used in many fields of biochemistry, biophysics and cell biology. Immunomagnetic microsphere separation and microparticle based flow cytometry play an important role in experimental protocols and analytical procedures (19, 20). Recently, microspheres have also served as powerful tools for studying complex cell biological phenomena such as collagen phagocytosis in fibroblasts (21, 22), viscoelastic responses of endothelial cells upon thrombin stimulation measured by magnetic microsphere-based microrheology (23), and re-organization of lipid rafts and signalling complexes upon various stimuli (24–30).

Microsphere-based techniques allow the localized stimulation of cells with precise control of timing and location, providing a versatile tool for kinetic assays of signal spreading and downstream propagation. Previous studies have reported focal erbB1 (epidermal growth factor receptor [EGFR]) activation induced by EGF-linked microspheres and Shc translocation to these sites (31, 32). On erbB1/green fluorescent protein transfected Chinese hamster ovary and A431 cells, the internalization of the EGF coated microspheres occurred rapidly but required erbB1 activation. Using other types of microspheres Verveer et al. (33) reported that focal stimulation generated phosphorylated erbB1 patches with rapid lateral spreading of erbB1 activation over the entire cell. The latter phenomenon was attributed to dissociation of activated erbB1 from the dimers and activation of erbB1 molecules in the absence of ligand, i.e., via re-dimerization outside of the microsphere-stimulated area. Our own studies have not confirmed this phenomenon (31, 32, 34). Another group (35) also failed to observe long-range erbB1 activation by focal (bead) EGF stimulation. They demonstrated that EGF-biotin–coupled streptavidin magnetic beads elicited a localized polymerization of actin even under conditions of receptor overexpression.

In the present study, we used EGF and trastuzumab coupled paramagnetic microspheres to explore the activation of erbB1 and erbB2 coexpressed on carcinoma cells. Our preparations are biologically active and allow specific stimulation and activation of erbB1 and erbB2. In the different cell lines, the localized stimulus applied via microspheres remained local, inducing an effort by the cell to internalize the microspheres; lateral spreading of receptor phosphorylation was not observed. Higher levels of erbB1 expression in the cells led to a more efficient transactivation of coexpressed erbB2, but erbB2 binding by trastuzumab beads did not induce erbB1 transactivation. In the trastuzumab-resistant JIMT-1 cells, where high MUC4 expression is thought to prevent trastuzumab binding and thus to prevent erbB2 activation and internalization, trastuzumab coupled to microspheres efficiently induced phosphorylation of erbB2 tyrosine residues. Thus, the high ligand density and the steric effect attributable to the microsphere technique can be used in combination with trastuzumab in solution to test for the presence and functionality of spatially hindered receptors.


  1. Top of page
  2. Abstract
  6. Acknowledgements


Six carcinoma cell lines, SKBR-3, JIMT-1(18), A431, an A431-derived cell line stably transfected with the erbB2-mYFP fusion gene (A4-erbB2-mYFP F4 line, expressing erbB2 with a C-terminal monomeric YFP fusion) (36), HeLa, and erbB1 transfected Chinese hamster ovary and MCF7 cells were used. Cells were cultured in a humidified atmosphere with 5% CO2 at 37°C in Dulbecco's Minimal Essential Medium (Sigma, Schnelldorf, Germany) supplemented with 10% fetal calf serum (Sigma) and antibiotics. Cells were propagated every 3 to 4 days. For microscopy experiments, cells were seeded onto 12-mm-diameter glass coverslips 2 days in advance and used at a confluency of 30% to 50%. Before stimulation experiments, the cultures were starved in serum-free Dulbecco's Minimal Essential Medium for 24 h.


The primary mouse monoclonal antibodies were against the activated forms of erbB1 (clone74, Transduction Laboratories, Lexington, KY, USA) and erbB2 (Ab18, clone PN2A, Lab Vision, Fremont, CA, USA) and against phosphotyrosine residues (PY99, cs-7020, Santa Cruz Biotechnology, Santa Cruz, CA, USA). Trastuzumab (Herceptin) and another anti-erbB2 antibody, 7C2, were a kind gift from Genentech, Inc. (South San Francisco, CA, USA). The antibody 528 against erbB1 was prepared from the supernatant of the corresponding hybridoma line (catalog no. HB-8509, American Type Culture Collection, Rockville, MD, USA). These antibodies were conjugated with sulfoindocyanine-succinimidyl ester (Cy5, monofunctional, Amersham, Piscataway, NJ, USA) or Alexa-488 (Invitrogen, Carlsbad, CA, USA) according to the manufacturers' instructions. Cy3- and Cy5-labeled secondary polyclonal antibodies were from Jackson ImmunoResearch (West Grove, PA, USA).

