ATS
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Peter S. Dahlberg
Michael A. Maddaus
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dahlberg, P. S.
Right arrow Articles by Ferrin, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dahlberg, P. S.
Right arrow Articles by Ferrin, L. J.
Related Collections
Right arrow Esophagus - cancer

Ann Thorac Surg 2004;78:1790-1800
© 2004 The Society of Thoracic Surgeons

Original article: general thoracic

ERBB2 Amplifications in Esophageal Adenocarcinoma

Peter S. Dahlberg, MD, PhDa,*, Blake A. Jacobson, PhDb, Ganesh Dahal, PhDa, James M. Fink, MD, PhDc, Robert A. Kratzke, MD, PhDb, Michael A. Maddaus, MDa, Lance J. Ferrin, MD, PhDd

a Department of Surgery, Division of Cardiovascular and Thoracic Surgery, Hennepin County Medical Center, Minneapolis, Minnesota, USA
b Department of Medicine, University of Minnesota, Hennepin County Medical Center, Minneapolis, Minnesota, USA
c Department of Pathology, Hennepin County Medical Center, Minneapolis, Minnesota, USA
d Yakima Gastroenterology, Yakima, Washington, USA

Accepted for publication May 7, 2004.

* Address reprint requests to Dr Dahlberg, Department of Surgery, Division of Cardiovascular and Thoracic Surgery, University of Minnesota, MMC 207, 420 Delaware St SE, Minneapolis, MN 55455, USA
dahlb002{at}umn.edu

Presented at the Fortieth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 26–28, 2004.


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results
 Comment
 DISCUSSION
 Acknowledgments
 References
 
BACKGROUND: ERBB2 (v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, Her-2-neu) gene amplification and overexpression has been reported in several types of cancer. The purpose of this study was to (1) determine the frequency of ERBB2 amplification (in comparison to other proto-oncogenes) in tumors from patients with esophageal adenocarcinoma, (2) characterize structural details of an ERBB2 amplicon in the esophageal adenocarcinoma cell line OE19 (contains a 100-fold ERBB2 amplification), and (3) test whether growth of the OE19 cell line is sensitive to the ERBB2 inhibitor trastuzumab (Herceptin; Genetech, Inc, San Francisco, CA).

METHODS: First, we determined the frequency, by Southern blotting techniques, of amplification of ERBB2 and 13 other proto-oncogenes in a panel of 25 esophageal adenocarcinoma tumors. Then, in a second panel of 10 tumor specimens, expression levels of the ERBB2 gene and of several other genes that flank ERBB2 on chromosome 17 were determined by microarray analysis. Next we characterized the ERBB2 amplicon in the esophageal adenocarcinoma cell line OE19 using cytogenetic methods and a Rec-A protein assisted restriction endonuclease mapping technique. Finally, an in vitro growth inhibition assay was used to measure the sensitivity of OE19 and OE33 cells to treatment with trastuzumab (humanized antibody to ERBB2).

RESULTS: ERBB2 was the most frequently amplified proto-oncogene among 25 esophageal adenocarcinoma tumors tested (greater than 10-fold amplification in 3 of 25 (12%) tumors tested). The OE19 cell line contains a 100-fold amplification of the ERBB2 gene, and highly expresses its messenger ribonucleic acid. Transcripts from genes that flank ERBB2 including GRB7, a protein linked to metastasis in esophageal cancer, also showed high levels of expression. In OE19 cells, the ERBB2 amplicon was localized to a homogeneously staining region of chromosome 14. Southern blots from the Rec-A protein assisted restriction endonuclease cleavage mapping experiments in OE19 showed a strong band of 210 kilobases in size, demonstrating that the main amplicon was a tandem repeat. In the in vitro growth inhibition assay, trastuzumab inhibited the OE19 and OE33 cells growth by 49% and 20%, respectively, at a saturating concentration of 20 µg/mL.

CONCLUSIONS: ERBB2 is the most frequently amplified proto-oncogene in esophageal adenocarcinoma among the genes that we tested. In the OE19 esophageal adenocarcinoma cell line, the ERBB2 amplicon is translocated onto chromosome 14, is amplified 100-fold at the deoxyribonucleic acid level, and is highly overexpressed at the messenger ribonucleic acid level. Finally, the growth of this cell line was inhibited by incubation with trastuzumab. These results demonstrate that a substantial number of esophageal adenocarcinomas have amplified copies of the ERBB2 gene, and that they may be responsive to ERBB2 targeted therapies such as trastuzumab.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results
 Comment
 DISCUSSION
 Acknowledgments
 References
 
Esophageal cancer is the ninth most prevalent malignancy in the world. Alarmingly, the incidence of esophageal adenocarcinoma has increased fourfold since 1970 [1], and esophageal adenocarcinoma now accounts for greater than 50% of all cases of esophageal cancer in the United States [2].

Overexpression of the growth factor receptor gene v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, Her-2-neu (ERBB2) has been implicated in the pathogenesis of esophageal adenocarcinoma and is found in 15% to 25% of tumor specimens (Fig 1) [3, 4]. However, the extent of altered transcription of ERBB2 and its contribution to a tumor cell's growth may depend on whether altered transcription was the result of transcriptional upregulation or the result of a gene amplification event. The latter mechanism is seen frequently in breast cancer and is associated with a favorable response to anti-ERBB2 therapy with trastuzumab (Herceptin; Genetech, Inc, San Francisco, CA). Therefore, we have chosen to study, in detail, the mechanisms and consequences of ERBB2 amplification in esophageal adenocarcinoma tumors. The purpose of this study was to (1) determine the frequencies of amplification of ERBB2 (in comparison to other proto-oncogenes) in esophageal adenocarcinoma tumors, (2) to characterize structural details of an ERBB2 amplicon in the esophageal adenocarcinoma cell line OE19, and (3) to test whether growth of the cell line is sensitive to the ERBB2 inhibitor trastuzumab.



