Advertisement

Selection of Potentially Metastatic Subpopulations Expressing c-erbB-2 from Breast Cancer Tissue by Use of an Extravasation Model

      Overexpression of the c-erbB-2 gene-encoded p185c–erbB-2 is correlated with early onset of metastasis in breast cancer patients. Furthermore, the detection of blood-borne epithelium-derived clustered cells expressing p185c–erbB-2 was related to advanced stages in breast cancer. To further elucidate the receptor's function in the metastatic process of human breast cancers, we analyzed disaggregated cells and cell clusters from freshly dissected breast cancer tissues. We studied whether their capability of extravasation is correlated with their expression of c-erbB-2. A model for the venular wall was constructed by growing human umbilical vein endothelial cells (HUVECs) on porous membranes coated with basement membrane extracellular matrix. In four control breast cancer cell lines (SK-BR-3, MCF-7, MDA-MB-468, and MDA-MB-468, the latter transfected with a full-length c-erbB-2 cDNA vector) producing different levels of the c-erbB-2 receptor, the expression level correlated positively with the invasiveness of the cells. The invasive, predominantly clustered cells from 14 of 23 tumors were positively stained for p185c–erbB-2 by immunocytochemistry. Furthermore, we show that the invasive cell populations express the metastasis-associated proteins matrix metalloproteinase MMP-2, CD44, and integrins αvβ3 and α6. In this first study on the behavior of cells and cell clusters from disaggregated operated cancers in an extravasation model, we could demonstrate the presence of c-erbB-2-expressing cell subpopulations within the individual breast cancers that are presumably of high metastatic potential.
      Breast cancer metastasis is the major cause of death for patients with breast carcinomas. Metastasis is viewed as a highly selective competition, favoring the survival of a subpopulation of metastatic tumor cells that pre-exists within the heterogeneous primary tumor.
      • Fidler IJ
      • Hart IR
      Biological diversity in metastatic neoplasms: origins and implications.
      • Liotta LA
      • Steeg PS
      • Stetler-Stevenson WG
      Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation.
      Early micrometastasis in patients with small resectable cancer poses a great problem for the treatment of cancer.
      • Schlimok G
      • Funke I
      • Holzmann B
      • Göttlinger G
      • Schmidt G
      • Häuser H
      • Swierkot S
      • Warnecke HH
      • Schneider B
      • Koprowski H
      • Riethmüller G
      Micrometastatic cancer cells in bone marrow: in vitro detection with anti-cytokeratin and in vivo labeling with anti-17-1A monoclonal antibodies.
      Hence, it is of major importance to understand the molecular mechanisms of cellular processes essential for cancer metastasis as a basis for new therapeutic approaches.
      The c-erbB-2 (HER-2/neu) proto-oncogene encodes a receptor tyrosine kinase, p185c–erbB-2 or p185neu, that shares extensive sequence homology with epidermal growth factor receptor (EGFR).
      • Coussens L
      • Yang-Feng TL
      • Liao Y-C
      • Chen E
      • Gray A
      • McGrath J
      • Seeburg PH
      • Libermann TA
      • Schlessinger J
      • Francke U
      • Levinson A
      • Ullrich A
      Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with Neu oncogene.
      Like EGFR, c-erbB-2 is expressed in various fetal and adult epithelia and is believed to play an important role in growth and development.
      • Press MF
      • Cordon-Cardo C
      • Slamon DJ
      Expression of the HER-2/neu proto-oncogene in normal human adult and fetal tissues.
      In cancer, oncogenic amplification and/or overexpression of c-erbB-2 is found in many different human primary tumors.
      • Brandt BH
      • Vogt U
      • Schlotter CM
      • Jackisch C
      • Werkmeister R
      • Thomas M
      • von Eiff M
      • Bosse U
      • Assmann G
      • Zänker KS
      Prognostic relevance of aberrations in the erbB oncogenes from breast, ovarian, oral and lung cancers: double-differential polymerase chain reaction (ddPCR) for clinical diagnosis.
      In human breast cancer, numerous studies have reported c-erbB-2 amplification and overexpression,
      • Yu D
      • Hung M-C
      The HER-2/neu gene in human cancers.
      and some of them found a positive correlation with earlier relapse and poorer overall survival of the patient.
      • 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.
      • Thor AD
      • Schwartz LH
      • Koerner FC
      • Edgerton SM
      • Skates SJ
      • Yin S
      • McKenzie SJ
      • Panicali DL
      • Marks PJ
      • Fingert HJ
      Analysis of c-erbB-2 expression in breast carcinomas with clinical follow-up.
      • Wright C
      • Angus B
      • Nicholson S
      • Sainsbury JR
      • Cairns J
      • Gullick WJ
      • Kelly P
      • Harris AL
      • Horne CH
      Expression of c-erbB-2 oncoprotein: a prognostic indicator in human breast cancer.
      In addition to the reported clinical correlations, experimental approaches using animal and in vitro models have provided evidence that the c-erbB-2 oncogene plays an important role in cancer metastasis.
      • Yu D
      • Hung M-C
      The HER-2/neu gene in human cancers.
      Yu and co-workers found that expression of c-erbB-2 promotes the invasion steps in the metastatic cascade in human lung and breast cancer cells, such as increased motility, migration through the extracellular matrix, and secretion of enzymes degrading the basement membrane.
      • Yu D
      • Wang S-S
      • Dulski K
      • Tsai C-M
      • Nicolson GL
      • Hung M-C
      c-erbB-2/neu overexpression enhances metastatic potential of human lung cancer cells by induction of metastasis-associated properties.
      • Tan M
      • Yao J
      • Yu D
      Overexpression of the c-erbB-2 gene enhanced intrinsic metastasis potential in human breast cancer cells without increasing their transformation abilities.
      In a recently published study by Verbeek et al, overexpression of c-erbB-2 was related to random cell migration.
      • Verbeek BS
      • Adriaansen-Slot SS
      • Vroom TM
      • Beckers T
      • Rijksen G
      Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts.
      It was also shown that p185c–erbB-2 interacts with members of the laminin receptor family (α6β4 and α6β1 integrins) and that this interaction might also contribute to generate a locomotive phenotype of carcinoma cells.
      • Falcioni R
      • Antonini A
      • Nisticò P
      • Di Stefano S
      • Crescenzi M
      • Natali PG
      • Sacchi A
      α6β4 and α6β1 integrins associate with ErbB-2 in human carcinoma cell lines.
      In a previous study in our laboratory we observed that a high-risk group of breast cancer patients expressed p185c–erbB-2within the primary tumor and on blood-borne epithelium-derived cells.
      • Brandt B
      • Roetger A
      • Heidl S
      • Jackisch C
      • Lellé RJ
      • Assmann G
      • Zänker KS
      Isolation of blood-borne epithelial derived c-erbB-2 oncoprotein positive clustered cells from the peripheral blood of breast cancer patients.
      The onset of progressive disease was related to the occurrence of blood-borne epithelium-derived cells positively stained for c-erbB-2 receptor protein. Here we report on the evaluation of the biological features of p185c–erbB-2-positive cells and cell clusters from fresh breast cancer tissue by in vitro extra-vasation experiments. For our study, we designed an in vitro model for the venular wall that consists of an endothelial monolayer of human umbilical vein endothelial cells (HUVECs) growing on porous membranes covered with extracellular matrix basement membrane. Thus, the transendothelial penetration by breast cancer cells followed by invasion of the underlying basement membrane can be examined. We evaluated our in vitro model using four breast cancer cell lines expressing different levels of p185c–erbB-2. One of them was transfected with the human c-erbB-2 cDNA. We then tested cells from disaggregated surgical breast tissue in our model to explore the invasion capacity of cells from benign and malignant breast tissues and the role of c-erbB-2 for their invasive potential. For the breast cancer cell lines we found that the level of c-erbB-2 expression correlated positively with the cells' extravasation potential. Our results for fresh breast cancer tissues provide evidence that c-erbB-2 expression is characteristic of cell populations with high locomotive capability that pre-exist within the primary tissues. Further immunocytochemical analysis of the invasive cell populations revealed the expression of proteins that are likely to be involved in the metastatic invasion process (matrix metalloproteinase MMP-2, CD44, and integrins αvβ3 and α6). We conclude that we can select cell subpopulations in breast cancer tissues that are presumably of high metastatic potential.

