The genetic mechanisms that control proliferation of childhood musculoskeletal malignancies, notably Ewing’s tumor (ET) and rhabdomyosarcoma (RMS), remain largely unknown. Most human cancers appear to overexpress at least one of the G1 cyclins (cyclins D1, D2, D3, E1, and E2) to bypass normal regulation of cell cycle G1 progression. We compared the gene expression profiles of 7 ET and 13 RMS primary tumor samples and found overexpression of cyclin D1 in all 7 ET samples. In contrast, RMS samples expressed higher levels of cyclin D2, cyclin D3, and cyclin E1. This was confirmed by quantitative reverse transcription-polymerase chain reaction and Western blot. The relative roles of RAS-extracellular signal-regulated kinase 1/2 and phosphatidylinositol 3′-kinase (PI3K)-AKT pathways in the regulation of D-type cyclin expression in these tumors were then assessed. Inhibition of either pathway reduced expression of cyclins D1, D2, and D3 in RMS lines, whereas only PI3K inhibitors blocked cyclin D1, D2, and D3 expression in ET lines. Furthermore, PI3K-AKT appeared to regulate D-type cyclin transcription in RMS lines through FKHR and FKHRL1. Finally, the role of the ET-associated EWS-FLI1 fusion gene in regulating D cyclin expression was studied. Inhibition of EWS-FLI1 expression in the TC71 ET line decreased cyclin D1 levels but increased cyclin D3 levels. In contrast, induction of EWS-FLI1 expression in the RD RMS cell line increased cyclin D1 expression but decreased cyclin D3 expression. Our results demonstrate distinct regulation of D-type cyclins in ET and RMS and indicate that EWS-FLI1 can modulate the expression of D-type cyclins independent of cellular backgrounds.

Ewing’s tumor (ET) and rhabdomyosarcoma (RMS) are malignant tumors of the musculoskeletal system in children, adolescents, and young adults that often share a similar small round cell morphology but have different histogenetic origins, genetic features, and clinical behaviors (1). ET arises in either bone or soft tissues, exhibits primitive neural features (2), and likely originates from a pluripotent neural stem cell. It is characterized by the presence of specific chromosomal translocations that fuse the 5′ portion of the 22q12 EWS gene with the 3′ portions of different ETS family genes (3). Among the observed fusion genes, EWS-FLI1 is present in nearly 85% of ETs (4), whereas the EWS-ERG gene occurs in the majority of the remaining 15% of ETs (5). RMS arises in soft tissues and is thought to originate from primitive skeletal muscle precursor cells (6). The embryonal subtype [embryonal RMS (ERMS)] constitutes approximately two thirds of all RMS and has frequent loss of heterozygosity at the 11p15 locus. The alveolar subtype [alveolar RMS (ARMS)] is characterized by PAX3-FKHR and PAX7-FKHR gene fusions, which are present in approximately 60% and 20% of ARMS tumors, respectively (6, 7). Despite extensive genetic characterization of ET and RMS, relatively little is known about the mechanisms underlying their uncontrolled proliferation.

Many human cancers harbor genetic alterations that target key regulators of cell cycle G1 progression. Analysis of the retinoblastoma and p53 pathways has demonstrated frequent p53 mutations and loss of expression of various cyclin-dependent kinase (CDK) inhibitors in ET and RMS cell lines, but these alterations are not observed in the majority of primary ET and RMS tumors (8, 9, 10, 11, 12, 13). Moreover, no mutations in retinoblastoma have been reported in ET or RMS tumors (9, 14). Among positive regulators of G1 progression, high levels of cyclin D1 have been reported in ET cell lines (10, 15). Gene expression profiling of ET, RMS, neuroblastoma, and Burkitt lymphoma cell lines as well as primary tumors with cDNA microarray further demonstrated high levels of cyclin D1 expression in ET and neuroblastoma but not in RMS or Burkitt lymphoma (16). According to these data (ref. 16; Fig. 3 B) expression of cyclin D1 in nearly half of the RMS samples is similar to that of Burkitt lymphoma, which is known to express cyclin D3 and/or D2, but not cyclin D1 (17). Therefore, whereas overexpression of cyclin D1 may be a mechanism of cell cycle progression in ET, the roles of cyclin D1, as well as those of two other D cyclins, cyclin D2 and D3, in RMS in this process are less clear.

The D-type cyclins function as key sensors for mitogenic growth factors, and their expression levels appear to be rate-limiting for cell cycle G1 progression (18). In accordance with their important roles, many human cancers overexpress at least one of the D-type cyclins by either amplification of D cyclin genes (CCND13) or aberrant mitogenic signaling that leads to up-regulation of their expression (19, 20). On the other hand, analysis of D cyclin promoters has revealed marked differences in their regulatory elements, which suggests that transcription of these genes is independently regulated (21, 22). Indeed, cell- and tissue-specific patterns of D-type cyclin expression have been reported (23, 24, 25). ET and RMS, which have different cells of origin and different genetic abnormalities, might therefore use D-type cyclins differently.

In this study, we analyzed the expression of D cyclins in ET and RMS primary tumors as well as tumor-derived cell lines. We observed that whereas cyclin D1 was the predominant D-type cyclin expressed in ET, RMS instead expressed cyclin D2 and cyclin D3 at high levels. We further demonstrated that both the RAS-extracellular signal-regulated kinase (ERK) 1/2 and the phosphatidylinositol 3′-kinase (PI3K)-AKT pathways could regulate the expression of all three D cyclins in RMS cell lines, whereas the PI3K-AKT pathway appeared to be the major regulatory cascade for D cyclin expression in ET cell lines. Moreover, inhibition of EWS-FLI1 expression in an ET cell line and induction of EWS-FLI1 expression in a RMS cell line reversed their cyclin D1 and D3 expression patterns, suggesting that this chimeric transcription factor mediates, at least in part, the observed overexpression of cyclin D1 and low expression of cyclin D3 in ET.

Tumors and Cell Lines.

Frozen primary tumor tissues were from Children’s Hospital Los Angeles. Fusion status was tested by reverse transcription-polymerase chain reaction (RT-PCR). Among the tumors for gene profiling, the seven ETs expressed EWS-FLI1 or EWS-ERG, the eight ARMS tumors expressed PAX3-FKHR or PAX7-FKHR, and the five ERMS tumors showed typical histology. Additional primary tumors (three ET and three RMS tumors) were used for data verification. ET96 and ET00 express EWS-FLI1, ET98 expresses EWS-ERG, RA97 expresses PAX3-FKHR, and RA96 expresses PAX7-FKHR. ET cell lines, including RDES, 6647, TC135, TC71, TC32, TTC466, and A4573, all express EWS-FLI1, with the exception of TTC466 (EWS-ERG). TC135, TC71, and TC32 have type I fusion; RDES and 6647 have type II fusion; and A4573 has type III fusion. RD, TC442, and Rh18 are ERMS cell lines, whereas Rh28, Rh30, and TC487 are ARMS cell lines expressing PAX3-FKHR. All cell lines were from the cell bank at Children’s Hospital Los Angeles [with the exception of RDES (American Type Culture Collection, Manassas, VA)] and were cultured in RPMI 1640 containing 10% fetal bovine serum (FBS; Invitrogen, Carlsbad, CA).