Preparation of Chemically Coupled Trastuzumab and EGF Microspheres

Natural murine EGF was obtained from IC Chemikalien (Ismaning, Germany). Carboxy-functionalized superparamagnetic 1-μm microspheres (SERA-MAG, Seradyn, IN, USA) were activated with 0.1 M sulpho-N-hydroxysulphosuccinimide (Pierce, Rockfort, IL, USA), 0.1 M 1-ethyl(-3-[3-dimethyl-aminopropyl])carbodiimide hydrochloride (Pierce) in 0.1 M 2-(N-morpholino)ethane sulfonic acid (MOPS; Sigma) buffer, at pH 6, for 1 h at room temperature. After two washes in 0.1 M MOPS, the microspheres were equilibrated in coupling buffer (0.1 M Na-phosphate, pH 8). The coupling reaction was carried out overnight at 4°C with 50 μg EGF or 250 μg trastuzumab in 150 μl of coupling buffer for 30 μl of the 5% microsphere slurry with constant agitation. The microspheres were then washed twice with coupling buffer and thoroughly with phosphate buffered saline (PBS). The remaining reactive groups were capped with 1 M ethanolamine for 2 h at room temperature followed by further washings in PBS. The microspheres were aliquoted and stored in PBS with 0.1% Na-azide. The presence of trastuzumab on the surface of microspheres was confirmed using fluorescently labeled polyclonal goat anti-human antibody. Uncoupled microspheres did not bind to the cells.

Stimulation and Immunofluorescence

To avoid nonspecific interaction of the microspheres, they were blocked with 1 mg/ml bovine serum albumin (BSA; Sigma) in Tyrode's buffer or PBS. Cells were washed in Tyrode's buffer containing 0.01 M glucose and 0.1-1 mg/ml BSA and maintained in this buffer in a humid chamber. Uncoupled, EGF-linked, and trastuzumab-linked magnetic microspheres or free trastuzumab, dye-conjugated trastuzumab or EGF in solution were used as stimulating ligands. For stimulation, the magnetic microspheres or free ligand were diluted in 50 μl PBS; the final concentration was approximately 10 to 30 microspheres per cell. To provide a specific time of interaction and initiation of stimulation, the coverslips were positioned on a magnet at time 0. The same ligands were also used in control experiments without the magnet. After a specified reaction time, between 2 and 45 min at 37°C, cells were rinsed with PBS and fixed with ice-cold methanol for 20 min at −20°C. After rehydrating the samples with PBS, cells were blocked and permeabilized for 20 min at room temperature using 1 mg/ml BSA (Sigma) and 0.1% Triton X-100 (Fluka, Buchs, Switzerland) in PBS. Primary antibodies against P-erbB1, P-erbB2, and phosphotyrosine (PY) were diluted in 100 μl PBS with 1mg/ml BSA and 0.1% Triton X-100 to a final concentration of 1 μg/ml. After 30 min at RT, cells were washed three times with 1 mg/ml BSA and 0.1% Triton X-100 in PBS for 5 min. Incubation with secondary Cy3-goat anti-mouse immunoglobulin (GAMIG) was performed as with the primary ones. Samples were washed again and mounted in 5 μl Mowiol (0.1 M Tris-HCl, pH 8.5, 25 w/v% glycerol, and 10% Mowiol 4-88; Hoechst Pharmaceuticals, Frankfurt, Germany) on precleaned microscopic slides.

Confocal Laser Scanning Microscopy

For confocal laser scanning microscopy, Zeiss (Göttingen, Germany) LSM 510 systems using Plan-Apochromat 63×/1.4 numerical aperture, oil DIC objective were used. YFP was excited with a 514-nm Ar ion laser and detected through a 530-nm bandpass filter. Cy3 was excited with a 543-nm HeNe laser and detected through a 560 longpass filter. Cy5 was excited with a 633-nm HeNe laser and detected through a 650 longpass filter. Pinholes were set to obtain 1-μm optical slices except for high Z-resolution sectioning (Fig. 2) where 0.5-μm slices were acquired in 0.4-μm steps. Reflection images of the microsphere distribution were taken using a 633-nm HeNe laser without filtering the reflected light, and the pinhole set to 10.3 Airy units. Then 512 × 512 pixel images were taken with pixel times of 6.4 μs and 2× line averaging in the case of high-resolution images. All images were obtained in multitrack mode to avoid crosstalk with other channels. With the given settings, channel crosstalk was negligible as determined using single-labeled samples.

Image Analysis

Reconstruction of orthogonal sections was performed using the LSM system software (version 3.2).

The cross-correlation coefficient characterizing the coincidence of signals in two channels of an image was calculated with a program custom written in the LabView environment (37, 38). Briefly, for a pair of images x and y, the cross-correlation coefficient is defined as

  • equation image(1)

where xi,j and yi,j are fluorescence pixel values at coordinates i,j in images x and y. Only those pixels were used for summation that were above the detection threshold in both images. Threshold values were determined as the highest histogram values from images of unlabeled samples taken with identical instrument parameters. The theoretical maximum is C = 1 for identical images, and a value close to 0 implies random spatial localization of the two labels relative to each other.

Other image processing tasks were performed using SCIL-Image (TNO, Delft, The Netherlands). For determining erbB phosphorylation in stimulated and nonstimulated areas, the background-corrected fluorescence of immunolabeled P-erbB was used as an input channel. The histogram mean of cells from images of unlabeled samples taken with identical instrument parameters were taken as background values. Binary masks on cells were created by thresholding reflection images. Threshold values were chosen manually to exclude all background pixels; occasional holes in cell masks were filled with dilation-erosion cycles using a 3 × 3 mask and a connectivity of 8 pixels. Additional binary masks for separating stimulated and nonstimulated areas were generated by thresholding reflection images of microspheres when microsphere-coupled ligands were used and by thresholding Cy-5-trastuzumab images when trastuzumab in solution was used as stimulus. For calculating the proportion of pixels above the resting level of phosphorylation, overall maximum values from corresponding nonstimulated but equivalently labeled cells were used as threshold.

