Cancer Research Cell Death Mechanisms and Cancer Therapy Genetics and Biology of Brain Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
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
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 Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Walentinsson, A.
Right arrow Articles by Levan, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Walentinsson, A.
Right arrow Articles by Levan, G.
[Cancer Research 61, 8263-8273, November 15, 2001]
© 2001 American Association for Cancer Research

Molecular Biology and Genetics

Independent Amplification of Two Gene Clusters on Chromosome 4 in Rat Endometrial Cancer

Identification and Molecular Characterization1

Anna Walentinsson2, Khalil Helou, Ville Wallenius, Hans J. Hedrich, Claude Szpirer and Göran Levan

Department of Cell and Molecular Biology-Genetics, Göteborg University, SE-40530 Gothenburg, Sweden [A. W., K. H., G. L.]; Research Centre for Endocrinology and Metabolism, Department of Internal Medicine, Sahlgrenska University Hospital, Göteborg University, S-413 45 Gothenburg, Sweden [V. W.]; Laboratory of Animal Science, Medizinische Hochschule, DE-30623 Hannover, Germany [H. J. H.]; and IBMM, Université Libre de Bruxelles, BE-6041 Gosselies, Belgium [C. S.]


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The BDII rat is genetically predisposed to hormone-dependent endometrial adenocarcinoma and was used to model human cancer. Tumors arising spontaneously in strain crosses involving BDII rats were analyzed by means of comparative genome hybridization. The most common aberration was amplification of the proximal region of rat chromosome 4, centered around bands q12-q22. The copy numbers of 15 cancer-related genes from the region were examined in tissue cultures of 11 endometrial carcinomas (10 endometrial adenocarcinomas and 1 endometrial squamous cell carcinoma) and one peritoneal mesothelioma. Amplification in rat chromosome 4 was detected in six tumors (50%), five of which carried two separate amplified regions, situated at 4q12-q13 and 4q21-q22, interrupted by a nonamplified segment at 4q13-q21.1. The genes Cdk6 (cyclin-dependent kinase 6) and Met (hepatocyte growth factor receptor) were located in the core of each amplified region and were amplified most recurrently and at the highest levels among the genes tested. Using fluorescence in situ hybridization on tumor metaphases, it was observed that the amplified Cdk6 and Met sequences were situated on typical homogeneously staining regions (HSRs). In three tumors, both genes were amplified in the same HSRs, whereas in two tumors, the amplified sequences of each gene were situated in separate HSRs. In addition, Cdk6 and Met amplification was consistently associated with a corresponding increase in gene expression, suggesting that the two genes might represent the targets for the amplifications. In the sixth tumor, which carried amplified sequences of Met but not of Cdk6, coexpression of Met and the normally silent hepatocyte growth factor gene (Hgf; the ligand of Met) was observed. This finding suggests that an autocrine signaling circuit might be operating in this particular tumor. Taken together, our findings suggest that up-regulation of Cdk6 and/or Met may contribute to the development of endometrial cancers in the BDII rat.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Lines of evidence suggest that the development of neoplastic disease requires multiple genetic lesions, ranging from single nucleotide alterations to gross chromosomal changes, occurring sequentially in a cell lineage (1) . DNA amplification represents one major molecular pathway that is thought to play a pivotal role in tumor development, in view of the fact that it provides a mechanism by which tumor cells can trigger enhanced expression of genes whose products are involved in cell proliferation (2) . In human cancers, the number of reports on gene amplification, often in relation to progressive tumor growth and poor prognosis, is continually growing (3, 4, 5) . In the majority of cases in which the amplified chromosome region has been identified and characterized, a proto-oncogene appears to be the target on which selection acts (4) . In some tumors, the architecture of the amplified sequences has been found to be quite simple and consisting of continuous repeats of not much more than the target gene (6 , 7) . However, a more common situation is that the amplified sequences are structurally rather complex and sometimes also internally rearranged, encompassing two or more coamplified genes, including the target gene (8 , 9) . Hence, the prevalence of gene amplification in diverse tumor types, as well as its biological and clinical significance in neoplastic development, makes amplified chromosomal regions interesting targets for detailed genetic analysis. Identification and characterization of the amplified genes can provide valuable insights into the pathogenesis of cancer and may also yield molecular markers for the evaluation of prognosis and therapy.

Carcinoma of the uterine corpus, also known as endometrial carcinoma, is the most frequently diagnosed malignancy of the female reproductive tract and the fourth most common cancer among women (10) .3 Still, the molecular genetic features of this tumor have yet to be described in significant detail. The inbred BDII rat strain is genetically prone to spontaneous hormone-dependent EAC4 (11) and may serve as a genetic model system of this tumor type. In a series of uterine tumors (mostly EACs) developed in F1, F2, and backcross progeny from crosses between BDII rats and rats of either of two nonsusceptible strains (BN and SPRD), a previous cytogenetic investigation disclosed repeatedly occurring manifestations of gene amplification in the form of HSRs. By using CGH, we could conclude that most commonly, the amplified sequences originated from a region in proximal RNO4, centered around bands q12-q22 (12) . Similar amplification in the proximal part of RNO4 has been detected previously in a subset of 7, 12-dimethylbenz[a]anthracene-induced rat sarcomas (13, 14, 15) , suggesting that involvement of this chromosome region represents a major pathway in a variety of tumor types. In the present investigation, we have undertaken a detailed qualitative and quantitative characterization of the amplification of RNO4-derived sequences. Our findings suggest that at least two subregions are involved in independent amplifications in these tumors. Each of the regions contains cancer-related genes, which may interact in the development of endometrial cancers.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Tumor Material.
Females of the inbred rat strain BDII/Han are predisposed to cancer in the endometrium of the uterus (EAC; Refs. 11 and 16 ). Virgin females (>90%) will develop this neoplasm, usually before 24 months of age. When BDII/Han rats were interbred with rats from the two nonsusceptible strains BN/Han and SPRD-Cu3/Han, a large fraction of the F1, F2, and backcross animals spontaneously developed tumors. Most of these tumors were pathologically characterized as EAC, but in addition, some other types of uterine and nonuterine tumors were present (see Ref. 12 ). The main material for the present investigation was 12 tissue cultures derived from 11 endometrial tumors (10 classified as EACs and 1 classified as ESCC) and from 1 tumor classified as peritoneal mesothelioma. DNA was extracted according to standard procedures using phenol/chloroform extraction.

Chromosome Preparations and Cytogenetic Analysis.
Chromosome preparations were made from cultured normal rat embryo fibroblasts, 2n = 42 (for single- and dual-color FISH mapping), as well as from the 12 tumor tissue cultures (for DNA sequence amplification and cytogenetic analyses). Spreads with elongated chromosomes (mainly prometaphases) were prepared from the rat embryo fibroblasts by treatment of cells with ICRF-145 [(4,4'-(1,2-ethanediyl)bis(2,6-piperazinedione), 25 µM; Funakoshi, Tokyo, Japan] during the final 60 min and Colcemid (0.05 µg/ml; Life Technologies, Inc., Grand Island, NY) during the final 20 min before harvest (17) . Tumor tissue cultures were treated with Colcemid but not with ICRF-145. The cells were harvested by mitotic shake-off, pelleted by centrifugation, and resuspended in 0.075 M KCl at room temperature for 30 min. Methanol-acetic acid fixation was carried out as described previously (18) . The tumor tissue metaphases were subjected to cytogenetic analysis after G-banding (18) , and 10–20 cells from each tumor were karyotyped. CGH was performed as described previously (12) .

Physical Localization of PAC Clones and DNA Sequence Amplification Analysis by FISH.
To obtain probes suitable for FISH, genomic DNA clones were isolated by screening of a rat PAC library, RPCI-31 (BAC-PAC resources; Roswell Park Cancer Institute, Buffalo, NY), using as probes purified PCR-generated 200-1200-bp fragments corresponding to genes of interest or full-length cDNAs (for Hgf and Met). The isolation and verification of the PACs have been described in detail elsewhere (19) . The PACs were used as probes in single- and dual-color FISH, which was performed essentially according to published procedures (19 , 20) .

Southern Blot.
Genomic DNA (15 µg) from the tumor tissue cultures, as well as from normal rat BDII liver (included as a control), was digested with EcoRI. Purified 200-1200-bp PCR products corresponding to the genes, as well as the Hgf and Met cDNAs, were radioactively labeled with [{alpha}-32P]dCTP by the random priming method (21) and subsequently hybridized one by one to the filter. After washing at high stringency, the filter was exposed to X-ray film (Hyperfilm-MP; Amersham, Buckinghamshire, United Kingdom). The film was developed, and the approximate level of amplification for each gene was calculated by comparison of the hybridization signal intensities (on the X-ray films) between tumor and normal control BDII DNA, using the dedicated software Quantity One, ver 4.2.2 (Bio-Rad Laboratories, Hercules, CA).5 To compensate for minor variations in sample loading, signals from a 146-bp ß-actin control probe (PCR amplified using the primers 5'-cacggcattgtaaccaactg-3' and 5'-ctgggtcatcttttcacggt-3') were used for normalization.