View larger version (28K):
[in this window]
[in a new window]
  		Fig 1. ERBB2 signaling; normally ERBB2, pairs with other epidermal growth factor receptor family receptors that have bound their respective ligands. A phosphorylation cascade is activated and results in activation of multiple intracellular pathways including mitogen activated protein kinase and phosphoinositide 3-kinase, which promote cell cycle progression and inhibit apoptosis. When ERBB2 is overexpressed as the result of a gene amplification event, the receptor can homodimerize, and it is constitutively active. (EGF = epidermal growth factor; EGFR = epidermal growth factor receptor; MAPK = mitogen activated protein kinase; P = phosphate; PI3K = phosphoinositide 3-kinase.)

 

   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results
 Comment
 DISCUSSION
 Acknowledgments
 References
 
Tumors and Cell Lines
OE19 [5] is an esophageal adenocarcinoma cell line obtained from a moderately differentiated, stage III, gastrointestinal junction tumor. OE33 [6] is a poorly differentiated, Barrett's-associated, esophageal adenocarcinoma cell line generated from a stage II lower esophageal tumor. Fourteen gastric cancer cell lines were used in this study; KATO-III, Hs746T, AGS, SNU-1, SNU-5, SNU-16, RF-48, RF-1, NCI-N87, SNU-601, SNU-620, SNU-638, SNU-668, and SNU-719 (ATCC, Manassas, VA). GM06990, a lymphoblastoid cell line obtained from the Coriell Institute for Medical Research (Camden, NJ) was used to isolate control deoxyribonucleic acid (DNA) for Southern blotting. Cell lines were grown to confluence at 37°C in RPMI 1640 supplemented with 2 mmol/L glutamine and 10% fetal bovine serum in 5% CO2.

The institutional review board at the University of Minnesota approved the protocols by which tissue was obtained from endoscopic procedures or surgical resections. Samples weighing approximately 100 to 500 mg were dissected from esophageal tumors, normal gastric or normal esophageal mucosa, or metastatic lymph nodes. They were flash frozen in liquid nitrogen and stored at –80°C. Tissue samples from adjacent areas of tumor were prepared and examined by a pathologist to confirm histologic type and a predominance of tumor cells (> 50%). Results from two sets of tumors are described. The first set of 25 samples was used for Southern analysis of ERBB2 DNA copy number. A second set of 10 samples was used for the microarray experiments that addressed ERBB2 messenger ribonucleic acid (RNA) levels.

Southern Blotting
Genomic DNA was extracted from cell cultures or from 100 to 200 mg frozen tissues using Qiagen genomic-tip 100/G kits (Qiagen, Valenicia, CA) according to manufacturer's instructions. The concentration of each DNA sample was determined spectrophotometrically at 260 nm.

For amplification studies DNA copy number of selected genes in tumor specimens or cell lines were determined by Southern analysis. Five µg of DNA was digested with EcoR1, Pst1, or HindIII then electrophoresed in 0.8% SeaKemGold agarose (Bio Whitaker Molecular Applications, Rockland, ME), blotted onto nylon membrane (GeneScreen Plus, NEN Life Sciences Products, Inc, Boston, MA), and hybridized with appropriate probes following standard protocols [7]. For probe DNA, image clones specific to epidermal growth factor receptor (EGFR), ERBB2, v-Ki-ras2 Kirsten rat sarcoma viral oncogene (KRAS), fibroblast growth factor receptor 2 (FGFR2), Mdm2, transformed 3T3 cell double minute 2, p53 binding protein (MDM2), cyclin e1 (CCNE1), cyclin D1 (CCND1), GATA binding protein 4 (GATA4), v-myc myelocytomatosis viral oncogene homolog//alias c-myc (MYC), v-myb myeloblastosis viral oncogene alias c-myb (MYB), NCDA3, CMET, zinc finger protein 146 alias OZF (ZNF146), and tumor necrosis factor receptor superfamily, member 6b, decoy alias DCR (TNFRSF6B) were purchased from American Type Culture Collection (ATCC, Manasses, VA). The probes were prepared by the random primer method [8] or as polymerase chain reaction (PCR) products of the entire proto-oncogene insert and radio labeled with [32P]-deoxy cytosine triphosphate. The hybridization signals were quantified using ImageQuant software on a Storm 860 (Sunnyvale, CA) phosphoimager. Scans of the gels stained with ethidium bromide were integrated to determine the relative amount of DNA in each lane, and the proto-oncogene band intensity in each lane was corrected for small variations in the amount of DNA loaded. We measured the precision of our Southern blotting measurements by examining ERBB2 spiked DNA samples. The standard deviation was 11% of the mean when comparing duplicate lanes on the same gel, and 30% when comparing on different gels. The method was shown to be linear up to a copy number of 100. Amplification numbers represent the copy number of the proto-oncogene per haploid genome equivalent, or fold amplification.

Amplification Analysis by Quantitative Real Time Polymerase Chain Reaction
Frozen cells of the two esophageal adenocarcinoma lines, selected tumors, and a control DNA sample were used to verify the copy number of ERBB2 using quantitative real time polymerase chain reaction with a universal TaqMan cytosine-adenine (CA) repeat probe [9, 10]. This probe was used in combination with unique primers that amplified a 250 base pair portion of an ERBB2 intron containing a CA repeat. The protocol for quantitative real time PCR was carried out using a Perkin-Elmer Biosystems (Foster City, CA) Prism 7700 system according to the manufacturer's instructions. In addition to the test samples, the assay included a no-template control, and a positive control.

Cytogenetic Analysis
The karyotypes of OE19 and OE33 were determined using standard cytogenetic techniques [11]. Fluorescence in situ hybridization (FISH) was performed utilizing the Vysis PathVysion ERBB2 DNA Probe Kit (Vysis, Downers Grove, IL) according to the manufacturer's recommendations. The ERBB2 DNA Probe Kit contains a probe to ERBB2 and a control probe to the chromosome 17 centromere. Karyotype analysis, FISH, and reverse G banding was accomplished utilizing a PSI (Perceptive Scientific Instruments, Inc, League City, TX) PowerGene System.