Expression Analysis with the U95Av2 Microarray.

Tissues that had >90% tumor cells were chosen, and total RNA was extracted (RNA STAT-60; Tel-Test Inc., Friendswood, TX), cleaned (RNeasy mini kit; Qiagen, Santa Clarita, CA), and quantitated. Synthesis of cDNA, biotin-labeled cRNA, target hybridization, washing, staining, and scanning followed the Affymetrix (Santa Clara, CA) manual. To test the chip-to-chip variation, the same biotin-labeled cRNA from one ET sample was hybridized to three chips.

Gene expression data were first analyzed with MAS 5.0 (Affymetrix), and the average difference of each probe set was then used to generate the text file for analysis with GeneSpring5.0 (Silicon Genetics, Redwood City, CA). Data were normalized before importing into GeneSpring. For each chip, the signal strength was normalized to the median of all of the measurements in that chip. For each gene, the signal strength was normalized to the median of every measurement for that gene throughout all of the chips imported. All of the negative values were forced to zero. The “trust” of each gene is calculated by multiplying the median of the chip by the median of the gene (500 = most trustworthy, 150 = moderately trustworthy, and 50 = least trustworthy).

Real-Time Quantitative Reverse Transcription-Polymerase Chain Reaction.

Total RNA was extracted with RNA STAT-60 when cells reach 70–80% confluence. One microgram of total RNA was digested with DNase I (Invitrogen) and reverse transcribed with Superscript II (Invitrogen) following the supplier’s instructions. Polymerase chain reaction (PCR) primers were designed with Primer Express 2.0 (Applied Biosystems, Foster City, CA). The primer sequences were as follows: D1F, 5′-CGCACGATTTCATTGAACACTT-3′; D1R, 5′-CGGATTGGAAATACTTCACAT-3′; D2F, 5′-TTGTCTCAAAGCTTGCCAGGA-3′; D2R, 5′-CGACTTGGATCCGTCACGTT-3′; D3F, 5′-CCTCTGTGCTACAGATTATACCTTTGC-3′; D3R, 5′-TTGCACTGCAGCCCCAAT-3′; E1F, 5′-TGAAGAAATGGCCAAAATCGA-3′; E1R, 5′-AACCCGGTCATCATCTTCTTTGT-3′; E2F, 5′-AGCCCAGCCAGACGGAAT-3′; E2R, 5′-CAGATAATACAGGTGGCCAACAAT-3′; and β-actin, 5′-GCACCCCGTGCTGCTGAC-3′ (forward) and 5′-CAGTGGTACGGCCAGAGG-3′ (reverse). One twentieth of the cDNA was amplified in a 25-μl reaction volume containing 0.5 μmol/L of each primer and QuantiTect SYBR Green master mix (Qiagen), using SmartCycler (Cepheid, Sunnyvale, CA). Standard curves were constructed by four serial 10-fold dilutions of cDNA. PCR conditions were as follows: 95°C for 900 seconds; 40 cycles of 95°C for 15 seconds, 55°C for 30 seconds (but 60°C for β-actin), 72°C for 30 seconds; and a final denaturing stage to generate a melting curve that correlates with the size and GC content of the product. Single-band PCR product was verified on 1.5% agarose gels. The reproducibility of the measurements was evaluated by triplicates. The mean ± SD of the cyclin to β-actin ratios was calculated for sample-to-sample comparison.

Western Blot Analysis.

Protein lysate preparation and immunoblotting were performed as described previously (26). Primary monoclonal antibodies used include cyclin D1, D2, D3, E1, and FLI1 from BD Bioscience and β-actin from Sigma (St. Louis, MO). Polyclonal antibodies include cyclin E2, CDK2, CDK4, and FKHRL1 from Santa Cruz Biotechnology (Santa Cruz, CA) and phospho-AKT (Ser473), AKT, phospho-ERK1/2 (Thr202/Tyr204), ERK1/2, phospho-FKHR (Ser256), phospho-FKHR (Thr24)/FKHRL1 (Thr32), FKHR, and phospho-AFX (Ser193) from Cell Signaling Technology (Beverley, MA). Secondary horseradish peroxidase-conjugated goat antimouse IgG, goat antirabbit IgG, and donkey antigoat IgG were from Santa Cruz Biotechnology. The blotted proteins were visualized using enhanced chemiluminescence kit (Amersham Pharmacia Biotech, Piscataway, NJ) and quantitated with FluorChem 8900 (Alpha Innotech, San Leandro, CA).

Immunoprecipitation.

Immunoprecipitation was performed as described previously (26). Briefly, protein lysates were first incubated with normal rabbit IgG and protein A+G. After spinning, the supernatant was quantitated. One milligram of protein lysates was incubated with rabbit antihuman CDK4 and then precipitated with protein A+G. After six washes, the complexes were denatured in SDS-reducing buffer for subsequent Western blot analysis.

Kinase Inhibitor Studies.

ET and RMS cells were starved in RPMI 1640 containing 0.5% FBS for 48 hours before the medium was changed and U0126 [mitogen-activated protein/ERK kinase (MEK) 1/2 inhibitor; 20 μmol/L], LY294002 (PI3K inhibitor; 50 μmol/L; Cell Signaling Technology), or an equivalent DMSO control was added. At various times after treatment, total RNA and protein were harvested for real-time RT-PCR and Western blot analysis.

Immunofluorescence Localization of FKHR and FKHRL1.

RD cells were seeded in a 4-well chamber slide and cultured in RPMI 1640 containing 0.5% FBS 48 hours and then treated with LY294002 or DMSO or left untreated. Two hours after treatment, immunofluorescence assay was performed as described previously (27). Cy3-conjugated goat antirabbit IgG (Jackson ImmunoResearch, West Grove, PA) was used for FKHR or FKHRL1 detection. Cell nuclei were counterstained with DAPI. Images were captured on a SKY-300V spectral imager (Applied Spectral Imaging, Carlsbad, CA).

DNA Constructs and Transfection and Selection of Clones.

EWS-FLI1 type I fusion cDNA was cloned into pcDNA4/TO (Invitrogen), a tetracycline-inducible expression vector, and named pcDNA/TO-EF. Antisense EWS-FLI1 type I fusion cDNA was cloned into pcDNA3 (Invitrogen) and named pcDNA-ASEF. Effectene (Qiagen) was used in all transfections according to the manufacturer’s manual. Single-clone-transfected TC71 cells were selected and maintained in medium containing 150 μg/ml G418 (Invitrogen). For the tetracycline-regulated system (T-Rex; Invitrogen), RD cells were first stably transfected with pcDNA6/TR, and monoclones were selected and maintained in tetracycline-free media containing 5 μg/ml blasticidin (Invitrogen). The inducibility of each clone was tested by transient transfection with pcDNA4/TO/LacZ, and then clones were stained for β-galactosidase using a β-galactosidase staining kit (Invitrogen). Two of 40 pcDNA6/TR clones were chosen for second stable transfection with pcDNA/TO-EF. Tetracycline-free media containing 5 μg/ml blasticidin and 400 μg/ml Zeocin (Invitrogen) were used to select and maintain monoclones. Tetracycline (1 μg/ml) was added to growth media for induction of EWS-FLI1 (tested by RT-PCR and Western blot).