RT-PCR.
Cytoplasmic RNA was extracted from the tumor tissue cultures and from cultured rat embryonic cells using the RNeasy Mini kit (Qiagen, Valencia, CA), and cDNA was prepared according to standard procedures using the Superscript preamplification system (Life Technologies, Inc.). With the primer pairs listed in Table 1Citation , each of the 15 genes was coamplified with ß-actin (primers 5'-cacggcattgtaaccaactg-3' and 5'-ctgggtcatcttttcacggt-3', generating a 146-bp fragment) using the tumor cDNAs (100 ng of each) as template. PCR was performed according to standard procedures; however, [F]-dUTP R110 (0.4 µM; PE Applied Biosystems, Foster City, CA) was added to each 10-µl reaction mixture. Thermal cycling was performed by initial denaturation at 94°C for 3 min, followed by 28 cycles consisting of 94°C for 1 min, 55°C-60°C for 1 min, and 72°C for 2 min and ending by an extension step of 72°C for 7 min. After PCR, a 2-µl aliquot was collected and subjected to electrophoresis on Tris/acetate/EDTA (TAE)-buffered agarose gels. For a semiquantitative analysis, the remaining 8-µl aliquot was ethanol precipitated and subsequently separated by PAGE in a 377 automated fluorescent DNA sequencer (PE Applied Biosystems). The data were collected and analyzed with the Genescan Analysis Software (PE Applied Biosystems).


View this table:
[in this window]
[in a new window]

  		Table 1 Oligonucleotide sequences of primer pairs used in RT-PCR and size in bps of corresponding amplified fragmentsa

 
Western Blot of the Rat Met Protein.
The Western blot experiments were performed under reducing conditions, mainly as described previously (13) . In brief, proteins were prepared from the cultured tumor cells by sonication. A total amount of 20 µg of protein/sample was run on NuPage 4–12% Bis-Tris gels, using a Novex Xcell II system (Novex, San Diego, CA), and then transferred to PVDF membranes (Novex) by electroblotting. The membranes were blocked overnight by incubation in 10% dry milk in a Tris-buffered saline buffer (50 mM Tris-HCl, 137 mM NaCl, and 0.1% Tween 20). The blots were then probed with a rabbit polyclonal antimouse Met antibody (SP260; Santa Cruz Biotechnology, Santa Cruz, CA) and then with an antirabbit horseradish peroxidase-linked F(ab')2 fragment (Amersham). The membranes were washed and exposed to the enhanced chemiluminescence detection system (Amersham).

DNA Sequencing of Exons 17–19 of the Rat Met Gene.
To evaluate the occurrence of mutations in Met exons 17, 18, and 19 encoding the tyrosine kinase domain of the Met, a total of 100 ng of cDNA from each of the tumor tissue cultures was subjected to PCR amplification, using the primers 5'-ccaccccaatgttctctcac-3' and 5'-ggtggtgaacttttgcgtct-3'. The resulting 382-bp fragment was purified by gel band purification using GFX spin columns (Amersham Pharmacia Biotech, Piscataway, NJ), and DNA sequencing was performed using the Prism BigDye Terminator chemistry (PE Applied Biosystems). The sequencing products were ethanol precipitated and then separated by PAGE in a 377 automated fluorescent DNA sequencer (PE Applied Biosystems). DNA sequence analysis was carried out with the Sequencing Analysis software, ver 3.3 (PE Applied Biosystems). The fragment was sequenced in both directions, using one or the other of the two PCR primers. cDNA from cultured rat embryonic cells was used as control.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Cytogenetic and CGH Findings in Rat Uterine Tumors.
Tissue cultures were established from 11 malignant rat uterine tumors (10 EACs and 1 ESCC) derived from F1, F2, and backcross animals in the two crosses. The tissue cultures were subjected to cytogenetic analysis, which revealed that most of the EAC tumors displayed chromosome numbers in the triploid/hypotetraploid region (2n = 53–69), but RUT2 and RUT7 were near diploid, and NUT51 and NUT82 were near tetraploid. The chromosome number of the ESCC culture, RUT5, was also in the near-tetraploid region. For comparison, a tissue culture from a peritoneal mesothelioma was included in the analysis (designated RUT29) and was found to be hyperdiploid (2n {approx} 44). Cytogenetic manifestations of gene amplification, represented by HSRs of various sizes, were present in 11 of the 12 tumor cultures (Fig. 1)Citation . By means of CGH analysis, the approximate position of the regions, from which the amplified sequences were derived, could be determined for each of the cultured tumors (Table 2Citation ; criterion used was a green:red fluorescence ratio >2.5). As shown, chromosome regions involved in amplification were tentatively identified in all tumor cultures displaying HSRs, and the data set suggests that at least eight different chromosome regions were involved in amplification in this material, some of them more than one time. The region most commonly exhibiting amplification was RNO4q12-q22 (Fig. 2)Citation . This was in accord with the CGH findings in the primary tumors, from which the tissue cultures were derived. In fact, moderate- to high-level gain of sequences from RNO4 was the most common alteration, affecting 11 (58%) of 19 primary malignant EACs or totally 12 (55%) of 22 tumors, including three additional tumors classified as uterine sarcoma, uterine carcinoma, and peritoneal mesothelioma, respectively (12) . Furthermore, in 10 (45%) of the 22 primary tumors, the fluorescence ratio curves were indicative of high-level copy number increases confined to RNO4q12-q22, which was suggestive of gene amplification (defined as >=5-fold increase in copy number; Ref. 2 ; Fig. 2Citation ).



View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Cytogenetic manifestations of gene amplification, represented by HSRs, were seen in 11 of 12 tumors. The chromosomal origin of the amplified sequences was determined by CGH analysis (see Table 2Citation ). The figure shows a metaphase spread in G-banding from the near-tetraploid NUT82 tumor. Most of the chromosomes look normal, but markers are apparent, including the two long HSRs (arrowheads). Both HSRs were derived from sequences in the proximal region of RNO4.

 

View this table:
[in this window]
[in a new window]

  		Table 2 Signs of gene amplification in the rat tumor tissue cultures studied, as determined by cytogenetic analysis and CGH

 


View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Examples of CGH fluorescence ratio profiles of RNO4 from the rat uterine tumor material. The 12 sets of profiles represent nine different tumors (ST, solid tumor; TC, tissue culture). A region centered around 4q12-q22 exhibits high-level gain of copy number, indicative of gene amplification. Note that there is a tendency of bimodality in some of the profiles.

 
Construction of a Physical Framework Map of the RNO4 Proximal Region.
To be able to define and delimit the amplified RNO4 segment, and to pinpoint putative tumor-associated loci, a detailed molecular cytogenetic characterization of the region was required. Exploiting comparative mapping data from HSA7q21 and 7q31–36 and mouse chromosomes 5 (0–12 cM) and 6 (0–15 cM), we were able to identify gene candidates for mapping and amplification analysis. Using dual-color FISH on rat metaphase and prometaphase chromosomes, as well as on interphase chromatin in resting cell nuclei, the chromosomal sublocalization and relative order of 15 loci were determined, as well as the distances between them, as described in detail elsewhere (19) . We found that all of the selected genes mapped on RNO4 as predicted, and the following map could be constructed (cytogenetic band position is given in parenthesis): (4cen)–Cdk5 (q11.2)–8–10 Mb–Hgf (q12.1)–7–9 Mb–Dmtf1 (q12-q13)–300–500 kb–Abcb1 (q12-q13)–4–6 Mb–Cyp51 (q13)–300–500 kb–Cdk6 (q13)–4–7 Mb–Tac1 (q13-q21)–50–200 kb–Asns (q13-q21)–4–6 Mb–Cav1 (q21.1-q21.2)–100–300 kb–Met (q21.2)–500–1000 kb–Wnt2 (q21.2-q21.31)–200–500 kb–Cftr (q21.2-q21.31)–10-15 Mb–Smoh (q22)–10-15 Mb–Braf (q22-q23)–3–5 Mb–Arhgef5 (q23.3)–(4qter; see also Fig. 4Citation ). If certain evolutionary events are taken into account, both gene order and distances between loci conform exceptionally well with the corresponding homologous human genes on HSA7q (19) .



