Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Translational Medicine Conference in Israel
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 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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nestl, A.
Right arrow Articles by Sleeman, J. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nestl, A.
Right arrow Articles by Sleeman, J. P.
[Cancer Research 61, 1569-1577, February 15, 2001]
© 2001 American Association for Cancer Research

Molecular Biology and Genetics

Gene Expression Patterns Associated with the Metastatic Phenotype in Rodent and Human Tumors1

Andrea Nestl2, Oliver D. Von Stein, Kurt Zatloukal, Wolf-Gerolf Thies, Peter Herrlich, Martin Hofmann3 and Jonathan P. Sleeman4

Forschungszentrum Karlsruhe, Institute of Toxicology and Genetics [A. N., O. D. V. S., W-G. T., P. H., M. H., J. P. S.] and University of Karlsruhe, Institute of Genetics [P. H.], D-76021 Karlsruhe, Germany, and University of Graz, Department of Pathology, Division of Experimental Cell Research and Oncology, A-8036 Graz, Austria [K. Z.]


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Using subtractive technology, we have generated metastasis-associated gene expression profiles for rat mammary and pancreatic adenocarcinomas. Several genes whose expression is thought to be related to tumor progression such as c-Met, urokinase-type plasminogen activator receptor, ezrin, HMG-1, oncomodulin, cathepsin, and caveolin were thereby isolated. Half of the metastasis-associated clones showed no significant homology to genes with known function. Notably, several of the metastasis-associated clones were also expressed in metastatic lines but not in nonmetastatic lines of other tumor models. Furthermore, in situ hybridization using selected clones documents the relevance of these results for human cancer because strong expression in tumor cells including metastases was detected in human colorectal cancer samples and, to a lesser extent, in mammary cancer samples. These data support the concept that tumors express a "metastatic program" of genes.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
A generally accepted principle is that tumor cells need to acquire altered cellular properties to metastasize. The nature and number of such properties will determine the range of secondary sites to which the tumor cells can spread. During progression to metastatic cancer, the development of heterogeneity within the primary tumor, driven largely by genomic instability (1) , generates populations of tumor cells with the properties needed for invasion, migration, and metastatic colony formation in other organs (2) . Not every tumor cell develops all these properties because the ability to enter the circulatory system or even to form micrometastases is insufficient for progressive metastatic growth (reviewed in Refs. 2 and 3 ). The critical factors determining overt metastasis formation are whether disseminated tumor cells die, survive at distant sites but remain dormant, or have the properties required to grow progressively as metastases. Thus, tumor cells that have the necessary properties to metastasize are selected for during their metastatic journey.

The mechanisms leading to the dissemination of tumor cells, the routes metastasizing tumor cells take through the organism, and the selection pressures tumor cells have to overcome are similar for many different types of cancer (reviewed in Ref. 3 ). It is therefore reasonable to assume that tumor cells from different tumor types require common properties and that they express similar metastasis-promoting genes. Attempts have been made to identify such genes using differential screening techniques, each identifying one gene or a few such genes (4, 5, 6, 7, 8, 9) . However, many outstanding fundamental questions remain to be answered. How many genes are involved in the metastatic process? Are any of these genes absolutely required for metastasis formation? Conversely, can the same property required for metastasis be provided by several different genes or groups of genes? Due to the effects of genomic instability, to what extent are genes that do not play a role in the metastatic process up-regulated during progression to the metastatic phenotype? To begin to find answers to these important questions, it would be necessary to describe and compare the repertoire of genes specifically expressed in metastasizing cells but not in their nonmetastasizing counterparts.

New technologies allow global descriptions of complex transcriptional changes associated with different tumor properties (10, 11, 12) . We have used SSH5 coupled with a sensitive high-throughput screening protocol (13) to define the profile of genes whose expression is up-regulated during progression from a locally growing tumor to metastatic competence. Using two rat tumor progression model systems, we have isolated and identified 268 different cDNAs representing individual genes whose expression is exclusive to or up-regulated in the metastatic phenotype (279 total clones, 11 of which were found in both screens). These target cDNAs were further screened for expression in other tumorigenic cell lines of both rat and human origin. In addition, human primary tumors and metastases of different tissue origin were monitored for expression of selected cDNAs by in situ hybridization studies. From our data, we conclude that subsets of genes are commonly up-regulated during metastatic progression and that several of the genes we have newly identified as being up-regulated in metastasizing cells may be useful diagnostically or as therapeutic targets in the treatment of cancer.


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Tumor Samples and Cell Lines.
The rat pancreatic cell lines BSp73-1AS, BSp73-10AS, and BSp73-ASML (14) ; the rat prostatic tumor cell lines G, AT-1, AT-3 AT-6, Mat-Lu, and Mat-LyLu (15) ; and the rat mammary carcinoma cell lines MTPa, MTC, MTLY, MTLN-2, and MTLN-3 (16) were cultured as described previously. NM081 and MT450, nonmetastasizing and metastasizing rat mammary carcinoma cell lines, respectively (17) , were both maintained in DMEM supplemented with 10% FCS and antibiotics. Two human cell lines, MDA-MB-231 and MDA-MB-468, were obtained from American Type Culture Collection (American Type Culture Collection numbers: MDA-MB-231, HTB-26; MDA-MB-468, HTB-132). All human tumor samples and corresponding normal tissue used for the in situ hybridization were obtained from the Institute of Pathology, University of Graz (Graz, Austria).

Subtractive Library Construction and Screening.
The construction and screening method of the subtracted mammary carcinoma library used in this study was essentially identical to that used in Ref. 13 , using the Clontech SSH kit (PCR-Select cDNA Subtraction Kit; Clontech Laboratories, Palo Alto, CA). Briefly, the subtracted cDNA library was amplified and TA cloned. Clones were picked, and their inserts were amplified by colony PCR. Equal amounts of the PCR reactions were electrophoresed on duplicate high-density gels. Gels were blotted and individually probed with driver and tester cDNA to identify differentially expressed clones.

PCR Amplification.
All clones were amplified by colony PCR using 96-well microtiter plates as follows. Five µl of each bacterial culture were added to 95 µl of sterile water and denatured for 5 min at 94°C. Amplification of 5 µl of this lysate was performed in a total volume of 100 µl using standard PCR buffer (Eurobio), 250 µM deoxynucleotide triphosphates, 10 pM each of primer PN1 and PN2, and 1 unit of Taq (Eurobio). The primers used were the inner primers of the Clontech SSH-Kit: (a) PN1, 5'-TCGAGCGGCCGCCCGGGCAGGT-3'; and (b) PN2, 5'-AGGGCGTGGTGCGGAGGGCGGT-3'.

Cycle conditions were as follows: 30 cycles of 94°C for 20 s, 68°C for 12 s, and 72°C for 30 s. Products were purified from agarose gels using DNAeasy (Biozyme) and used for Northern analysis.

