| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Tumor Biology |
Department of Experimental Pathology and Laboratory Medicine [A. A., M. A. T.], Molecular Biology Institute [S. M. W., M. A. T., C. T. D.], Jonsson Comprehensive Cancer Center [M. A. T., C. T. D.], and Department of Pediatrics, Gwynne Hazen Cherry Memorial Labs [C. T. D.], University of California at Los Angeles, Los Angeles, California 90095
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
C) have a greatly reduced capability to form colonies in soft agar and tumors in severe combined immunodeficient mice. The rate of tumor formation for NIH 3T3 that express EWS/FLI1 C is 50 days, whereas EWS/FLI1 forms tumors within 22 days. In addition, cells expressing the EWS/FLI1 C mutant failed to completely demonstrate the round-cell histology that is seen in both Ewing’s tumor cell lines and NIH 3T3 cells expressing full-length EWS/FLI1. Northern and microarray analyses were performed to assess the effect of loss of the FLI1 COOH terminus on transcriptional modulation of EWS/FLI1 target genes. We found that although EWS/FLI1 C up-regulates smaller numbers of genes (21 genes) compared with EWS/FLI1 (34 genes), 41% of the EWS/FLI1 targets were also up-regulated by EWS/FLI1 C. On the other hand, EWS/FLI1 C is unable to down-regulate genes (3 genes) as efficiently as EWS/FLI1 (39 genes) with only one target gene repressed by both fusion constructs. Our study indicates that the EWS/FLI1 transcription factor has strong transcriptional activating as well as repressing properties and suggests that transcriptional activation and repression of target genes may occur through biochemically different mechanisms. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
EWS/FLI1 is the most prevalent fusion and is found in 
85% of ES/PNET patients (3
, 4)
. Alternative and less frequent EWS fusion partners in ES/PNET include: ERG on chromosome 21 q22.2, ETV-1/ER81 on chromosome 7 p22, E1AF/PEA3 on chromosome 17 q12, and FEV on chromosome 2 (5, 6, 7, 8)
. ES/PNET fusion genes can be subdivided into two groups according to amino acid sequence similarity in their ETS DBDs. EWS/FLI1, EWS/ERG, and EWS/FEV form one highly related group (Group 1). EWS/ETV1 and EWS/E1AF form the second group (Group 2).
Extensive experimental data exist that demonstrate the oncogenic potential of EWS/ETS fusion genes from both groups. Expression of EWS/ETS proteins promote anchorage-independent growth in vitro and/or accelerate tumorigenesis of NIH 3T3 cells in vivo (9, 10, 11) . The transforming activities of EWS/FLI1 require both the EWS and the DBD of FLI1 (12, 13, 14) . Antagonizing EWS/FLI1 through antisense oligonucleotides or dominant negative FLI1 constructs inhibits growth of ES/PNET tumor-derived cell lines both in vitro and in vivo (15, 16, 17, 18, 19, 20) . These data, together with the observation that EWS fusion genes and fusion proteins are present in virtually all Ewing’s cases, provide strong evidence that formation of EWS/ETS chimeric fusion gene is a necessary event in the development of ES/PNET.
Recently, we have shown that NIH 3T3 cells that express EWS/FLI1 or EWS/ETV1 form tumors in immunodeficient mice that display a round-cell histology reminiscent of ES/PNET (11) . This result suggests that despite their structural differences, Group 1 and Group 2 EWS/ETS fusions both promote tumorigenesis through similar genetic pathways. However, there appears to be a quantitative difference in tumorigenic potential between these two EWS/ETS fusion types: EWS/FLI1 was able to induce tumor formation at a significantly faster rate than EWS/ETV1. Surprisingly, initial experiments suggested that the enhanced tumorigenic effect of EWS/FLI1 mapped to the COOH-terminal 89-amino-acids of FLI1 rather than the FLI1 DBD (11) .