Cell Proliferation Assays.

For growth curve analysis, the same number of cells was seeded in 12-well plates and counted on 3 consecutive days before reaching confluence. In the [3H]thymidine incorporation assay, the same number of cells was seeded in 48-well plates with 0.5 ml of medium. Each sample had 4 wells (1 well for the background radioactivity and 3 wells for cell counts). Twenty-four hours after seeding, cells were counted, and 1 μCi of [3H]thymidine (ICN, Valeant Pharmaceuticals, Costa Mesa, CA) was added to the medium (with the exception of the background control). After a 4-hour incubation at 37°C, cells were washed three times with cold phosphate-buffered saline and dissolved in Solvable (200 μl/well; NEN, Perkin Elmer International, Wellesley, MA). Cell lysates (150 μl) were then transferred to the scintillation vial containing 5 ml of scintillation solution (Fisher Chemicals, St. Louis, MO). [3H]Thymidine activity was measured with a scintillation counter (Beckman). The background reading was subtracted before the mean and SDs of the triplicates were calculated. Cell cycle distribution was determined by DNA content as described before. Fluorescence-activated cell sorting (FACS) was performed on FACStar (Becton Dickinson, San Jose, CA) and analyzed with Mac cycle software.

Primary Ewing’s Tumor and Rhabdomyosarcoma Tissues Have Different Expression Patterns of D-Type Cyclins and Cyclin E1.

We studied the gene expression profiles of 7 ET, 8 ARMS, and 5 ERMS tumors using Affymetrix U95Av2 arrays, on which there are 65 probe sets for G1 cyclins, CDKs, CDK-activating kinase (CAK, a complex formed by CDK7, cyclin H, and MAT1), CDK inhibitors, the retinoblastoma protein family (RB1, RBL1/p107, and RBL2/p130), the E2F transcription factors (E2F1 to E2F6, TFDP1, and TFDP2), p53, MDM2, and MDM4. For each probe set, we compared the average of the relative expression value in RMS with ET. Only the probe sets for the cyclin D1 gene (CCND1) had >2-fold higher average expression value in ET compared with RMS, whereas the probe sets for the cyclin D2 gene (CCND2) the cyclin D3 gene (CCND3), and the cyclin E1 gene (CCNE1) had >2-fold higher average expression value in RMS (Fig. 1,A). Expression levels of the CDK inhibitor, CDKN1C/p57, were markedly high in six RMS samples but were low in the other RMS samples and most ET samples (data not shown). Levels of proliferating cell nuclear antigen (PCNA) and MKI-67 were only slightly higher (<1-fold) in RMS samples than in ET samples (data not shown). Thus, the observed differences in D-type cyclin expression were not due to different proliferation rates between ET and RMS samples. Based on the expression of 10 probe sets for D-type and E-type cyclins, hierarchical clustering was able in every case to distinguish 7 ET samples from 13 RMS samples (Fig. 1,A). This differential expression pattern of G1 cyclins was more striking when the eight ARMS samples were compared with the seven ET samples (Fig. 1 A). In contrast, cyclin E2 gene (CCNE2) expression was variable in both tumor types.

Cyclin D1 Is the Major D Cyclin Expressed in Ewing’s Tumor, Whereas Cyclins D2 and D3 Are the Major D Cyclins Expressed in Rhabdomyosarcoma.

We assayed G1 cyclin expression in three primary ET tumors, three RMS tumors, seven ET cell lines, and six RMS cell lines (see Materials and Methods for fusion gene status) by quantitative real-time RT-PCR and Western blot. Consistent with the microarray data, ET tumors had higher cyclin D1 mRNA and protein levels, whereas RMS tumors had higher levels of cyclin D2 and D3 (Fig. 1,B and C). More striking differences in the expression of D cyclins were observed when the ET and RMS cell lines were compared (Fig. 2,A and B). Cyclin D1 expression in ET cell lines was previously reported to be variable (9) or consistently high (10). Although we also observed some variation in cyclin D1, we found this variation correlated with EWS-FLI1 and EWS-ERG fusion gene expression in the ET cell lines (Fig. 2,B). TC135, which had the lowest EWS-FLI1 expression level of the ET cell lines tested, showed the lowest level of cyclin D1 but a high level of cyclin D3 expression. A4573, which expresses an EWS-FLI1 type III fusion gene, showed similar cyclin D1 expression but has higher cyclin D2 and D3 expression than the other ET lines (Fig. 2 A and B).

Compared with ET, RMS tumors and cell lines generally had higher cyclin E1 mRNA (Figs. 1,B and 2,A) and proteins (Figs. 1,C and 2,B). Cyclin E2 appears to be the major E-type cyclin expressed in RMS at the mRNA level (Figs. 1,B and 2,A), but not at the protein level (Figs. 1,C and 2 B). Cyclin E2 protein appeared to be high in samples with low cyclin E1 levels. No substantial differences in the expression of CDKs (CDK2, CDK4, and CDK6) or CDK inhibitors (CDKN2A/p16, CDKN2B/p15, CDKN2C/p18, CDKN2D/p19, CDKN1A/p21, CDKN1B/p27, and CDKN1C/p57) were observed between the RMS lines and the ET lines (data not shown).

We next immunoprecipitated cyclin D-CDK4 complexes with anti-CDK4 antibodies and then tested for the abundance of each D-type cyclin. These studies confirmed that the major D cyclins expressed in the RMS lines (Rh18 and TC487) and ET lines (TC32 and TC466) were also the major cyclins associated with their catalytic partner, CDK4 (Fig. 2 C).

MEK Inhibition Decreases D-Type Cyclin Expression in Rhabdomyosarcoma but not in Ewing’s Tumor.

We next assessed whether differences in D-type cyclin expression between ET and RMS were accompanied by differences in upstream mitogenic signaling pathways. In many cell types, ERK1 and ERK2 relay mitogenic signals from membrane growth factor receptors to the nucleus (28, 29). Activation of ERK1/2 is sufficient to induce cyclin D1 transcription even in the absence of growth factors (30). Whether cyclin D2 and D3 expression is regulated by ERK1/2 is unknown. Constitutive activation of ERK1/2 was observed in all RMS tumors and cell lines tested (data not shown). In contrast, ERK1/2 activities varied in ET tumors and cell lines; no correlation between ERK1/2 activities and cyclin D1 expression was observed (data not shown), consistent with a previous report (15). We then blocked ERK1/2 activation in ET (TC71 and 6647) and RMS (Rh18 and RD) cell lines with the MEK1/2 inhibitor, U0126 (31, 32). ET and RMS cells were serum-starved in 0.5% FBS medium before and during treatment with either U0126 or DMSO vehicle control. Substantial inhibition of ERK1/2 activation (especially p44 ERK1) was observed by 4 hours and continued until at least 24 hours after U0126 treatment in all of the ET and RMS cells tested (Fig. 3,A). MEK inhibition, however, had no effect on D-type cyclin RNA or protein expression levels in ET cells (Fig. 3,A and B). Cyclin D2 protein was virtually undetectable in ET cell lines. In contrast, MEK inhibition in RMS cells led to decreased expression of all three D-type cyclins (Fig. 3,A and B) by 8 hours, with maximum levels at 24 hours after treatment (Fig. 3 A).