Northern Analysis.
Poly(A)+ RNA was isolated from cell lines by standard methods, and 2-µg aliquots were size-fractionated on a 1.0% formaldehyde-agarose gel. Ethidium bromide staining was used to ensure equivalent loading in each lane. Gels were blotted onto Hybond N+ membrane filters (Amersham). The filters were then cross-linked (UV Stratalinker 2400; Stratagene) and hybridized at 65°C in Church buffer [1 mM EDTA, 0.5 M NaHPO4 (pH 7.2), and 7% SDS]. Probes were generated by PCR amplification of the target clones, followed by gel purification and 32P-labeling of the cDNA fragment (ReadyPrime; Amersham). Unincorporated label was removed before hybridization using an Elutip (Schleicher & Schüll) according to the manufacturer’s specifications. After hybridization with the labeled probes, membranes were washed twice in 2x SSC and 0.1% SDS and twice in 1x SSC and 0.1% SDS at 64°C, after which they were exposed to film for between 2 h and 8 days. Filters were stripped and subsequently probed with another gene to be analyzed. Alternatively, stripped filters were finally hybridized with a glyceraldehyde-3-phosphate dehydrogenase probe to ensure the presence of equivalent amounts of poly(A)+ RNA in each lane of a filter and to ensure that the hybridization patterns of different filters could be compared.

Computational Analysis.
All computational analyses concerning the identity of the clones obtained through the subtracted libraries were performed using the blastn sequence similarity search.6 The clones were compared with the sequences contained in the human and rodent public domain databases, including all of the nonredundant GenBank, European Molecular Biology Laboratory, DDBJ, and PDB sequences, and the DDBJ EST divisions. Before comparative blast screens, we developed a program that automatically recognizes and removes the flanking nested primers to compare only the inserts and thus avoids false blastn results. In the first round of comparison, all of the sequences were compared with each other to identify fragments isolated several times (18) . In the next round, the sequences were compared with human and rodent sequences in the public domain databases (19) .

In Situ Hybridization.
In situ hybridization was performed to detect S7 and CD24 expression in human tumor material. The inserts to be tested were cloned into the bidirectional TA vector pGEM-Teasy (Promega) and transcribed in vitro using SP6 or T7 RNA polymerase (Life Technologies, Inc.). Sections (4 µm thick) of paraffin-embedded human tumor samples were mounted on silanized glass slides, deparaffinized, and postfixed in 4% paraformaldehyde in PBS for 20 min at room temperature. After rinsing with TBS [50 mM Tris-HCl (pH 7.5) and 150 mM NaCl], sections were treated with 0.2 M HCl for 10 min, and, after washing in TBS, incubated in 20 µg/ml proteinase K (Sigma, St. Louis, MO) in TBS containing 2 mM CaCl2 for 15 min at 37°C. The reaction was stopped with TBS for 5 min at 4°C. After treatment with 0.5% acetic anhydride in 100 mM Tris (pH 8.0) for 10 min, sections were rinsed with TBS, dehydrated in graded ethanol, and air dried. Hybridization with fluorescence-labeled (RNA color kit; Amersham Life Science) sense and antisense RNA probes was performed according to the manufacturer’s instructions. After H&E counterstaining, the sections were analyzed by light microscopy. The in situ hybridizations were performed with 10 mammary carcinomas (invasive ductal carcinomas including lymph node metastases from four of the tumors), 5 samples with benign fibrocystic disease, and 10 colorectal carcinomas (including the corresponding nonneoplastic mucosa from the resection margins and lymph node metastases from four tumors). The carcinomas were staged according to UICC guidelines (20) .


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Choice of Experimental System.
We set out to define sets of genes up-regulated or exclusively expressed in metastatic human tumors. Instead of starting directly from human tumor material, we chose to analyze defined clonal rodent tumor cell lines (14 , 16) . The disadvantage of this approach is that contributions to metastasis by stromal cells and transcripts induced by the interaction of tumor cells with their environment are lost. This is outweighed by numerous advantages. Tumor cell lines often exhibit a reproducible metastatic or nonmetastatic phenotype that can be retested at any stage of the analysis (e.g., Ref. 17 ). Moreover, tumor cell lines are accessible to genetic manipulation and functional tests in experimental animals. Rat tumor cells have the advantage of being able to be passaged in syngeneic animals, whereas human tumor cells have to be passaged in the rather artificial setting of an immunodeficient host. Furthermore, the expected high degrees of homology between rodent and human sequences should permit subsequent isolation of human homologues of candidate tumor progression genes and evaluation of their expression in primary human tumor material.

For the molecular comparisons, we used two rat carcinoma models. The rat pancreatic adenocarcinoma model is comprised of several clones that differ in their metastatic potential in vivo and have been derived from a common primary tumor (14) . For example, BSp73-1AS cells form primary tumors but do not metastasize, whereas BSp73-ASML cells are highly metastatic and, after s.c. injection, disseminate via the lymphatics to finally colonize the lungs. The rat mammary adenocarcinoma cell system 13762NF (16) is composed of a number of cell lines derived from a parental mammary tumor and its corresponding spontaneous lung and lymph node metastases. For example, the cell line MTPa has been reported to be nonmetastatic in vivo in syngeneic animals, whereas MTLY is highly metastatic, giving rise to multiple metastases in the lymph nodes and lungs (16) .

Before subtractions, the phenotypes of the cell lines were verified by s.c. injection of each cell line into syngeneic animals (Table 1)Citation . In all cases, 100% of animals developed tumors after injection of 5 x 105 cells. MTPa and BSp73-1AS did not metastasize at all; the MTLN cell lines and BSp73-ASML grew in both the lymph nodes and lungs, whereas MTLY and BSp73-10AS appeared to prefer the lymphoid organs. Despite extensive investigation of many organs, metastases were only observed in the lymph nodes or lungs.


View this table:
[in this window]
[in a new window]
  		Table 1 Verification of the in vivo metastatic potential of the rat cell lines used

A total of 5 x 105 viable cells in PBS were injected s.c. into the flank of each animal. Three to nine animals were used for each cell line, and the animals were maintained until the animals became moribund, or the primary tumor had reached the legal limit as set by German animal protection regulations. The animals were then sacrificed, and autopsy was performed. Metastases were noted and scored.

 
Subtractive Cloning of Differentially Expressed Genes.
From the cell lines examined and documented in Table 1Citation , we decided to compare the gene expression profiles of pairs of tumor cells with the most extreme metastatic behavior. Transcripts of the nonmetastatic mammary tumor cell line MTPa were subtracted from those of the metastatic cell line MTLY (this library was denoted ML). The nonmetastatic pancreatic cell line BSp73-1AS was subtracted from its metastatic counterpart BSp73-ASML (this library was denoted PL). Both subtractive libraries were generated and screened as described previously (10 , 13) . A total of 1985 and 5000 clones were screened from the mammary and pancreatic tumor libraries, respectively, resulting in the isolation of 487 and 494 differentially expressed cDNA clones, which, through sequence analysis, equated to 160 and 119 individual genes, respectively7 (Table 2Citation A). Eleven of these genes were found in both the mammary and pancreatic tumor library screens. Thus, a total of 268 different cDNAs were isolated in the two screens. A complete list of all clones obtained will be accessible via an internet web page.8


View this table:
[in this window]
[in a new window]
  		Table 2 Statistical sequence analysis

 
As indicated in Table 2Citation , the clones of both carcinoma-specific libraries were grouped according to their degree of sequence homology to the described public domain human and rodent genes (18 , 19) . Sixty-five clones identified from the ML subtraction and 43 clones identified from the PL subtraction represent described rat genes (see Table 2BCitation ); an additional 25 ML and 16 PL clones share significant homologies with known genes from other species, and some of them show homology to rat genes and could therefore encode new rat family members. All sequences with a homology of less than 70% to previously described genes can be separated into those that demonstrate significant homology to already existing ESTs and those that show little or no significant match to any sequence in the public databases. Clones with less than 70% homology to previously described genes have been classified as novel. On this basis, 70 of 160 (ML) and 60 of 119 (PL) clones represent previously unidentified sequences.7

By probing Northern blots with randomly selected clones, we convincingly verified that the SSH method does indeed identify differentially expressed genes (Fig. 1)Citation . The normalization step of SSH enriches for rare transcripts and thereby allows the identification of both high and low abundance transcripts. As indicated in Fig. 1Citation , clones C and B isolated from the PL and ML, respectively, represent genes expressed at low levels. Conversely, clones F (PL) and L (ML) represent transcripts of high abundance.