View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 4. On the left, the ideogram of RNO4 is shown, including higher magnification of the proximal part. The cytogenetic localization of the 15 genes analyzed is indicated. The notches on the thin line immediately to the right of the ideogram shows the approximate physical positions of the genes (not completely to scale; for details, see Ref. 19 ). The diagrams in the right part of the figure show the copy number level for each studied locus in the six tumors ({circ} connected by ----). Copy number classes refer to the number of copies per DCSE; thus, when there is no amplification or reduction, this value will be 2; when there is amplification, the value should be >=5. A value <=1.5 is considered to be indicative of copy number reduction (deletion). Division into classes: A = <1; B = 1–1.5; C = 1.6–4; D = 5–10; E = 11–25; F = 26–40; G = 41–55; H = 56–70; I = 71–90; and J = >90 copies/DCSE, of which class C represents the normal range. The diagrams generated resemble the patterns seen in the CGH analysis but give much greater detail. Note that there are some regions which actually show reduction in copy numbers (shaded) suggestive of chromosomal deletion, perhaps occurring during the actual amplification process. The data strongly suggest the involvement of two distinct targets for gene amplification on RNO4 in these tumors.

 
Gene Copy Number Changes in the RNO4 Proximal Region.
To map the amplified region on RNO4 in detail and to be able to define the minimal region/regions of common amplification, the copy number for each of the 15 genes distributed across bands q11-q23 of RNO4 was monitored in the 12 tumor tissue cultures by conventional Southern blot analysis. The PCR-generated probes were hybridized one by one to filters containing equal quantities of digested DNA from each of the tumor samples, along with the normal control sample (liver DNA from a BDII animal). Examples of the results are shown in Fig. 3Citation . After hybridization, the observed bands were densitometrically scanned, and the approximate level of amplification was determined for each probe by calculating the ratio between the signal intensity obtained from the individual tumor sample with that obtained from the normal control sample. To distinguish gene amplification from low-level RNO4 copy number increase (e.g., trisomy), as well as from hyper-diploidy, only gene copy number increases >=2.5-fold were scored as significant, corresponding to >=5 copies/DCSE.



View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. Examples from the Southern blot analysis of gene amplification for loci in the proximal part of RNO4 in the rat uterine tumors studied. Filters with EcoRI-restricted genomic DNA from 12 tumor tissue cultures and control liver DNA (BDII) were sequentially subjected to hybridization with probes corresponding to each of the 15 genes. In addition, a probe corresponding to ß-actin was hybridized to the filter as a loading control. Subsequently, the relative amount of each gene was determined in the tumor DNAs as described. Results are shown for 3 of the 15 genes studied (Cdk6, Asns, and Met). There are very clear signs of gene amplification (signal intensity increased by >=2.5-fold compared with control levels in normal DNA) for the Met gene, representing the 4q21-q22 amplicon, in six of the tumors (RUT5, 7, 13, 29, NUT51, and 82). For five of them (all except RUT7), there was amplification also of Cdk6, representing the 4q12-q13 amplicon. The Asns gene, situated between Cdk6 and Met, was not amplified in any of the tumors. Furthermore, both Cdk6 and Asns display much reduced signal in RUT7, and the Asns signal is slightly reduced in RUT5 and NUT51.

 
Significantly elevated copy numbers (amplification) for some of the 15 genes studied were detected in 6 of the 12 tumor cultures (RUT5, 7, 13, 29, NUT51, and 82; Fig. 4Citation ). For five of these tumors (RUT7, 13, 29, NUT51, and 82), the Southern blot data coincided with the CGH findings, which displayed fluorescence ratios >2.5 in the RNO4q12-q22 region (Table 2)Citation . The CGH analysis of the sixth tumor, RUT5, did not disclose any significant DNA copy number change affecting RNO4. However, the resolution of CGH is limited, and gain of DNA copy number affecting a small region (<=2–4 Mb) may be undetectable (22) . Moreover, in the CGH analysis of RUT13, it was observed that the average fluorescence ratio gave double peaks in the RNO4q12–22 region. Accordingly, individual metaphases in the composite (FITC/rhodamine) CGH pictures of RUT13 sometimes showed two distinct regions of elevated green:red signal ratio in the proximal part of RNO4 (Fig. 5A)Citation . A similar hybridization pattern was occasionally seen in individual metaphases of another three of the analyzed tumors: RUT29, NUT51, and 82. Thus, these observations suggested that two adjacent or possibly overlapping amplicons might be present in these four tumors. The copy numbers detected by Southern blot analysis corroborate the interpretation that there were two distinct amplified regions, which had been derived from bands 4q12-q13 and 4q21-q22, respectively. As is shown in Fig. 4Citation , the more distal region (4q21-q22) was amplified in all six tumors, whereas the more proximal region (4q12-q13) was amplified in five of them (all except RUT7). The two amplified regions were disconnected by a segment containing the genes Asns and Tac1, both located at the 4q13-q21 transition. Thus, the copy numbers of Asns (Fig. 3)Citation and Tac1 were in the normal range (~2 copies/DCSE) in RUT29 but slightly elevated in RUT13 (~3 copies/DCSE) and NUT82 (3–4 copies/DCSE). In RUT5 and NUT51, the analysis showed that there was a decrease in copy number (~1 copy/DCSE) of these two genes.



View larger version (58K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 5. A, results from CGH analysis of the RUT13 rat EAC tissue culture. FITC-labeled tumor DNA (green signals) and rhodamine-labeled normal DNA (red signals) were cohybridized onto normal rat metaphase chromosomes. The insert shows the RUT13 average FITC:rhodamine fluorescence ratio profile for RNO4 (12 metaphases were analyzed). In RUT13 (and in four additional tumors), two discrete regions displaying high-level amplification were detected in the proximal region of RNO4, involving bands q12-q13 and q21-q22, respectively. The presence of two amplicons is evident by the two well-separated regions of green (tumor specific) hybridization on both RNO4 homologues in the metaphases (arrowheads), as well as by the bimodal shape of the RNO4 ratio profile. B and C, dual-color FISH in the RUT13 rat EAC tissue culture. FITC-labeled PAC455(Cdk6) (green signals) and rhodamine-labeled PAC252(Met) (red signals) were cohybridized onto RUT13 metaphase spreads and interphase nuclei. In B, hybridization to the "native gene sites" at 4q13 (Cdk6) and 4q21.2 (Met) can be seen on one intact RNO4 (arrowhead). In addition, extensive amplification signals are seen in chromosomally integrated HSRs. The amplified Cdk6 sequences are confined to one HSR, whereas the amplified Met sequences are present in two separate HSRs. The Cdk6 probe yielded a pattern of compact and uniform hybridization, as compared with the discontinuous, ladder-like Met hybridization pattern. Because the two genes were not coamplified in the same HSR, the two amplicons must have originated from two independent amplification processes. In interphase nuclei (C), the Cdk6 and Met signals were clustered in one and two distinct restricted areas, respectively. D–G, dual-color FISH in the NUT82 rat EAC tissue culture. FITC-labeled PAC455(Cdk6) (green signals) and rhodamine-labeled PAC252(Met) (red signals) were cohybridized onto NUT82 metaphase spreads and interphase nuclei. In contrast to the situation in RUT13, the signals in interphase nuclei of NUT82 are overlapping (D). In the metaphase shown (E–G), three intact RNO4 copies are present (arrowheads), all showing hybridization at the native gene sites at 4q13 (Cdk6; E) and 4q21.2 (Met; F). In the composite picture (G), it can be seen that the amplified Cdk6 and Met sequences are confined to the same two HSRs, and in this case, the Cdk6 hybridization pattern is more discontinuous and ladder like, whereas the Met hybridization pattern appears to be essentially uniform.

 
For both amplified regions, it was obvious that the amplicons were not uniform; instead, they varied in extension and in amplification level among different tumors. Among the gene probes exhibiting significantly elevated copy numbers in the RNO4q12-q13 region, Cdk6 displayed the highest level of amplification, ranging from 10-fold in RUT29 to >50-fold in NUT82 (Fig. 3)Citation . In four (RUT13, 29, NUT51, and 82) of the five tumors containing amplified Cdk6 sequences, Cyp51 was coamplified; however, the amplification level was considerably lower than that of Cdk6, ranging from 5-fold in RUT29 to ~30-fold in RUT13 and NUT82. Slightly elevated copy numbers were also seen for Abcb1 in NUT82 (~5 copies/DCSE) and for both Abcb1 and Dmtf1 in RUT29 (~6 copies/DCSE).