This present work extends this initial finding and demonstrates crucial biological roles of the ETS COOH-terminal domain in EWS/ETS fusion genes. We show that the loss of the COOH-terminal domain greatly attenuates the ability of EWS/FLI1 to promote anchorage-independent growth. Cells that express EWS/FLI1 COOH-terminal deletion mutants demonstrate a progressive decrease in tumorigenic potency and loss of round-cell histology. Finally, deletion of the FLI1 COOH terminus profoundly effects transcriptional repression of specific EWS/FLI1 target genes.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
C, and FEF C DerivativesC was created by PCR amplification of the parental FEF (type 4). We used a 5' primer that contained an EcoRI restriction site, Flag sequence, and the first 20 bp of the EWS coding region; and a 3' primer that matched the sequence of the last 30 bp found at the end of the FLI1 DBD, followed by a HindIII site. Constructs FEF 3, and FEF (2+3) were made the same way, using the same 5' primer, but with 3' primers matching the last 20 bp of region 2 in the COOH terminus, or the last 20 bp of region 1, respectively. Constructs FEF 1, FEF 2, FEF (1+2), and FEF (1+3) were created using a two-stage PCR process. In the first step, two fragments were amplified. The first fragment, a 5' EWS/FLI C fragment, was amplified as described above. The second fragment was created by using complementary 3' PCR primers to specific regions in the COOH terminus and corresponding 5' primer for that region. These two PCR fragments were then joined by amplifying them in a reaction using vector-specific primers that flanked EWS and the FLI1 sequences. All of our 3' primers contained a HindIII restriction sites. The PCR products were then digested with EcoRI and HindIII, gel purified, and subcloned into the retroviral expression vector SR
MSV-tkneo HindIII. The inserts were then subcloned into pBluescript II KS+ and sequenced.
Retrovirus Production and Cell Culture
For production of retroviruses, we used 293T cells. Cells were cultured in DMEM supplemented with 5% calf serum and glutamine. 293Ts were transfected with recombinant SR MSV-Tk Neo expression vectors and 
(-) packaging plasmids by the calcium phosphate precipitation method. After 48 h, the media containing the retrovirus were harvested. Infections were carried out by adding the recombinant viral stocks to polyclonal NIH 3T3 cells; the infected cells were then selected with 450 µg of G418 per ml of culture media for 7–10 days.
The monoclonal cell lines were created by using a limiting dilution protocol. In brief, cells were seeded in 96-well culture plates at 5 cells/ml or 0.5 cells/ml, and fed with conditioned media from the parental lines. Wells containing single clones were expanded to establish the cell lines. Tumor-derived cell lines were established by plating minced tumors in DMEM/10% calf serum.
Growth in Semisolid Media and Tumorigenesis Assay
Five x 103 polyclonal NIH 3T3 cells expressing FEF, FEF C, and empty vector (Tk Neo) were seeded in semisolid agar as described previously (12)
, photographs were taken at day 14. One million NIH 3T3 cells that had been selected for the FEF, FEF C, and FEF C deletion constructs were s.c. injected into the nape region of the neck of CB17 SCID mice. All of the mice were female and of about 5–6 weeks of age at the time of injection. Mice were observed until a grossly visible tumor of 2 cm in diameter was present. The animals were killed at this time, and the tumors were excised; a small piece of the tumor was fixed in 10% buffered formalin for paraffin embedding and H&E staining.
Northern and Western Blot Analyses
RNA and protein were harvested from infected cells immediately after antibiotic selection. Total RNA was prepared from lysates using STAT-60 (TEL-TEST, Inc.), 8 µg of RNA of each sample was electrophoresed in a formaldehyde-1% gel, and transferred to nitrocellulose membrane by standard methods. Northern blot hybridization and probe preparations were carried through as described previously (12)
. For immunoblots, cells were harvested and lysed in NP40 lysis buffer (50 mM Tris, 300 mM NaCl, 10% glycerol, 1 mM EDTA). Protein concentration was determined by Bradford assay against BSA standards. Sixty µg of protein in SDS loading buffer was loaded for each cell line, Coomassie Blue staining of the SDS gel, and Ponceau S staining of the nitrocellulose membrane after transfer was used to assess equal loading. Monoclonal anti-Flag (M2) was obtained from Sigma Chemical Co. The secondary antibody was horseradish peroxidase-conjugated goat-antimouse (Jackson Labs). The antibody-bound proteins were detected with ECL (Enhanced ChemiLuminescence) Western blotting detection reagents (Amersham).
Affymetrix Genechip Analysis
Genechips.
Mu11kSubA Genechips were purchased from Affymetrix, Inc. The Mu11kSubA set contains 6500 probe sets from mRNA transcripts and EST clusters. Oligonucleotide probes are synthesized on the arrays complementary to a portion of each gene. On the murine 11K set, there are 20 oligonucleotide probe pairs for each sequence.