Immunohistochemistry
Immunohistochemical analysis of ERBB2 protein expression in cell lines OE19 and OE33 was performed. Cell lines were grown for 5 days, then harvested, and paraffin embedded. The HercepTest (DakoCytomation, Carpenteria, CA) was used for the analysis, which was performed according to the manufacturer's instructions. The level of ERBB2 expression was scored on a scale of 0 to 3, where 0 = no staining, 1+ = a faint (barely) perceptible staining in part of the cell membrane, 2+ = a weak to moderate complete membrane staining, and 3+ = a strong complete membrane. The slides were viewed and scored by a pathologist. The observation was blinded with regard to the FISH data and ERBB2 status with trastuzumab.

Rec-A Protein Assisted Restriction Endonuclease Cleavage
Rec-A protein assisted restriction endonuclease cleavage (12) was used to map large segments of human DNA contained in the ERBB2 amplicon in OE19. Schematics of the method are shown in Figures 2A and 2B. A unique EcoRI restriction site, 5'AGAACCTGCAAGTAATCCGGGGACGAATTCTGCACAAGTGACCACTGAGAA 3' (the targeted EcoR1 site is italicized), was used in these experiments. Agarose (FMC Bioproducts) embedded DNA was prepared from freshly harvested or frozen OE19 [13, 14]. Rec-A protein assisted restriction endonuclease cleavage was performed on DNA in 50 µ L agarose plugs (20 µg of genomic DNA) according to Ferrin and Camerini-Otero [14, 15]. The efficiency of cleavage was typically 60% to 80% since protection from methylation is not complete.



View larger version (22K):
[in this window]
[in a new window]
  		Fig 2. ERBB2 amplicon mapping using Rec-A protein assisted restriction endonuclease (RARE) cleavage of genomic deoxyribonucleic acid (DNA). (A) Strategy. If multiple copies of an amplified gene are present as a tandem repeat within a contiguous stretch of DNA, cutting at a single site (indicated by the clear vertical line) will yield fragments with a molecular weight determined by the length of the repetitive unit. (B) The RARE cleavage. The technique takes advantage of the ability of Rec-A protein from E coli to pair an oligonucleotide at a unique site of homology to form a three-stranded structure called a synaptic complex (oligonucleotide and two strands of native DNA). This acts as a sequence-specific "masking tape" for DNA and protects the site from methylation during subsequent treatment with EcoR1 methylase. After removal of the synaptic complex, digestion with EcoR1 results in cutting of genomic DNA at a single site. (Modified from Ferrin LJ, et al, Science; 1991;254:1494–7 [12], with permission.)

 
Rec-A protein assisted restriction endonuclease cleavage products were analyzed on a 1.2% SeaKemGold agarose gel (Bio Whitaker Molecular Applications) in a CHEF II pulse-field gel electrophoresis apparatus (Biorads Laboratories, Richmond, CA) with 0.5x TBE running buffer at 12°C with 80 to 250 seconds switch time, 4.5 v/cm volts for 36 hours. The DNA products from the gels were transferred onto a nylon membrane (GeneScreen Plus) and hybridized with an ERBB2 probe (cDNA 1475 to 2239 from human genome) [16]. Molecular weights of the DNA products that hybridized with the probe were determined in reference with the corresponding lambda or chromosome markers run simultaneously on the gel.

Ribonucleic Acid Isolation
Sample total RNA was prepared by a Tri Reagent (Sigma, St. Louis, MO) technique. Frozen tissue was ground to a cornmeal texture using mortar and pestle cooled with liquid nitrogen, and then it was homogenized in 3 mL of Tri Reagent. The RNA was isolated by phenol cholorform extraction followed by isopropanol precipitation. It was washed in 75% ethanol and resuspended in RNAse-free water. The sample was further purified by a cleanup procedure utilizing RNeasy (Qiagen, Valencia, CA) kits and following the manufacturer's directions. The RNA concentration was measured by spectrophotometry, and its integrity was assessed by formaldehyde-agarose gel electorphoresis.

Microarray
Double-stranded cDNA was synthesized from the purified RNA using the SuperScript Choice kit (Invitrogen, Carlsbad, CA), a polydeoxythymidine primer, and instructions provided by Affymetrix (Santa Clara, CA). Amounts of total cellular RNA ranging from 5 to 40 µg were used for the synthesis. The DNA was extracted in phenol choloroform and precipitated in ethanol. Finally, biotin-labeled cRNA was generated from the cDNA using the Enzo BioArray High Yield RNA Transcript Labeling Kit (Enzo Diagnostics). The RNA was purified using an Rneasy mini column (Quiagen, Valencia, CA), concentrated by ethanol precipitation, and fragmented. Samples were hybridized to Affymetrix U95 gene chips using the manufacturer's protocols. Analysis of microarray data was performed on GeneData Expressionist Suite v4.0 (GeneData, San Francisco, CA). Normalization of all chips was based on a reference experiment that was computed as a feature-wise average experiment from all of the chips included in the study.

Sensitivity of Cell Growth to Trastuzumab
For growth analysis, cells were plated at a density of 5 x 104 cells/mL in 24-well cell culture polystyrene plates (Corning Inc, Corning, NY). After 24 hours of incubation at 37°C, the media was changed to media with trastuzumab or control murine monoclonal antibody 1770 (IgG2a, from David L. Dunn, MD, PhD, University of Minnesota), which binds to a smooth strain of Salmonella lipopolysaccharide. The media was replaced after three days and the cell growth was monitored under inverted microscope daily. When the cells in the control wells were 60% confluent, they were harvested by washing with phosphate buffered saline (PBS) followed by trypsinization and the cells were counted in a hemocytomer. The percentage of viable cells was calculated as described by Harlow and Lane [17].