Determination of Receptor Numbers and Trastuzumab Dissociation Constants

SKBR-3, JIMT-1, A431, and A4-erbB2-mYFP cells were cultured as described above and harvested upon reaching half confluence, washed once in ice-cold PBS, and resuspended at 5 × 105 cells in 100 μl PBS. Cells were incubated for 10 min on ice with Alexa-488 conjugated antibodies. Kd values were determined using concentrations of 2.5 to 50 μg/ml; for determining expression levels, saturating concentrations of 50 μg/ml were used. After staining, cells were centrifuged for 4 min at 800g, washed in PBS, and fixed with 1% formaldehyde. Intensity histograms of 2 × 104 cells were measured with a Becton Dickinson (Franklin Lakes, NJ, USA) FACScan flow cytometer taking FL1 data. Surface expression of erbB2 on SKBR-3 was quantified using QIFIKIT (DakoCytomation, Glostrup, Denmark) according to the manufacturer's instructions and was found to be 1.1 × 106 molecules per cell. The relative numbers of cell surface proteins on other cells were calculated by comparing the FL1 fluorescence intensities with that of erbB2 on SKBR-3, taking the labeling ratios of the antibodies into consideration. ErbB1 levels were determined similarly using the anti-erbB1 monoclonal antibody 528. For determining Kd-s, background corrected histogram means, normalized to the maximal intensities were used from each sample. The bound fraction was estimated from the relative labeling intensities and known total receptor numbers. Scatchard plots with bound/free ratio against bound antibodies were used to calculate the Kd-s.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Trastuzumab and EGF Microspheres Are Functionally Active

To test the functionality of Trastuzumab- and EGF-coupled microspheres, we first used a cell line derived from the A431 epidermoid carcinoma by stably transfecting it with the erbB2-mYFP chimera gene. Native A431 cells express ∼2 × 106 erbB1 and ∼4 × 104 erbB2 receptors as measured by flow cytometry. The transfected line A4-erbB2-mYFP exhibits a somewhat downregulated expression of erbB1 (1.2 × 106) and 9.5 × 105 erbB2 per cell. Applying EGF microspheres, erbB1 could be focally activated (Fig. 1a) as previously described for these microspheres (31, 32). Activation was detected as total tyrosine phosphorylation (Mab PY99; including activation of SHC) (32) and as specific phosphorylation of erbB1 (Mab clone74). The correlation between the phosphorylation signal and location of microspheres was high, as characterized by the cross-correlation coefficients (C) 0.68 and 0.57 (Fig. 1b) for phospho-erbB1 and phosphotyrosine, respectively. (Not all microspheres lay in the same plane as the confocal immunofluorescence images shown in Figure 1. Thus, activation signals coincided with a subset of microsphere loci in this case, whereas the correlation coefficients were calculated from stacks of confocal images through the cells.) Trastuzumab microspheres on the same cells also evoked local total tyrosine phosphorylation (Mab PY99, Fig. 1a) and specific erbB2 phosphorylation (Mab 18). The correlation of microsphere position with erbB2 and generic tyrosine phosphorylation was high, as seen from cross-correlation coefficients 0.68 and 0.47, respectively (Fig. 1b). When uncoupled microspheres were used as control, no activation above the level of baseline phosphorylation was detected. It is noteworthy that the correlation between microspheres and specific erbB phosphorylation was always higher than that between microspheres and generic phosphotyrosine labeling, implying that signal spread from the activated receptors may be attributed to other molecular species, including adapters like SHC, but possibly also to other downstream effectors such as RAS.

thumbnail image

Figure 1. Transactivation of erbB2 by EGF microspheres and failure of trastuzumab microspheres to transactivate erbB1 on A4-erbB2-mYFP cells expressing high levels of erbB1 and erbB2. a: Images of A431 cells, stably transfected with an erbB2-mYFP construct, are shown after 10 min of stimulation with EGF- or trastuzumab-coupled microspheres. Green images, YFP signal; red images, activated erbB1 (left), activated erbB2 (middle), and tyrosine phosphorylation (right). Reflection gray-scale images show the location of individual and clustered microspheres as bright spots. On the overlay images, only overlap of the microspheres localized in the confocal fluorescence plane can show coincidence. Scale bar = 10 μm. Upper row, EGF-coupled microspheres; lower row, trastuzumab-coupled microspheres. b: Cross-correlation values (± standard deviation) for EGF (white columns) and trastuzumab (gray columns) microsphere-covered areas and phosphorylation of erbB1 (left), erbB2 (middle), and tyrosine residues (right).