View larger version (74K):
[in this window]
[in a new window]
  		Fig. 1. Northern analysis of a number of positively identified clones from the ML and PL screens. To verify that the sequences isolated in the two SSH screens were indeed differentially expressed between the nonmetastasizing cell line and its metastasizing counterpart, we tested randomly selected cDNAs by direct Northern analysis. Each lane was loaded with 2 µg of poly(A)+ RNA. The poly(A)+ RNA was isolated from MTPa and MTLY cells (ML) or from BSp73-1AS, BSp73-10AS, and BSp73-ASML cells (PL). The blots were hybridized with the test cDNAs (B–L) or with glyceraldehyde-3-phosphate dehydrogenase (A), which serves as a loading control. For the ML library: B, HB2; C, mirf 6; D, novel clone 56; E, B 23; F, IQGAP2; G, novel clone 163; H, novel clone 154; I, novel clone 155; J, novel clone 159; K, cytokeratin 19; and L, novel clone 165. For the PL library: B, novel clone 75; C, novel clone 100, D, novel clone 135; E, caveolin-1; F, testin 2; G, uPAR; H, novel clone 72; I, ezrin; J, HMG-1; K, Src SH3-binding protein; and L, calgizzarin.

 
Cancer cells from different tumor types likely require common properties during metastasis, and therefore recurrent patterns of metastasis-related gene expression should be found. Indeed, within the list of metastasis-related cDNAs for which sequences are already in the databases, we find gene products that have been described previously by others as associated with tumor progression (21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51) .8 Furthermore, several genes were identified as being metastasis associated in both the PL and ML subtractions (Fig. 2)Citation . Moreover, several genes isolated in SAGE screens to identify human colonic and pancreatic tumor-specific genes were also found in the ML SSH and PL SSH libraries (Fig. 2)Citation .



View larger version (36K):
[in this window]
[in a new window]
  		Fig. 2. Sequence comparison between different screens. A illustrates a comparison of the genes isolated in the ML and PL screens presented in this study. Only 10 genes were isolated in both screens. B demonstrates the comparison between the SAGE screen for tumor-related genes in colon and pancreatic cancer (11) and our SSH analysis, based on the sequences published on the SAGE web site (http://welchlink.welch.jhu.edu). ESTs have been excluded in B because EST accession numbers have not been made accessible in the databases. C lists the identity of the genes that were found in more than one of the screens analyzed in A and B.

 
Expression of cDNA Sequences in Several Metastatic Tumor Cell Lines.
To determine to what extent metastatic tumor cells of different origin share a basic set of gene products expressed in the metastatic phenotype, we analyzed the expression of randomly selected clones in several tumor progression models (Fig. 3)Citation . For this purpose, two additional tumor systems were used, namely, a pair of mammary tumor cell lines and a group of related prostatic cancer cell lines (15 , 17) . The metastatic behavior of these tumor cell lines was verified by in vivo passaging (Table 1)Citation . Additionally, two human MDA-MB cell lines reported to be metastatic in immunocompromised mice (52 , 53) were included in the analysis.



View larger version (70K):
[in this window]
[in a new window]
  		Fig. 3. Comparison of expression profiles of novel genes and known tumor-associated molecules in cell lines of differing metastatic potential. The figure is composed of Northern blots from four different rat tumor systems and two human breast carcinoma cell lines. Except for the NM081/MT450 pair, all of the tumor systems used are comprised of clonal cell lines derived from a common primary tumor that differ only in their capacity to metastasize (see Table 1Citation ). This metastatic potential is indicated. Each lane on the Northern blots was loaded with 2 µg of poly(A)+ mRNA. The blots were hybridized with various differentially expressed cDNAs isolated in the mammary (A) and pancreatic (B) SHH screens. The hybridization probes used were randomly chosen from the metastasis-associated genes isolated in the ML and PL screens and represent 10% of the isolated genes in each case. A rough classification (correlative index) is given to describe the correlation between expression of the gene and the metastatic potential of the cell lines investigated. Genes whose expression is up-regulated in all metastatic cell lines and absent or significantly reduced in all nonmetastatic lines were scored +++. Genes that are clearly differentially but not exclusively expressed in metastatic cells were arbitrarily scored ++ or +, depending on the strength of the correlation of expression with metastatic potential. C, representative example of ethidium bromide-stained RNA gels before Northern blotting to demonstrate equivalence of loading.

 
By direct Northern analysis, we found that several of the selected clones were differentially expressed in metastasizing cell lines from tumor models other than the model from which the clones were originally isolated. cDNA clones whose expression correlates with metastatic potential in more than one tumor model are the most likely candidates for general relevance in the metastatic process. Genes shown to be differentially expressed in all four systems tested were CD24, S7, B23, oncomodulin, and HB2. Genes clearly differentially expressed but not correlating with metastatic potency in all tumor models tested were choline kinase, megakaryocyte potentiating factor, metastasis-associated factor, c-Met, mirf 6, cytokeratin 8, uPAR, HMG-1, and the novel clones 23, 25, 56, 62, 63, 82, 154, 162, and 163. The 26 clones used for this analysis represent only 10% of all of the clones identified and were randomly selected. The remaining 242 as yet untested genes are expected to show a similar degree of correlation between expression and metastatic potential.

Selected Examples of the Metastatic Gene Expression Pattern in Human Primary and Metastatic Carcinomas.
The ultimate test for the validity of our differential isolation approach is the exclusive or strongly enhanced expression of candidate clones in human cancer, which we tested by in situ hybridization analysis. We concentrated our study on two cDNA clones, CD24 and S7, which showed the best cross-hybridization to human sequences (Fig. 3)Citation and are highly expressed in the metastatic rat cell lines tested.

In situ hybridizations were performed on tumor sections from primary tumors as well as from corresponding metastases and nonneoplastic tissue from colorectal and breast carcinomas. In the colorectal carcinomas, CD24 mRNA expression was weakly detectable in normal colonic mucosa but highly abundant in tumor cells and, to a lesser extent, in the cells of the tumor stroma (Fig. 4)Citation . No marked difference in the expression levels was found between the primary tumors and the corresponding lymph node metastases. S7 was very weakly expressed in normal colorectal mucosa and was highly abundant in colorectal tumor cells. There were no marked differences in the expression levels of S7 between the primary tumors and the corresponding lymph node metastases. There was, however, a striking difference in the expression between S7 and CD24 in that CD24 was not only overexpressed in the invasive carcinoma but was also overexpressed in the adjacent noninvasive areas with mild atypia (compare Fig. 4, j–lCitation ). Furthermore, the expression of CD24 in the tumor stroma was more pronounced than the expression of S7.