   Results
Top
 Abstract
 Introduction
 Material and Methods
 Results
Comment
 DISCUSSION
 Acknowledgments
 References
 
Oncogene Amplification in Esophageal Adenocarcinoma
Amplification of several protooncogens has been reported in tumors of gastrointestinal tract origin. At the time we initiated our experiments ten proto-oncogenes had been identified in gastric adenocarcinoma, six in esophageal adenocarcinoma. More recently, amplification of an additional 3 genes has been reported in esophageal adenocarcinoma specimens [18]. We measured the copy number of 14 proto-oncogenes (the 11 proto-oncogenes implicated in upper gastrointestinal tract cancers and 3 additional oncogenes that have been described in pancreatic and colon tumors) in a panel of 25 esophageal adenocarcinoma tumors. The results are shown in Table 1 and in Figure 3. Overall, a greater than twofold amplification of at least 1 oncogene was observed in 13 of 25 (52%) tumors. There were 5 tumors (12%) having at least 1 proto-oncogene with a 10-fold or greater amplification. The most commonly amplified proto-oncogene was ERBB2, which was amplified at greater than twofold in 5 of 25 (20%) tumors and at greater than 10-fold in 3 specimens (12%). The only other proto-oncogene that we found to be amplified at greater than 10-fold in any of the samples was v-Ki-ras2 Kirsten rat sarcoma viral oncogene (KRAS) (8%; 2 out of 25). High-level amplification was seen as frequently in the well-differentiated tumors as it was in poor or moderately differentiated tumors.


View this table:
[in this window]
[in a new window]
  		Table 1. Proto-Oncogene Amplification

 


View larger version (41K):
[in this window]
[in a new window]
  		Fig 3. ERBB2 copy number in 25 esophageal cancers and 2 cell lines. Deoxyribonucleic acid, isolated from tumor specimens and cell lines, was electrophoresed through agarose, transferred onto a nylon membrane, and hybridized with an ERBB2 probe. Densitometry was performed to determine copy number (per haploid genome equivalent; ie, normal would be 1.0). Labels correspond to those in Table 1. High level amplification of ERBB2 was in samples 7, 11c, and 31 as well as in both esophageal adenocarcinoma cell lines, OE19 and OE33.

 
Protooncogene amplification was also examined in 2 esophageal and 14 gastric adenocarcinoma cell lines (Table 1). High-level amplification (> 10-fold) was seen in 5 of the 16 cell lines tested (31%); 3 had ERBB2 amplifications (including both esophageal cell lines), 1 had a v-myc myelocytomatosis viral oncogene homolog//alias c-myc (MYC) amplification, and 2 had fibroblast growth factor receptor 2 (FGFR2) amplifications. A single gastric cell line, SNU-16 had high-level amplification of 2 protooncogenes (FGFR2 and MYC). The highest levels of amplification were seen with the ERBB2 protooncogene (100-fold in esophageal cell line OE19) and with FGFR2 [95-fold in SNU-16]). The copy numbers in Table 1 were determined by Southern blotting. More recently, we have used real time PCR to determine gene copy number using a universal TaqMan CA repeat probe [10, 11]. Good agreement between the two techniques has been obtained, and the PCR is a simpler technique requiring less starting material than the Southern analysis.

Gene Amplification and Gene Expression
We used microarray analysis to analyze ERBB2 gene expression in a second panel of 10 different esophageal adenocarcinoma tumors and in the 2 esophageal adenocarcinoma cell lines with ERBB2 amplifications. High levels of ERBB2 mRNA were seen in 2 of the 10 tumors tested and in the esophageal adenocarcinoma cell lines OE19 and OE33 (Fig 4). Transcripts from genes that flank ERBB2 including GRB7, a protein linked to metastasis in esophageal cancer, also showed high levels of expression. The ERBB2 gene was amplified sevenfold in the highest expressing tumor (esophageal adenocarcinoma 12 in the figure) when tested using real time-PCR (data not shown). The high expressing cell line OE19 has a 100-fold amplification of the ERBB2 gene, and OE33 has a 14-fold amplification. ERBB2 protein expression in OE19 and OE33 was also measured by immunohistochemical staining of cell blocks with anti-ERBB2 antibodies using the HercepTest. Expression was 3+ in both cell lines in comparison to 0+ in KATOIII cells, which do not contain an amplified ERBB2 gene (Fig 5).



View larger version (16K):
[in this window]
[in a new window]
  		Fig 4. ERBB2 microarray expression. Relative expression levels of ERBB2 were determined by microarray analysis using Affymetrix U95 chips. Each chip has 3 oligonucleotides representing ERBB2 transcripts. Sample groups from left to right are specimens from normal esophagus, esophageal adenocarcinoma, esophageal adenocarcinoma metastasis, normal stomach, and the 2 esophageal adenocarcinoma cell lines OE19 and OE33. Expression values identified by squares, circles, and triangles represent messenger RNA levels of ERBB2 measured by 3 different Affymetrix ERBB2 oligonucleotide probes. (EAC = esophageal adenocarcinoma; ESO = normal esophagus; MET = metastasis; STO = normal stomach.)

 


View larger version (148K):
[in this window]
[in a new window]
  		Fig 5. Immunohistochemical staining of ERBB2 protein in cell line OE19. Cell line OE19 was stained with antibody to ERBB2 using a commercially available method (HercepTest). Cells stained 3+ indicating high-level protein expression. Cells from the KATO-III gastric adenocarcinoma cell line that does not have an ERBB2 amplification were scored as 0 (figure not shown).

 
Mapping of the OE19 ERBB2 Amplicon
We determined structural details of the ERBB2 amplicon in OE19 cells. Since amplified genes are typically found within the nucleus on chromosomal cytogenetic structures known as homogeneously staining regions or on extra chromosomal double minute chromosomes, the karyotype of the OE19 cell line was determined and searched for either of these features. OE19 contained a hypotriploid karyotype with numerous numerical and structural chromosomal abnormalities including marker chromosomes. No double minutes were identified. A homogeneously staining region was identified on the short arm of one chromosome 14 in all of the cells analyzed (Fig 6). A second homogeneously staining region was identified on the short arm of one chromosome 15 in a small percentage of the cells.