PI3K Inhibition Decreases D Cyclin Expression in both Rhabdomyosarcoma and Ewing’s Tumor Cells.

D-type cyclins are also posttranscriptionally regulated, although most work has focused on cyclin D1 regulation. PI3K-dependent signaling pathways up-regulate cyclin D1 translation and down-regulate cyclin D1 degradation in response to mitogenic stimuli (33, 34). We evaluated the role of the PI3K-AKT pathway in D-type cyclin regulation in ET and RMS. ET and RMS cells were starved in 0.5% FBS medium before and during treatment with either the PI3K inhibitor (LY294002) or DMSO control. LY294002 treatment markedly reduced AKT phosphorylation and D cyclin proteins in both ET and RMS cells (Fig. 3,C). No decrease in D cyclin RNAs was observed in LY294002-treated ET cells (Fig. 3,D), implying that PI3K blockade inhibited D cyclin expression at the posttranscriptional level. Cyclin D2 protein was virtually undetectable in ET cells. In contrast, significant decreases in all three D cyclin RNAs were observed in LY294002-treated RMS cells (Fig. 3 D), suggesting that both transcriptional and posttranscriptional regulation of D cyclin expression was likely affected by PI3K blockade.

PI3K Inhibition in Rhabdomyosarcoma Cells Is Associated with Increased Nuclear Localization of FOXO1a/FKHR and FOXO3a/FKHRL1.

AKT directly phosphorylates FOXO forkhead transcription factors, FOXO1a/FKHR, FOXO3a/FKHRL1, and FOXO4/AFX (35, 36), leading to nuclear exclusion and inhibition of FOXO factor-mediated gene expression (27, 37). Recent publications have demonstrated that the three above-mentioned FOXO factors repress the transcription of cyclin D1 and D2 (38, 39). Our gene expression data from primary ET and RMS tumors indicated that FKHR and FKHRL1 were expressed at substantially higher levels in RMS compared with ET (data not shown). We thus tested whether phosphorylation and nuclear localization of FOXO factors differed in RMS cells after PI3K inhibition.

Compared with DMSO-treated RD cells, the decrease in phosphorylation of FKHRL1 Thr32 and FKHR Thr24 was evident at 2 hours after LY294002 treatment and continued for 24 hours when corrected for loading using β-actin as the standard (Fig. 4,A). No substantial changes in phosphorylation of FKHR Ser256 and AFX Ser293 were observed, likely due to the serum-deprived growth conditions. The subcellular localization of FKHR and FKHRL1 with or without LY294002 treatment was then detected by immunofluorescence. Two patterns of FOXO localization (nuclear plus cytoplasmic and nuclear alone) were observed (Fig. 4,B). Compared with untreated (Fig. 4,B, Blank) and DMSO-treated RD cells, increased nuclear localization of FKHR and FKHRL1 was observed in RD cells 2 hours after LY294002 treatment. This included increased numbers of cells with exclusive nuclear localization and decreased cytoplasmic staining in cells with both nuclear and cytoplasmic staining (Fig. 4 B).

Antisense Blockade of EWS-FLI1 Expression in Ewing’s Tumor Reduces Cyclin D1 Expression but Increases Cyclin D3 Expression.

Because ET is characterized by the presence of specific gene fusions, we wished to determine whether EWS-ETS fusion proteins affect differential D-type usage in ET. An antisense EWS-FLI1 type 1 fusion construct was stably transfected into the TC71 ET line. Compared with untransfected and blank vector-transfected TC71 controls, expressing antisense EWS-FLI1 in ASEF9 cells decreased EWS-FLI1 protein (Fig. 5,A). This was associated with decreased cyclin D1 and increased cyclin D3 expression at both the RNA (data not shown) and protein levels (Fig. 5,A). Despite this inhibition, cyclin D1 remained the dominant D cyclin. Expression of cyclin D2 remained low. No alterations in cyclin E1 expression were observed (Fig. 5,A). No other alterations in cell cycle-related proteins were detected. To test the effects of EWS-FLI1 inhibition on cell proliferation, we assessed cell proliferation by cell counting and observed a significant decrease in the number of ASEF9 cells compared with controls (Fig. 5,B). To determine the proliferation ratios, S-phase DNA synthesis activity was measured by [3H]thymidine incorporation. Compared with TC71 controls, [3H]thymidine incorporation was 2-fold less in ASEF9 cells by 28 hours (Fig. 5,C). A 50% increase in ASEF9 cells in G0-G1 phase was detected by flow cytometry (Fig. 5 D). Similar results were observed in additional clones stably expressing antisense EWS-FLI1 (data not shown). These data confirm that EWS-FLI1 is essential for ET cell proliferation but also confirm that expression of this chimeric oncoprotein specifically correlates with differential expression of cyclin D1 and D3.

EWS-FLI1 Increases Cyclin D1 but Decreases Cyclin D3 Expression in Rhabdomyosarcoma.

To further analyze the roles of EWS-FLI1 on D-type cyclin expression and the effect of ERK1/2 activation in this process we established a tetracycline-controlled EWS-FLI1 expression system in the RMS cell line RD (RD/trexEF). ERK1/2 phosphorylation and D-type cyclin expression were assessed at different time points after EWS-FLI1 induction. Fig. 6,A and B shows that expression of EWS-FLI1 protein was first detectable 6 hours after induction, and levels peaked by 12 hours. Increases in cyclin D1 protein were first observed at 6 hours after induction and continued to increase thereafter. Although no changes in cyclin D2 were observed, expression of cyclin D3 was markedly reduced after induction of EWS-FLI1. In fact, cyclin D1 replaced cyclin D3 as the major D cyclin expressed in RD after induction of EWS-FLI1. This reversal of D-type cyclin expression after EWS-FLI1 induction was also observed at the RNA level by real-time RT-PCR (data not shown). ERK1/2 activities were transiently up-regulated at early time points, likely due to changing the medium at 0 hour, and then decreased as EWS-FLI1 protein began to accumulate (Fig. 6 A and B). Thus, changes in cyclin D1 and D3 expression after EWS-FLI1 induction appeared to be independent of ERK1/2 activation. These data corroborate those of the antisense studies indicating that EWS-FLI1 specifically regulates cyclin D1 and D3 usage in tumor cells.

We also assessed whether EWS-FLI1 induction and its effects on cyclin D1 and D3 expression altered proliferation of RD/trexEF cells by comparing cell cycle distribution with flow cytometry. RD cells arrested in G1 phase 24 hours after EWS-FLI1 induction (Fig. 6 C). Similar results were obtained with additional RD/trexEF clones (data not shown). EWS-FLI1 protein is required for ET cell proliferation; alone, however, it is not sufficient to promote cell proliferation.