      Materials and Methods

      Cell Lines and Culture

      HUVECs were isolated from human umbilical cord veins as described by Gimbrone et al
      • Gimbrone Jr, MA
      • Cotran RS
      • Folkman J
      Human vascular endothelial cells in culture: growth and DNA synthesis.
      with modifications according to Friedl et al.
      • Friedl P
      • Tatje D
      • Czapla R
      An optimized culture medium for human endothelial cells from human umbilical veins.
      The cells were grown on gelatin-coated flasks and passaged four to six times in a medium containing equal volumes of Iscove's modified Dulbecco's medium (IMDM) and Ham's F12 nutrient mixture (Life Technologies, Eggenstein, Germany) supplemented with 10 μg/ml sodium heparin (Boehringer Ingelheim, Heidelberg, Germany), 5 μg/ml transferrin, 2.5 ng/ml basic fibroblast growth factor (bFGF; Sigma, Deisenhofen, Germany), 5 μmol/L β-mercaptoethanol, 2 mmol/L l-glutamine (Life Technologies), 100 U/ml penicillin, 100 μg/ml streptomycin, 250 ng/ml amphotericin B (Sigma), and 15% fetal calf serum (FCS; PAA Laboratories, Linz, Austria).
      Breast cancer cell lines MCF-7, SK-BR-3, and MDA-MB-468 were obtained from the American Type Culture Collection (ATCC, Rockville, MD) and cultured in Dulbecco's modified Eagle's medium (DMEM; ICN, Eschwege, Germany) supplemented with 2 mmo/L l-glutamine, antibiotic drugs as above, and 10% FCS. For culture of transfectants, cells were grown under the same conditions except for addition of 400 μg/ml Geneticin (G418) (Sigma) to the culture medium.

      Cell Transfection and Selection

      MDA-MB-468 cells were stably transfected with the plasmid vector pCVN/HER-2 (generously provided by A. Ullrich, Martinsried, Germany) using the LIPOFECTIN-Reagent (Life Technologies) as described in the technical protocol. For control experiments, cells were transfected with plasmid vector pCVN lacking the c-erbB-2 cDNA insert. Two days after transfection, cells were harvested and selected with 400 μg/ml G418. For a second selection step, G418-resistant cells were positively sorted using immunomagnetic beads (Dynabeads, Dynal, Hamburg, Germany) coated with p185c–erbB-2-specific antibody c-neu (Ab-5, Oncogene Research Products, Cambridge, MA). Sorted MDA-MB-468/HER-2 cells were >99% p185c–erbB-2-positive as detected by FACS analysis (FACScalibur, Becton-Dickinson, Heidelberg, Germany).

      Patients, Tissue Collection, and Disaggregation

      Human mammary tissue specimens were received from the operating room in sterile tubes containing DMEM with 10% FCS and antibiotic drugs and were kept on ice until disaggregation. We examined 20 primary breast carcinomas (5 stage I and 15 stage IIA/IIIB) and 3 lymph node metastases (2 stage IIIB and 1 stage IV). Tumor histology was classified according to conventional criteria, and all identifiable lymph nodes were histologically examined. Mammary tissues from five patients with benign breast diseases (four with cystic mastopathia and one with fibroadenoma) were chosen as negative controls. Breast tissue was mechanically disaggregated by means of automated tissue disaggregator Medimachine (DAKO, Hamburg, Germany). For this purpose, tumor tissue was cut into small pieces and placed into a disposable disaggregation chamber (Medicon; DAKO) together with 1.5 ml of serum-free invasion medium (DMEM with 2 mmol/L l-glutamine, antibiotic drugs, and 0.1% bovine serum albumin (BSA; Sigma). The Medicon was inserted into the motor unit of the machine and run for 2 minutes at a pre-fixed rotation speed of approximately 80 rpm. An aliquot of the cell suspension containing single cells and cell clusters up to 30 cells was examined microscopically for cell viability by Trypan blue exclusion. Cells were counted in a Neubauer hematocytometer.

      Invasion Assays

      Originally, our in vitro invasion assay was based on the procedure of Albini et al.
      • Albini A
      • Iwamoto Y
      • Kleinman HK
      • Martin GR
      • Kozlowski JM
      • McEwan RN
      A rapid in vitro assay for quantitating the invasive potential of tumor cells.
      Cell culture inserts with 8-μm porous polyethylene terephthalate (PET) membranes were placed in 12-well plastic tissue culture plates (Becton Dickinson Labware, Franklin Lakes, NY). The membranes were coated with basement membrane extracellular matrix (ECM; Harbor Bio-Products, Norwood, MA) at a concentration of 125 μg/cm2 by drying an appropriate ECM dilution overnight under a laminar flow hood. Dried ECM was rehydrated with 500 μl of HUVEC culture medium (see above) for 1 hour. HUVECs were seeded onto the coated membranes in a concentration of 2 × 105 cells/well. After culturing for 2 days at 37°C in a humidified atmosphere of 5% CO2, HUVECs formed confluent monolayers, which was verified by panoptic staining (Diff-Quik; Baxter Health Care Co., Miami, FL). Cell culture inserts were used for up to 3 days after endothelial cells reached confluence.
      For the invasion assays of breast cancer cell lines, cells were harvested with 0.25% trypsin/2 mmol/L EDTA (Life Technologies) and adjusted to a density of 2 × 105 cells/ml with serum-free invasion medium (see above), and 2 × 105 cells were placed onto the HUVEC monolayer on the ECM-coated membrane. The invasion assays of primary breast cancer cells were performed applying approximately 106 disaggregated cells to the membrane. The invasion medium was placed into the wells under the bottom sides of the membranes as well. Invasion assays were incubated for 48 hours at 37°C in 5% CO2.
      At the end of the invasion assay, the HUVEC monolayer and noninvading cells on the upper surface of the membrane were removed by cotton swabs and thorough rinsing with phosphate-buffered saline (PBS; pH 7.4). Invading cells on the bottom side of the membrane were fixed in 4% paraformaldehyde and characterized using double immunocytochemistry (see below).