Preparation of Biotin-labeled cRNA for Hybridization.
NIH 3T3 cells were virally infected with corresponding retroviral stocks FEF, FEF C, and Tk Neo in two independent experiments. Biotin-labeled cRNA was prepared according to Affymetrix protocols. Briefly, total RNA was isolated from NIH 3T3 cells expressing FEF, FEF C, and empty vector (Tk Neo) from experiments 1 and 2 according to the RNeasy total RNA isolation protocol (Qiagen). Twenty µg of the total RNA were used to synthesize first-strand cDNA (Life Technologies, Inc.). T7-(dT)24 oligomer [5'-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGG-(dT)24-3'] (Ambion MEGAScript T7 kit) was used for priming first-strand cDNA synthesis instead of the oligo (dT), second-strand cDNA was synthesized according to standard protocols. Biotin-11-cytidine-5'-triphosphate (Bio-11-CTP;ENZO) and Biotin-16-Uridine-5'-triphosphate (Bio-16-UTP;ENZO) were then incorporated in an in vitro transcription reaction using T7 polymerase (Ambion MEGAScript T7 kit Cat. No.1334). Murine Genechips were hybridized with Biotin-labeled cRNA probes from each replicate experiment at the University of California, San Diego, GeneChip Core facility. The biotin-labeled probes were bound to R-phycoerythrin steptavidin (Molecular Probes P/N S-866), and scanned on Hewlett Packard GeneArray scanner. The results were analyzed using GeneSpring software (Silicon Genetics, Inc.). We defined a threshold of 20 for the negative and low values (31
, 32)
. Only genes that scored P (present) call on both replicates as foreground signals were considered in the computation analyses. The averages of intensities for all experiments were then calculated and were used to determine fold differences. In our analyses, we also compared the duplicates to each other. For example, when the level of expression of all genes in FEF C experiment 1 were compared with FEF C experiment 2, we found that 5.6% of the genes are different from each other by at least 2-fold in their intensity values. This difference was less in the two Tk Neo trials (3.2%), and more in the FEF duplicates (8.2%). Therefore, we felt justified to compute a simple intensity average for each gene, and report only the genes that are up- or down-regulated by 3-fold or more (Supplementary Table 1 and 2 can be viewed online at http://cancerres.aacrjournals.org). Furthermore, we defined a second criteria for uniquely modulated genes: these genes must be up- or down-regulated by 3-fold compared with Tk Neo, and the difference between FEF and FEF C must be at least 2-fold. For example, a gene is considered uniquely up-regulated by FEF when FEF/TK is equal or more than three and FEF/FEF C is more than 2-fold (Supplementary tables).
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Whereas the FLI1 and ETV1 ETS DBDs are only 63% identical, both of these DBDs can bind the same consensus sequence in vitro (6
, 21)
. The ETV1 DBD can also replace the biological activity of the FLI1 DBD in transformation assays (11)
. On the contrary, the COOH terminus of ETV1 is significantly shorter than the COOH terminus of FLI1, and, more importantly, it does not share any substantial homology with the FLI1 COOH terminus.
|
C) was created, and its ability to transform cells was assessed in comparison to the full-length Flag-tagged EWS/FLI1 (FEF). The FEF C construct does not contain the last 89 amino acids after the FLI1 ETS DNA-binding domain. Polyclonal NIH 3T3 populations that express FEF C were generated by retroviral-mediated gene transfer (FEF C-3T3). As an initial indication of biological activity, FEF C-3T3 cells were plated in soft agar, and their growth compared with similar NIH 3T3 cell populations expressing full-length FEF or empty vector control (Tk Neo; Fig. 2A
). FEF C-3T3 cells exhibit markedly attenuated anchorage-independent growth, forming 90% fewer colonies in soft agar than cells expressing FEF.
|
C-3T3 cells was also assessed in vivo. Equal numbers of NIH 3T3 cells expressing FEF, FEF C, or empty vector were injected s.c. into CB-17-SCID mice. The rate of tumor formation for tumors of 2.0 cm in diameter was monitored over a 90-day period. Although cells expressing either mutant or wild-type FEF constructs all form tumors, the mean time to tumor formation in FEF C-3T3 cells (50 days) differs significantly from that of FEF-3T3 cells (22 days; Fig. 2B
).