To compare the molecular mechanisms involved in regulating the proliferation of ET and RMS tumor cells, we studied the expression profile of genes involved in cell cycle G1 progression and identified different expression patterns of D-type cyclins in these two childhood tumor types. Both primary tumor and tumor-derived cell line studies demonstrated that cyclin D1 was the major D-type cyclin in ET, whereas cyclins D2 and D3 predominated in RMS. These observations were confirmed at both the RNA and protein levels. In addition, we found that whereas signaling through both RAS-ERK1/2 and PI3K-AKT cascades regulated D-type cyclin expression in RMS, D-type cyclin expression in ET appeared to be regulated at the posttranscriptional level by PI3K-AKT. Our data further indicate that expression of the EWS-FLI1 fusion protein can influence the levels of cyclin D1 and D3 in both ET and RMS cellular backgrounds and that this is independent of ERK1/2 activation.

Most studies of D-type cyclin regulation have focused on cyclin D1. In our study, blocking sustained ERK1/2 activation by MEK1/2 inhibition abolished the expression of all three D-type cyclins in RMS cell lines but not in ET lines. Thus, in RMS, RAS-ERK1/2 signaling controls cyclin D1 and cyclin D2 and D3 expression. RAS-ERK1/2 signaling induces cyclin D1 promoter transcription through induction of specific transcription factors (40, 41). Although cyclin D promoters have marked differences in their regulatory elements, consensus binding sites for a number of transcription factors implicated in transcriptional regulation of cyclin D1 expression, including activator protein-1 and Sp1, are also present on cyclin D2 and D3 promoters (21, 22, 41). MEK1/2 inhibition in our studies may have blocked the function of these or other transcription factors in RMS cells, leading to the observed decreases in expression of all three D-type cyclins. To elucidate how the different cellular backgrounds of RMS and ET might contribute to the differential responses to MEK1/2 inhibition, we reviewed gene expression profiles of nuclear factors in ET and RMS primary tumor samples. This demonstrated that Id2 (inhibitor of DNA binding 2), a helix-loop-helix protein, was much more highly expressed in ET compared with RMS (data not shown). Id2 is a downstream target of the EWS-ETS fusion proteins and is overexpressed in ET tumor and cell lines (42, 43). Deregulated expression of Id proteins (Id1–3) occurs in several human tumor types and has been implicated in the regulation of tumor growth, angiogenesis, invasiveness, and metastasis (44, 45). Id2 blocks RAS-ERK1/2 signaling to serum response element by down-regulating transcriptional activity mediated by ternary complex factors (46). High levels of Id2 in ET may block regulation of cyclin D expression by RAS-ERK1/2 signaling.

The PI3K-AKT pathway also regulates cyclin D1 expression. PI3K activation of p70 S6 kinase increases cyclin D1 translation (33). Moreover, glycogen synthase kinase 3β phosphorylates cyclin D1 and promotes its proteasomal degradation (34, 47). AKT inhibits this by phosphorylating glycogen synthase kinase 3β at Ser9(48, 49). Sustained AKT activation was observed in all of the ET and RMS samples tested, and PI3K inhibition reduced expression of all three D-type cyclins in both ET and RMS cells. D-type cyclin levels in RMS therefore appear to represent a functional balance between signaling through the RAS-ERK1/2 and PI3K-AKT pathways. Although the mechanism(s) by which PI3K-AKT signaling regulates D-type cyclin expression in these tumors remains unknown, there appear to be distinct differences between ET and RMS. In ET, PI3K inhibition decreased cyclin D expression at the posttranscriptional level, whereas in RMS, such decreases appear to be at both the transcriptional and posttranscriptional levels. When we tested the roles of FKHR and FKHRL1 in regulating D-type cyclin transcription in RMS lines, we found decreased phosphorylation and increased FKHR and FKHRL1 nuclear localization shortly after blocking PI3K-AKT signaling in the RD RMS line. Although only one of the two AKT phosphorylation sites (e.g., Thr24) on FKHR was blocked by LY294002 treatment in serum-starved RD cells, such a blockade is sufficient to increase the nuclear localization of FKHR. Both FKHR/FOXO1a and FKHRL1/FOXO3a repress cyclin D1 and D2 transcription (38, 39). However, the common FOXO DNA binding element, the insulin response sequences, is not required for the repression of CCND1 and CCND2 promoter activities (39). Whether FKHR and FKHRL1 repress D-type cyclin transcription indirectly by interacting with other transcription factors or directly by binding to new DNA response elements is not clear. In addition to cyclin D1 and D2, we found that blocking PI3K-AKT signaling also appeared to decrease cyclin D3 RNA.

EWS-FLI1 expression correlates with expression of cyclin D1 (10, 50, 51), but whether EWS-FLI1 directly or indirectly regulates cyclin D1 expression is unknown. EWS-FLI1 binds DNA in a sequence-specific manner (52). Potential EWS-FLI1 binding sites can be identified in D-type cyclin promoters (21, 22). EWS-FLI1 binds directly to the cyclin D1 promoter region (42), indicating that cyclin D1 may be a direct target of this fusion protein. When we induced EWS-FLI1 expression in RD cells, both cyclin D1 and D3 were among the early response genes following EWS-FLI1 induction. We also showed3 that ectopically expressing EWS-FLI1 in RD cells could transactivate the cyclin D1 promoter and that this transactivation requires EWS-FLI1 DNA binding. Whether EWS-FLI1 directly regulates cyclin D3 expression remains to be determined.

Forced expression of EWS-FLI1 up-regulates ERK1/2 activity in NIH3T3 cells (53), indicating that EWS-FLI1 may indirectly regulate D-type cyclin expression through ERK1/2. However, our data and the data of others (15) indicate that RAS-ERK1/2 activation may not be essential for D-type cyclin expression in ET. Although the mechanism by which EWS-FLI1 activates ERK1/2 is not clear, such regulation may depend on cellular background (i.e., in ET versus RMS). No increases in ERK1/2 activation were observed when we induced EWS-FLI1 expression in the RMS line. Instead, with accumulation of EWS-FLI1 protein, ERK1/2 activation in RD/trex cells actually declined. If EWS-FLI1 modulates D-type cyclin expression through ERK1/2 signaling, it would be difficult to explain why induction of EWS-FLI1 in RD cells increased cyclin D1, decreased cyclin D3, and had no effects on cyclin D2 expression. Expression of cyclin D1 in ET therefore appears to be transcriptionally regulated by EWS-FLI1 and posttranscriptionally regulated by the PI3K-AKT signaling.

Ectopic expression of EWS-FLI1 in primary human fibroblasts and primary mouse embryo fibroblasts induces p53- or p16INK4A-dependent growth arrest (54, 55). The early-passage RD line used in our study has missense p53 mutations but does express p16INK4A (data not shown).4 To elucidate the role of EWS-FLI1 in cell proliferation, we chose TC71, an ET cell line that has a nonsense mutation in the p53 gene and lacks CDKN2A/p16 expression. We found that blocking EWS-FLI1 expression in TC71 cells could still lead to cell cycle G1 arrest. Therefore, whereas EWS-FLI1 itself may not be sufficient to accelerate cell proliferation, it is required for the proliferation of ET cells in which proapoptotic or growth arrest pathways have been inactivated. Although the underlying mechanisms by which EWS-FLI1 mediates its proliferative effects are largely unknown, it is likely that they involve the up-regulation of positive cell cycle regulators such as cyclin D1.

In conclusion, our results indicate that different D-type cyclins control proliferation of ET and RMS cells. This selective usage of D-type cyclins is due to, at least in part, the different downstream effects of common mitogenic signaling pathways as well as the expression of the EWS-FLI1 fusion gene in ET.