View larger version (192K):
[in this window]
[in a new window]
  		Fig. 4. Detection of CD24 and S7 expression in human colorectal cancer by in situ hybridization. Sections of formaldehyde-fixed and paraffinembedded human colorectal carcinomas, corresponding nonneoplastic mucosa, and lymph node metastases were hybridized with fluorescence-labeled sense and antisense RNA and subsequently counterstained with H&E. a–f, CD24 expression in a poorly differentiated adenocarcinoma of the colon, staged as pT4b; pN2; pV1 (a–c, antisense; e and f, sense control). Note the marked overexpression of CD24 in the carcinoma (b; CA, carcinoma; M, muscularis propria) as compared with the nonneoplastic mucosa at the resection margin (a). g–i, S7 expression in the same samples shown in a–f. Note the marked overexpression of S7 in the primary cancer as compared with the nonneoplastic mucosa. Furthermore, there is even a slightly stronger hybridization signal in the metastasis than in the primary tumor (compare h and i). j–m, detailed analysis of CD24 and S7 expression in noninvasive areas with nuclear atypia (stars in j and k) and invasive carcinoma (j, S7 antisense; k, CD24 antisense) and the corresponding nonneoplastic mucosa at the resection margin (l, CD24 antisense; m, CD24 sense). Note that CD24 is already overexpressed in noninvasive lesions with nuclear atypia (compare k and j), whereas overexpression of S7 is strictly restricted to the invasive carcinoma (j). Magnification: a–i, x 60; j–m, x 120.

 
In the breast carcinomas, CD24 expression levels were essentially comparable in the nonneoplastic mammary parenchyma and in the carcinoma cells of the primary tumors as well as the corresponding metastases (Fig. 5)Citation . In general, the hybridization signals obtained for S7 were much weaker than those obtained for CD24. In comparison with nonneoplastic mammary gland parenchyma and primary carcinoma, a moderate overexpression of S7 was seen in lymph node metastases (Fig. 5)Citation .



View larger version (117K):
[in this window]
[in a new window]
  		Fig. 5. Detection of CD24 and S7 expression in human breast cancer by in situ hybridization. Sections of formaldehyde-fixed and paraffin-embedded human invasive ductal breast carcinomas (b and e), the corresponding nonneoplastic mammary parenchyma (a and d), and a lymph node metastasis (c and f) were hybridized with fluorescence-labeled sense and antisense RNA and subsequently counterstained with H&E. a–c, CD24 antisense probe; d–f, S7 antisense probe. Note that CD24 expression is clearly detected in nonneoplastic lobules, and there is no striking difference in the expression level between nonneoplastic parenchyma, primary carcinoma, and metastasis. S7 expression is not seen in nonneoplastic parenchyma and primary carcinoma but becomes detectable in the metastasis. a–f, x120.

 
Thus, genes whose expression we identified as being metastasis associated in several rodent tumor systems also show a tumor stage-dependent expression behavior in the human cancers investigated, particularly those of colorectal origin, proving the validity of our approach.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
It is clear from many studies that tumor cells that can metastasize have gained properties not possessed by their nonmetastasizing counterparts. These additional properties must be reflected in changed gene expression patterns. However, it is completely unknown how many and which genes are differentially expressed between metastasizing and nonmetastasizing cells from a given tumor. Furthermore, there is no indication as to what extent these differentially expressed genes are absolutely required for metastasis formation by other tumor types or to what extent there is redundancy in the gene expression patterns that lead to metastatic competence. Here we present the first reported attempt to describe and compare the complete sets of genes expressed in metastasizing but not in nonmetastasizing tumor cells.

From our experiments, we conclude that metastasizing tumor cells from different tumor types up-regulate many common genes. Firstly, many of the known genes isolated in the PL and ML screens have previously been associated with tumor progression. Secondly, the Northern blots in Fig. 3Citation show that many of the differentially expressed clones from the PL and ML screens are also expressed in a metastasis-specific fashion in other model systems. Thirdly, the two genes selected for in situ hybridization analysis are strongly up-regulated in human tumor material (Figs. 4Citation and 5)Citation . Fourthly, several genes were identified as being metastasis associated by both the PL and ML subtractions.

We also conclude that metastasizing tumor cells up-regulate large pools of genes. Genes identified as being metastasis associated in both the PL and ML subtractions were comparatively few in number (a total of 11). This could be interpreted to mean that mammary and pancreatic cancer cells use widely divergent subsets of genes to metastasize or that there is extensive redundancy in the requirement of genes whose expression is associated with metastasis formation, either because the same property can be provided by many different genes or groups of genes or because many genes are up-regulated but not necessary for metastasis formation. However, other data suggest that the yield of clones differentially expressed in our subtractions was not exhaustive. Randomly selected clones identified as being expressed in a metastasis-specific manner in the ML but not PL subtraction were used to screen Northern blots of the pancreatic tumor cells (Fig. 3)Citation . Here, 12 of 16 clones proved to be up-regulated in the metastatic BSp73-ASML but not in the BSp73-1AS nonmetastatic pancreatic carcinoma cells. Similarly, seven of seven clones from the PL screen that were not isolated in the ML screen were nevertheless more highly expressed in the metastasizing MTLY mammary carcinoma cells than in the nonmetastasizing MTPa cells. Thus, although we isolated many genes as being expressed in a metastasis-specific manner in the PL (119 genes) and ML (160 genes) screens, we anticipate that many more genes are also differentially expressed in metastatic cells of the rat mammary and pancreatic carcinoma models. This suggests that metastasizing tumor cells up-regulate a large pool of genes compared with their nonmetastasizing counterparts.

On the basis of our current data, we cannot draw any firm conclusions about the degree of redundancy in gene expression required for metastatic competence. There clearly is redundancy in the genes that are up-regulated during metastatic progression because several of the clones from the PL and ML screens were not significantly up-regulated in a metastasis-specific manner in all tumor models or were even down-regulated (e.g., uPAR, novel genes 23 and 82; see Fig. 3Citation ). This is to be expected: for every qualitative genetic change that is advantageous for metastatic progression, there will be many others that are either deleterious for the tumor cell or neutral in terms of conferring metastatic properties on the tumor cell. Furthermore, some genes are likely to be up-regulated in a cell type-specific manner.

Several laboratories have made attempts to screen for cancer-associated transcripts using different methods [differential display, representational differentational analysis, SAGE, and SSH (10 , 54, 55, 56, 57) ]. Of those techniques capable of efficient expression profiling, SAGE has the advantage that even moderate changes in transcription can be detected, but it suffers from the fact that the sequence information it produces is only 13-bp long. Not only is this problematic for nucleotide sequence comparison algorithms, but it is also not ideal for use in techniques such as in situ hybridization and Northern blotting. On the other hand, subtractive cloning techniques such as SSH yield cDNA fragments of 200–800 bp, which overcomes the problems associated with SAGE. Furthermore, SSH enriches differentially expressed genes with high efficiency irrespective of their relative abundance. The combination of SSH with high-throughput screening methods, for example, using microarrays (58) , provides a powerful method for the rapid identification of differentially expressed genes. SSH suffers from the drawback that it does not readily detect transcriptional differences less than 5-fold. SAGE and SSH have never been compared directly using the same cell system. However, the data presented in this study verify that SSH is a valuable tool for identifying differentially expressed genes in tumors and other cellular contexts.