      Immunocytochemistry

      Membranes with invasive cells were double stained by applying a combined immunogold-enzymatic technique according to Riesenberg et al
      • Riesenberg R
      • Oberneder R
      • Kriegmair M
      • Epp M
      • Bitzer U
      • Hofstetter A
      • Braun S
      • Riethmüller G
      • Pantel K
      Immunocytochemical double staining of cytokeratin and prostate specific antigen in individual prostatic tumor cells.
      with slight modifications. All antibodies were diluted in PBS containing 10% AB serum (Biotest, Dreieich, Germany) and 0.1% acetylated BSA (BSA-C; Aurion, Wageningen, The Netherlands), and each incubation step was followed by three 3-minute washes with PBS.
      After fixation (see above), cells were permeabilized by incubation in 0.1% Triton X-100 in PBS for 10 minutes. Cells were blocked for 20 minutes in 10% AB serum with 0.1% BSA-C. Cells were incubated with rabbit polyclonal antibody c-erbB-2 oncoprotein (DAKO), followed by incubation with 5-nm-gold-conjugated goat anti-rabbit antibody (Paesel & Lorei, Hanau, Germany). After a second fixation step with 2% glutaraldehyde in PBS and a second blocking step, mouse monoclonal biotinylated antibody to human cytokeratin 8 (Progen, Heidelberg, Germany) was applied. Samples were then incubated with alkaline-phosphatase-conjugated streptavidin (Jackson ImmunoResearch, West Grove, PA).
      For detection of metastasis-associated proteins, mouse monoclonal antibodies to matrix metalloproteinase MMP-2 (Oncogene Research Products), to CD44 (phagocytic glycoprotein-1 or hyaluronic acid receptor; Chemicon, Temecula), and integrin αvβ3 (vitronectin receptor; Chemicon), and to integrin α6 (component of the laminin receptors; Novocastra Laboratories, New Castle, UK) were applied. Samples were then incubated with rabbit anti-mouse immunoglobulins (DAKO) as a link antibody, followed by the next incubation step with a complex of calf intestinal alkaline phosphatase and mouse monoclonal anti-alkaline phosphatase (APAAP; DAKO).
      The antibody binding to cytokeratin, MMP-2, CD44, and integrins αvβ3 and α6, respectively, was visualized immunoenzymatically by using the DAKO Newfuchsin substrate system (DAKO) according to the manufacturer's protocol. Silver enhancement of colloidal gold particles was performed with a silver enhancement kit (IntenSE) purchased from Amersham (Braunschweig, Germany). After rinsing the samples with distilled water, freshly prepared silver enhancement mixture was applied for ∼20 minutes while monitoring the reaction under the microscope. To abrogate the enhancement reaction, membranes were rinsed with distilled water. Samples were counterstained using Mayer's hematoxylin (Merck, Darmstadt, Germany). Finally, membranes were separated from the insert assembly and mounted with Aquamount improved (BDH Laboratory Supplies, Poole, UK) on microscope glass slides. To exclude nonspecific staining, unrelated rabbit IgG (Sigma) and mouse myeloma proteins (MOPC21; Sigma) served as a negative control for immunocytochemistry. Enzymatic and immunogold staining with silver enhancement was viewed by light microscopy and epipolarization using a fluorescent microscope (Laborlux S, Leica, Wetzlar, Germany) with an IGS filter (Leica).

      Transmission Electron Micrography

      For transmission electron microscopy (TEM), PET membranes with HUVEC monolayers were washed once with PBS and fixed with 2% glutaraldehyde. The samples were embedded in Epon 812, and ultra-thin sections were mounted on 200-mesh copper grids. The specimens were examined with a Philips EM at 60 kV.

      Western Blot Analysis

      For preparation of cell lysates, confluent cell monolayers were washed three times with PBS and scraped from 75-cm2 tissue culture flasks in ice-cold lysis buffer containing 150 mmol/L NaCl, 5 mmol/L EDTA, 10 mmol/L Tris/HCl (pH 7.2), 0.1% SDS, 0.1% sodium deoxycholate, 1% Triton X-100, and protease inhibitors (cocktail tablets Complete; Boehringer, Mannheim, Germany). Breast tissue was homogenized in the same lysis buffer. After 15 minutes on ice, the lysate was centrifuged at 4°C and 12,000 × g for 1 hour. Protein concentration was determined using the BCA protein assay reagent (Pierce, Rockford, IL), and defined amounts of protein were electrophoresed on a 7.5% SDS-polyacrylamide gel. After electrophoresis, proteins were transferred onto a polyvinylidenfluoride membrane (Roth, Karlsruhe, Germany), and nonspecific binding sites were blocked with 10% nonfat dry milk (De-Vau-Ge Gesundkostwerk, Lüneburg, Germany) in PBS with 0.1% Tween 20. Blots were probed with antibody c-neu (Ab-3) for detection of p185neu (Oncogene Research Products), followed by incubation with horseradish-peroxidase-conjugated goat anti-mouse antibody (Amersham). Both antibodies were used at a concentration of 0.1 μg/ml. Bands were visualized with the enhanced chemiluminescence (ECL) system (Amersham) with an exposure time of 3 minutes.

      Results

      Evaluation of an in Vitro Extravasation Model Using Breast Cancer Cell Lines with Different c-erbB-2 Expression