Fig. 1.

Expression of D-type cyclins and E-type cyclins in ET and RMS primary tumor samples. A, hierarchical clustering of primary tumors samples and G1 cyclins. Each row represents the relative expression level of each probe set across different samples. Its gene name, GenBank locus, and the ratio of average expression levels between ET and RMS are indicated on the right. Each column represents an individual sample. The blue line underneath the first three columns indicates the triplicates of the same ET sample. The orange lines underneath other columns indicate the eight ARMS samples. The relative expression level is represented by the red (high expression) and green (low expression) pseudo-color, and the trust of the data is indicated by the saturation of the color (unsaturated color, low trust; saturated color, high trust). In the scale below, the horizontal axis represents the scale for normalized expression value, and the vertical axis is the trust scale. B, relative expression of D- and E-type cyclin mRNAs in ET (ET98, ET96, and ET00), ERMS (Re94), and ARMS (Ra96 and Ra97) primary tumor samples. Real-time quantitative RT-PCR was performed as described in Materials and Methods. For each sample, the average ± SD of the cyclin to β-actin ratio from at least three independent PCR reactions is shown. C, Western blots detecting D- and E-type cyclins in ET and RMS tumors. β-Actin is shown as the loading control.

Fig. 1.

Expression of D-type cyclins and E-type cyclins in ET and RMS primary tumor samples. A, hierarchical clustering of primary tumors samples and G1 cyclins. Each row represents the relative expression level of each probe set across different samples. Its gene name, GenBank locus, and the ratio of average expression levels between ET and RMS are indicated on the right. Each column represents an individual sample. The blue line underneath the first three columns indicates the triplicates of the same ET sample. The orange lines underneath other columns indicate the eight ARMS samples. The relative expression level is represented by the red (high expression) and green (low expression) pseudo-color, and the trust of the data is indicated by the saturation of the color (unsaturated color, low trust; saturated color, high trust). In the scale below, the horizontal axis represents the scale for normalized expression value, and the vertical axis is the trust scale. B, relative expression of D- and E-type cyclin mRNAs in ET (ET98, ET96, and ET00), ERMS (Re94), and ARMS (Ra96 and Ra97) primary tumor samples. Real-time quantitative RT-PCR was performed as described in Materials and Methods. For each sample, the average ± SD of the cyclin to β-actin ratio from at least three independent PCR reactions is shown. C, Western blots detecting D- and E-type cyclins in ET and RMS tumors. β-Actin is shown as the loading control.

Close modal
Fig. 2.

Expression of D- and E-type cyclins in seven ET (6647, RDES, TC135, TC71, TC32, TTC466, and A4573) and six RMS (RD, TC442, Rh18, Rh28, Rh30, and TC487) cell lines. A, relative expression of D- and E-type cyclin mRNAs in ET and RMS cell lines. For each sample, the average ± SD of the cyclin to β-actin ratio from at least three independent real-time quantitative RT-PCR reactions is shown. B, Western blots detecting expression of cyclin D, cyclin E, and fusion protein EWS-FLI1 or EWS-ERG in ET and RMS lines. EWS-FLI1 type III fusion (A4573) and type II fusion (RDES and 6647) have larger fusion proteins than the type I fusion. β-Actin is shown as the loading control. C, Western blots detecting D-type cyclin protein levels in cell lysates immunoprecipitated with anti-CDK4.

Fig. 2.

Expression of D- and E-type cyclins in seven ET (6647, RDES, TC135, TC71, TC32, TTC466, and A4573) and six RMS (RD, TC442, Rh18, Rh28, Rh30, and TC487) cell lines. A, relative expression of D- and E-type cyclin mRNAs in ET and RMS cell lines. For each sample, the average ± SD of the cyclin to β-actin ratio from at least three independent real-time quantitative RT-PCR reactions is shown. B, Western blots detecting expression of cyclin D, cyclin E, and fusion protein EWS-FLI1 or EWS-ERG in ET and RMS lines. EWS-FLI1 type III fusion (A4573) and type II fusion (RDES and 6647) have larger fusion proteins than the type I fusion. β-Actin is shown as the loading control. C, Western blots detecting D-type cyclin protein levels in cell lysates immunoprecipitated with anti-CDK4.

Close modal
Fig. 3.

MEK1/2 inhibition blocked ERK1/2 activation and decreased D-type cyclin expression in serum-starved RMS cells but not in serum-starved ET cells, whereas PI3K inhibition blocked AKT activation and decreased cyclin D expression in both serum-starved RMS cells and serum-starved ET cells. A, Western blots detecting ERK1/2 phosphorylation at Thr202/Tyr204 (p-ERK) and the expression of D-type cyclins at different hours after U0126 (U) or DMSO (D) treatment of 6647 and RD cells in 0.5% FBS culture medium. Total ERK1/2 (t-ERK) is shown as the loading control. B, relative expression of cyclin D mRNAs in ET (6647 and TC71) and RMS (Rh18 and RD) cell lines after 12 hours of treatment with U0126, DMSO, or drug-free culture medium with 0.5% FBS (Blank). The mean ± SD of the cyclin D to β-actin ratio from at least three independent real-time quantitative RT-PCRs is shown. C, Western blots detecting AKT phosphorylation at Ser473 (p-AKT) and the expression of D-type cyclins at different hours after LY294002 (LY) or DMSO (D) treatment of 6647 and Rh18 cells in 0.5% FBS culture medium. Total AKT (t-AKT) is shown as the loading control. D, expression of cyclin D RNAs in ET (6647 and TC71) and RMS (Rh18, Rh28, and Rh30) cell lines after 12 hours of treatment with LY294002, DMSO, or drug-free culture medium with 0.5% FBS (Blank). The mean ± SD of at least three independent real-time quantitative RT-PCRs is shown.

Fig. 3.

MEK1/2 inhibition blocked ERK1/2 activation and decreased D-type cyclin expression in serum-starved RMS cells but not in serum-starved ET cells, whereas PI3K inhibition blocked AKT activation and decreased cyclin D expression in both serum-starved RMS cells and serum-starved ET cells. A, Western blots detecting ERK1/2 phosphorylation at Thr202/Tyr204 (p-ERK) and the expression of D-type cyclins at different hours after U0126 (U) or DMSO (D) treatment of 6647 and RD cells in 0.5% FBS culture medium. Total ERK1/2 (t-ERK) is shown as the loading control. B, relative expression of cyclin D mRNAs in ET (6647 and TC71) and RMS (Rh18 and RD) cell lines after 12 hours of treatment with U0126, DMSO, or drug-free culture medium with 0.5% FBS (Blank). The mean ± SD of the cyclin D to β-actin ratio from at least three independent real-time quantitative RT-PCRs is shown. C, Western blots detecting AKT phosphorylation at Ser473 (p-AKT) and the expression of D-type cyclins at different hours after LY294002 (LY) or DMSO (D) treatment of 6647 and Rh18 cells in 0.5% FBS culture medium. Total AKT (t-AKT) is shown as the loading control. D, expression of cyclin D RNAs in ET (6647 and TC71) and RMS (Rh18, Rh28, and Rh30) cell lines after 12 hours of treatment with LY294002, DMSO, or drug-free culture medium with 0.5% FBS (Blank). The mean ± SD of at least three independent real-time quantitative RT-PCRs is shown.