Because the data presented here represent the first extensive description of genes expressed specifically in metastasizing cells, comparison with previous differential screens is informative but needs to be interpreted with caution. The only other extensive descriptions of genes up-regulated in tumors used differential hybridization to arrayed cDNAs, SAGE, or SSH. These studies either compared normal tissue with tumor material or ER-positive cells with ER-negative cells, and no investigation of metastasis-related gene expression has been made (11 , 58, 59, 60) When the differentially expressed sequences in these analyses are compared with the sequences identified in the ML and PL screens, few of the sequences are identical. This is illustrated by a comparison of human pancreatic tumor-specific genes identified by SAGE with metastasis-specific genes expressed in rat pancreatic tumor cells identified by SSH (Fig. 2)Citation . At first sight, this would suggest that many of the genes identified by SAGE as being up-regulated in the tumor material do not contribute to the metastatic phenotype but rather are concerned with tumorigenesis. However, it should be borne in mind that the different screens do not constitute an exhaustive list of up-regulated genes. Furthermore, these comparisons are preliminary because ESTs are not always made accessible in the databases, and identities may be overlooked if the ESTs are short.

With respect to molecular functions of the genes we report to be expressed in a metastasis-specific fashion, several of the known sequences isolated in the PL and ML screens have previously been reported to be involved in the process of metastasis. For example, c-Met is the receptor for hepatocyte growth factor/scatter factor. A large body of evidence implicates this receptor-ligand pair in the processes of tumorigenesis, invasion, migration, and metastasis (reviewed in Ref. 61 ). Interestingly, another gene identified as being up-regulated in metastasizing cells is ezrin, a cytoplasmic substrate for c-Met required for hepatocyte growth factor-mediated morphogenesis whose expression is associated with the metastatic phenotype (62 , 63) . We also isolated uPAR in our screens, which binds urokinase-type plasminogen activator and is believed to be essential for activation of numerous proteases involved in invasion (64) . As a further example, elevated caveolin protein levels have been associated with human prostate carcinoma progression (65) , and we report here that caveolin is a gene up-regulated in metastatic cells.

The data presented here would suggest that some of the clones we have identified as being expressed in a tumorigenesis-related fashion could also prove to serve as new markers for further elucidation of the molecular events in the dysplasia-carcinoma sequence. Others may serve as targets for therapeutic intervention. An impressive correlation between expression and metastatic potential was seen with CD24, S7, and several novel clones. CD24 is expressed on B-cell precursors, neutrophils, neurons, and keratinocytes (66 , 67) , and its expression has not previously been associated directly with metastasis. However, it is expressed on leukemic cells of the B-cell lineage and on various solid tumors such as small cell lung carcinoma, neuroblastoma, rhabdomyosarcoma, and renal cell carcinoma (68, 69, 70, 71) . CD24 was also identified by a SSH screen to identify genes differentially expressed between ER-positive and -negative cells lines (58) . Significantly, reduction or absence of CD24 expression on acute lymphoblastic lymphoma cells predicts a better prognosis for patients (72) . As a surface protein able to bind to P-selectin, CD24 could promote metastasis formation by mediating binding of circulating tumor cells to endothelium (32) .

Although several ribosomal components have been reported to be up-regulated in cancer, S7 has not previously been associated with tumor growth and metastasis. It is a component of the translation machinery, and its RNA binding properties are required for the correct folding of the 16S rRNA (73) . Our observations call for a novel function for S7. Enhanced S7 expression in tumor cells is not simply a correlate of enhanced proliferation rate because only very weak expression is observed in the highly proliferative colonic mucosa. In colorectal carcinomas, we observed that S7 is homogeneously expressed and up-regulated specifically in invasive carcinomas and metastases. Moreover, S7 expression is also clearly enhanced in metastases of the mammary carcinomas we investigated. Our data suggest that S7 may be useful in certain instances as a marker for tumor staging and indicate that additional experiments are warranted to investigate whether S7 plays a functional role in invasive tumor growth.

In conclusion, this study is the first reported attempt to describe the profile of genes expressed specifically in metastasizing tumor cells using a direct comparative method and proves that SSH is a useful tool in identifying differentially expressed genes. We demonstrate that a large number of genes are up-regulated during progression to metastatic competence, and we have confirmed the association of the expression of certain genes with the metastatic phenotype. More significantly, we have connected the expression of already characterized genes that have not previously been linked with tumor metastasis with the metastatic phenotype. Most importantly, we have isolated totally novel metastasis-associated genes. These genes represent a rich source of candidates for application in the diagnosis and therapy of human cancer.

Note Added in Proof:
While this manuscript was under review, Clark et al.(Nature 406: 532–535, 2000) reported the isolation of genes using microarray screening whose expression correlates with metastatic potential.


   ACKNOWLEDGMENTS
 
We thank Andrea Fuchsbichler for able assistance with the in situ hybridizations and Christian Ahrens for preliminary data analysis.


   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 in part by a grant from the Mildred-Scheel Stiftung and Grant S7401-MOB from the Austrian Science Fund (to K. Z.) Back

2 Present address: LYNX Therapeutics GmbH, Im Neuenheimer Feld 515, D-69120 Heidelberg, Germany. Back

3 Present address: LION Bioscience AG, Im Neuenheimer Feld 515, D-69120 Heidelberg, Germany. Back

4 To whom requests for reprints should be addressed, at Forschungszentrum Karlsruhe, Institute of Toxicology and Genetics, P. O. Box 3640, D-76021 Karlsruhe, Germany. Phone: 49-7247-826069; Fax: 49-7247-823354; E-mail: sleeman{at}itg.fzk.de Back

5 The abbreviations used are: SSH, suppression subtractive hybridization; ML, mammary carcinoma-specific library; PL, pancreatic carcinoma-specific library; poly(A)+ RNA, polyadenylated RNA; EST, expressed sequence tag.; SAGE, serial analysis of gene expression; uPAR, urokinase-type plasminogen activator receptor; ER, estrogen receptor. Back

6 See http://www.ncbi.nlm.nih.gov/BLAST/for details. Back

7 It is possible that some of the novel sequences identified originate from the same gene. We have assumed that each novel EST we identified corresponds to one gene. The most recent database search was performed on November 24, 2000. Back