      As a system that mimics the in vivo situation in blood capillaries, we used an in vitro model consisting of a porous PET membrane coated with extracellular matrix and a monolayer of HUVECs (Figure 1, A–C). Confluence of HUVEC monolayers was verified by panoptic staining (Figure 1A). The size proportions of the HUVEC monolayer (1), basement membrane layer (2), and PET membrane (3) are revealed by TEM (Figure 1C).
      Figure thumbnail gr1
      Figure 1The model for extravasation. A: Density of the HUVEC monolayer grown for 48 hours on the upper side of the ECM-coated membrane is demonstrated by panoptic staining. Scale bar, 100 μm. B:Schematic drawing of the model. C: Cross section of the model vessel wall observed by transmission electron microscopy. Scale bar, 10 μm. 1, HUVEC monolayer; 2, extracellular matrix basement membrane; 3, PET membrane.
      To evaluate the selection capacity of the model, a panel of breast cancer cell lines (SK-BR-3, MCF-7, and MDA-MB-468) expressing different levels of p185c–erbB-2 was tested for their extravasation potential. As shown in Figure 2, MDA-MB-468 cells express no detectable levels of p185c–erbB-2, and only very few cells penetrated the HUVEC monolayer and the basement membrane after incubation for 48 hours. SK-BR-3 cells expressing high p185c–erbB-2 levels showed high transendothelial invasiveness, and MCF-7 cells expressing moderate amounts of p185c–erbB-2 showed increased invasiveness compared with MDA-MB-468 cells. To ensure that differences in the extravasation capacity were due to differences in the c-erbB-2 expression levels, and not to a different genetic background, we transfected low-invasive MDA-MB-468 cells with the plasmid vector pCVN/HER2, containing the full-length wild-type human c-erbB-2 cDNA,
      • Hudziak RM
      • Schlessinger J
      • Ullrich A
      Increased expression of the putative growth factor receptor p185HER2 causes transformation and tumorigenesis of NIH 3T3 cells.
      to generate the MDA-MB-468/HER-2 transfectants. As shown in Figure 2B, their c-erbB-2 expression level is between the one of MCF-7 and SK-BR-3. We used the MDA-MB-468/neo cell line, which was established by transfecting MDA-MB-468 cells with the empty pCVN vector, as a control. As for parental MDA-MB-468 cells, p185c–erbB-2 is not detectable in MDA-MB-468/neocells. Again, we compared the c-erbB-2 expression level with the number of cells on the bottom side of the membrane and found significantly more cells from the MDA-MB-468/HER2 cell line with high c-erbB-2 expression, than MDA-MB-468/neo cells, which exhibited the same low extra-vasation capacity as the parental cells (Figure 2A).
      Figure thumbnail gr2
      Figure 2Extravasation potential of breast cancer cell lines and expression of c-erbB-2. A: A total of 2 × 105 tumor cells were placed onto the HUVEC monolayer on the ECM-coated membrane. The columns show numbers of invasive cells of indicated breast cancer cell lines after a 48-hour incubation. Each column represents an average counting from five separate experiments.B: Immunoblot analysis for the c-erbB-2-encoded p185 proteins in the cell lysates of the indicated cell lines. Protein (20 μg) from each sample was electrophoresed on a 7.5% SDS-polyacrylamide gel, transferred to PVDF membrane, and probed with a monoclonal anti-human c-neu antibody (Ab-3). The position of p185c–erbB-2(p185) is indicated.
      For visualization of c-erbB-2 expression in the highly motile SK-BR-3 and MDA-MB-468/HER-2 cells, we characterized the cells on the bottom side of the porous membrane by double immunocytochemistry for epithelial marker cytokeratin 8 and p185c–erbB-2(Figure 3). In the case of cells from disaggregated breast cancer tissue (see below), cytokeratin analysis was useful to characterize invasive cells with regard to their histological origin. As shown in Figure 3, A and C), for both c-erbB-2-overexpressing cell lines (wild-type SK-BR-3 and MDA-MB-468/HER-2), double immunocytochemistry revealed an accumulation of p185c–erbB-2 in membrane protrusions (arrows) indicated by the black silver stain, whereas cytokeratin (red color) was mainly localized in the core of the cells. In Figure 3, B and D), the same cells are viewed by epipolarization, which enhances the sensitivity of the silver staining and makes the stain shine with a fluorescent-like bluish glow against a dark background.
      Figure thumbnail gr3
      Figure 3Transendothelial invasive cells from breast cancer cell lines stained by double immunocytochemistry to cytokeratin and p185c–erbB-2. A: An invasive SK-BR-3 cell was fixed on the bottom side of the membrane. Double immunocytochemistry was performed using a mouse monoclonal biotinylated antibody to human cytokeratin 8 and a rabbit polyclonal antibody to c-erbB-2 oncoprotein. C: An invasive MDA-MB-468/HER-2 cell stained as described above. Membrane protrusions of both cell types contain less cytokeratin (red color from the Newfuchsin dye reaction) but accumulations of the c-erbB-2 receptor visualized by the dark brown staining from the silver enhancement procedure (arrows). B and D:When viewing the same cells by epipolarization, the silver granules appear in a fluorescent-like bluish glow against a dark background. Scale bars, 10 μm.

      Extravasation Capacity and c-erbB-2 Expression of Cells from Disaggregated Surgical Breast Tissues