Download figure to PowerPoint

EGF-Bound erbB1 Transactivates erbB2 but Trastuzumab-Bound erbB2 does not Transactivate erbB1

Measuring the activation of erbB1 and erbB2 upon stimulation allowed the assessment of the degree of transactivation of erbB1 to erbB2 and of erbB2 to erbB1. In the latter case, trastuzumab-coupled microspheres provided a unique new tool because erbB2 has no physiologic ligand, although trastuzumab induces specific erbB2 phosphorylation. We noted quite strong erbB2 phosphorylation upon EGF-microsphere stimulation (Fig. 1a, top row, middle panels), for which the cross-correlation between microsphere locus and P-erbB2 was 0.43 (Fig. 1b). In the reverse direction, however, very little activation of erbB1 was seen upon trastuzumab-microsphere stimulation (Fig 1a, lower row, left panels). A P-erbB1 specific immunofluorescent signal was not observed above the level in starved, resting cells, and correlated poorly with microsphere location (C = 0.13). In a few random cells, erbB1 was spontaneously activated without added ligand (see membrane of a cell at the top of the panel), a property characteristic of A431 cells. Because on these cells erbB1 and erbB2 were expressed at roughly the same levels, the asymmetric results indicate a much higher affinity of EGF-bound erbB1 for erbB2 than that of the trastuzumab-bound erbB2 for erbB1. A possible explanation could be that bivalent trastuzumab binding to erbB2 favors erbB2 dimer formation and sterically hinders its association with erbB1, although possibly not by obstructing the dimerization loop, but rather by preventing the formation of a multimer of at least one erbB1 and an active erbB2 dimer (39).

Stimulation by EGF and Trastuzumab Microspheres Evokes a Well-Localized Activation Process

As shown in Figure 1, trastuzumab- and EGF-coupled microspheres caused a local activation of erbB receptors, which was evident also by the high degree of correlation between microsphere position and receptor phosphorylation. In contrast to a previous report (33) but in confirmation of our previous data (31, 32), we observed only highly localized activation of the receptors in the immediate proximity to the microsphere. There was no spreading of phosphotyrosine-erbB1 activation over the plasma membrane in any cell line expressing erbB1 and stimulated with EGF-linked magnetic microspheres regardless of whether the microspheres were applied with or without a magnet. These included A431, HeLa, and erbB1 transfected Chinese hamster ovary and MCF7 cells with receptor densities from 2 × 106 to 6 × 104; see also Lidke et al. (34) (some cell lines not shown). Transactivation of erbB2 induced by EGF-linked magnetic microspheres was also spatially well localized. ErbB2-mYFP was recruited with erbB1 to EGF microspheres from the surrounding membrane, but P-erbB2 was almost exclusively at the periphery of EGF microspheres. Similarly, trastuzumab microspheres recruited erbB2-mYFP to their neighborhood and initiated phosphorylation of the receptor, which was then followed by apparent efforts to internalize the microspheres (Fig. 2). The colocalization of red (phosphotyrosine) and green (erbB2-mYFP) signals in Figure 2 was almost complete around the microspheres. The erbB2-mYFP molecules and the surrounding membrane concentrated at the periphery of the microspheres, with a signal-free area corresponding to the interior of the microsphere (see the orthogonal sections in Fig. 2). Clusters of erbB2-mYFP were also observed outside the microsphere-covered areas (arrows in Fig. 2) but there was no sign of phosphorylation in these clusters. Stated quantitatively, in microsphere-covered areas the ratio of background corrected YFP/phosphotyrosine signal was 1.2 ± 0.2, whereas in areas outside the microspheres that were above background for YFP, the corresponding ratio was 6.8 ± 1.4. These observations of local activation by and internalization of trastuzumab microspheres were very similar to those using EGF-magnetic microspheres published previously (31, 32, 34, 35).

thumbnail image

Figure 2. Trastuzumab microspheres are internalized after erbB2 activation. Stimulation with Herceptin microspheres for 10 min resulted in focal aggregation and phosphorylation of erbB2 molecules. Three-dimensional reconstruction of the image stacks shows microspheres engulfed by the membrane. Orthogonal sections of a part of an A4-erbB2-mYFP cell are shown. The x-y section, as indicated by the blue lines in the x-z and y-z projections, shows the membrane surface of the cell where the microspheres were bound. Green channel, erbB2-mYFP; red channel, phosphotyrosine immunofluorescence. Microspheres are surrounded by aggregated erbB2-mYFP receptors, overlapping completely with the phosphotyrosine signal. Membrane areas devoid of microspheres contain clusters of erbB2-mYFP that are negative for phosphotyrosine (PY) and therefore appear as faint green patches (arrows). The hollow area indicated by the cross-hairs is the center of the nonfluorescent microsphere.

Download figure to PowerPoint

Trastuzumab Microspheres are Effective even if Trastuzumab in Solution has no Effect