View larger version (17K):
[in this window]
[in a new window]
  		Fig 6. Karyotype analysis of esophageal cancer cell line OE19. The karyotype of OE19 was complex, which is typical of carcinoma cell lines. There were numerous deletions and additions. Of note, there were two relatively normal appearing chromosome 17 seconds. No double minute chromosomes were apparent, but there was an obvious homogeneous staining region found on chromosome 14 (circled). (mar = marker chromosomes.)

 
Fluorescence in situ hybridization was performed on metaphase chromosome spreads of OE19 cells utilizing probes for ERBB2 (mapping to chromosome 17) and the chromosome 17 centromere. Reverse G-banding of the FISH labeled metaphases identified the homogeneously staining regions on chromosomes 14 and 15 as the sight of ERBB2 amplification (Fig 7A). Examination of interphase nuclei revealed a single large cluster (two clusters in a small percentage of the cells) of ERBB2 signals that were too close together to accurately count. There appeared to be greater than 50 signals per interphase cell.



View larger version (79K):
[in this window]
[in a new window]
  		Fig 7. Fluorescence in situ hybridization of ERBB2 deoxyribonucleic acid (DNA) in esophageal cancer cell lines OE19 (A) and OE33 (B). Fluorescence in situ hybridization was performed utilizing the Vysis PathVysion ERBB2 DNA Probe Kit. The ERBB2 DNA appears pink and a control probe to the chromosome 17 is green. Reverse banding techniques identified the site of amplification as the homogenously staining region seen in the OE19 karyotype on chromosome 14.

 
We also determined the location of the ERBB2 amplicon in OE33 cells. OE33 contained a hypopentaploid karyotype with numerous numerical and structural chromosomal abnormalities. Each metaphase contained three normal chromosome 17 seconds and one, two, or three derivative chromosome 17 seconds, each containing a homogeneously staining region replacing most of the long arm. No double minutes were identified (data not shown).

Fluorescence in situ hybridization was performed on metaphase chromosome spreads of OE33 cells utilizing probes for ERBB2 and the chromosome 17 centromere. The results are shown in Figure 7B. Reverse G-banding of the FISH labeled metaphases identified the homogeneously staining regions on the derivative chromosome 17 seconds as the sight of ERBB2 amplification. Examination of interphase nuclei revealed multiple large clusters of ERBB2 signals that were too close together to accurately count. There appeared to be greater than 50 signals per interphase cell.

Rec-A Protein Assisted Restriction Endonuclease Cleavage Mapping
The Rec-A protein assisted restriction endonuclease cleavage mapping was used to identify the size of the amplicon repeat in OE19. An oligonucleotide corresponding to the 51 nucleic acids spanning the exon 10 EcoR1 site in the ERBB2 gene was synthesized. After protection of the site the genomic DNA was treated with EcoR1 methylase to block all other sites. The synaptic complex was removed and the DNA digested with EcoR1. Cleavage occurs only at the targeted site. Genomic DNA was then subjected to pulsed field electrophoresis through agarose and probed with an ERBB2 probe from exon 10. The resulting gel, shown in Figure 8, had a prominent band at 210 kb (kilobase) and a faint band at 420 kb (probably represented a doublet of the main amplicon). These results suggest that the main amplicon contained ERBB2 gene in a tandem repeat of 210 kb, indicating that it contains sizable amounts of DNA from outside the 29 kb ERBB2 gene.



View larger version (37K):
[in this window]
[in a new window]
  		Fig 8. Rec-A protein assisted restriction endonuclease (RARE) cleavage mapping of OE19 esophageal carcinoma cell genomic deoxyribonucleic acid (DNA). The RARE cleavage mapping was used to identify the size of the amplicon repeat in OE19 cells. An oligonucleotide corresponding to the 51 nucleic acids spanning the exon 10 EcoR1 site in the ERBB2 gene was synthesized. After protection of the site with Rec-A protein, the genomic DNA was treated with EcoR1 methylase to block all other EcoR1 sites. The synaptic complex was removed and the DNA digested with EcoR1. Cleavage occurs only at the targeted site that was protected from methylation. Genomic DNA was then subjected to pulsed field electrophoresis through agarose and probed with an ERBB2 probe from exon 10. The gel has a single prominent band at 210 kb and a faint band at 420 kb (probably represented a doublet of the main amplicon). These results are consistent with an ERBB2 amplicon of 210 kb configured as a tandem repeat. These results indicate that the amplicon contains sizable amounts of DNA from outside the 29 kb ERBB2 gene.

 
Trastuzumab Sensitivity
Finally, the dependence of OE19 (100-fold ERBB2 amplification) and OE33 (14-fold amplification) cell growth on ERBB2 protein expression was tested by measuring the ability of trastuzumab to inhibit cell growth in vitro. Trastuzumab is a humanized murine monoclonal antibody directed against the extra cellular domain of the ERBB2 protein. Incubation of OE19 and OE33 cells in saturated (20 µg/mL) concentrations of trastuzumab resulted in 49% and 20% inhibition of proliferation, respectively (Fig 9), whereas control antibody 1770 had no effect. Growth of the gastric carcinoma cell line KATO-III, which does not have ERBB2 amplification, was not sensitive to trastuzumab. The SK-BR-3 breast cancer cells (ERBB2 amplified) showed a similar degree of sensitivity to trastuzumab as did OE19.