      To investigate whether cells from fresh breast cancer tissues are capable of extravasation in our model and whether c-erbB-2 plays an important role for this phenomenon, we tested 23 mechanically disaggregated malignant tumor tissues (20 primary breast cancers and three lymph node metastases; Table 1). Using double immunocytochemistry (see above), invasive cells were characterized with regard to their epithelial origin and c-erbB-2 expression; 12 of 16 primary breast cancer tissues contained c-erbB-2-positive, predominantly clustered cells capable of invasion after 48 hours. There was no tumor that contained only single invasive cells. We never observed invasive cell clusters that were positive for p185c–erbB-2 and negative for cytokeratin. In case a tumor contained c-erbB-2-expressing invasive cell clusters, all of the cells within the invasive clusters were positive for p185c–erbB-2.
      Table 1Extravasation after 48 Hours and c-erbB-2 Expression of Invasive Cells and Fresh Breast Tissues
      Number of samples
      Tissue samplesInvasive entitiesc-erbB-2 expression of invasive entitiesp185c-erbB-2 immunoblot (tissue samples positive)
      Primary breast carcinomas (n = 20)164++128
      9+
      3(+)
      Lymph node metastases (n = 3)31++22
      1+
      1(+)
      Controls, tissues from benign breast diseases (n = 5)11(+)00
      Invasive entities = invasive clustered or single cells that stained positive for epithelial marker cytokeratin 8. (+), 1 to 10; +, 10 to 100; ++, >100 invasive entities.
      In Figure 4, invasive cells/cell clusters from surgical breast cancers characterized by double immunocytochemistry are shown. Figure 4A illustrates a single cell on the bottom side of the PET membrane. In Figure 4, C and E), typical invasive cell clusters are shown. The black silver stain on the membrane visualizes the c-erbB-2 expression. In Figure 4, B, D, and F, the cells are viewed with epipolarization (see above). In some cases we observed that the invasive cells detached from the bottom of the membrane, attached to the plastic surface of the culture well, and proliferated for several days (not shown).
      Figure thumbnail gr4
      Figure 4Transendothelial invasive cells and cell clusters from disaggregated breast cancer tissues stained by double immunocytochemistry to cytokeratin and p185c–erbB-2. A: Typical single double-positive cell on the bottom side of the membrane. Scale bar, 10 μm. C and E: Double-positive cell clusters. Scale bars, 20 μm. Co-localization of antibody binding to p185c–erbB-2 and cytokeratin is visualized by the presence of black grains on the background of a red-colored product. B,D, and F: The same cells are viewed using an epipolarization filter, which enables the discrimination of the specific luminous silver stain from nonspecific background staining.
      All carcinoma tissues that expressed detectable p185c–erbB-2 (eight primary breast carcinomas and two lymph node metastases) as determined by Western blotting on a piece of the tissue we also used for disaggregation (see Figure 5 and Table 1) contained invasive cell clusters or cells. Each of the three lymph node metastases contained invasive cells, and in two cases invasive cells/cell clusters were positive for p185c–erbB-2, too. Initially, it seems to be paradoxical that four mammary carcinoma tissues, which were negative for the c-erbB-2 protein, presented p185c–erbB-2-positive invasive cells/cell clusters in the extravasation assay. This phenomenon can be explained by the presence of only very few c-erbB-2-expressing cell populations within a heterogeneous tumor, not detectable by Western blotting, that were capable of transendothelial migration. Interestingly, four breast carcinomas that did not contain invasive cells/cell clusters were pathologically staged lymph node negative. Those carcinomas did not express c-erbB-2. For several reasons, quantification of invasive cells from fresh breast tissue was substantially more difficult than for the cell lines (see previous section). First, there were remarkable differences in cell viability after disaggregation (between 10% and 60%). Second, tumors consist of varying proportions of carcinoma cells and surrounding stromal cells, eg, fibroblasts. And third, we observed the vast majority of the invasive cells to be organized in dense clusters (see Figure 4, C and E), which made determination of cell numbers very difficult. In a semiquantitative analysis, we classified the tissues we had tested in the three categories: (+), representing 1 to 10; +, representing 10 to 100; and ++, representing >100 invasive entities, considering a cell cluster or single cell expressing the epithelial marker cytokeratin as one entity (see Table 1). Regarding the correlation between c-erbB-2 expression and motility, as observed using the cell lines (Figure 2), a similar trend was also evident for the tumor tissues applied to our extravasation model. There were five tumors with a high expression of p185c–erbB-2 as determined by Western blot analysis (breast carcinomas 10, 17, 18, and 20 and lymph node metastasis 21 in Figure 5), and four of them contained >100 invasive entities. The five tumors expressing medium or low levels of the receptor (breast carcinomas 3, 6, 13, and 14 and lymph node metastasis 23 in Figure 5) contained 10 to 100 invasive entities, whereas 1 to 10 invasive entities occurred only in tumors without detectable expression of p185c–erbB-2.
      Figure thumbnail gr5
      Figure 5Western blot analysis of expression of p185c–erbB-2 in fresh breast tissues. Protein (20 μg) from a piece of the tissue we also used for disaggregation was analyzed with a monoclonal anti-human c-neu antibody (Ab-3). The position of p185c–erbB-2(p185) is indicated.Lanes 1 to 20, protein from primary breast cancers;lanes 21 to 23, protein from lymph node metastases;lanes 24 to 28, protein from benign breast tissues;lane 29, protein from control MDA-MB-468/HER-2 cells. Breast carcinomas 3, 6, 10, 13, 14, 17, 18, and 20 and lymph node metastases 21 and 23 expressed detectable levels of p185c–erbB-2 after a 3-minute exposure using a chemiluminescence immunoblot procedure.
      We used cells from five benign tissues (four cystic mastopathias and one fibroadenoma tissue), which should not have an invasive potential as a control. Unexpectedly, there were two single cells in one of the five control benign breast tissues that were found on the bottom side of the PET membrane. This could be explained by a leakage in our model vessel wall or by the presence of a few premalignant cells in this tissue, which was histologically classified as a fibroadenoma. The first explanation is more likely, because the two invasive entities were single cells, which should pass a membrane leakage more easily than a cluster. This is also supported by the fact that neither of the cells expressed c-erbB-2. In addition, all benign samples were p185c–erbB-2-negative as shown in Figure 5. As mentioned above, in contrast to this benign tissue, there was no malignant tumor that contained only single invasive cells, but at least a combination of single and clustered cells or solely clustered cells that managed to migrate through the model vessel wall.
      For further analysis of the invasive c-erbB-2-expressing entities from fresh breast cancer tissue with regard to their biological relevance in the metastatic process, we performed immunocytochemistry with monoclonal antibodies directed against matrix metalloproteinase MMP-2, CD44 (hyaluronic acid receptor), and integrins αvβ3 and α6, which are suggested to be involved in the metastatic invasion process.
      • Liotta LA
      • Steeg PS
      • Stetler-Stevenson WG
      Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation.
      • Stetler-Stevenson WG
      • Aznavoorian S
      • Liotta LA
      Tumor cell interactions with the extracellular matrix during invasion and metastasis.
      • Sy MS
      • Liu D
      • Schiavone R
      • Ma J
      • Mori H
      • Guo Y
      Interactions between CD44 and hyaluronic acid: their role in tumor growth and metastasis.
      • Imhof BA
      • Piali L
      • Gisler RH
      • Dunon D
      Involvement of α6 and αv integrins in metastasis.
      • Brooks PC
      • Strömblad S
      • Sanders LC
      • von Schalscha TL
      • Aimes RT
      • Stetler-Stevenson WG
      • Quigley JP
      • Cheresh DA
      Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin αvβ3.
      In four of four tumors with p185c–erbB-2-expressing invasive cell clusters, the invasive entities also expressed CD44; in three of four, they expressed αvβ3 integrin; and in two of four, they expressed MMP-2 and α6 integrins.
      In Figure 6, A–D, examples of transendothelial invasive cell clusters characterized by double immunocytochemistry to p185c–erbB-2 and MMP-2, CD44, and αvβ3 and α6 integrins, respectively, are shown. Again, the brown/black silver stain visualizes the c-erbB-2 expression, whereas the red color reveals the expression of the respective metastasis-associated protein.
      Figure thumbnail gr6
      Figure 6Transendothelial invasive cell clusters from breast cancer tissues stained by double immunocytochemistry to p185c–erbB-2 and MMP-2, CD-44, αvβ3, and α6 integrins, respectively. The red color from the Newfuchsin dye reaction indicates the expression of the respective metastasis-associated protein, whereas expression of p185c–erbB-2 is visualized by the brown/black staining from the silver enhancement reaction. Scale bars, 20 μm. A:Double-positive cell cluster for MMP-2 and p185c–erbB-2.B: Double-positive cell cluster for CD44 and p185c–erbB-2. C: Double-positive cell cluster for αvβ3 integrin and p185c–erbB-2. D:Double-positive cell cluster for α6 integrins and p185c–erbB-2. The small inset in the lower right corner represents a cell cluster that was incubated with unrelated rabbit IgG and mouse myeloma proteins as negative controls for double immunocytochemistry.