Close modal
Fig. 4.

Decreased FKHR and FKHRL1 phosphorylations and increased nuclear localizations of FKHR and FKHRL1 in RD cells after PI3K blockade. A, Western blots detecting FKHR phosphorylation at Thr32 (pFKHRt) and Ser256 (pFKHRs), FKHRL1 phosphorylation at Thr32 (pFKHRL1t), and AFX phosphorylation at Ser193 (pAFXs). B, immunofluorescence detecting subcellular localization of Cy3-labeled FKHR (C-F) and Cy3-labeled FKHRL1 (C-FL) in untreated (Blank), DMSO-treated, or LY294002-treated RD cells. The combined images of DAPI nuclear stain with Cy3-labeled FKHR (C-F & D) or Cy3-labeled FKHRL1 (C-FL & D) are also shown.

Fig. 4.

Decreased FKHR and FKHRL1 phosphorylations and increased nuclear localizations of FKHR and FKHRL1 in RD cells after PI3K blockade. A, Western blots detecting FKHR phosphorylation at Thr32 (pFKHRt) and Ser256 (pFKHRs), FKHRL1 phosphorylation at Thr32 (pFKHRL1t), and AFX phosphorylation at Ser193 (pAFXs). B, immunofluorescence detecting subcellular localization of Cy3-labeled FKHR (C-F) and Cy3-labeled FKHRL1 (C-FL) in untreated (Blank), DMSO-treated, or LY294002-treated RD cells. The combined images of DAPI nuclear stain with Cy3-labeled FKHR (C-F & D) or Cy3-labeled FKHRL1 (C-FL & D) are also shown.

Close modal
Fig. 5.

Inhibition of EWS-FLI1 expression in TC71 cells results in decreased cyclin D1 and increased cyclin D3 levels, with cell cycle G1 arrest. A, comparison of EWS-FLI1, cyclin D1, cyclin D2, cyclin D3, and cyclin E1 protein levels in untransfected (Blank), pcDNA3 vector-transfected (Vector), and pcDNA-ASEF-transfected monoclone 9 (ASEF9). For detection of cyclin D2 expression, 3× more total cellular proteins were used. B, growth curve analysis. C, [3H]thymidine incorporation assay. D, FACS analysis of cell cycle distribution in controls (Blank and Vector) and ASEF9 cells.

Fig. 5.

Inhibition of EWS-FLI1 expression in TC71 cells results in decreased cyclin D1 and increased cyclin D3 levels, with cell cycle G1 arrest. A, comparison of EWS-FLI1, cyclin D1, cyclin D2, cyclin D3, and cyclin E1 protein levels in untransfected (Blank), pcDNA3 vector-transfected (Vector), and pcDNA-ASEF-transfected monoclone 9 (ASEF9). For detection of cyclin D2 expression, 3× more total cellular proteins were used. B, growth curve analysis. C, [3H]thymidine incorporation assay. D, FACS analysis of cell cycle distribution in controls (Blank and Vector) and ASEF9 cells.

Close modal
Fig. 6.

Inducible expression of EWS-FLI1 in RD/trexEF clone 7. A, Western blots detecting EWS-FLI1 protein, G1 cyclins and ERK1/2 phosphorylation at Thr202/Tyr204 (p-ERK) at different hours after treatment with either ethanol vehicle control (C) or 1 μg/ml tetracycline (T). Total ERK (t-ERK) was used as the loading control. B, Western blot band intensities in A were quantitated with FluorChem 8900. The intensity ratios of EWS-FLI1/t-ERK (EF/ERK), cyclin D1/t-ERK (D1/ERK), cyclin D3/t-ERK (D3/ERK), phospho-p44ERK/p44ERK (pERK1/ERK1), and phospho-p42ERK/p42ERK (pERK2/ERK2) in ethanol control (-C)- or tetracycline (-T)-treated RD/trexEF clone 7 are plotted. C, FACS analysis comparing cell cycle distribution in untreated RD cells (Blank) and ethanol-treated (ETOH-c) RD cells at 24 hours after induction with tetracycline (T-induced).

Fig. 6.

Inducible expression of EWS-FLI1 in RD/trexEF clone 7. A, Western blots detecting EWS-FLI1 protein, G1 cyclins and ERK1/2 phosphorylation at Thr202/Tyr204 (p-ERK) at different hours after treatment with either ethanol vehicle control (C) or 1 μg/ml tetracycline (T). Total ERK (t-ERK) was used as the loading control. B, Western blot band intensities in A were quantitated with FluorChem 8900. The intensity ratios of EWS-FLI1/t-ERK (EF/ERK), cyclin D1/t-ERK (D1/ERK), cyclin D3/t-ERK (D3/ERK), phospho-p44ERK/p44ERK (pERK1/ERK1), and phospho-p42ERK/p42ERK (pERK2/ERK2) in ethanol control (-C)- or tetracycline (-T)-treated RD/trexEF clone 7 are plotted. C, FACS analysis comparing cell cycle distribution in untreated RD cells (Blank) and ethanol-treated (ETOH-c) RD cells at 24 hours after induction with tetracycline (T-induced).

Close modal

Grant support: National Cancer Institute Director’s Challenge grant CA88199-01 (T. Triche), the Las Madrinas Endowment for Molecular Pathology (T. Triche), and the Fannie Rippel Foundation (P. Sorensen).

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.

Requests for reprints: Timothy J. Triche, Department of Pathology and Laboratory Medicine, Children’s Hospital Los Angeles, Keck School of Medicine, University of Southern California, Box 43, 4650 Sunset Boulevard, Los Angeles, CA 90027. Phone: 323-669-4516; Fax: 323-667-1123; E-mail: [email protected]

3

Unpublished data.

4

Other later passages of RD have been reported to lack p16INK4A (56).

We are grateful to Dr. Lingtao Wu for reviewing the manuscript and to Carol Peebles for assistance in this project. We thank Betty Schaub, Nilmini Waidyaratne, and Violette Shahbazian for their technical support.