The SKBR-3 cell line is a well-known trastuzumab sensitive model of erbB2 overexpressing mammary carcinoma. This cell line expresses 1.9 × 105 erbB1 and 1.1 × 106 erbB2 molecules per cell as measured by flow cytometry. Although proliferation of SKBR-3 and several other breast tumor lines are inhibited by trastuzumab, a good in vitro model of trastuzumab resistance was unavailable until the recent establishment of the JIMT-1 line (18). These cells express ∼8 × 104 ErbB1 and ∼6 × 105 ErbB2 molecules (as detected by antibodies 528 and 2C4). However, if fluorescently labeled trastuzumab is used to detect ErbB2 levels on JIMT-1, the apparent level is considerably lower, ∼8 × 104. This result is indicated in Figure 3a by the apparent reduced labeling of JIMT-1 compared with SKBR-3. Saturation curves (Fig. 3b) of A488-trastuzumab binding to these cells yielded maximal fluorescence values for SKBR-3 that were 10 times those of JIMT-1. Kd values of trastuzumab derived from Scatchard plots were 2.5 and 25 nM for SKBR-3 and JIMT-1, respectively. Thus, trastuzumab binding was hindered significantly on JIMT-1 cells, although stimulation of these cells with trastuzumab microspheres (Fig. 4a) led to ready phosphorylation of erbB2, as in the case of SKBR-3. The cross-correlation coefficients between the microspheres and the P-erbB2 signal in the images shown were 0.73 and 0.57 for SKBR-3 and JIMT-1, respectively. Increased phosphorylation was confirmed using anti-PY staining (not shown). However, the application of trastuzumab in solution led to hardly any activation of erbB2 (Fig. 4; note that for JIMT-1, the P-erbB2 signal is displayed at four times amplification). In SKBR-3 cells, the level of erbB2 phosphorylation increased in areas where trastuzumab or trastuzumab microspheres were bound, as opposed to nonstimulated areas of the plasma membrane (Fig. 5a), and the number of pixels above the maximal resting erbB2 phosphorylation level also increased (Fig. 5b). In the case of JIMT-1, trastuzumab in solution increased erbB2 phosphorylation, albeit to a much lesser extent than trastuzumab microspheres (Fig. 5a). The same conclusion was derived from the increased number of pixels above resting activation levels (Fig. 5b), and was consistent with the level of binding of trastuzumab that could be demonstrated by flow cytometry (Fig 3b). JIMT-1 cells can also internalize soluble trastuzumab to some extent (17), which may correspond to the small extent of activation and phosphorylation of erbB2 that we observed. The present data also suggest that the function of erbB2 is intact in JIMT-1 but that the binding site of trastuzumab is blocked as proposed in (17). This inhibition can apparently be overridden by applying a higher density of ligand and, possibly, as a result of the steric effects arising from the microspheres. A general finding was that the application of trastuzumab in solution led to lesser activation of all cell lines than that achieved by trastuzumab microspheres. The higher local ligand concentration also may account for this finding.

thumbnail image

Figure 3. Trastuzumab shows different affinity toward erbB2 on JIMT-1 and SKBR-3 cells. a: Trastuzumab-specific fluorescence is much lower on JIMT-1 cells than on SKBR-3 cells. Confocal images of both cell lines were taken with identical instrumental settings. Images are from 120- × 120-μm2 areas. The optical slice was 1 μm. b: The dissociation constant of trastuzumab determined from saturation curves was 10 times higher for JIMT-1 than for SKBR-3.

Download figure to PowerPoint

thumbnail image

Figure 4. Activation of erbB1 and erbB2 by trastuzumab- and EGF-coupled microspheres and trastuzumab in solution on SKBR-3 and JIMT-1 cells. SKBR-3 and JIMT-1 cells were stimulated with trastuzumab-coupled microspheres, trastuzumab in solution, or EGF-coupled microspheres for 10 min. Red channel, activated erbB1 (left) or activated erbB2 (right). Reflection gray-scale images show distribution of the microspheres as bright spots. Green channel, trastuzumab in solution binding to the cells. The P-erbB1signal in all panels and the P-erbB2 in the case of JIMT-1 cells stimulated with trastuzumab in solution or EGF microspheres are displayed at four times amplification. Scale bars = 10 μm.

Download figure to PowerPoint

thumbnail image

Figure 5. Activation and transactivation of erbB2 on SKBR-3 and JIMT-1 cells quantitated by digital image processing. a: ErbB2 phosphorylation quantitated by immunofluorescence and compared for SKBR-3 and JIMT-1 stimulated with trastuzumab coupled to microspheres, trastuzumab in solution, and EGF microspheres. The total background-corrected signal was averaged and plotted for the whole cell (white columns) and for areas under (gray columns) and outside (black columns) the microspheres or fluorescence signal from directly labeled trastuzumab. b: Proportion of pixels above resting ErbB2 phosphorylation levels determined for trastuzumab- or microsphere-covered areas (white columns) and for areas outside the stimulated regions (black columns). For comparison of cells treated with soluble trastuzumab, maximal resting P-ErbB2 levels were determined from nonstimulated cells and from cells treated with non-coupled microspheres for comparing microsphere-stimulated cells. Averages from at least 10 cells from three independent experiments are plotted with standard deviations represented as error bars.