The genes Cav1 and Met (Fig. 3)Citation were consistently coamplified in all six tumors containing amplicons corresponding to the distal region. The levels of amplification of the two genes were closely correlated to each other and ranged from 6–7-fold in RUT7 to >75-fold in RUT13. Although the exact copy numbers were more difficult to assess when there was high-level amplification, the data suggested that the Cav1 copy number was slightly lower than that of Met in most tumors. The difference in Cav1 and Met copy number was most pronounced in RUT29, in which the degree of Cav1 amplification was ~33-fold, as compared with the >50-fold amplification of Met. In addition to Cav1 and Met amplification, five tumors (RUT7, 13, 29, NUT51, and 82) also contained amplified sequences of the Wnt2 gene. The estimated copy numbers of Wnt2 ranged from ~4-fold in RUT7 to ~30-fold in RUT13. Furthermore, amplification of Cftr was detected in four tumors (RUT7, 13, NUT51, and 82), and the estimated copy numbers were almost identical to those observed for Wnt2. Of the four tumors harboring amplified Cftr sequences, elevated levels of Smoh were seen in RUT7 (~6 copies/DCSE) and in RUT13 (~5 copies/DCSE). Interestingly, low-level amplification (~6 copies/DCSE) of Smoh was also observed in RUT29, a tumor presenting with a normal Cftr copy number (2 copies/DCSE). Although not conclusive, these results suggest that this particular tumor might contain three distinct amplicons.

Reduction in copy number from the normal 2 copies/DCSE was detected for some gene probes in the present tumor material (Fig. 4)Citation . A common observation was that copy numbers <1.5 copies/DCSE were scored for gene probes flanking amplified regions, e.g., Hgf, Dmtf1, and Abcb1 were all present at a lower copy number than normal in RUT13, and the same was true for Wnt2, Cftr, and Smoh in RUT5. Likewise, decreased copy numbers of Tac1, Asns, and Smoh were seen in NUT51. The lowest copy numbers of all were recorded in RUT7 for the genes Cdk6, Tac1, and Asns (~0.5 copies/DCSE). As mentioned previously, this tumor contained only one amplicon, with low-level amplification, corresponding to the distal region. Thus, the data strongly suggest that in RUT7, there is an interstitial deletion of bands 4q13-q21 adjacent to the amplification encompassing bands 4q21-q22.

Chromosomal Sublocalization of Amplified Sequences.
To obtain additional information on the chromosomal sublocalization of the amplified loci, as well as on the intercellular heterogeneity within each tumor, the isolated PAC clones were used one by one as probes in FISH on chromosome slides prepared from the six tumor tissue cultures containing RNO4 amplification. In all tumors, distinct hybridization signals at the "native" site of the gene in the RNO4q12-q22 region were obtained on all intact RNO4 chromosomes. In concordance with the results from Southern blot analysis, signs of amplification of some genes were seen in all six tumors. For these gene probes, clusters of extensive hybridization were detected in resting cell nuclei, and in metaphases, the additional signals were located at chromosomal sites separate from the native gene sites at RNO4, usually on typical HSRs. The distribution of the signals over the HSRs varied from tightly clustered, forming compact "chromosome paint-like" structures to rather dispersed on elongated HSRs. Although the exact number of signals could not be determined in cases of high-level amplification, the signal counts generally correlated well with the copy numbers estimated from the Southern blot. It was also noticed that the intercellular heterogeneity within each tumor was considerable, in copy number and organization of probe signals, and, to a lesser extent, in number and size of the HSRs.

To study the chromosomal sublocalization of the amplified sequences derived from the two amplified subregions, cohybridization of differentially labeled PACs representing Cdk6 (located at 4q13) and Met (located at 4q21.2), respectively, was performed onto metaphase spreads from the five tumors containing two amplicons in RNO4. In RUT13 and RUT29, the two probes hybridized to separate HSRs. In accordance with the Southern blot data, RUT13 contained high-level amplification of sequences corresponding to both Cdk6 and Met. Notably, the Met signals were more numerous than those of Cdk6. In the majority of metaphase spreads of this tumor, the Cdk6 probe gave rise to a uniform and compact hybridization pattern on one HSR. The amplified Met sequences in the same tumor were consistently located on two HSRs, which were distinct from those harboring amplified sequences of Cdk6. In contrast to the uniform Cdk6 signal pattern, the Met signals were unevenly distributed on these HSRs, forming regions of very compact hybridization adjacent to regions with virtually no hybridization, in a ladder-like fashion (Fig. 5B)Citation . Furthermore, as would be expected from the chromosomal sublocalization of signals, the Met signals were clustered to two restricted areas of the interphase nuclei, whereas the Cdk6 signals in most cells were confined to one cluster (Fig. 5C)Citation . In RUT29, the signal counts revealed that Cdk6 was present in ~20 copies, which was located on one HSR, whereas high-level amplification of Met was seen on two HSRs. The distribution of the Met signals on the HSRs was ladder like and resembled that observed in RUT13.

In contrast to the situation in RUT13 and RUT29, the Cdk6 and Met probes yielded hybridization on the same HSRs in RUT5, NUT51, and NUT82. In RUT5, the two probes cohybridized to two HSRs, one large that was situated in a large subtelocentric chromosome and one small that was situated in the distal part of RNO4. It was also observed that the RUT5 cell population was homogenous with respect to the Cdk6 copy number and heterogeneous with respect to the Met copy number. On the basis of the signal counts, the number of Cdk6 copies was 50–60 in every cell, whereas ~15–20 Met copies were present in 90% of the cells while the remaining 10% presented with high-level Met amplification. In the latter fraction of cells, the Met signals formed very compact, ladder-like structures on the larger one of the two HSRs. In contrast, the pattern of Cdk6 signals was quite uniform and not very compact on both HSRs in RUT5. In NUT82, the probes corresponding to Cdk6 and Met yielded numerous hybridization signals dispersed over the whole length of two long HSRs. The Cdk6 signal distribution on the HSRs was typically ladder like, whereas the distribution of Met signals was uniform (Fig. 5, E–G)Citation . As could be expected from the analysis of metaphases, the Cdk6 and Met signals were clustered on top of each other in interphase nuclei (Fig. 5D)Citation . A similar hybridization pattern was observed in NUT51, which also contained two long HSRs. However, the signal counts suggested that the Cdk6 and Met copy numbers were somewhat lower than those observed in NUT82.

mRNA Expression.
DNA amplification in tumors is usually accompanied by an elevated expression of the amplified gene (3) . Thus, RT-PCR was used to examine the mRNA levels in the tumor tissue cultures. One by one, the genes were coamplified with ß-actin, using equal amounts of each tumor cDNA as template. We found that ß-actin was expressed at equal levels in all tumors. On the basis of the relative amounts of the fragments representing the tested gene and ß-actin, respectively, the expression in each tumor was classified as very weak ([+]), weak (+), moderate (++), or strong (+++; Table 3Citation ). When detectable fragments were absent, the outcome was scored as no expression (-).


View this table:
[in this window]
[in a new window]

  		Table 3 mRNA and Met protein expression in the rat tumor tissue cultures studied

 
Detectable amounts of mRNA were present for all of the 15 genes in at least one of the 12 tumors (Table 3)Citation . Three genes (Hgf, Abcb1, and Tac1) were expressed merely in four tumors or less, e.g., Tac1 mRNA was present in very low amounts only in RUT25, whereas moderate and low amounts of Hgf mRNA were detected in RUT7 and in RUT30, respectively (Fig. 6)Citation . Another five genes (Cdk5, Dmtf1, Asns, Smoh, and Braf) displayed quite low and rather uniform levels of expression among the tumors, e.g., Dmtf1 was weakly expressed in all tumors, except in RUT3, in which the level of expression was slightly reduced. A similar pattern of weak but ubiquitous expression was seen for Smoh. Furthermore, the Asns expression was weak in RUT6, RUT12, and NUT82; absent in RUT5, 7, 13, 29, and 30; and very weak in the rest of the tumors. Arhgef5 was expressed in all tumors, at moderate levels in RUT7, 12, 25, 29, and 30 and at low or very low levels in the rest of the tumors.



View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 6. Analysis of mRNA levels in rat tumor tissue cultures. cDNA from the 12 tumor cultures, as well as from rat embryonic cells (data not shown), was used as template for PCR amplification. A 146-bp ß-actin fragment was coamplified with each of the 15 analyzed genes, as a relative comparison. The PCR products were analyzed on agarose gels. The panels display the results from RT-PCR of Hgf (335 bp), Cdk6 (344 bp), Tac1 (312 bp), Cav1 (392 bp), Met (317 bp), and Wnt2 (262 bp). The relative levels of expression were estimated from the gel photographs, but in some cases, more exact values were obtained by analysis in the ABI 377 instrument using the Genescan software. The results from the analysis are shown in Table 3Citation .