View larger version (9K):
[in this window]
[in a new window]
  		Fig 9. Trastuzumab sensitivity of cancer cell lines. The dependence of cell growth on ERBB2 protein expression was tested by measuring the ability of trastuzumab (Herceptin) to inhibit cell growth in vitro. Cells were incubated in media with a saturating dose of 20 µg/mL of trastuzumab or with control antibody. Cells were counted after 96 hours. OE19 is an esophageal adenocarcinoma cell line with a 100-fold ERBB2 amplification, OE 33 is an esophageal adenocarcinoma cell line with a 14-fold ERBB2 amplification, SK-BR-3 and BT-474 are breast cancer cell lines with about a 10-fold ERBB2 amplification, and KATO-III is a gastric adenocarcinoma cell line that does not have an ERBB2 amplification. Mean ± standard error of 4 to 6 independent experiments are shown. Growth of all cell lines with ERBB2 amplifications were significantly inhibited by treatment with Herceptin (p < 0.05 Duncan multiple range test).

 

   Comment
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
DISCUSSION
 Acknowledgments
 References
 
ERBB2 in Cancer Biology
The ERBB2 gene is located on chromosome 17q11.2. It has 26 exons spread more than 29 kb of genomic sequence [19, 20], and it shares homology with the epidermal growth factor receptor as well as with two other members of the ERBB family, ERBB3 and ERBB4. Activation of intrinsic tyrosine kinase activity of ERBB2 occurs following heterodimerization or homodimerization with other family members. Signaling leads to activation of cellular pathways that regulate cell growth, and apoptosis including mitogen activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K) pathways [21].

The ERBB2 [22, 23] gene is amplified in several tumors including cancers of the breast [24, 25], prostate [26], pancreas [27], lung [28], and ovary [29]. Little, however, is known about the structural details of individual ERBB2 amplicons, about the mechanisms by which cells acquire them, or about whether tumors with structurally different amplicons have different biological behaviors. One study [30] of breast cancer specimens showed a core region of 280 kb around the ERBB2 gene that was amplified. Other studies of tumors with ERBB2 amplifications have reported high levels of expression of genes located near ERBB2, including PNMT, GRB7, and NLM64 [31–33], implying amplification, or at least overexpression, of megabase segments of DNA.

Our results show that ERBB2 is the most frequent gene amplified at a high level in esophageal adenocarcinoma (among the oncogenes that have been studied), and that tumors and cell lines with ERBB2 amplifications also expressed ERBB2 mRNA at high levels [34]. Data from other recent studies of gene amplification in esophageal adenocarcinoma that looked at a slightly different set of proto-oncogenes also support the important role of ERBB2 in a sizable subset of esophageal adenocarcinoma tumors [18]. Miller and colleagues [18] examined 97 tumor samples and found that 19 (21.8%) had ERBB2 amplifications (> onefold). Of the 13 genes they examined (7 are represented in our study), ERBB2 was the most commonly amplified gene in both studies. The KRAS was found to be amplified at similar but low rates, whereas cyclin e1 (CCNE1) was reported only by Miller and colleagues [18] to be amplified in any of the specimens. Some of these discrepancies could be due to the use of slightly different criteria for determining amplification status (> twofold in our studies vs > onefold in Miller and colleagues) in the studies.

ERBB2 Amplicons
Cell lines provide plentiful sources of DNA for study of amplified genes; however, only a few esophageal adenocarcinoma cell lines are readily available. Both of the 2 lines that we have in our laboratory have high-level ERBB2 gene amplifications. Amplified genes may integrate at the normal chromosomal location of the oncogene, at a different normal chromosomal site, or they may be carried as extra chromosomal small circular bits of DNA called double minute chromosomes. The expression of an amplified oncogene, and therefore its ability to transform a cell, may be highly dependent on its location and on the promoter driving its expression. In other words, all ERBB2 amplicons may not be created equally. We have studied ERBB2 amplicons in 5 malignant cell lines (2 esophageal adenocarcinoma, 1 gastric, 2 breast) using karyotype and FISH analysis. None of the cell lines had double minutes: Amplifications were found on the normal ERBB2 chromosome 17 location in OE33 and at different chromosomal locations in OE19. Mapping studies, using Rec-A protein assisted restriction endonuclease mapping in the esophageal adenocarcinoma cell line OE19, identified the repetitive unit of the ERBB2 amplicon as a 210 kilobase stretch of genomic DNA.

We realize the limitations of working with cell lines. However, all amplification events identified in cell lines have also been seen in human tumors so the data from these experiments is likely to be relevant to the study of human cancer. Additional studies are being planned to map amplicons from resected esophageal adenocarcinoma specimens and to correlate the details of the amplicon (such as chromosomal location or presence of coamplified sequences) with the tumor's biological behavior.

Targeting ERBB2 as an Antineoplastic Strategy
If a primary oncogenic pathway in a tumor can be identified, then the natural next step is to treat the patient with therapy targeted against the activated proto-oncogene. Trastuzumab (Herceptin), an ERBB2 antagonist, has been used clinically to treat patients with breast cancers that over express ERBB2 [26]. In our studies the drug also inhibited growth of the OE19 esophageal adenocarcinoma cell line. It was slightly less effective against OE33, which has an ERBB2 amplification of a lesser copy number, and it was not effective against the gastric carcinoma cell line KATO-III, which does not have an ERBB2 amplification. Presumably, in these in vitro experiments, trastuzumab inhibits cell growth by inhibiting activation of pathways such as MAPK or PI3K that are important in controlling cell growth and apoptosis. When administered to patients, the drug may also have additional antitumor effects related to its ability to activate complement or to bind to immune cell Fc receptors. It has not been tested in patients with esophageal adenocarcinoma tumors with ERBB2 overexpression; however, these data suggest the drug may be effective in the treatment of this group of patients.

Future Directions
In our future studies we will look for ERBB2 amplification-specific gene signatures in esophageal adenocarcinoma specimens. This approach has already been reported in the study of breast cancers [35], although the authors used a liberal definition of amplification (copy number > 2) and therefore may not have been describing specimens that were truly ERBB2-driven malignancies. We also plan on testing targeted therapy directed against genes involved in pathways such as MAPK and PI3K that are downstream from ERBB2 membrane-associated signaling. Small molecule inhibitors, as well as inhibitory RNAs, will be tested alone and in combination with trastuzumab for synergistic effects. Our long-term vision is that one day this subset of esophageal adenocarcinoma tumors with ERBB2 amplifications will be readily identifiable and that patients will receive therapy directed at this growth factor receptor pathway, which is the primary force driving proliferation of the patient's malignancy.