Download figure to PowerPoint

Transactivation of erbB2 Via Ligand Bound erbB1 Depends on Receptor Density

Because erbB2 could be activated by trastuzumab-linked microspheres in SKBR-3 and JIMT-1, we examined whether transactivation of erbB1 and erbB2 occurred on these cell lines. A lack of trastuzumab microspheres failed to activate erbB1 (Fig. 4), a result consistent with that obtained on A4erbB2-mYPF cells. The cross-correlations of bead location and P-erbB1 signal in the images shown were 0.26 and 0.16 for SKBR-3 and JIMT-1, respectively. Trastuzumab in solution was also unable to evoke erbB1 phosphorylation (Fig. 4), whereas EGF-linked microspheres were effective in directly activating erbB1 on SKBR-3 and JIMT-1, albeit at very low levels. The signal intensity of P-erbB1 immunofluorescence was roughly proportional to the level of erbB1 expression on the two cell lines, whereas the fraction of pixels under EGF microspheres that showed erbB1 phosphorylations above resting levels were 59 ± 22% and 43 ± 19% (n > 10) for SKBR-3 and JIMT-1 respectively, comparable to the results with trastuzumab microspheres. The very high degree of P-erbB2 activation seen in the case of A4erbB2-mYFP cells could not be detected at the EGF-linked microsphere loci on SKBR-3 and JIMT-1 cells. Nevertheless, phospho-erbB2 was generated underneath these microspheres (Fig. 4) and the P-erbB2 signal intensity in SKBR-3 and JIMT-1 reached 40% and 25%, respectively, of that generated by trastuzumab microspheres (Fig. 5a). In addition, the proportion of pixels under microspheres that exhibited erbB2 phosphorylation above resting levels was significantly higher than outside the microspheres (Fig. 5b). The differences in the activation of erbB2 through erbB1 in the cell lines examined appeared to be proportional to the differences in erbB1 expression levels. We conclude that erbB2 transactivation in cells with similar erbB2 levels is dependent on the expression level of erbB1.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Several lines of evidence indicate that cross-talk and transmodulation among members of the erbB receptor family (7, 11, 12, 40) and the immediate molecular environment of erbB receptors (17, 41–43) are important factors in the signaling processes that ultimately determine the proliferation and survival state of many epithelial cells. It is believed that interactions among members of the erbB family are sensitively regulated by their cell surface expression levels and the corresponding interactions with the palette of ligands that efficiently activate signaling pathways originating from each specific erbB kinase. Our microsphere-based stimulation studies support this supposition, inasmuch as (a) strong transactivation from ligand-bound erbB1 to erbB2 on cells expressing high levels of both receptors was detected and (b) a lesser degree of erbB2 (trans)activation occurred despite effective activation of erbB1 by EGF beads on cells with similar levels of erbB2 but an order of magnitude lower density of erbB1. We did not detect transphosphorylation of erbB1 by trastuzumab-stimulated erbB2, indicating that activation by this antibody may not promote heterodimer formation and transphosphorylation.

EGF- and trastuzumab-coupled paramagnetic microspheres are excellent tools for applying spatially restricted and temporally synchronized specific stimuli to plasma membranes. On the cells examined, receptor activation, measured by in situ immunofluorescence of specific and nonspecific phosphorylation, was restricted to the loci of the microspheres and did not propagate to other membrane domains rich in the same receptor. Activation initiates internalization of the receptors, which, however, is hindered by covalent linkage of the ligands to the microspheres, such that a membrane pocket is formed about them. The signals specific for the YFP-tagged receptor and for phosphotyrosine colocalized, in agreement with earlier findings using EGF-coupled microspheres (31,32).

We have compared a trastuzumab resistant cell line, JIMT-1, that expresses erbB2 receptors at a level 50% that of the trastuzumab sensitive line SKBR-3. Trastuzumab binds poorly to JIMT-1 with a Kd 10 times that of SKBR-3. Although trastuzumab in solution caused a high degree of erbB2 activation in SKBR-3, the phosphorylation of erbB2 induced in JIMT-1, although detectable, was minimal. Nonetheless, activation of erbB2 by trastuzumab-coupled magnetic microspheres was equally effective in both cell lines. We recently reported (17) that JIMT-1 overexpresses MUC4, a sialomucin complex with tandem EGF homologue domains potentially responsible for enhanced metastatic activity and escape from the immune surveillance (43) and presumed to be responsible for the decreased trastuzumab binding. Here we have demonstrated that applying paramagnetic microspheres, the MUC4 coat on the surface of JIMT-1 cells can be perturbed such that erbB2 surface receptors are activated. Because none of our cell lines were activated using control paramagnetic microspheres lacking the ligand, we conclude that the trastuzumab coupled to the microsphere was responsible for this activation, with the high ligand density and perhaps steric effects of the microspheres being the basis for overcoming the hindered accessibility of erbB2.


  1. Top of page
  2. Abstract
  6. Acknowledgements

We thank Dr. Janine Post for construction of the erbB2-mYFP plasmid and Dr. Diane Lidke for pertinent preliminary experiments. We thank Gudrun Heim and Daniela Reichhardt for expert technical assistance in these experiments.