8 http://igtmv1.fzk.de/www/itg/sleeman/sleeman.html. Back

Received 7/31/00. Accepted 12/15/00.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Tlsty T. D. Cell-adhesion-dependent influences on genomic instability and carcinogenesis.. Curr. Opin. Cell Biol., 10: 647-653, 1998.[Medline]
  2. Fidler I. J. Tumor heterogeneity and the biology of cancer invasion and metastasis.. Cancer Res., 38: 2651-2660, 1978.[Abstract/Free Full Text]
  3. Sleeman J. P. The lymph node as a bridgehead in the metastatic dissemination of tumors.. Recent Results Cancer Res., 157: 55-81, 2000.[Medline]
  4. Pencil S. D., Toh Y., Nicolson G. L. Candidate metastasis-associated genes of the rat 13762NF mammary adenocarcinoma.. Breast Cancer Res. Treat., 25: 165-174, 1993.[Medline]
  5. Daigneault L., Beaulieu R., Filion M., Gaboury L., Royal A., Babai F. Cloning and identification of genes differentially expressed in metastatic and non-metastatic rat rhabdomyosarcoma cell lines.. Clin. Exp. Metastasis, 13: 345-356, 1995.[Medline]
  6. Salesiotis A. N., Wang C. K., Wang C. D., Burger A., Li H., Seth A. Identification of novel genes from stomach cancer cell lines by differential display.. Cancer Lett., 91: 47-54, 1995.[Medline]
  7. van Groningen J. J., Bloemers H. P., Swart G. W. Identification of melanoma inhibitory activity and other differentially expressed messenger RNAs in human melanoma cell lines with different metastatic capacity by messenger RNA differential display.. Cancer Res., 55: 6237-6243, 1995.[Abstract/Free Full Text]
  8. Ishiguro T., Nakajima M., Naito M., Muto T., Tsuruo T. Identification of genes differentially expressed in B16 murine melanoma sublines with different metastatic potentials.. Cancer Res., 56: 875-879, 1996.[Abstract/Free Full Text]
  9. Hashimoto Y., Shindo-Okada N., Tani M., Takeuchi K., Toma H., Yokota J. Identification of genes differentially expressed in association with metastatic potential of K-1735 murine melanoma by messenger RNA differential display.. Cancer Res., 56: 5266-5271, 1996.[Abstract/Free Full Text]
  10. Diatchenko L., Lau Y. F., Campbell A. P., Chenchik A., Moqadam F., Huang B., Lukyanov S., Lukyanov K., Gurskaya N., Sverdlov E. D., Siebert P. D. Suppression subtractive hybridization: a method for generating differentially regulated or tissue-specific cDNA probes and libraries.. Proc. Natl. Acad. Sci. USA, 93: 6025-6030, 1996.[Abstract/Free Full Text]
  11. Zhang L., Zhou W., Velculescu V. E., Kern S. E., Hruban R. H., Hamilton S. R., Vogelstein B., Kinzler K. W. Gene expression profiles in normal and cancer cells.. Science (Washington DC), 276: 1268-1272, 1997.[Abstract/Free Full Text]
  12. Sgroi D. C., Teng S., Robinson G., LeVangie R., Hudson J. R., Jr., Elkahloun A. G. In vivo gene expression profile analysis of human breast cancer progression. Cancer Res., 59: 5656-5661, 1999.[Abstract/Free Full Text]
  13. von Stein O. D., Thies W. G., Hofmann M. A high throughput screening for rarely transcribed differentially expressed genes.. Nucleic Acids Res., 25: 2598-2602, 1997.[Abstract/Free Full Text]
  14. Matzku S., Komitowski D., Mildenberger M., Zoller M. Characterization of BSp73, a spontaneous rat tumor, and its in vivo selected variants showing different metastasizing capacities.. Invasion Metastasis, 3: 109-123, 1983.[Medline]
  15. Isaacs J. T., Isaacs W. B., Feitz W. F., Scheres J. Establishment and characterization of seven Dunning rat prostatic cancer cell lines and their use in developing methods for predicting metastatic abilities of prostatic cancers.. Prostate, 9: 261-281, 1986.[Medline]
  16. Neri A., Nicolson G. L. Phenotypic drift of metastatic and cell-surface properties of mammary adenocarcinoma cell clones during growth in vitro.. Int. J. Cancer, 28: 731-738, 1981.[Medline]
  17. Kim U. Pathogenesis and characteristics of spontaneously metastasizing mammary carcinomas and the principle of metastasis.. J. Surg. Oncol., 33: 151-165, 1986.[Medline]
  18. Tatusova T. A., Madden T. L. BLAST 2 Sequences, a new tool for comparing protein and nucleotide sequences.. FEMS Microbiol. Lett., 174: 247-250, 1999.[Medline]
  19. Altschul S. F., Madden T. L., Schaffer A. A., Zhang J., Zhang Z., Miller W., Lipman D. J. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.. Nucleic Acids Res., 25: 3389-3402, 1997.[Abstract/Free Full Text]
  20. Osterreichische Gesellschaft für Pathologie Histologische Tumorklassifikation, 2nd ed. Wien, New York: Springer Verlag, 1994.
  21. Di Renzo M. F., Poulsom R., Olivero M., Comoglio P. M., Lemoine N. R. Expression of the Met/hepatocyte growth factor receptor in human pancreatic cancer.. Cancer Res., 55: 1129-1138, 1995.[Abstract/Free Full Text]
  22. Ferno M., Bendahl P. O., Borg A., Brundell J., Hirschberg L., Olsson H., Killander D. Urokinase plasminogen activator, a strong independent prognostic factor in breast cancer, analysed in steroid receptor cytosols with a luminometric immunoassay.. Eur. J. Cancer, 32A: 793-801, 1996.
  23. Sivaparvathi M., Sawaya R., Gokaslan Z. L., Chintala S. K., Rao J. S., Chintala K. S. Expression and the role of cathepsin H in human glioma progression and invasion.. Cancer Lett., 104: 121-126, 1996.[Medline]
  24. Nasu Y., Timme T. L., Yang G., Bangma C. H., Li L., Ren C., Park S. H., DeLeon M., Wang J., Thompson T. C. Suppression of caveolin expression induces androgen sensitivity in metastatic androgen-insensitive mouse prostate cancer cells.. Nat. Med., 4: 1062-1064, 1998.[Medline]
  25. Wauters C. C., Smedts F., Gerrits L. G., Bosman F. T., Ramaekers F. C. Keratins 7 and 20 as diagnostic markers of carcinomas metastatic to the ovary.. Hum. Pathol., 26: 852-855, 1995.[Medline]
  26. Yokoyama N., Shirai Y., Hatakeyama K. Immunohistochemical detection of lymph node micrometastases from gallbladder carcinoma using monoclonal anticytokeratin antibody.. Cancer (Phila.), 85: 1465-1469, 1999.[Medline]
  27. Schoenfeld A., Luqmani Y., Sinnett H. D., Shousha S., Coombes R. C. Keratin 19 mRNA measurement to detect micrometastases in lymph nodes in breast cancer patients.. Br. J. Cancer, 74: 1639-1642, 1996.[Medline]
  28. Rosel M., Claas C., Seiter S., Herlevsen M., Zoller M. Cloning and functional characterization of a new phosphatidyl-inositol anchored molecule of a metastasizing rat pancreatic tumor.. Oncogene, 17: 1989-2002, 1998.[Medline]