 
Strong expression was restricted to those genes displaying significantly elevated copy numbers in the Southern blot analysis. The amount of Cdk6 mRNA corresponded well with the Cdk6 gene copy number in that the expression was strongest in the four tumors having the highest Cdk6 copy numbers (RUT5, 13, NUT51, and 82; Fig. 6Citation ). Furthermore, the expression of Cyp51 was strongest in RUT13, a tumor having a >30-fold Cyp51 amplification, and weakest in RUT5 and 7. The latter two tumors actually displayed a reduced Cyp51 copy number (<=1.5 copies/DCSE). A close correlation between mRNA levels and copy numbers was also observed for the Met gene, located in the distal amplicon. The Met expression was strongest in four tumors exhibiting the highest Met copy numbers by Southern blot analysis (RUT13, 29, NUT51, and 82), as well as in RUT7 showing 7-fold Met amplification (Fig. 6)Citation . Notably, the Met expression in RUT5, a tumor displaying 23-fold Met amplification, was moderate and resembled that of RUT3, a tumor known to have trisomy for RNO4 (12) . The pattern of Cav1 expression among the tumors was quite similar to that of Met, in that the strongest expression was detected in RUT7, 13, 29, and NUT82. However, a very weak Cav1 expression was seen in NUT51, a tumor having >40-fold amplification of Cav1 (Fig. 6)Citation .

The Wnt2 expression was strongest in RUT2 and NUT82 and slightly weaker in RUT7, RUT13, and NUT51 (Fig. 6)Citation . Notably, the Wnt2 expression was very low in RUT29, a tumor shown to have 21-fold amplification of Wnt2. In RUT5, the Wnt2 expression was absent, which was in accordance with the observed Wnt2 copy number reduction (<=1.5 copies/DCSE). Furthermore, the Cftr expression was strong in RUT13, moderate in RUT6, 7, NUT51, and 82, and very weak in RUT5 and RUT29. Thus, when comparing the correspondence between mRNA expression and gene copy number, the correlation was higher for Cftr than for Wnt2.

Expression of the Met Protein.
In addition to the assessment of Met mRNA levels in the tumor tissue cultures, the expression status of Met, the protein encoded by the Met proto-oncogene, was analyzed. Using an antimouse Met antibody, Western blot analysis of the six tumors harboring Met amplification and the remaining six nonamplified tumors was carried out. Undetectable or minimally detectable levels of Met protein were seen in all nonamplified tumors, except in RUT3, which displayed a low Met protein abundance. This tumor was known to contain a considerable fraction of cells having trisomy for RNO4 (12) . In the six tumors with Met amplification, the levels of Met protein corresponded well with the levels of Met mRNA, as well as the Met gene copy numbers. Large amounts of Met protein were detected in RUT13, 29, NUT51, and 82, whereas RUT7 and RUT5 displayed moderate and low amounts of Met protein, respectively (Fig. 7)Citation .



View larger version (38K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 7. Western blot analysis of Met expression in rat tumor tissue cultures. The membrane was probed with a rabbit polyclonal antimouse Met antibody. Two main bands are visible, corresponding to the single-chain Mr 170,000 Met precursor and the Mr 140,000 native ß-chain (generated by cleavage of the Mr 170,000 precursor). The amount of Met protein is closely correlated to the mRNA expression levels (see Fig. 6Citation and Table 3Citation ) and is particularly high in the cultures that exhibit Met gene amplification.

 
Mutation Analysis of the Rat Met Gene.
In human cancers, activation of the MET proto-oncogene by point mutations in the tyrosine kinase domain has been reported previously in hereditary and sporadic forms of papillary renal carcinoma (23) and in lymph node metastases of head and neck squamous cell carcinomas (24) . To determine whether Met was activated by mutation in the present set of rat tumors, DNA specimens of the 12 cultured tumors were screened for DNA sequence alterations in exons 17–19, containing the tyrosine kinase domain. Comparison of the obtained DNA sequences with rat Met sequences in the GenBank DNA sequence database revealed a C->T alteration at the third position of codon 1172 in exon 17 (AAT->AAC), which was present in all tumors. However, because this mutation did not result in an amino acid change (both AAT and AAC code for aspargine), it was interpreted as a silent polymorphism. In fact, this sequence variant appeared to be typical for all three strains used in the present investigation. The generated rat Met sequence was submitted to GenBank and was given the accession no. AF352173.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
To study the genetic aspects of the development of EAC, the EAC-prone BDII rat was used to model human cancer. Using cytogenetics and CGH, the pattern chromosomal changes was analyzed in a set of uterine tumors (mostly EACs), which developed spontaneously in progeny from crosses between BDII rats and rats from two nonsusceptible strains (12) . Amplification of the proximal region of RNO4 emerged as the most common aberration, both in primary tumors and in tissue cultures derived from them. To enable positional localization of candidate oncogenes, we isolated an array of large insert genomic PAC clones using PCR-generated probe fragments representing genes predicted to be located across the region (19) . Using the FISH and Southern blot techniques, we mapped amplified RNO4 regions in detail in a material of tissue cultures of 10 EACs, one ESCC, and one peritoneal mesothelioma. Six of the tumors carried different RNO4-derived amplification units, and we were able to identify two independent regions of common amplification encompassing 4q13 and 4q21.1-q21.2, respectively, suggesting that selection acts on at least two target genes. A similar situation has been reported in human breast carcinomas (25 , 26) , in which coamplification of two syntenic, yet separate, regions on HSA17q was observed.

The more proximal amplified segment comprised up to four of the genes tested (Dmtf1, Abcb1, Cyp51, and Cdk6), of which Cdk6, located at 4q13, was amplified most recurrently. The level of Cdk6 amplification was also markedly higher than the amplification levels of the coamplified genes in tumor amplicons containing additional genes. Significantly, the elevated Cdk6 copy numbers were associated with elevated levels of Cdk6 mRNA, which was in accord with the typical relationship between gene amplification and overexpression. The human homologue, CDK6, has been localized to chromosome 7q21-q22 and has been postulated to be an important player in cell cycle control, providing a link between growth factor stimulation and onset of cell cycle progression (27) . In fact, Harbour et al. (28) showed that CDK6 is one of the CDKs causing a sequential phosphorylation of the RB1 protein that will progressively block active growth suppression by RB1 in the G1 stage and thereby facilitate entry into S phase. Despite the fact that CDK6 has features that are putatively oncogenic, reports describing CDK6 alterations in human cancer are sparse. Some cases of tumor-associated CDK6 overexpression, in the absence of CDK6 amplification, have been described, e.g., CDK6 overexpression was detected in T-cell lymphoblastic lymphoma/leukemia (29) , as well as in splenic marginal zone lymphomas, carrying a specific translocation between HSA2 and 7, in which the HSA7 breakpoint was found to be situated upstream the CDK6 transcription start site (30) . In glioblastomas, overexpression of CDK6 could only be detected in advanced tumors but not in corresponding tumors of lower grade (31) , suggesting that CDK6 up-regulation might promote progression of these tumors. The only documentation of CDK6 overexpression in association with CDK6 amplification comes from human gliomas (32) . The authors suggested that the amplification was restricted to CDK6, because no evidence could be found of coamplification of additional genes on HSA7, including EGFR at 7p11-p12, MET at 7q31, as well as expressed sequence tags mapping near the CDK6 locus.

The amplified sequences in the second region encompassed up to five of the genes tested (Cav1, Met, Wnt2, Cftr, and Smoh), and the core of the amplification was centered in the vicinity of Cav1 and Met at 4q21.1–21.2. These two genes were coamplified in all six tumors harboring RNO4 amplification; however, Met copy numbers were usually slightly higher than those of Cav1. Furthermore, the correlation between gene amplification and gene expression was closer for Met than for Cav1. In fact, the Met mRNA and protein levels were greatly elevated whenever there was Met amplification. The only exception was RUT5, which displayed Met mRNA and protein levels similar to those in the RNO4-trisomic RUT3, despite the fact that it had >20-fold amplification of Met. It should be noted that RUT5 was classified as an ESCC, whereas the remaining tumors (except RUT29) were EACs. In humans, CAV1 and MET are colocated at 7q31 and have both been assigned functions, which make them supposedly cancer related. The CAV1 product, caveolin, acts as a scaffolding protein, which organizes and concentrates specific lipids and lipid-modified signaling molecules within specialized membrane invaginations, called caveoles (33) . Lines of observations suggest that caveolin has an inhibitory effect on these signaling molecules (reviewed in Ref. 34 ), which is in accord with a role of CAV1 as a putative tumor suppressor gene (35) . The MET proto-oncogene was originally identified as a transforming gene activated by translocation in a chemically transformed human osteosarcoma cell line (36) . MET encodes a transmembrane growth factor receptor tyrosine kinase, whose ligand, a soluble cytokine, is HGF/SF (37) . Although expressed in many cell types, the MET receptor tyrosine kinase is found at highest levels in epithelial cells. In contrast, HGF/SF is predominantly expressed in mesenchymal cells and is thought to act on MET-expressing cells in a paracrine fashion (38) . The signaling mediated by HGF/SF-MET interactions has been postulated to play an important role in proliferation and motility of epithelial cells in a variety of tissues, including the endometrial epithelium (39 , 40) . In fact, in vitro studies have shown that HGF/SF can stimulate invasion of endometrial carcinoma cell lines that express MET (41) .