      Discussion

      It is widely accepted that metastasis is not a random, but a highly selective, process that favors the survival and growth of a few subpopulations of cells that pre-exist within the parent neoplasm.
      • Fidler IJ
      Critical factors in the biology of human cancer metastasis: twenty-eighth G. H. A. Clowes Memorial Award Lecture.
      As discussed by Liotta et al,
      • Liotta LA
      • Steeg PS
      • Stetler-Stevenson WG
      Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation.
      estimating the size of the metastatic subpopulation has clinical significance, as a prognostic assay based on a sample of the primary tumor would be highly inaccurate if the aggressive subpopulation was only a small proportion of the total number of tumor cells. In previous studies in our laboratory, we isolated clustered cells positive for cytokeratin/p185c–erbB-2 from the peripheral blood of breast cancer patients.
      • Brandt B
      • Roetger A
      • Heidl S
      • Jackisch C
      • Lellé RJ
      • Assmann G
      • Zänker KS
      Isolation of blood-borne epithelial derived c-erbB-2 oncoprotein positive clustered cells from the peripheral blood of breast cancer patients.
      Therefore, we were interested in the metastatic behavior of cells from fresh tumor tissue, especially in the step of extravasation, which follows the tumor cells' passage through the bloodstream. We developed a model consisting of a HUVEC monolayer growing on porous membranes coated with extracellular matrix, which mimics the venular walls in vitro. By using fresh tumor tissue in this model, we were also able to examine whether the previous findings about the role of c-erbB-2 in breast cancer metastasis using breast cancer cell lines that had been cultivated extensively in vitro could be extrapolated.
      By applying several breast carcinoma cell lines to the model, our extravasation assay was evaluated. We found a strong positive correlation between the level of p185c–erbB-2 expression and the numbers of cells that had penetrated the barrier after 48 hours. It should be noted that all cell lines used for our experiments derive from metastatic breast carcinomas and that increased expression of c-erbB-2 augmented the invasive capacity in our model. This was clearly demonstrated by the c-erbB-2 cDNA transfection of the low-invasive MDA-MB-468 cell line. Simultaneously, we demonstrated that our in vitro model is suitable to identify cancer cells that are likely to succeed extravasation. Furthermore, our results from testing the breast cancer cell lines support the data of Yu et al
      • Yu D
      • Wang S-S
      • Dulski K
      • Tsai C-M
      • Nicolson GL
      • Hung M-C
      c-erbB-2/neu overexpression enhances metastatic potential of human lung cancer cells by induction of metastasis-associated properties.
      and Tan et al
      • Tan M
      • Yao J
      • Yu D
      Overexpression of the c-erbB-2 gene enhanced intrinsic metastasis potential in human breast cancer cells without increasing their transformation abilities.
      who found that the introduction of the human c-erbB-2 gene into lung and breast cancer cells promotes the invasion steps in the metastatic cascade, such as increased motility, migration through extracellular matrix, and secretion of basement-membrane-degrading enzymes. We extended their results in that we demonstrated that the migration through an endothelial monolayer grown on extracellular matrix is also enhanced by the increased expression of c-erbB-2.
      Our finding that all tumor tissues that express c-erbB-2 contained invasive entities (Table 1) and that the invasive cell number tends to be related to the level of c-erbB-2 expression underlines the important role of the c-erbB-2 receptor for the transendothelial invasiveness of the cells and correlates with our results for the breast carcinoma cell lines. We observed a high ratio of p185c–erbB-2-positive, highly motile cells and cell clusters that derived from disaggregated breast cancer tissues. Thus, we assume an important functional role of the c-erbB-2 receptor in the metastatic steps that require cell migration, eg, the extravasation process.
      Recent studies have shown that overexpression of c-erbB-2 is sufficient to induce cell migration.
      • Verbeek BS
      • Adriaansen-Slot SS
      • Vroom TM
      • Beckers T
      • Rijksen G
      Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts.
      It was reported that a critical level of p185c–erbB-2 seems to be necessary to achieve transformation,
      • DiFiore PP
      • Pierce JH
      • Kraus MH
      • Segatto O
      • King CR
      • Aaronson SA
      erbB-2 is a potent oncogene when overexpressed in NIH/3T3 cells.
      which can be explained by a model in which there is an equilibrium between monomeric and dimeric forms of c-erbB-2.
      • Di Marco E
      • Pierce J
      • Knickley C
      • DiFiore P
      Transformation of NIH 3T3 cells by overexpression of the normal coding sequence of the rat neu gene.
      As the quantity of p185c–erbB-2increases by overexpression, the equilibrium is shifted to the dimeric state resulting in constitutive activation of the tyrosine kinase and inappropriate cellular signaling, subsequently leading to a locomotive phenotype.
      • Di Marco E
      • Pierce J
      • Knickley C
      • DiFiore P
      Transformation of NIH 3T3 cells by overexpression of the normal coding sequence of the rat neu gene.
      Thus, it was not necessary to add an agonist of the erbB receptors (eg, epidermal growth factor, EGF) to our extravasation assay to stimulate cell motility.
      The process of cell migration requires the coordinated activation of both growth factor and adhesion receptor signaling.
      • Verbeek BS
      • Adriaansen-Slot SS
      • Vroom TM
      • Beckers T
      • Rijksen G
      Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts.
      It has been described that signals downstream of c-erbB-2 can modulate integrin-mediated processes.
      • Dougall WC
      • Qian X
      • Peterson NC
      • Miller MJ
      • Samanta A
      • Green MI
      The neu-oncogene: signal transduction pathways, transformation mechanisms and evolving therapies.
      The reported interaction of p185c–erbB-2 with members of the integrin family
      • Falcioni R
      • Antonini A