The protein expression levels of FEF and FEF C in G418 selected cells were confirmed by immunoblotting analyses with anti-Flag antibody (Fig. 2C)
. The level of FEF C mutant protein detected is approximately equal to the full-length, indicating that loss of the FLI1 COOH terminus does not destabilize EWS/FLI1 protein. To confirm that expression of the FEF C was not lost during the long period to formation of tumor, Western blot analyses was repeated on the tumor cell lines derived from the excised tumors (Fig. 2C)
. FEFC protein is highly expressed in tumor-derived cell lines, as is full-length FEF. These data indicate that the differences in in vitro and in vivo behavior of these cell lines are attributable to altered biological characteristics of FEF C and not simply differences in protein expression levels or stability.
Interstitial Deletions within the FLI1 COOH Terminus Attenuate EWS/FLI1 Tumorigenicity.
Amino acid sequence alignment of the FLI1 COOH terminus shows high homology across diverse species ranging from 97% overall identity between human and mouse to 83% identity between human and Xenopus laevis (Fig. 3)
. In addition to the results of the FEF C transformation assays, this evolutionary conservation across species further suggests the biological importance of this domain.
|
. Six FEF deletion mutants were created that contained all possible combinations of the three subdivided COOH-terminal regions (Fig. 4)
. These constructs were then transduced into NIH 3T3 cells using replication-deficient retroviruses. Polyclonal NIH 3T3 populations were then injected s.c. into SCID mice and the time to tumor formation was assessed.
|
. Attenuation of tumorigenicity was incremental and generally paralleled the amount of the COOH terminus that was deleted rather than the absence of any particular region. For example, mutants containing a deletion of any single COOH terminus region (FEF 1, FEF 2, or FEF 3) form tumors at the same rate as full-length FEF. Mutants with deletions of two COOH-terminal regions form tumors at rates that are intermediate between full-length FEF and FEF C. As before, expression levels of FEF interstitial mutants were assessed in input cell lines and tumor explant cell lines by anti-Flag immunoblot (Fig. 5)
. Protein levels of all of the mutants were comparable with full-length FEF, which indicated that the observed differences in tumorigenic rates among NIH 3T3 populations are attributable to differing biological properties of the EWS/FLI1 COOH-terminal interstitial mutants.
|
Histological analyses were performed on the tumors generated from NIH 3T3 cells stably expressing complete or partial FEF COOH-terminal mutants to investigate what effect the COOH terminus has on tumor cell morphology. Control cell lines, containing either empty vector (Tk Neo) or full-length FEF, displayed the expected morphologies: empty vector tumor cells were densely packed and spindle shaped, creating the appearance of fibrosarcoma; FEF cells displayed a round-cell appearance reminiscent of ES/PNET (Fig. 6A)
. All of the FEF mutants with partial deletions of the COOH terminus also demonstrated small-round-cell histology (data not shown). By contrast, complete deletion of the COOH terminus (FEF C) promoted tumors that displayed biphenotypic morphology with both round and spindle components (Fig. 6A)
. This biphenotypic appearance was recapitulated in culture of FEF C tumor-derived cell lines (data not shown).
|
C to give rise to two morphologies or (b) there are genetically distinct subpopulations of NIH 3T3 cells that can respond differently to FEF C. To discern between these two possibilities, monoclonal NIH 3T3 cell lines expressing full-length FEF, FEF C, or control empty vector were isolated by limiting dilution. We tested two FEF and TK Neo monoclones and eight monoclones of C. Monoclones were then implanted into SCID mice and the resultant tumors examined histologically. Both the full-length FEF and empty vector monoclone tumors maintained their native pure morphologies. FEF monoclones lead to round-cell tumors, and empty vector to spindle-cell tumors, consistent with the parental polyclonal populations tested in the same assay (data not shown). The FEF C monoclonal tumors, however, were split. Of the five FEF C monoclones that formed tumors within the duration of the assay, four displayed round-cell morphologies (e.g., Fig. 6A
, FEF C monoclone 1), whereas one formed spindle-cell tumors (Fig. 6A
, FEF C monoclone 2). There appears to be no correlation between the rate of tumor formation and the histological appearance of these tumors. Anti-Flag immunoblot demonstrates equal levels of FEF C protein in spindle-compared with round-cell tumor cell lines, which indicates that the difference in tumor-cell morphology was not attributable to variation in FEF C expression (Fig. 6B)
.
Mutation of the FLI1 COOH Terminus Alters the EWS/FLI1 Target Gene Repertoire.