   DISCUSSION
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
 DISCUSSION
Acknowledgments
 References
 
DR THOMAS K. WADDELL (Toronto, Ontario, Canada): I'm not so familiar with this area and I'm afraid I'm not going to be able to quote you the exact trial, but it's my impression that there have been studies looking at Herceptin where, in fact, the response to treatment wasn't correlated with the level of the ERBB2. In particular, you had an opportunity to examine this, I guess, with two cell lines that expressed different amounts of ERBB2. I missed it, but I think the OE33 had levels about 14, and the 19 was about levels of 100. So in your in vitro assays was there a correlation between the level of expression of ERBB2 and the sensitivity to Herceptin?

DR DAHLBERG: At the saturating concentrations, OE33 is about 30% to 40% inhibited by treatment with Herceptin as opposed to upwards of 50% in most experiments with the OE19. So there is a correlation between the dose of the gene and its Herceptin sensitivity.

DR WADDELL: I don't mean to ask a question that's sort of rude, but the conclusion of your abstract is that fine mapping of this type of mechanism of mutation is important for clinical reasons, but is that really true? Couldn't we just skip all the business about how the gene gets amplified and simply look at the response of the tumor in vitro, with respect to the overall level of expression? Does it really matter how the gene gets amplified and what gets amplified with it?

DR DAHLBERG: Well, I don't know. Like you pointed out before, the expression levels of ERBB2 don't always correlate real well with the effect of treatment with Herceptin in some of the clinical trials that have been done. So there might be something informative about the analysis of coexpressed genes that will tell you which tumors really will respond to Herceptin and which may respond to a different type of targeted therapy.

In terms of the fine detail mapping we may also get a better idea of what is driving theses tumors. Questions like are the promoters that are driving expression of ERBB2 normal or have they been mutated can be answered. Targeting promoters may turn out to be a more effective strategy to shut down ERBB2 signaling than therapies directed against the expressed receptor. What is also apparent from some of our preliminary results is that amplicons evolve, and by understanding those details we might stumble onto some novel treatment strategies.

DR WADDELL: So it may be relevant to disease causation as much as anything else.

DR DAHLBERG: Right.


   Acknowledgments
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
 DISCUSSION
 Acknowledgments
References
 
We would like to thank Jose Jessurun, MD, for his help with histopathology, Arek Dudek, MD, for providing us with trastuzumab, and Curtis M. Nelson for technical assistance with microarray and in vitro Herceptin experiments.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results
 Comment
 DISCUSSION
 Acknowledgments
 References
 