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A, McGuire WL. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1987; 235: 177182.
  • 2
    Baselga J, Tripathy D, Mendelsohn J, Baughman S, Benz CC, Dantis L, Sklarin NT, Seidman AD, Hudis CA, Moore J, et al. Phase II study of weekly intravenous recombinant humanized anti-p185HER2 monoclonal antibody in patients with HER2/neu-overexpressing metastatic breast cancer. J Clin Oncol 1996; 14: 737744.
  • 3
    Cobleigh MA, Vogel CL, Tripathy D, Robert NJ, Scholl S, Fehrenbacher L, Wolter JM, Paton V, Shak S, Lieberman G, et al. Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2-overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 1999; 17: 26392648.
  • 4
    Slamon D, Pegram M. Rationale for trastuzumab (Herceptin) in adjuvant breast cancer trials. Semin Oncol 2001; 28( suppl 3): 1319.
  • 5
    Vogel CL, Cobleigh MA, Tripathy D, Gutheil JC, Harris LN, Fehrenbacher L, Slamon DJ, Murphy M, Novotny WF, Burchmore M, et al. First-line Herceptin monotherapy in metastatic breast cancer. Oncology 2001; 61( suppl 2): 3742.
  • 6
    Sliwkowski MX, Lofgren JA, Lewis GD, Hotaling TE, Fendly BM, Fox JA. Nonclinical studies addressing the mechanism of action of trastuzumab (Herceptin). Semin Oncol 1999; 26( suppl 12): 6070.
  • 7
    Nagy P, Jenei A, Damjanovich S, Jovin TM, Szöllösi J. Complexity of signal transduction mediated by ErbB2: clues to the potential of receptor-targeted cancer therapy. Pathol Oncol Res 1999; 5: 255271.
  • 8
    Baselga J. Herceptin alone or in combination with chemotherapy in the treatment of HER2-positive metastatic breast cancer: pivotal trials. Oncology 2001; 61( suppl 2): 1421.
  • 9
    Motoyama AB, Hynes NE, Lane HA. The efficacy of ErbB receptor-targeted anticancer therapeutics is influenced by the availability of epidermal growth factor-related peptides. Cancer Res 2002; 62: 31513158.
  • 10
    Austin C, De Maziere A, Pisacane P, van Dijk S, Eigenbrot C, Sliwkowski M, Klumperman J, Scheller R. Endocytosis and sorting of ErbB2 and the site of action of cancer therapeutics trastuzumab and geldanamycin. Mol Biol Cell 2004; 15: 52685282.
  • 11
    Yarden Y, Sliwkowski MX. Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2001; 2: 127137.
  • 12
    Vereb G, Nagy P, Park JW, Szöllősi J. Signaling revealed by mapping molecular interactions: implications for ErbB-targeted cancer immunotherapies. Clin Appl Immunol Reviews 2002; 2: 169186.
  • 13
    Hudziak RM, Lewis GD, Winget M, Fendly BM, Shepard HM, Ullrich A. p185HER2 monoclonal antibody has antiproliferative effects in vitro and sensitizes human breast tumor cells to tumor necrosis factor. Mol Cell Biol 1989; 9: 11651172.
  • 14
    Nagata Y, Lan KH, Zhou X, Tan M, Esteva FJ, Sahin AA, Klos KS, Li P, Monia BP, Nguyen NT, et al. PTEN activation contributes to tumor inhibition by trastuzumab, and loss of PTEN predicts trastuzumab resistance in patients. Cancer Cell 2004; 6: 117127.
  • 15
    Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors modulate in vivo cytoxicity against tumor targets. Nat Med 2000; 6: 443446.
  • 16
    Carraway KL, Price-Schiavi SA, Komatsu M, Jepson S, Perez A, Carraway CA. Muc4/sialomucin complex in the mammary gland and breast cancer. J Mammary Gland Biol Neoplasia 2001; 6: 323337.
  • 17
    Nagy P, Friedlander E, Tanner M, Kapanen AI, Carraway KL, Isola J, Jovin TM. Decreased accessibility and lack of activation of ErbB2 in JIMT-1, a Herceptin-resistant, MUC4-expressing breast cancer cell line. Cancer Res 2005; 65: 473482.
  • 18
    Tanner M, Kapanen AI, Junttila T, Raheem O, Grenman S, Elo J, Elenius K, Isola J. Characterization of a novel cell line established from a patient with Herceptin-resistant breast cancer. Mol Cancer Ther 2004; 3: 15851592.
  • 19
    Greenfield LJJr, Sun F, Neelands TR, Burgard EC, Donnelly JL, MacDonald RL. Expression of functional GABAA receptors in transfected L929 cells isolated by immunomagnetic bead separation. Neuropharmacology 1997; 36: 6373.
  • 20
    Cook EB, Stahl JL, Lowe L, Chen R, Morgan E, Wilson J, Varro R, Chan A, Graziano FM, Barney NP. Simultaneous measurement of six cytokines in a single sample of human tears using microparticle-based flow cytometry: allergics vs. non-allergics. J Immunol Methods 2001; 254: 109118.
  • 21
    Segal G, Lee W, Arora PD, McKee M, Downey G, McCulloch CA. Involvement of actin filaments and integrins in the binding step in collagen phagocytosis by human fibroblasts. J Cell Sci 2001; 114: 119129.
  • 22
    Arora PD, Silvestri L, Ganss B, Sodek J, McCulloch CA. Mechanism of cyclosporin-induced inhibition of intracellular collagen degradation. J Biol Chem 2001; 276: 1410014109.