EWS/FLI1 is thought to mediate its oncogenic effect by acting as a transcription factor that modulates the expression of a set of target genes. The fact that the FEF C mutant has attenuated tumorigenic potency and altered histological appearance suggests that FEF C may not be able to regulate target genes as effectively as full-length FEF.
Northern analyses were performed to provide an initial estimate of the effect of deleting the FLI1 COOH terminus on the transcriptional activation of selected FEF target genes. Total RNA Northern blots generated from polyclonal NIH 3T3 populations stably expressing each of the FEF mutants were hybridized to cDNA probes from five previously described FEF up-regulated target genes (Refs. 10
, 23, 24, 25
; Fig. 7
). In general, the FEF-deletion mutants similarly modulated all five genes that were up-regulated by full-length FEF. Although the FEF mutants were not able to up-regulate target genes to as high a level as full-length FEF, most target genes were induced to levels above empty vector negative control (Tk Neo). FEF (2+3), containing only region 1 of the COOH terminus, proved the exception because it was unable to transcriptionally activate three of the five FEF target genes. These results suggest that although the FEF mutants have decreased tumorigenic potencies, they are for the most part still able to transcriptionally activate EWS/FLI1 target genes.
|
C, microarray studies were performed. Heterogeneous-labeled cRNA probes were generated from polyclonal NIH 3T3 cells expressing full-length FEF, FEF C, and empty vector, and were hybridized to prefabricated oligonucleotide arrays representing 6500 murine genes (Affymetrix-Mu11KsubA). Expression levels of each of the 6500 genes in the NIH 3T3 populations were quantified by the intensity of hybridization on the microarrays. Genes that were modulated >3-fold by either FEF or FEF C over empty vector control were identified and compared (Fig. 8)
. The Genechip analyses are the average of two independent experiments (see "Material and Methods" for details). Each cRNA probe was prepared twice from two freshly transduced NIH 3T3 cell populations and hybridized onto two Affymetrix chips from the same murine 11KsubA set. As an internal control for our Genechip analysis, we evaluated the level of activation of known EWS/FLI1 target genes. mFNG and murine E2-C, two previously identified target genes, were present in the Mu11ksubA gene set. In our analyses, we identified both of these genes to be activated >2-fold by FEF. Although FEF C also activated mFNG and E2-C, activation was lower than with full-length FEF, which matched our Northern blot analysis (Fig. 7)
.
|
C up-regulated smaller numbers of genes (21 genes) compared with FEF (34 genes), but the overlap between the two populations is significant. Fourteen (41%) of FEF-activated genes were also activated by FEF C. By contrast, the cohort of down-regulated genes presents a very different picture. Whereas FEF down-regulated 39 genes by 3-fold, FEF C down-regulated only 3 genes. More significantly, only 1 (2.5%) of the genes repressed by FEF was also down-regulated by FEF C (Fig. 8)
. It should be noted that because these experiments used stable expressing cell lines, the observed down-regulated target genes could have resulted either from direct repression by EWS/FLI1 or indirectly through activation of a repressor(s) by EWS/FLI1. |
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One possible explanation is that any stretch of amino acids after the DBD can rescue the transforming ability of FEF C, perhaps by altering the secondary structure of the entire protein. However, artificial replacement of the FLI1 COOH terminus by the COOH terminus of ETV1 resulted in a chimeric protein (EWS/FLI/ETV) that demonstrated the same attenuated NIH 3T3 tumor formation as that seen with FEF C (11)
. These observations strongly suggest that the specific FLI1 COOH-terminal sequence contains a unique biological function necessary for efficient cellular transformation.
In addition to tumor rate, the loss of pure-round-cell histology in FEF C-3T3 tumors provides another indication of biological activity associated with the FLI1 COOH terminus. We have shown previously that this histology is not attributable simply to rapid tumor growth but has been seen in virtually all of our EWS/ETS NIH 3T3 tumor assays. The biphenotypic spindle/round-cell morphology seen with FEF C tumors can be separated into the two histological components via clonal isolation and is not related to the level of FEF C protein expression. This suggests that the differing morphologies are caused by clonal variation within the polyclonal NIH 3T3 populations. We hypothesize that the round-cell morphology requires modulation of a specific set of target genes by the EWS/FLI1 oncoprotein. Full-length FEF can activate all of these genes, irrespective of the variation in the NIH 3T3 genetic background. However, FEF C may activate only a subset of these required genes and, therefore, requires a preexisitng genetic background to establish the round-cell morphology. An alternative explanation would be that because of the attenuated biology of FEF C, additional mutations need to occur subsequent to transduction of the fusion in order for the round-cell morphology to be apparent. Because this morphology is strikingly close to that seen in human ES/PNET, these differing FEF C monoclones may prove useful reagents to identify those genes dictating ES/PNET histology.