  29. Bernaert D., Brewer L. M., MacManus J. P., Galand P. Immunocytochemical detection of the onco-developmental protein oncomodulin in pre-neoplastic and neoplastic hepatocellular lesions during hepatocarcinogenesis in rats.. Int. J. Cancer, 43: 719-727, 1989.[Medline]
  30. Amundadottir L. T., Nass S. J., Berchem G. J., Johnson M. D., Dickson R. B. Cooperation of TGF {alpha} and c-Myc in mouse mammary tumorigenesis: coordinated stimulation of growth and suppression of apoptosis.. Oncogene, 13: 757-765, 1996.[Medline]
  31. LaCasse E. C., Baird S., Korneluk R. G., MacKenzie A. E. The inhibitors of apoptosis (IAPs) and their emerging role in cancer.. Oncogene, 17: 3247-3259, 1998.[Medline]
  32. Aigner S., Ramos C. L., Hafezi-Moghadam A., Lawrence M. B., Friederichs J., Altevogt P., Ley K. CD24 mediates rolling of breast carcinoma cells on P-selectin.. FASEB J., 12: 1241-1251, 1998.[Abstract/Free Full Text]
  33. Kondo T., Minamino N., Nagamura-Inoue T., Matsumoto M., Taniguchi T., Tanaka N. Identification and characterization of nucleophosmin/B23/numatrin which binds the anti-oncogenic transcription factor IRF-1 and manifests oncogenic activity.. Oncogene, 15: 1275-1281, 1997.[Medline]
  34. Nygaard S. J., Haugland H. K., Kristoffersen E. K., Lund-Johansen M., Laerum O. D., Tysnes O. B. Expression of annexin II in glioma cell lines and in brain tumor biopsies.. J. Neurooncol., 38: 11-18, 1998.[Medline]
  35. Gill Z. P., Perks C. M., Newcomb P. V., Holly J. M. Insulin-like growth factor-binding protein (IGFBP-3) predisposes breast cancer cells to programmed cell death in a non-IGF-dependent manner.. J. Biol. Chem., 272: 25602-25607, 1997.[Abstract/Free Full Text]
  36. Pomerantz J., Schreiber-Agus N., Liegeois N. J., Silverman A., Alland L., Chin L., Potes J., Chen K., Orlow I., Lee H. W., Cordon-Cardo C., DePinho R. A. The Ink4a tumor suppressor gene product, p19Arf, interacts with MDM2 and neutralizes MDM2’s inhibition of p53.. Cell, 92: 713-723, 1998.[Medline]
  37. Tanaka M., Adzuma K., Iwami M., Yoshimoto K., Monden Y., Itakura M. Human calgizzarin; one colorectal cancer-related gene selected by a large scale random cDNA sequencing and Northern blot analysis.. Cancer Lett., 89: 195-200, 1995.[Medline]
  38. Kobayashi T., Hirayama Y., Kobayashi E., Kubo Y., Hino O. A germline insertion in the tuberous sclerosis (Tsc2) gene gives rise to the Eker rat model of dominantly inherited cancer.. Nat. Genet., 9: 70-74, 1995.[Medline]
  39. Xiang Y. Y., Wang D. Y., Tanaka M., Suzuki M., Kiyokawa E., Igarashi H., Naito Y., Shen Q., Sugimura H. Expression of high-mobility group-1 mRNA in human gastrointestinal adenocarcinoma and corresponding non-cancerous mucosa.. Int. J. Cancer, 74: 1-6, 1997.[Medline]
  40. Hernandez-Alcoceba R., Saniger L., Campos J., Nunez M. C., Khaless F., Gallo M. A., Espinosa A., Lacal J. C. Choline kinase inhibitors as a novel approach for antiproliferative drug design.. Oncogene, 15: 2289-2301, 1997.[Medline]
  41. Bang Y. J., Pirnia F., Fang W. G., Kang W. K., Sartor O., Whitesell L., Ha M. J., Tsokos M., Sheahan M. D., Nguyen P., et al Terminal neuroendocrine differentiation of human prostate carcinoma cells in response to increased intracellular cyclic AMP.. Proc. Natl. Acad. Sci. USA, 91: 5330-5334, 1994.[Abstract/Free Full Text]
  42. Hecht J. T., Hogue D., Wang Y., Blanton S. H., Wagner M., Strong L. C., Raskind W., Hansen M. F., Wells D. Hereditary multiple exostoses (EXT): mutational studies of familial EXT1 cases and EXT-associated malignancies.. Am. J. Hum. Genet., 60: 80-86, 1997.[Medline]
  43. Wu C. G., Habib N. A., Mitry R. R., Reitsma P. H., van Deventer S. J., Chamuleau R. A. Overexpression of hepatic prothymosin {alpha}, a novel marker for human hepatocellular carcinoma.. Br. J. Cancer, 76: 1199-1204, 1997.[Medline]
  44. Fulchignoni-Lataud M. C., Olchwang S., Serre J. L. The fragile X CGG repeat shows a marked level of instability in hereditary non-polyposis colorectal cancer patients.. Eur. J. Hum. Genet., 5: 89-93, 1997.
  45. Kondoh G., Yomogida K., Dohmae K., Nozawa M., Koga M., Nonomura N., Miki T., Okuyama A., Nishimune Y. Coexpression of multiple Sertoli cell and Leydig cell marker genes in the spontaneous testicular tumor of F344 rat: evidence for phenotypical bifurcation of the interstitial cell tumor.. Jpn. J. Cancer Res., 88: 839-845, 1997.[Medline]
  46. Berta G. N., Ghezzo F., D’Avolio A., Zulian P., Carbone V., Racca S., Vercellino V., Di Carlo F. Enhancement of calcyclin gene RNA expression in squamous cell carcinoma of the oral mucosa, but not in benign lesions.. J. Oral Pathol. Med., 26: 206-210, 1997.[Medline]
  47. Kojima T., Oh-eda M., Hattori K., Taniguchi Y., Tamura M., Ochi N., Yamaguchi N. Molecular cloning and expression of megakaryocyte potentiating factor cDNA.. J. Biol. Chem., 270: 21984-21990, 1995.[Abstract/Free Full Text]
  48. Manni A., Grove R., Kunselman S., Aldaz M. Involvement of the polyamine pathway in breast cancer progression.. Cancer Lett., 92: 49-57, 1995.[Medline]
  49. Lamb R. F., Ozanne B. W., Roy C., McGarry L., Stipp C., Mangeat P., Jay D. G. Essential functions of ezrin in maintenance of cell shape and lamellipodial extension in normal and transformed fibroblasts.. Curr. Biol., 7: 682-688, 1997.[Medline]
  50. Bleuel K., Popp S., Fusenig N. E., Stanbridge E. J., Boukamp P. Tumor suppression in human skin carcinoma cells by chromosome 15 transfer or thrombospondin-1 overexpression through halted tumor vascularization.. Proc. Natl. Acad. Sci. USA, 96: 2065-2070, 1999.[Abstract/Free Full Text]
  51. Celis J. E., Ostergaard M., Basse B., Celis A., Lauridsen J. B., Ratz G. P., Andersen I., Hein B., Wolf H., Orntoft T. F., Rasmussen H. H. Loss of adipocyte-type fatty acid binding protein and other protein biomarkers is associated with progression of human bladder transitional cell.. Cancer Res., 56: 4782-4790, 1996.[Abstract/Free Full Text]
  52. Zhang R. D., Fidler I. J., Price J. E. Relative malignant potential of human breast carcinoma cell lines established from pleural effusions and a brain metastasis.. Invasion Metastasis, 11: 204-215, 1991.[Medline]
  53. Hendrix M. J., Seftor E. A., Seftor R. E., Trevor K. T. Experimental co-expression of vimentin and keratin intermediate filaments in human breast cancer cells results in phenotypic interconversion and increased invasive behavior.. Am. J. Pathol., 150: 483-495, 1997.[Abstract]