      • Nisticò P
      • Di Stefano S
      • Crescenzi M
      • Natali PG
      • Sacchi A
      α6β4 and α6β1 integrins associate with ErbB-2 in human carcinoma cell lines.
      may also contribute to generate a more invasive phenotype in carcinoma cells.
      This is the first study of the behavior of cells and cell clusters from disaggregated fresh breast cancers in an extravasation model. Our finding of predominantly clustered carcinoma cells that managed to migrate through our model vessel wall is consistent with data from Friedl et al
      • Friedl P
      • Noble PB
      • Walton PA
      • Laird DW
      • Chauwin PJ
      • Tabah RJ
      • Black M
      • Zänker KS
      Migration of coordinated cell clusters in mesenchymal and epithelial cancer explants in vitro.
      that suggests that locomoting highly polarized clustered cells are prime candidates for precursors of metastasis formation. It also backs up our data that mainly cytokeratin/p185c–erbB-2-positive clustered cells are found in the peripheral blood of breast cancer patients, which we assumed to be the metastasis-forming entities.
      • Brandt B
      • Roetger A
      • Heidl S
      • Jackisch C
      • Lellé RJ
      • Assmann G
      • Zänker KS
      Isolation of blood-borne epithelial derived c-erbB-2 oncoprotein positive clustered cells from the peripheral blood of breast cancer patients.
      Our discovery of c-erbB-2 expression in the majority (14 of 19, Table 1) of transendothelial invasive cell clusters and single cells from surgical breast cancers supports the assumption that p185c–erbB-2 indicates even small subpopulations with high invasion potential within the primary tumor. This is substantiated by our finding that four tumors with no detectable expression of c-erbB-2 by Western blotting contained cells/cell clusters with high locomotive potential that expressed p185c–erbB-2. Moreover, the immunocytochemical detection of several presumably metastasis-associated proteins (MMP-2, CD44, and integrins αvβ3 and α6),
      • Liotta LA
      • Steeg PS
      • Stetler-Stevenson WG
      Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation.
      • Stetler-Stevenson WG
      • Aznavoorian S
      • Liotta LA
      Tumor cell interactions with the extracellular matrix during invasion and metastasis.
      • Sy MS
      • Liu D
      • Schiavone R
      • Ma J
      • Mori H
      • Guo Y
      Interactions between CD44 and hyaluronic acid: their role in tumor growth and metastasis.
      • Imhof BA
      • Piali L
      • Gisler RH
      • Dunon D
      Involvement of α6 and αv integrins in metastasis.
      • Brooks PC
      • Strömblad S
      • Sanders LC
      • von Schalscha TL
      • Aimes RT
      • Stetler-Stevenson WG
      • Quigley JP
      • Cheresh DA
      Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin αvβ3.
      which are expressed by the invasive cells/cell clusters selected in our extravasation model, further supports this view.
      As mentioned above, metastasis results from the preferential survival and growth of a few subpopulations of cells that pre-exist within the parent tumor.
      • Fidler IJ
      Critical factors in the biology of human cancer metastasis: twenty-eighth G. H. A. Clowes Memorial Award Lecture.
      Clinical data support this view.
      • Gusterson B
      • Gelber R
      • Goldhirsch A
      • Price K
      • Save-Soderberg J
      • Anbazhagan R
      • Styles J
      • Rudenstam C-M
      • Golouh R
      • Reed R
      • Martinez-Tello F
      • Tiltman A
      • Torhorst J
      • Grigolato P
      • Bettelheim R
      • Neville A
      • Burki K
      • Castiglione M
      • Collins J
      • Lindtner J
      • Senn H-J
      Prognostic importance of c-erbB-2 expression in breast cancer.
      Overexpression of c-erbB-2 even on a focal basis in cancer tissues has been taken as indicating patients who are poorly responsive to adjuvant chemotherapy.
      • Gusterson B
      • Gelber R
      • Goldhirsch A
      • Price K
      • Save-Soderberg J
      • Anbazhagan R
      • Styles J
      • Rudenstam C-M
      • Golouh R
      • Reed R
      • Martinez-Tello F
      • Tiltman A
      • Torhorst J
      • Grigolato P
      • Bettelheim R
      • Neville A
      • Burki K
      • Castiglione M
      • Collins J
      • Lindtner J
      • Senn H-J
      Prognostic importance of c-erbB-2 expression in breast cancer.
      The detection of those entities in a primary tumor that are less sensitive to chemotherapy may be achieved by our assay. In addition, the more detailed characterization of the metastatic subpopulations could be essential for devising new therapeutic approaches.
      Our extravasation assay allows an analysis of individual breast tumorsin vitro and may help to identify those patients who are prime candidates for more aggressive chemotherapeutic regimens, whereas other patients can be protected from the harm of an aggressive chemotherapy. Thus, by creating this assay we made a contribution to the individualization of the clinical management of breast cancer.
      Thus, our findings may close a gap in that we selected the same cells with regard to their extravasation capacity we would have later isolated from the patients' bloodstream. Moreover, our experimental results complement the findings of Pantel et al
      • Pantel K
      • Schlimok G
      • Braun S
      • Kutter D
      • Lindemann F
      • Schaller G
      • Funke I
      • Izbicki JR
      • Riethmüller G
      Differential expression of proliferation-associated molecules in individual micrometastatic carcinoma cells.
      and Niehans et al
      • Niehans GA
      • Singleton PS
      • Dykovski D
      • Kiang D
      Stability of Her-2/neu expression over time and at multiple metastatic sites.
      who reported that p185c–erbB-2-positive cells from breast cancer patients lodged in the bone marrow or spread to distant organs. Our view is also supported by the finding that the four breast tumors that did not contain invasive entities in our model were derived from patients without detectable lymph node metastasis. Nevertheless, because of the small sample number in this study and the short follow-up time, we were not able to detect further correlations between the tumor staging and/or prognosis and invasive potential of the cells from the disaggregated tumors.