  • 23
    Bausch AR, Ziemann F, Boulbitch AA, Jacobson K, Sackmann E. Local measurements of viscoelastic parameters of adherent cell surfaces by magnetic bead microrheometry. Biophys J 1998; 75: 20382049.
  • 24
    Tanimura N, Nagafuku M, Minaki Y, Umeda Y, Hayashi F, Sakakura J, Kato A, Liddicoat DR, Ogata M, Hamaoka T, et al. Dynamic changes in the mobility of LAT in aggregated lipid rafts upon T cell activation. J Cell Biol 2003; 160: 125135.
  • 25
    Kovacs B, Maus MV, Riley JL, Derimanov GS, Koretzky GA, June CH, Finkel TH. Human CD8+ T cells do not require the polarization of lipid rafts for activation and proliferation. Proc Natl Acad Sci USA 2002; 99: 1500615011.
  • 26
    Shah M, Patel K, Fried VA, Sehgal PB. Interactions of STAT3 with caveolin-1 and heat shock protein 90 in plasma membrane raft and cytosolic complexes. Preservation of cytokine signaling during fever. J Biol Chem 2002; 277: 4566245669.
  • 27
    Ko KS, Arora PD, McCulloch CA. Cadherins mediate intercellular mechanical signaling in fibroblasts by activation of stretch-sensitive calcium-permeable channels. J Biol Chem 2001; 276: 3596735977.
  • 28
    Dai Z, Luo X, Xie H, Peng HB. The actin-driven movement and formation of acetylcholine receptor clusters. J Cell Biol 2000; 150: 13211334.
  • 29
    MacGillivray MK, Cruz TF, McCulloch CA. The recruitment of the interleukin-1 (IL-1) receptor-associated kinase (IRAK) into focal adhesion complexes is required for IL-1beta–induced ERK activation. J Biol Chem 2000; 275: 2350923515.
  • 30
    Wiederkehr-Adam M, Ernst P, Muller K, Bieck E, Gombert FO, Ottl J, Graff P, Grossmuller F, Heim MH. Characterization of phosphopeptide motifs specific for the Src homology 2 domains of signal transducer and activator of transcription 1 (STAT1) and STAT3. J Biol Chem 2003; 278: 1611716128.
  • 31
    Brock R, Jovin TM. Heterogeneity of signal transduction at the subcellular level: microsphere-based focal EGF receptor activation and stimulation of Shc translocation. J Cell Sci 2001; 114: 24372447.
  • 32
    Brock R, Jovin TM. Quantitative image analysis of cellular protein translocation induced by magnetic microspheres: application to the EGF receptor. Cytometry 2003; 52A: 111.
  • 33
    Verveer PJ, Wouters FS, Reynolds AR, Bastiaens PI. Quantitative imaging of lateral ErbB1 receptor signal propagation in the plasma membrane. Science 2000; 290: 15671570.
  • 34
    Lidke DS, Lidke KA, Rieger B, Jovin TM, Arndt-Jovin DJ. Reaching out for signals: Filopodia sense EGF and respond by directed retrograde transport of activated receptors. J Cell Biol 2005; 170: 619626.
  • 35
    Kempiak SJ, Yip SC, Backer JM, Segall JE. Local signaling by the EGF receptor. J Cell Biol 2003; 162: 781787.
  • 36
    Lidke DS, Nagy P, Heintzmann R, Arndt-Jovin DJ, Post JN, Grecco H, Jares-Erijman EA, Jovin TM. Quantum dot ligands provide new insights into erbB/HER receptor-mediated signal transduction. Nat Biotechnol 2004; 22: 198203.
  • 37
    Vereb G, Matkó J, Vámosi G, Ibrahim SM, Magyar E, Varga S, Szöllősi J, Jenei A, Gáspár RJr, Waldmann TA, et al. Cholesterol-dependent clustering of IL-2Ralpha and its colocalization with HLA and CD48 on T lymphoma cells suggest their functional association with lipid rafts. Proc Natl Acad Sci USA 2000; 97: 60136018.
  • 38
    Vereb G, Szöllősi J, Damjanovich S, Matkó J. Exploring membrane microdomains and functional protein clustering in live cells with flow and image cytometric methods. In: GeddesCD, LakowiczJR, editors. Reviews in fluorescence. New York: Kluwer Academic/Plenum; 2004. p 99120.
  • 39
    Bagossi P, Horváth G, Vereb G, Szöllősi J, Tőzsér J. Molecular modeling of nearly full-length ErbB2 receptor. Biophys J 2005; 88: 13541363.
  • 40
    Citri A, Skaria KB, Yarden Y. The deaf and the dumb: the biology of ErbB-2 and ErbB-3. Exp Cell Res 2003; 284: 5465.
  • 41
    Camirand A, Lu Y, Pollak M. Co-targeting HER2/ErbB2 and insulin-like growth factor-1 receptors causes synergistic inhibition of growth in HER2-overexpressing breast cancer cells. Med Sci Monit 2002; 8: BR521BR526.
  • 42
    Nagy P, Vereb G, Sebestyén Z, Horváth G, Lockett SJ, Damjanovich S, Park JW, Jovin TM, Szöllősi J. Lipid rafts and the local density or ErbB proteins influence the biological role of homo- and heteroassociations of ErbB2. J Cell Sci 2002; 115: 42514262.
  • 43
    Price-Schiavi SA, Andrechek E, Idris N, Li P, Rong M, Zhang J, Carraway CA, Muller WJ, Carraway KL. Expression, location, and interactions of ErbB2 and its intramembrane ligand Muc4 (sialomucin complex) in rat mammary gland during pregnancy. J Cell Physiol 2005; 203: 4453.