Hence, on the basis of the postulated functions of CAV1 and MET, we speculate that MET represents a more attractive candidate target for gene amplification than does CAV1. In fact, numerous reports have emerged that implicate MET in tumor development. Specifically, up-regulation of MET transcripts has been observed in a variety of human cancers, suggesting a broad role for this receptor. Although overexpression of MET has been detected in various carcinomas, including gastric (5) , ovarian (42) , breast (43) , thyroid (44) , colorectal (45) , and even in endometrial carcinomas (46) , as well as in mucoskeletal tumors (47) . MET amplification has only been reported from human gastric carcinomas (5) and gliomas (48 , 49) . Thus, the present study is unique in that it is the first to demonstrate amplification of Met, in addition to overexpression, in endometrial carcinomas. There have also been reports of activating point mutations in human MET in certain tumor types (23 , 24) . We did not find any Met point mutations in the present material. Actually, this would probably not be expected, at least not in tumors showing amplification. It is known that the DNA sequence of amplified genes is seldom altered; instead, it appears that it is the up-regulation of the unchanged wild-type gene that contributes to tumor development (50) .

Carcinomas are derived from epithelial tissues that normally express Met. Thus, it is not unexpected that the expression of Met mRNA was seen in all of the rat carcinomas studied. Somewhat surprisingly, the gene for the Met ligand, Hgf, was expressed (in absence Hgf amplification) in two tumors. In one of them (RUT30), the Hgf expression was quite weak, and in this tumor, the receptor gene appeared to be down-regulated (no Met protein was detectable in the Western blot; Fig. 7Citation ), making an autocrine stimulation mechanism less likely in this tumor. In contrast, the other tumor (RUT7) exhibited quite strong Hgf expression in conjunction with a 7-fold Met amplification and strong expression of both Met mRNA and Met protein. Thus, one might speculate that an autocrine feedback loop is operating in this particular tumor, as compared with the paracrine HGF/SF stimulation generally suggested for cells of epithelial origin. RUT7 was also the only tumor displaying amplification of Met but not of Cdk6. In fact, there was a reduced copy number and weak expression of Cdk6 in this tumor.

Although gene amplification is frequently observed in cancer, the mechanisms underlying the process have yet to be described in significant detail. Several models have been proposed to account for the formation of intrachromosomal amplification (reviewed in Refs. 2 and 51 ). An attractive model, which may, at least in part, explain the formation of the large and sometimes discontinuously ladder-like chromosomal configuration of the amplified sequences observed in the present tumor material, is the one proposed by Coquelle et al. (52) . According to this model, amplification relies on double-stranded chromosome breakage at a minimum of two sites, possibly fragile sites, which bracket the region to become amplified. The initial breakage occurs at a site telomeric to the gene destined to become amplified and is followed by a sister-chromatid fusion at the break, yielding a dicentric bridge, which undergoes a second breakage during mitotic segregation, at a site proximal to the gene. The new chromosome breakage will trigger a new round of fusion and breakage, and, if there is selection pressure favoring multiple copies of the target gene, these cycles of breakage-fusion-bridge events could eventually give rise to arrays of intrachromosomal amplification, containing large regular inverted repeats (52) . At later stages during the amplification process, when the selective pressure is relieved, the breakage-fusion-bridge cycles may be perpetuated by breakage at random sites along the bridge. This could explain a pattern of discontinuous "mixed ladder" organization of amplified sequences, such as that observed in NUT82 for the Cdk6 gene (Fig. 5E)Citation , in which the initial regular spacing is altered. There are several examples of the occurrence of fragile sites flanking amplified regions in tumors. Coamplification of several genes in band 11q13 is a common finding in many human cancers (9 , 53) . According to Coquelle et al. (52) , at least three fragile sites are located in the vicinity of this chromosome region, including FRA11A at 11q13.3, FRA11F at 11q14.2, and FRA11H at 11q13. Shuster et al. (53) suggested that FRA11A might be involved in the amplification process of 11q13 in oral squamous cell carcinomas. The amplified RNO4 region studied here shares homology with the long arm of HSA7. Several fragile sites have been suggested to be located there, including FRA7E at 7q21.2, FRA7F at 7q22, FRA7G at 7q31.2, and FRA7H at 7q32.3 (54) . CAV1 and MET were found to map near FRA7G (35 , 54) , whereas CDK6, at 7q21-q22, could be located in the vicinity of FRA7E and/or FRA7F.

The amplified regions delineated in the present set of tumors were, in a recurrent manner, flanked by regions displaying reductions in gene copy number, compared with the normal 2 copies/DCSE. Not only was the Asns gene never included in either of the amplicons, the Asns copy number was actually reduced in three of the six tumors with RNO4 amplification, implying that a region including the Asns gene might be deleted in cells from these tumors. Amplifications accompanied by adjacent deletions have also been observed in HSA11q13 in human head and neck carcinomas (55) and in HSA17q12-q21 in breast cancer (7) . Thus, it would be tempting to speculate that loss of chromosomal material adjacent to amplifications may occur as a part of the process leading to amplification. However, exactly how this could take place and whether or not these amplification-associated deletions represent pathways for tumor suppressor gene inactivation remains to be elucidated.

The identification of amplified genes in tumors may provide some insights into the pathogenesis of the neoplasms and may consequently be of a certain prognostic value. In many human tumors studied to date, gene amplification is often strongly associated with an aggressive behavior and poor outcome, e.g., MYCN amplification in neuroblastoma patients has been found to be correlated with an advanced stage and a rapid disease progression (56 , 57) and serves today as a valuable prognostic marker for this disease. Likewise, MET amplification and/or overexpression has been correlated with depth of tumor invasion and lymph node metastasis in gastric carcinomas (5) . In endometrial carcinoma, pathological features have been used for prognostic purposes, such as histological type, grade of differentiation, depth of myometrial invasion, and the occurrence of lymph node metastases (58) . In addition, other prognostic factors in endometrial carcinoma are currently being investigated, including estrogen and progesterone receptor status, c-ERBB-2 status, and tumor ploidy (59 , 60) . Interestingly, Wagatsuma et al. (46) showed that there was a correlation between MET expression and prognosis in patients with endometrial carcinoma.

The major reason to use an animal system to model a human complex disease, such as endometrial cancer, is that the heterogeneity in genetic and environmental factors, inherent in the human population, may be greatly reduced in the model. Because the susceptibility to spontaneously arising EAC tumors is clearly inherited from the BDII strain, and because the gene pool size and environmental variation is kept at a minimum, one might expect that the variation in genetic changes within the tumors would also be quite limited. However, gross chromosomal changes, detectable by cytogenetics and/or CGH, were still found to be quite variable among these tumors (12) . In the present study, 15 cases of gene amplification were detected in 11 tumor cultures, involving eight different chromosomes (Table 2)Citation . This gives a clear indication that in cancer development, there must be many possible pathways of genetic changes that may lead to the same end, even in a system of reduced variability, such as the one studied here. The challenge is to identify and characterize each of these pathways. In our material, one change stood out: amplification in the proximal part of RNO4, which was detected by CGH in five tumors, but another one (RUT5) was added after the higher resolution analysis with individual gene probes had been performed. Thus, it seems clear that amplifications involving this particular chromosome region are part of a major pathway toward malignancy in this model.

The six tumors exhibiting amplification in proximal RNO4 belonged to three different tumor types, according to the pathological analysis. Four of them (RUT7, 13, NUT51, and 82), however, represented the EAC tumor type that is typical for the BDII strain. Of these, RUT7 exhibited amplification only in the more distal region, and, as mentioned above, the finding of Hgf expression in this tumor provides a possible growth-stimulating mechanism involving an autocrine loop, which might contribute to the malignant phenotype. For the remaining tumors (three EACs, an ESCC, and a peritoneal mesothelioma), the combination of amplification and overexpression of Cdk6 and Met provides suggestive evidence for their interaction in a pathway leading to the malignant transformation, although at this stage, the details of this putative interaction is not clear. Obviously, it cannot be ruled out that other cancer-related genes might be situated inside the amplicons and affect the development of these tumors. However, the known functions of Cdk6 and Met in cell cycle progression and proliferation/motility, respectively, make them attractive candidates to play important roles in the genesis or progression of the rat tumors studied. It would be of interest to determine whether the corresponding pathways are involved in the development of human endometrial cancer and other human malignant tumors.


   ACKNOWLEDGMENTS
 
We thank Elisabet Magnusson and Brita Bjönness at the Department of Cell and Molecular Biology-Genetics, Göteborg University, for excellent technical assistance. We also thank Anna Danielsson for help with densitometrical analysis of Southern blot data and György Horvath, both at the Department of Oncology, Sahlgrenska University Hospital, Göteborg University, for comments on the manuscript.