  54. Liang P., Pardee A. B. Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction.. Science (Washington DC), 257: 967-971, 1992.[Abstract/Free Full Text]
  55. Sokolov B. P., Prockop D. J. A rapid and simple PCR-based method for isolation of cDNAs from differentially expressed genes.. Nucleic Acids Res., 22: 4009-4015, 1994.[Abstract/Free Full Text]
  56. Lisitsyn N., Wigler M. Cloning the differences between two complex genomes.. Science (Washington DC), 259: 946-951, 1993.[Abstract]
  57. Velculescu V. E., Zhang L., Vogelstein B., Kinzler K. W. Serial analysis of gene expression.. Science (Washington DC), 270: 484-487, 1995.[Abstract/Free Full Text]
  58. Yang G. P., Ross D. T., Kuang W. W., Brown P. O., Weigel R. J. Combining SSH and cDNA microarrays for rapid identification of differentially expressed genes.. Nucleic Acids Res., 27: 1517-1523, 1999.[Abstract/Free Full Text]
  59. Gress T. M., Muller-Pillasch F., Geng M., Zimmerhackl F., Zehetner G., Friess H., Buchler M., Adler G., Lehrach H. A pancreatic cancer-specific expression profile.. Oncogene, 13: 1819-1830, 1996.[Medline]
  60. Kang D. C., LaFrance R., Su Z. Z., Fisher P. B. Reciprocal subtraction differential RNA display: an efficient and rapid procedure for isolating differentially expressed gene sequences.. Proc. Natl. Acad. Sci. USA, 95: 13788-13793, 1998.[Abstract/Free Full Text]
  61. Vande Woude G. F., Jeffers M., Cortner J., Alvord G., Tsarfaty I., Resau J. Met-HGF/SF: tumorigenesis, invasion and metastasis. CIBA Found. Symp., 212: 119-130, 1997.[Medline]
  62. Crepaldi T., Gautreau A., Comoglio P. M., Louvard D., Arpin M. Ezrin is an effector of hepatocyte growth factor-mediated migration and morphogenesis in epithelial cells. J. Cell Biol., 138: 423-434, 1997.[Abstract/Free Full Text]
  63. Akisawa N., Nishimori I., Iwamura T., Onishi S., Hollingsworth M. A. High levels of ezrin expressed by human pancreatic adenocarcinoma cell lines with high metastatic potential.. Biochem. Biophys. Res. Commun., 258: 395-400, 1999.[Medline]
  64. Rabbani S. A., Xing R. H. Role of urokinase (uPA) and its receptor (uPAR) in invasion and metastasis of hormone-dependent malignancies.. Int. J. Oncol., 12: 911-920, 1998.[Medline]
  65. Yang G., Truong L. D., Timme T. L., Ren C., Wheeler T. M., Park S. H., Nasu Y., Bangma C. H., Kattan M. W., Scardino P. T., Thompson T. C. Elevated expression of caveolin is associated with prostate and breast cancer.. Clin. Cancer Res., 4: 1873-1880, 1998.[Abstract]
  66. Shirasawa T., Akashi T., Sakamoto K., Takahashi H., Maruyama N., Hirokawa K. Gene expression of CD24 core peptide molecule in developing brain and developing non-neural tissues.. Dev. Dyn., 198: 1-13, 1993.[Medline]
  67. Magnaldo T., Barrandon Y. CD24 (heat stable antigen, nectadrin), a novel keratinocyte differentiation marker, is preferentially expressed in areas of the hair follicle containing the colony-forming cells.. J. Cell Sci., 109: 3035-3045, 1996.[Abstract]
  68. Perri R. T., Royston I., LeBien T. W., Kay N. E. Chronic lymphocytic leukemia progenitor cells carry the antigens T65, BA-1, and Ia.. Blood, 61: 871-875, 1983.[Abstract/Free Full Text]
  69. Droz D., Zachar D., Charbit L., Gogusev J., Chretein Y., Iris L. Expression of the human nephron differentiation molecules in renal cell carcinomas.. Am. J. Pathol., 137: 895-905, 1990.[Abstract]
  70. Jackson D., Waibel R., Weber E., Bell J., Stahel R. A. CD24, a signal-transducing molecule expressed on human B cells, is a major surface antigen on small cell lung carcinomas.. Cancer Res., 52: 5264-5270, 1992.[Abstract/Free Full Text]
  71. Akashi T., Shirasawa T., Hirokawa K. Gene expression of CD24 core polypeptide molecule in normal rat tissues and human tumor cell lines.. Virchows Arch., 425: 399-406, 1994.[Medline]
  72. Lavabre-Bertrand T., Duperray C., Brunet C., Poncelet P., Exbrayat C., Bourquard P., Lavabre- Bertrand C., Brochier J., Navarro M., Janossy G. Quantification of CD24 and CD45 antigens in parallel allows a precise determination of B-cell maturation stages: relevance for the study of B-cell neoplasias. Leukemia (Baltimore), 8: 402-408, 1994.[Medline]
  73. Wimberly B. T., White S. W., Ramakrishnan V. The structure of ribosomal protein S7 at 1.9 A resolution reveals a ß-hairpin motif that binds double-stranded nucleic acids.. Structure, 5: 1187-1198, 1997.[Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Cell Physiol.Home page
M. L. Styers, A. K. O'Connor, R. Grabski, E. Cormet-Boyaka, and E. Sztul
Depletion of {beta}-COP reveals a role for COP-I in compartmentalization of secretory compartments and in biosynthetic transport of caveolin-1
Am J Physiol Cell Physiol, June 1, 2008; 294(6): C1485 - C1498.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
S. Yamashita, S. Takahashi, N. McDonell, N. Watanabe, T. Niwa, K. Hosoya, Y. Tsujino, T. Shirai, and T. Ushijima
Methylation Silencing of Transforming Growth Factor-{beta} Receptor Type II in Rat Prostate Cancers
Cancer Res., April 1, 2008; 68(7): 2112 - 2121.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
R. Monni, L. Haddaoui, A. Naba, I. Gallais, M. Arpin, P. Mayeux, and F. Moreau-Gachelin
Ezrin is a target for oncogenic Kit mutants in murine erythroleukemia
Blood, March 15, 2008; 111(6): 3163 - 3172.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
L. Yang, H. Avila, H. Wang, J. Trevino, G. E. Gallick, Y. Kitadai, T. Sasaki, and D. D. Boyd
Plasticity in Urokinase-Type Plasminogen Activator Receptor (uPAR) Display in Colon Cancer Yields Metastable Subpopulations Oscillating in Cell Surface uPAR Density--Implications in Tumor Progression
Cancer Res., August 15, 2006; 66(16): 7957 - 7967.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
L. Lugini, P. Matarrese, A. Tinari, F. Lozupone, C. Federici, E. Iessi, M. Gentile, F. Luciani, G. Parmiani, L. Rivoltini, et al.
Cannibalism of live lymphocytes by human metastatic but not primary melanoma cells.
Cancer Res., April 1, 2006; 66(7): 3629 - 3638.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
Y. Yu, E. Davicioni, T. J. Triche, and G. Merlino
The Homeoprotein Six1 Transcriptionally Activates Multiple Protumorigenic Genes but Requires Ezrin to Promote Metastasis
Cancer Res., February 15, 2006; 66(4): 1982 - 1989.
[Abstract] [Full Text] [PDF]