  1. Devesa SS, Blot WJ, Fraumeni J. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer. 1998;83:2049–2053[Medline]
  2. Chen X, Yand CS. Esophageal adenocarcinoma: a review and perspective on the mechanisms of carcinogenesis and chemoprevention. Carcinogenesis. 2001;22:1119–1129[Abstract/Free Full Text]
  3. Hardwick RH, Shepherd NA, Moorghen M, Newcomb PV, Alderson D. c-erbB-2 overexpression in the dysplasia/carcinoma sequence of Barrett's esophagus. J Clin Pathol. 1995;48:129–132[Abstract/Free Full Text]
  4. Ross JS, McKenna BJ. The HER-2/neu oncogenes in tumors of the gastrointestinal tract. Cancer Invest. 2001;19:554–568[Medline]
  5. http://www.biotech.ist.unige.it/cldb/cl3761.html
  6. http://www.biotech.ist.unige.it/cldb/cl3762.html
  7. Sambrook J, Rossell DW, Sambrook J. Molecular cloning: a laboratory manual. 3rd ed. Plainview, NY: Cold Spring Harbor Laboratory; 2001.
  8. Fienberg AP, Vogelstein B. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem. 1983;132:6–13[Medline]
  9. Ginzinger DG, Godfrey TE, Nigro J, et al. Measurement of DNA copy number at microsatellite loci using quantitative PCR analysis. Cancer Res. 2000;60:5405–5409[Abstract/Free Full Text]
  10. Nigro JM, Takahashi MA, Ginzinger DG, et al. Detection of 1p and 19q loss in oligodendroglioma by quantitative microsatellite analysis, a real-time quantitative polymerase chain reaction assay. Am J Pathol. 2001;158:1253–1262[Abstract/Free Full Text]
  11. Barch M, Knutsen T, Spurbeck J. The AGT cytogenetics laboratory manual, 3rd ed. 3rd ed. Philadelphia, PA: Lippincott-Raven; 1977.
  12. Ferrin LJ, Camerini-Otero RD. Selective cleavage of human DNA: RecA-assisted restriction endonuclease (RARE) cleavage. Science. 1991;254:1494–1497[Abstract/Free Full Text]
  13. Wang Y, Schwaltz DC. Chopped inserts: a convenient alternative to agarose/DNA inserts or beads. Nucleic Acid Res. 1993;21:2528[Free Full Text]
  14. Ferrin LJ, Camerini-Otero RD. Long-range mapping of gaps and telomeres with RecA-assisted restriction endonuclease (RARE) cleavage. Nature Genet. 1994;6:379–383[Medline]
  15. Ferrin LJ. Flexible genetic engineering using RecA protein. Mol Biotechnol. 2001;18:233–241[Medline]
  16. Wolff R, Gemmill R. Purifying and analyzing genomic DNA. Green ED, Birren B, Klapholz S, Myers RM, Hieter P. Genome analysis: a laboratory manual. 3rd ed. Plainville, NY: Cold Spring Harbor Laboratory Press; 1997. p. 186–207
  17. Harlow E, Lane D. Antibodies: a laboratory manual. 3rd ed. Cold Spring, NY: Cold Spring Harbor Laboratory; 1988.
  18. Miller CT, Moy JR, Lin L, et al. Gene amplification in esophageal adenocarcinomas and Barrett's with high dysplasia. Clin Cancer Res. 2003;9:4819–4825[Abstract/Free Full Text]
  19. Mendrola JM, Berger MB, King MC, Lemmon MA. The single transmembrane domains of ErbB receptors self-associate in cell membranes. J Biol Chem. 2002;277:4704–4712[Abstract/Free Full Text]
  20. Penuel E, Akita RW, Sliwkowski MX. Identification of a region within the ErbB2/HER2 intracellular domain that is necessary for ligand-independent association. J Biol Chem. 2002;277:28468–28473[Abstract/Free Full Text]
  21. Lenferink AEG, Busse D, Flasnagan WM, Yakes M, Arteaga CL. ErbB2/neu kinase modulates cellular p27kp1 and cyclin D1 through signaling pathways. Cancer Res. 2001;61:6583–6591[Abstract/Free Full Text]
  22. 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[Abstract/Free Full Text]
  23. di Fiore PP, Pierce JH, Kraus MH, Segatto O, King CR, Aaronson SA. erbB-2 is a potent oncogene when overexpressed in NIH/3T3 cells. Science. 1987;237:178–182[Abstract/Free Full Text]
  24. Skacel M, Ormsby AH, Pettay JD, et al. Aneusomy of chromosomes 7, 8, and 17 and amplification of HER-2/neu and epidermal growth factor receptor in Gleason score 7 prostate carcinoma: a differential fluorescent in situ hybridization study of Gleason pattern 3 and 4 using tissue microarray. Hum Pathol. 2001;32:1392–1397[Medline]
  25. Park J, Rhim JS, Park S, Kimm S, Kraus MH. Amplification, overexpression, and rearrangement of the erbB-2 protooncogene in primary human stomach carcinomas. Cancer Res. 1989;49:6605–6609[Abstract/Free Full Text]
  26. Chiang P, Beer DG, Wei W, Orringer MB, Kurnit DM. Detection of erbB-2 amplifications in tumors and sera from esophageal carcinoma patients. Clin Cancer Res. 1999;5:1381–1386[Abstract/Free Full Text]
  27. Safran H, Steinhoff M, Mangray S, et al. Overexpression of the HER-2/neu oncogene in pancreatic adenocarcinoma. Am J Clin Oncol. 2001;24:496–499[Medline]
  28. Hirsch FR, Franklin WA, Veve R, Varella-Garcia M, Bunn PA Jr. HER2/neu expression in malignant lung tumors. Semin Oncol. 2002;29:51–58
  29. Seki A, Yoshinouchi M, Seki N, Kodama J, Miyagi Y, Kudo T. Detection of c-erbB-2 and FGF-3 (INT-2) gene amplification in epithelial ovarian cancer. Int J Oncol. 2000;17:103–106[Medline]
  30. Kauraniemi P, Kuukasjarvi T, Sauter G, et al. Amplification of a 280-kilobase core region of the ERBB2 locus leads to activation of 2 hypothetical proteins in breast cancer. Am J Path. 2003;163:1979–1984[Abstract/Free Full Text]
  31. Nezu M, Sasaki H, Kuwahara Y, et al. Identification of a novel promoter and exons of the cERBB-2 gene. Biochem Biophys Res Commun. 1999;258:499–505[Medline]
  32. Kauraniemi P, Barlund M, Monni O, Kallioniemi A. New amplified and highly expressed genes discovered in the ERBB2 amplicon in breast cancer by cDNA microarrays. Cancer Res. 2001;61:8235–8240[Abstract/Free Full Text]
  33. Van De Vijver MJ, He YD, Van't Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002;347:1999–2009[Abstract/Free Full Text]
  34. Dahlberg PS, Ferrin LJ, Grindle SM, et al. Gene expression profiles in esophageal adenocarcinoma. Ann Thorac Surg. 2004;77:1008–1015[Abstract/Free Full Text]
  35. Berticci F, Borie N, Ginestier C, et al. Identification and validation of an ERBB2 gene expression signature in breast cancers. Oncogene. 2004;23:2564–2575[Medline]



This article has been cited by other articles:


Home page
 Cancer Res.Home page
K. Kunii, L. Davis, J. Gorenstein, H. Hatch, M. Yashiro, A. Di Bacco, C. Elbi, and B. Lutterbach
FGFR2-Amplified Gastric Cancer Cell Lines Require FGFR2 and Erbb3 Signaling for Growth and Survival
Cancer Res., April 1, 2008; 68(7): 2340 - 2348.
[Abstract] [Full Text] [PDF]

Home page
 The OncologistHome page
S. Ekman, M. Bergqvist, C.-H. Heldin, and J. Lennartsson
Activation of Growth Factor Receptors in Esophageal Cancer Implications for Therapy
Oncologist, October 1, 2007; 12(10): 1165 - 1177.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
Z. Hou, S. Sang, H. You, M.-J. Lee, J. Hong, K.-V. Chin, and C. S. Yang
Mechanism of Action of (-)-Epigallocatechin-3-Gallate: Auto-oxidation-Dependent Inactivation of Epidermal Growth Factor Receptor and Direct Effects on Growth Inhibition in Human Esophageal Cancer KYSE 150 Cells
Cancer Res., September 1, 2005; 65(17): 8049 - 8056.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to Personal Folders
Right arrow Download to citation manager
Right arrow Author home page(s):
Peter S. Dahlberg
Michael A. Maddaus
Right arrow Permission Requests
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Dahlberg, P. S.
Right arrow Articles by Ferrin, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Dahlberg, P. S.
Right arrow Articles by Ferrin, L. J.
Related Collections
Right arrow Esophagus - cancer

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
ANN THORAC SURG ASIAN CARDIOVASC THORAC ANN EUR J CARDIOTHORAC SURG
J THORAC CARDIOVASC SURG ICVTS ALL CTSNet JOURNALS