   FOOTNOTES
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by grants from the European Commission (contract ERBBio4CT960562), the Swedish Cancer Society, the Nilsson-Ehle Foundation, and the IngaBritt and Arne Lundberg Research Foundation. Back

2 To whom requests for reprints should be addressed, at Department of Cell and Molecular Biology-Genetics, Göteborg Univeristy, Box 462, S-405 30 Gothenburg, Sweden. Phone: 46-31-7733298; Fax: 46-31-7732599; Back

3 Internet address: http://www.cancer.org. Back

4 The abbreviations used are: EAC, endometrial adenocarcinoma; HSR, homogeneously staining region; CGH, comparative genome hybridization; RNO4, rat chromosome 4; ESCC, endometrial squamous cell carcinoma; FISH, fluorescence in situ hybridization; PAC, P1 artificial chromosome; RT-PCR, reverse transcription-PCR; HGF, hepatocyte growth factor; MET, HGF receptor; HSA7, human chromosome 7; DCSE, diploid chromosome set equivalent; CDK, cyclin-dependent kinase; SF, scatter factor. Back

5 Internet address: http://www.bio-rad.com. Back

Received 6/22/01. Accepted 9/19/01.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Vogelstein B., Kinzler K. W. The multistep nature of cancer. Trends Genet., 9: 138-141, 1993.[Medline]
  2. Schwab M. Oncogene amplification in solid tumors. Semin. Cancer Biol., 9: 319-325, 1999.[Medline]
  3. Schwab M., Amler L. C. Amplification of cellular oncogenes: a predictor of clinical outcome in human cancer. Genes Chromosomes Cancer, 1: 181-193, 1990.[Medline]
  4. Brodeur G. M., Hogarty M. D. Gene amplification in human cancers: biological and clinical significance Vogelstien B. Kinzler K. W. eds. . The Genetic Basis of Human Cancer, : 161-172, McGraw-Hill, Inc. New York 1998.
  5. Nakajima M., Sawada H., Yamada Y., Watanabe A., Tatsumi M., Yamashita J., Matsuda M., Sakaguchi T., Hirao T., Nakano H. The prognostic significance of amplification and overexpression of c-met and c-erb B-2 in human gastric carcinomas. Cancer, 85: 1894-1902, 1999.[Medline]
  6. Amler L. C., Schwab M. Amplified N-myc in human neuroblastoma cells is often arranged as clustered tandem repeats of differently recombined DNA. Mol. Cell. Biol., 9: 4903-4913, 1989.[Abstract/Free Full Text]
  7. Jarvinen T. A., Tanner M., Barlund M., Borg A., Isola J. Characterization of topoisomerase II {alpha} gene amplification and deletion in breast cancer. Genes Chromosomes Cancer, 26: 142-150, 1999.[Medline]
  8. Collins V. P. Gene amplification in human gliomas. Glia, 15: 289-296, 1995.[Medline]
  9. Schuuring E. The involvement of the chromosome 11q13 region in human malignancies: Cyclin D1 and EMS1 are two new candidate oncogenes-a review. Gene, 159: 83-96, 1995.[Medline]
  10. Press M. F. Gynecologic cancers. Cancer Suppl., 83: 1751-1756, 1998.
  11. Deerberg F., Kaspareit J. Endometrial carcinoma in BDII/Han rats: model of a spontaneous hormone-dependent tumor. J. Natl. Cancer Inst. (Bethesda), 78: 1245-1251, 1987.
  12. Helou K., Walentinsson A., Beckmann B., Johansson Å., Hedrich H. J., Szpirer C., Klinga-Levan K., Levan G. Analysis of genetic changes in rat endometrial carcinomas by means of comparative genome hybridization. Cancer Genet. Cytogenet., 127: 118-127, 2001.[Medline]
  13. Helou K., Wallenius V., Qiu Y., Öhman F., Ståhl F., Klinga-Levan K., Kindblom L-G., Jansson J-O., Levan G. Amplification and overexpression of the hepatocyte growth factor receptor (HGFR/MET) in rat DMBA sarcomas. Oncogene, 18: 3226-3234, 1999.[Medline]
  14. Walentinsson A., Sjöling Å., Helou K., Klinga-Levan K., Levan G. Genome-wide assessment of genetic alterations in DMBA-induced rat sarcomas: cytogenetic, CGH and allelotype analyses reveal recurrent DNA copy number changes in rat chromosome 1, 2, 4, and 7. Genes Chromosomes Cancer, 28: 184-195, 2000.[Medline]
  15. Helou K., Walentinsson A., Kost-Alimova M., Levan G. Hgfr/Met oncogene acts as target for gene amplification in DMBA-induced rat sarcomas: free chromatin FISH analysis of amplicon arrays in homogeneously staining regions. Genes Chromosomes Cancer, 30: 416-420, 2001.[Medline]
  16. Kaspareit-Rittinghausen J., Deerberg F., Rapp K. Mortality and incidence of spontaneous neoplasms in BDII/Han rats. Z. Verstierkd., 30: 209-216, 1987.
  17. Helou K., Walther L., Günther E., Levan G. Cytogenetic orientation of the rat major histocompatibility complex (MHC) on chromosome 20. Immunogenetics, 47: 166-169, 1998.[Medline]
  18. Islam M. Q., Levan G. A new fixation procedure for improved quality G-bands in routine cytogenetic work. Hereditas, 107: 127-130, 1987.[Medline]
  19. Walentinsson, A., Helou, K., and Levan, G. A dual-color FISH gene map of the proximal region of rat chromosome 4 and comparative analysis in human and mouse. Mamm. Genome, in press, 2001.
  20. Pinkel D., Landegent J., Collins C., Fuscose J., Segraves R., Lucas J., Gray J. Fluorescence in situ hybridization with human chromosome-specific libraries: detection of trisomy 21 and translocations of chromosome 4. Proc. Natl. Acad. Sci. USA, 85: 9138-9142, 1988.[Abstract/Free Full Text]
  21. Feinberg A. P., Vogelstein B. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem., 132: 6-13, 1983.[Medline]
  22. Piper J., Rutovitz D., Sudar D., Kallioniemi A., Kallioniemi O. P., Waldman F. M., Gray J. W., Pinkel D. Computer image analysis of comparative genomic hybridization. Cytometry, 19: 10-26, 1995.[Medline]
  23. Schmidt L., Duh F-M., Chen F., Kishida T., Glenn G., Choyke P., Scherer S. W., Zhuang Z., Lubensky I., Dean M., Allikmets R., Chidambaram A., Bergerheim U. R., Feltis J. T., Casadevall C., Zamarron A., Bernues M., Richard S., Lips C. J. M., Walther M. M., Tsui L-C., Geil L., Orcutt M. L., Stackhouse T., Lipan J., Slife L., Brauch H., Decker J., Niehans G., Hughson M. D., Moch H., Storkel S., Lerman M. I., Linehan W. M., Zbar B. Germline and somatic mutations in the tyrosine kinase domain of the MET proto-oncogene in papillary renal carcinomas. Nat. Genet., 16: 68-73, 1997.[Medline]
  24. Di Renzo M. F., Olivero M., Martone T., Maffe A., Maggiora P., Stefani A. D., Valente G., Giordano S., Cortesina G., Comogli O. P. M. Somatic mutations of the MET oncogene are selected during metastatic spread of human HNSC carcinomas. Oncogene, 19: 1547-1555, 2000.[Medline]
  25. Bärlund M., Tirkkonen M., Forozan F., Tanner M. M., Kallioniemi O., Kallioniemi A. Increased copy number at 17q22–q24 by CGH in breast cancer is due to high-level amplification of two separate regions. Genes Chromosomes Cancer, 20: 372-376, 1997.[Medline]
  26. Latham C., Zhang A., Nalbanti A., Månér S., Zickert P., Blegen H., Zetterberg A. Frequent co-amplification of two different regions on 17q in aneuploid breast carcinomas. Cancer Genet. Cytogenet., 127: 16-23, 2001.[Medline]
  27. Meyerson M., Harlow E. Identification of G1 kinase activity for cdk6, a novel cyclin D partner. Mol. Cell. Biol., 14: 2077-2086, 1994.[Abstract/Free Full Text]
  28. Harbour J. W., Luo R. X., Dei Santi A., Postigo A. A., Dean D. C. Cdk phosphorylation triggers sequential intramolecular interactions that progressively block Rb functions as cells move through G1. Cell, 98: 859-869, 1999.[Medline]
  29. Chilosi M., Doglioni C., Yan Z., Lestani M., Menestrina F., Sorio C., Benedetti A., Vinante F., Pizzolo G., Inghirami G. Differential expression of cyclin-dependent kinase 6 in cortical thymocytes and T-cell lymphoblastic lymphoma/leukemia. Am. J. Pathol., 152: 209-217, 1998.[Abstract]