      Acknowledgements

      We are grateful to Julia Karow, Institute of Molecular Medicine, Oxford, UK, and Frank Gebhardt, Institute of Clinical Chemistry and Laboratory Medicine, Münster, Germany, for critically reading the manuscript. We are indebted to Dr. David Troyer, Institute of Atherosclerosis Research, Münster, Germany, for transmission electron microscopy. We also thank Dr. Gerd Assmann, Institute of Clinical Chemistry and Laboratory Medicine, Münster, Germany, for financial support.

      References

        • Fidler IJ
        • Hart IR
        Biological diversity in metastatic neoplasms: origins and implications.
        Science. 1982; 217: 998-1003
        • Liotta LA
        • Steeg PS
        • Stetler-Stevenson WG
        Cancer metastasis and angiogenesis: an imbalance of positive and negative regulation.
        Cell. 1991; 64: 327-336
        • Schlimok G
        • Funke I
        • Holzmann B
        • Göttlinger G
        • Schmidt G
        • Häuser H
        • Swierkot S
        • Warnecke HH
        • Schneider B
        • Koprowski H
        • Riethmüller G
        Micrometastatic cancer cells in bone marrow: in vitro detection with anti-cytokeratin and in vivo labeling with anti-17-1A monoclonal antibodies.
        Proc Natl Acad Sci USA. 1987; 84: 8672-8676
        • Coussens L
        • Yang-Feng TL
        • Liao Y-C
        • Chen E
        • Gray A
        • McGrath J
        • Seeburg PH
        • Libermann TA
        • Schlessinger J
        • Francke U
        • Levinson A
        • Ullrich A
        Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with Neu oncogene.
        Science. 1985; 230: 1132-1139
      1. Nilsen-Hamilton M Growth Factors and Development. Academic Press, London1990
        • Press MF
        • Cordon-Cardo C
        • Slamon DJ
        Expression of the HER-2/neu proto-oncogene in normal human adult and fetal tissues.
        Oncogene. 1990; 5: 953-962
        • Brandt BH
        • Vogt U
        • Schlotter CM
        • Jackisch C
        • Werkmeister R
        • Thomas M
        • von Eiff M
        • Bosse U
        • Assmann G
        • Zänker KS
        Prognostic relevance of aberrations in the erbB oncogenes from breast, ovarian, oral and lung cancers: double-differential polymerase chain reaction (ddPCR) for clinical diagnosis.
        Gene. 1995; 159: 35-42
        • Yu D
        • Hung M-C
        The HER-2/neu gene in human cancers.
        in: Freireich EJ Stass SA Molecular Basis of Oncology. Blackwell Science, Cambridge1995: 131-162
        • 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: 177-182
        • Thor AD
        • Schwartz LH
        • Koerner FC
        • Edgerton SM
        • Skates SJ
        • Yin S
        • McKenzie SJ
        • Panicali DL
        • Marks PJ
        • Fingert HJ
        Analysis of c-erbB-2 expression in breast carcinomas with clinical follow-up.
        Cancer Res. 1989; 49: 7147-7152
        • Wright C
        • Angus B
        • Nicholson S
        • Sainsbury JR
        • Cairns J
        • Gullick WJ
        • Kelly P
        • Harris AL
        • Horne CH
        Expression of c-erbB-2 oncoprotein: a prognostic indicator in human breast cancer.
        Cancer Res. 1989; 49: 2087-2090
        • Yu D
        • Wang S-S
        • Dulski K
        • Tsai C-M
        • Nicolson GL
        • Hung M-C
        c-erbB-2/neu overexpression enhances metastatic potential of human lung cancer cells by induction of metastasis-associated properties.
        Cancer Res. 1994; 54: 3260-3266
        • Tan M
        • Yao J
        • Yu D
        Overexpression of the c-erbB-2 gene enhanced intrinsic metastasis potential in human breast cancer cells without increasing their transformation abilities.
        Cancer Res. 1997; 57: 1199-1205
        • Verbeek BS
        • Adriaansen-Slot SS
        • Vroom TM
        • Beckers T
        • Rijksen G
        Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts.
        FEBS Lett. 1998; 425: 145-150
        • Falcioni R
        • Antonini A
        • Nisticò P
        • Di Stefano S
        • Crescenzi M
        • Natali PG
        • Sacchi A
        α6β4 and α6β1 integrins associate with ErbB-2 in human carcinoma cell lines.
        Exp Cell Res. 1997; 236: 76-85
        • Brandt B
        • Roetger A
        • Heidl S
        • Jackisch C
        • Lellé RJ
        • Assmann G
        • Zänker KS
        Isolation of blood-borne epithelial derived c-erbB-2 oncoprotein positive clustered cells from the peripheral blood of breast cancer patients.
        Int J Cancer. 1998; 76: 824-828
        • Gimbrone Jr, MA
        • Cotran RS
        • Folkman J
        Human vascular endothelial cells in culture: growth and DNA synthesis.
        J Cell Biol. 1977; 60: 673-684
        • Friedl P
        • Tatje D
        • Czapla R
        An optimized culture medium for human endothelial cells from human umbilical veins.
        Cytotechnology. 1989; 2: 171-179
        • Albini A
        • Iwamoto Y
        • Kleinman HK
        • Martin GR
        • Kozlowski JM
        • McEwan RN
        A rapid in vitro assay for quantitating the invasive potential of tumor cells.
        Cancer Res. 1987; 47: 3239-3245
        • Riesenberg R
        • Oberneder R
        • Kriegmair M
        • Epp M
        • Bitzer U
        • Hofstetter A
        • Braun S
        • Riethmüller G
        • Pantel K
        Immunocytochemical double staining of cytokeratin and prostate specific antigen in individual prostatic tumor cells.
        Histochemistry. 1993; 99: 61-66
        • Hudziak RM
        • Schlessinger J
        • Ullrich A
        Increased expression of the putative growth factor receptor p185HER2 causes transformation and tumorigenesis of NIH 3T3 cells.
        Proc Natl Acad Sci USA. 1987; 84: 7159-7163
        • Stetler-Stevenson WG
        • Aznavoorian S
        • Liotta LA
        Tumor cell interactions with the extracellular matrix during invasion and metastasis.
        Annu Rev Cell Biol. 1993; 9: 541-573
        • Sy MS
        • Liu D
        • Schiavone R
        • Ma J
        • Mori H
        • Guo Y
        Interactions between CD44 and hyaluronic acid: their role in tumor growth and metastasis.
        Curr Top Microbiol Immunol. 1996; 213: 129-153
        • Imhof BA
        • Piali L
        • Gisler RH
        • Dunon D
        Involvement of α6 and αv integrins in metastasis.
        Curr Top Microbiol Immunol. 1996; 213: 195-203
        • Brooks PC
        • Strömblad S
        • Sanders LC
        • von Schalscha TL
        • Aimes RT
        • Stetler-Stevenson WG
        • Quigley JP
        • Cheresh DA
        Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin αvβ3.
        Cell. 1996; 85: 683-693
        • Fidler IJ
        Critical factors in the biology of human cancer metastasis: twenty-eighth G. H. A. Clowes Memorial Award Lecture.
        Cancer Res. 1990; 50: 6130-6138
        • DiFiore PP
        • Pierce JH
        • Kraus MH
        • Segatto O
        • King CR
        • Aaronson SA
        erbB-2 is a potent oncogene when overexpressed in NIH/3T3 cells.
        Science. 1990; 237: 178-182
        • Di Marco E
        • Pierce J
        • Knickley C
        • DiFiore P
        Transformation of NIH 3T3 cells by overexpression of the normal coding sequence of the rat neu gene.
        Mol Cell Biol. 1990; 10: 3247-3252
        • Dougall WC
        • Qian X
        • Peterson NC
        • Miller MJ
        • Samanta A
        • Green MI
        The neu-oncogene: signal transduction pathways, transformation mechanisms and evolving therapies.
        Oncogene. 1994; 9: 2109-2123
        • Friedl P
        • Noble PB
        • Walton PA
        • Laird DW
        • Chauwin PJ
        • Tabah RJ
        • Black M
        • Zänker KS
        Migration of coordinated cell clusters in mesenchymal and epithelial cancer explants in vitro.
        Cancer Res. 1995; 55: 4557-4560
        • Gusterson B
        • Gelber R
        • Goldhirsch A
        • Price K
        • Save-Soderberg J
        • Anbazhagan R
        • Styles J
        • Rudenstam C-M
        • Golouh R
        • Reed R
        • Martinez-Tello F
        • Tiltman A
        • Torhorst J
        • Grigolato P
        • Bettelheim R
        • Neville A
        • Burki K
        • Castiglione M
        • Collins J
        • Lindtner J
        • Senn H-J
        Prognostic importance of c-erbB-2 expression in breast cancer.
        J Clin Oncol. 1992; 10: 1049-1056
        • Pantel K
        • Schlimok G
        • Braun S
        • Kutter D
        • Lindemann F
        • Schaller G
        • Funke I
        • Izbicki JR
        • Riethmüller G
        Differential expression of proliferation-associated molecules in individual micrometastatic carcinoma cells.
        J Natl Cancer Inst. 1993; 85: 1419-1423
        • Niehans GA
        • Singleton PS
        • Dykovski D
        • Kiang D
        Stability of Her-2/neu expression over time and at multiple metastatic sites.
        J Natl Cancer Inst. 1993; 85: 1235-1240