Although the COOH terminus of FLI1 clearly conveys biological activity to the EWS/FLI1 fusion, the biochemical mechanisms responsible for this effect are presently unclear. One potential explanation is that the COOH terminus plays a role in site-specific DNA binding of EWS/FLI1. However, it has been previously demonstrated that the in vitro DNA binding activity of FLI1 with a truncated COOH terminus does not differ from full-length FLI1 (16)
, and the sequence-specific DNA binding properties of the constructs similar to those with FEF C are not different from their wild-type counterparts in vitro (1
, 26)
. Nevertheless, equivalency in DNA binding between full-length FEF and FEF C may not hold true in vivo. ETS proteins are known to form heteromeric complexes with other DNA-binding transcription factors at promoter and enhancer elements of target genes (reviewed in Ref. 27
). For example, synergistic interactions have been described between ETS1 and PAX5 at the MB-1 promoter (28)
. Interaction with core binding factor (CBF 2) has been shown to enhance in vivo DNA binding of ETS1 (29)
. It is possible that the FLI1 COOH terminus mediates interaction with other transcription factors that would enhance binding to regulatory regions of EWS/FLI1 target genes.
Alternatively, the FLI1 COOH terminus may act to enhance transcriptional activation of EWS/FLI1 target genes. Previous studies using model reporter gene assays have detected some transcriptional activation properties associated with both the FLI1 and ERG C-termini (1
, 26
, 30)
. These results were confirmed in similar assays using ß-galactosidase/FLI1 fusion genes (16)
. The incremental loss of biological activity seen in our partial COOH-terminal deletion mutants is reminiscent of transcriptional activation domains. To a certain extent, the fact that FEF is able to up-regulate some target genes to a higher level than that seen with FEF C could be a reflection of loss of transcription activation capability in the COOH terminus.
Our microarray experiments indicate that deletion of the COOH terminus results in a global reduction of genes that are transcriptionally modulated by EWS/FLI1. Not only are the total number of affected target genes reduced but the quantitative degree to which they are modulated is also decreased. Strikingly, this effect is not evenly distributed over up- and down-regulated target genes. Whereas FEF C is able to up-regulate up to 41% of the activated genes, it represses only 2.5% of the FEF down-regulated genes. These analyses were performed using a factor of three to determine the differential expression of genes. Interestingly, when we reduced the stringency of the screening to 2-fold, we observed a similar trend (data not shown). On the one hand, this may be simply a fortuitous finding and reflect a differential effect on EWS/FLI1 target genes involved in mediating gene repression. On the other hand, it may be that transcriptional activation and repression of target genes by EWS/FLI1 may occur through biochemically different mechanisms. Future identification of EWS/ETS direct target genes will help distinguish these two possibilities.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by NIH Grant CA32737 from the National Cancer Institute; by Training Grant CA09056 (to A. A.); by USPHS National Research Service Award GM07185 (to S. M. W.); by NIH Grant CA74929 (to M. A. T.); and by the Lymphoma Research Foundation of America. 
2 Supplementary data for this article is available at Cancer Research Online [http://cancerres.aacrjournals.org].
3 To whom requests for reprints should be addressed, at UCLA Medical Center, Department of Pediatrics, A2–312 MDCC, Los Angeles, CA 90095.
4 The abbreviations used are: ES, Ewing’s sarcoma; PNET, primitive neuroectodermal tumor; DBD, DNA binding domain; SCID, severe combined immunodeficient; FEF, Flag-tagged EWS/Fli-1.