  30. Corcoran M. M., Mould S. J., Orchard J. A., Ibbotson R. E., Chapman R. M., Boright A. P., Platt C., Tsui L. C., Scherer S. W., Oscier D. G. Dysregulation of cyclin dependent kinase 6 expression in splenic marginal zone lymphoma through chromosome 7q translocations. Oncogene, 18: 6271-6277, 1999.[Medline]
  31. Lam P. Y., Di Tomaso E., Ng H. K., Pang J. C., Roussel M. F., Hjelm N. M. Expression of p19INK4d, CDK4, CDK6 in glioblastoma multiforme. Br. J. Neurosurg., 14: 28-32, 2000.[Medline]
  32. Costello J. F., Plass C., Arap W., Chapman V. M., Held W. A., Berger M. S., Su Huang H. J., Cavenee W. K. Cyclin-dependent kinase 6 (CDK6) amplification in human gliomas identified using two-dimensional separation of genomic DNA. Cancer Res., 57: 1250-1254, 1997.[Abstract/Free Full Text]
  33. Engelman J. A., Zhang X., Galbiati F., Volonte D., Sotgia F., Pestell R. G., Minetti C., Scherer P. E., Okamoto T., Lisanti M. P. Molecular genetics of the caveolin gene family: implications for human cancers, diabetes, Alzheimer disease, and muscular dystrophy. Am. J. Hum. Genet., 63: 1578-1587, 1998.[Medline]
  34. Okamoto T., Schlegel A., Scherer P. E., Lisanti M. P. Caveolins, a family of scaffolding proteins for organizing "preassembled signaling complexes" at the plasma membrane. J. Biol. Chem., 273: 5419-5422, 1998.[Free Full Text]
  35. Engelman J. A., Zhang X. L., Lisanti M. P. Genes encoding human caveolin-1 and -2 are co-localized to the D7S522 locus (7q31.1), a known fragile site (FRA7G) that is frequently deleted in human cancers. FEBS Lett., 436: 403-410, 1998.[Medline]
  36. Cooper C. S., Park M., Blair D. G., Tainsky M. A., Huebner K., Croce C. M., Vande Woude G. F. Molecular cloning of a new transforming gene from a chemically transformed human cell line. Nature (Lond.), 311: 29-33, 1984.[Medline]
  37. Maggiora P., Gambarotta G., Olivero M., Giodano S., Di Renzo M. F., Comoglio P. M. Control of invasive growth by the HGF receptor family. J. Cell Physiol., 173: 183-186, 1997.[Medline]
  38. Gherardi E., Stoker M. Hepatocyte growth factor-scatter factor: mitogen, motogen, and met. Cancer Cells, 3: 227-232, 1991.[Medline]
  39. Weidner K. M., Sachs M., Birchmeier W. The Met receptor tyrosine kinase transduces motility, proliferation, and morphogenic signals of scatter factor/hepatocyte growth factor in epithelial cells. J. Cell Biol., 121: 145-154, 1993.[Abstract/Free Full Text]
  40. Sugawara J., Fukaya T., Murakami T., Yoshida H., Yajima A. Hepatocyte growth factor stimulated proliferation, migration, and lumen formation of human endometrial epithelial cells in vitro. Biol. Reprod., 57: 936-942, 1997.[Abstract]
  41. Bae-Jump V., Segreti E. M., Vandermolen D. Hepatocyte growth factor (HGF) induces invasion of endometrial carcinoma cell lines in vitro. Gynecol. Oncol., 73: 265-272, 1999.[Medline]
  42. Di Renzo M. F., Olivero M., Katsaros D., Crepaldi T., Gaglia P., Zola P., Sismondi P., Comoglio P. M. Overexpression of the Met/HGF receptor in ovarian cancer. Int. J. Cancer, 58: 658-662, 1994.[Medline]
  43. Beviglia L., Matsumoto K., Lin C. S., Ziober B. L., Kramer R. H. Expression of the c-Met/HGF receptor in human breast carcinoma: correlation with tumor progression. Int. J. Cancer, 74: 301-309, 1997.[Medline]
  44. Di Renzo M. F., Olivero M., Ferro S., Prat M., Bongarzone I., Pilotti S., Belfiore A., Costantino A., Vigneri R., Pierotti M. A. Overexpression of the c-MET/HGF receptor gene in human thyroid carcinomas. Oncogene, 7: 2549-2553, 1992.[Medline]
  45. Di Renzo M. F., Olivero M., Giacomini A., Porte H., Chastre E., Mirossay L., Nordlinger B., Bretti S., Bottardi S., Giordano S. Overexpression and amplification of the met/HGF receptor gene during the progression of colorectal cancer. Clin. Cancer Res., 1: 147-154, 1995.[Abstract]
  46. Wagatsuma S., Konno R., Sato S., Yajima A. Tumor angiogenesis, hepatocyte growth factor, and c-Met expression in endometrial carcinoma. Cancer (Phila.), 82: 520-530, 1998.[Medline]
  47. Wallenius V., Hisaoka M., Helou K., Levan G., Mandahl N., Meis-Kindblom J. M., Kindblom L-G., Jansson J-O. Overexpression of the hepatocyte growth factor (HGF) receptor (Met) and presence of a truncated and activated intracellular HGF receptor fragment in locally aggressive/malignant human musculoskeletal tumors. Am. J. Pathol., 156: 821-829, 2000.[Abstract/Free Full Text]
  48. Wullich B., Muller H. W., Fischer U., Zang K. D., Meese E. Amplified met gene linked to double minutes in human glioblastoma. Eur. J. Cancer, 29A: 1991-1995, 1993.
  49. Fischer U., Müller H-W., Sattler H-P., Feiden K., Zang K. D., Meese E. Amplification of the MET gene in glioma. Genes Chromosomes Cancer, 12: 63-65, 1995.[Medline]
  50. Schwab M. Amplification of oncogenes in human cancer cells. Bioessays, 20: 473-479, 1998.[Medline]
  51. Stark G. R., Debatisse M., Giulotto E., Wahl G. M. Recent progress in understanding mechanisms of mammalian DNA amplification. Cell, 57: 901-908, 1989.[Medline]
  52. Coquelle A., Pipiras E., Toledo F., Buttin G., Debatisse M. Expression of fragile sites triggers intrachromosomal mammalian gene amplification and sets boundaries to early amplicons. Cell, 89: 215-225, 1997.[Medline]
  53. Shuster M. I., Han L., Le Beau M. M., Davis E., Sawicki M., Lese C. M., Park N. H., Colicelli J., Gollin S. M. A consistent pattern of RIN1 rearrangements in oral squamous cell carcinoma cell lines supports a breakage-fusion-bridge cycle model for 11q13 amplification. Genes Chromosomes Cancer, 28: 153-163, 2000.[Medline]
  54. Huang H., Qian C., Jenkins R. B., Smith D. I. Fish mapping of YAC clones at human chromosomal band 7q31.2: identification of YACS spanning FRA7G within the common region of LOH in breast and prostate cancer. Genes Chromosomes Cancer, 21: 152-159, 1998.[Medline]
  55. Jin Y., Hoglund M., Jin C., Martins C., Wennerberg J., Akervall J., Mandahl N., Mitelman F., Mertens F. FISH characterization of head and neck carcinomas reveals that amplification of band 11q13 is associated with deletion of distal 11q. Genes Chromosomes Cancer, 22: 312-320, 1998.[Medline]
  56. Brodeur G. M., Seeger R. C., Schwab M., Varmus H. E., Bishop J. M. Amplification of N-myc in untreated human neuroblastomas correlates with advanced disease stage. Science (Wash. DC), 224: 1121-1124, 1984.[Abstract/Free Full Text]
  57. Seeger R. C., Brodeur G. M., Sather H., Dalton A., Siegel S. E., Wong K. Y., Hammond D. Association of multiple copies of the N-myc oncogene with rapid progression of neuroblastomas. N. Engl. J. Med., 313: 1111-1116, 1985.[Abstract]
  58. Boronow R. C., Morrow C. P., Creasman W. T., Disaia P. J., Silverberg S. G., Miller A., Blessing J. A. Surgical staging in endometrial cancer: clinical-pathologic findings of a prospective study. Obstet. Gynecol., 63: 825-832, 1984.[Medline]
  59. Chew S. H. Prognostic factors in endometrial carcinoma. Ann. Acad. Med. Singap., 28: 266-271, 1999.[Medline]
  60. Susini T., Rapi S., Massi D., Savino L., Amunni G., Taddei G. L., Massi G. Preoperative evaluation of tumor ploidy in endometrial carcinoma: an accurate tool to identify patients at risk for extrauterine disease and recurrence. Cancer (Phila.), 86: 1005-1012, 1999.[Medline]




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
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 Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Walentinsson, A.
Right arrow Articles by Levan, G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Walentinsson, A.
Right arrow Articles by Levan, G.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online