Received 1/31/01. Accepted 4/26/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Mol. Cell. Biol., 20: 81-90, 2000.This article has been cited by other articles:
|
|
 |
|
  G. H. S. Richter, S. Plehm, A. Fasan, S. Rossler, R. Unland, I. M. Bennani-Baiti, M. Hotfilder, D. Lowel, I. von Luettichau, I. Mossbrugger, et al. EZH2 is a mediator of EWS/FLI1 driven tumor growth and metastasis blocking endothelial and neuro-ectodermal differentiation PNAS, March 31, 2009; 106(13): 5324 - 5329. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Asano, M. Markiewicz, M. Kubo, G. Szalai, D. K. Watson, and M. Trojanowska Transcription Factor Fli1 Regulates Collagen Fibrillogenesis in Mouse Skin Mol. Cell. Biol., January 15, 2009; 29(2): 425 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Wakahara, T. Ohno, M. Kimura, T. Masuda, S. Nozawa, T. Dohjima, T. Yamamoto, A. Nagano, G. Kawai, A. Matsuhashi, et al. EWS-Fli1 Up-Regulates Expression of the Aurora A and Aurora B Kinases Mol. Cancer Res., December 1, 2008; 6(12): 1937 - 1945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Potikyan, R. O.V. Savene, J. M. Gaulden, K. A. France, Z. Zhou, E. S. Kleinerman, S. L. Lessnick, and C. T. Denny EWS/FLI1 Regulates Tumor Angiogenesis in Ewing's Sarcoma via Suppression of Thrombospondins Cancer Res., July 15, 2007; 67(14): 6675 - 6684. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Carrillo, E. Garcia-Aragoncillo, D. Azorin, N. Agra, A. Sastre, I. Gonzalez-Mediero, P. Garcia-Miguel, A. Pestana, S. Gallego, D. Segura, et al. Cholecystokinin Down-Regulation by RNA Interference Impairs Ewing Tumor Growth Clin. Cancer Res., April 15, 2007; 13(8): 2429 - 2440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kim, C. T. Denny, and R. Wisdom Cooperative DNA Binding with AP-1 Proteins Is Required for Transformation by EWS-Ets Fusion Proteins. Mol. Cell. Biol., April 1, 2006; 26(7): 2467 - 2478. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. van den Akker, S. Ano, H.-M. Shih, L.-C. Wang, M. Pironin, J. J. Palvimo, N. Kotaja, O. Kirsh, A. Dejean, and J. Ghysdael FLI-1 Functionally Interacts with PIASx{alpha}, a Member of the PIAS E3 SUMO Ligase Family J. Biol. Chem., November 11, 2005; 280(45): 38035 - 38046. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ano, R. Pereira, M. Pironin, I. Lesault, C. Milley, I. Lebigot, C. T. Quang, and J. Ghysdael Erythroblast Transformation by FLI-1 Depends upon Its Specific DNA Binding and Transcriptional Activation Properties J. Biol. Chem., January 23, 2004; 279(4): 2993 - 3002. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-C. Chang and V. B. Shidham Molecular Genetics of Pediatric Soft Tissue Tumors: Clinical Application J. Mol. Diagn., August 1, 2003; 5(3): 143 - 154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Deneen, H. Hamidi, and C. T. Denny Functional Analysis of the EWS/ETS Target Gene Uridine Phosphorylase Cancer Res., July 15, 2003; 63(14): 4268 - 4274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. C. Torchia, S. Jaishankar, and S. J. Baker Ewing Tumor Fusion Proteins Block the Differentiation of Pluripotent Marrow Stromal Cells Cancer Res., July 1, 2003; 63(13): 3464 - 3468. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. W. French, C. S. Malone, R. R. Shen, M. Renard, S. E. Henson, M. D. Miner, R. Wall, and M. A. Teitell Sp1 Transactivation of the TCL1 Oncogene J. Biol. Chem., January 3, 2003; 278(2): 948 - 955. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Spahn, R. Petermann, C. Siligan, J. A. Schmid, D. N. T. Aryee, and H. Kovar Interaction of the EWS NH2 Terminus with BARD1 Links the Ewing's Sarcoma Gene to a Common Tumor Suppressor Pathway Cancer Res., August 15, 2002; 62(16): 4583 - 4587. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Welford, S. P. Hebert, B. Deneen, A. Arvand, and C. T. Denny DNA Binding Domain-independent Pathways Are Involved in EWS/FLI1-mediated Oncogenesis J. Biol. Chem., November 2, 2001; 276(45): 41977 - 41984. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
, EWS amino acids 1–264; 
, FLI1 37 amino acids at the translocation breakpoint; 
, FLI1 DBD; 
, FLI1 COOH terminus. The regions in EWS/ETV1 are depicted as: 
, ETV DBD;
, ETV1 COOH terminus.
, Flag tag; 
3 and FEF/FEF