1
Arndt CA, Crist WM. Common musculoskeletal tumors of childhood and adolescence.
N Engl J Med
,
341
:
342
-52,  
1999
.
2
Franchi A, Pasquinelli G, Cenacchi G, et al Immunohistochemical and ultrastructural investigation of neural differentiation in Ewing sarcoma/PNET of bone and soft tissues.
Ultrastruct Pathol
,
25
:
219
-25,  
2001
.
3
Arvand A, Denny CT. Biology of EWS/ETS fusions in Ewing’s family tumors.
Oncogene
,
20
:
5747
-54,  
2001
.
4
Delattre O, Zucman J, Plougastel B, et al Gene fusion with an ETS DNA-binding domain caused by chromosome translocation in human tumours.
Nature (Lond)
,
359
:
162
-5,  
1992
.
5
Sorensen PH, Lessnick SL, Lopez-Terrada D, et al A second Ewing’s sarcoma translocation, t(21;22), fuses the EWS gene to another ETS-family transcription factor, ERG.
Nat Genet
,
6
:
146
-51,  
1994
.
6
Merlino G, Helman LJ. Rhabdomyosarcoma: working out the pathways.
Oncogene
,
18
:
5340
-8,  
1999
.
7
Barr FG. Gene fusions involving PAX and FOX family members in alveolar rhabdomyosarcoma.
Oncogene
,
20
:
5736
-46,  
2001
.
8
Kovar H, Auinger A, Jug G, et al Narrow spectrum of infrequent p53 mutations and absence of MDM2 amplification in Ewing tumours.
Oncogene
,
8
:
2683
-90,  
1993
.
9
Kovar H, Jug G, Aryee DN, et al Among genes involved in the RB dependent cell cycle regulatory cascade, the p16 tumor suppressor gene is frequently lost in the Ewing family of tumors.
Oncogene
,
15
:
2225
-32,  
1997
.
10
Dauphinot L, De Oliveira C, Melot T, et al Analysis of the expression of cell cycle regulators in Ewing cell lines: EWS-FLI-1 modulates p57KIP2 and c-Myc expression.
Oncogene
,
20
:
3258
-65,  
2001
.
11
Felix CA, Kappel CC, Mitsudomi T, et al Frequency and diversity of p53 mutations in childhood rhabdomyosarcoma.
Cancer Res
,
52
:
2243
-7,  
1992
.
12
Mulligan LM, Matlashewski GJ, Scrable HJ, Cavenee WK. Mechanisms of p53 loss in human sarcomas.
Proc Natl Acad Sci USA
,
87
:
5863
-7,  
1990
.
13
Iolascon A, Faienza MF, Coppola B, et al Analysis of cyclin-dependent kinase inhibitor genes (CDKN2A, CDKN2B, and CDKN2C) in childhood rhabdomyosarcoma.
Genes Chromosomes Cancer
,
15
:
217
-22,  
1996
.
14
De Chiara A, T’Ang A, Triche TJ. Expression of the retinoblastoma susceptibility gene in childhood rhabdomyosarcomas.
J Natl Cancer Inst (Bethesda)
,
85
:
152
-7,  
1993
.
15
Lawlor ER, Scheel C, Irving J, Sorensen PH. Anchorage-independent multi-cellular spheroids as an in vitro model of growth signaling in Ewing tumors.
Oncogene
,
21
:
307
-18,  
2002
.
16
Khan J, Wei JS, Ringner M, et al Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks.
Nat Med
,
7
:
673
-9,  
2001
.
17
Suzuki R, Kuroda H, Komatsu H, et al Selective usage of D-type cyclins in lymphoid malignancies.
Leukemia (Baltimore)
,
13
:
1335
-42,  
1999
.
18
Sherr CJ. Mammalian G1 cyclins.
Cell
,
73
:
1059
-65,  
1993
.
19
Hunter T, Pines J. Cyclins and cancer. II. Cyclin D and CDK inhibitors come of age.
Cell
,
79
:
573
-82,  
1994
.
20
Sherr CJ. Cancer cell cycles.
Science (Wash DC)
,
274
:
1672
-7,  
1996
.
21
Herber B, Truss M, Beato M, Muller R. Inducible regulatory elements in the human cyclin D1 promoter.
Oncogene
,
9
:
1295
-304,  
1994
.
22
Brooks AR, Shiffman D, Chan CS, Brooks EE, Milner PG. Functional analysis of the human cyclin D2 and cyclin D3 promoters.
J Biol Chem
,
271
:
9090
-9,  
1996
.
23
Matsushime H, Roussel MF, Ashmun RA, Sherr CJ. Colony-stimulating factor 1 regulates novel cyclins during the G1 phase of the cell cycle.
Cell
,
65
:
701
-13,  
1991
.
24
Inaba T, Matsushime H, Valentine M, et al Genomic organization, chromosomal localization, and independent expression of human cyclin D genes.
Genomics
,
13
:
565
-74,  
1992
.
25
Tam SW, Theodoras AM, Shay JW, Draetta GF, Pagano M. Differential expression and regulation of cyclin D1 protein in normal and tumor human cells: association with Cdk4 is required for cyclin D1 function in G1 progression.
Oncogene
,
9
:
2663
-74,  
1994
.
26
Wu L, Chen P, Shum CH, et al MAT1-modulated CAK activity regulates cell cycle G1 exit.
Mol Cell Biol
,
21
:
260
-70,  
2001
.
27
Biggs WH, III, Meisenhelder J, Hunter T, Cavenee WK, Arden KC. Protein kinase B/Akt-mediated phosphorylation promotes nuclear exclusion of the winged helix transcription factor FKHR1.
Proc Natl Acad Sci USA
,
96
:
7421
-6,  
1999
.
28
Marshall CJ. Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation.
Cell
,
80
:
179
-85,  
1995
.
29
Cobb MH, Goldsmith EJ. How MAP kinases are regulated.
J Biol Chem
,
270
:
14843
-6,  
1995
.
30
Lavoie JN, L’Allemain G, Brunet A, Muller R, Pouyssegur J. Cyclin D1 expression is regulated positively by the p42/p44MAPK and negatively by the p38/HOGMAPK pathway.
J Biol Chem
,
271
:
20608
-16,  
1996
.
31
Favata MF, Horiuchi KY, Manos EJ, et al Identification of a novel inhibitor of mitogen-activated protein kinase kinase.
J Biol Chem
,
273
:
18623
-32,  
1998
.
32
Davies SP, Reddy H, Caivano M, Cohen P. Specificity and mechanism of action of some commonly used protein kinase inhibitors.
Biochem J
,
351
:
95
-105,  
2000
.
33
Muise-Helmericks RC, Grimes HL, Bellacosa A, et al Cyclin D expression is controlled post-transcriptionally via a phosphatidylinositol 3-kinase/Akt-dependent pathway.
J Biol Chem
,
273
:
29864
-72,  
1998
.
34
Diehl JA, Cheng M, Roussel MF, Sherr CJ. Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization.
Genes Dev
,
12
:
3499
-511,  
1998
.
35
Brunet A, Bonni A, Zigmond MJ, et al Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor.
Cell
,
96
:
857
-68,  
1999
.
36
Kops GJ, de Ruiter ND, De Vries-Smits AM, et al Direct control of the Forkhead transcription factor AFX by protein kinase B.
Nature (Lond)
,
398
:
630
-4,  
1999
.
37
Brownawell AM, Kops GJ, Macara IG, Burgering BM. Inhibition of nuclear import by protein kinase B (Akt) regulates the subcellular distribution and activity of the forkhead transcription factor AFX.
Mol Cell Biol
,
21
:
3534
-46,  
2001
.
38
Schmidt M, Fernandez de Mattos S, van der Horst A, et al Cell cycle inhibition by FoxO forkhead transcription factors involves downregulation of cyclin D.
Mol Cell Biol
,
22
:
7842
-52,  
2002
.
39
Ramaswamy S, Nakamura N, Sansal I, Bergeron L, Sellers WR. A novel mechanism of gene regulation and tumor suppression by the transcription factor FKHR.
Cancer Cell
,
2
:
81
-91,  
2002
.
40
Murphy LO, Smith S, Chen RH, Fingar DC, Blenis J. Molecular interpretation of ERK signal duration by immediate early gene products.
Nat Cell Biol
,
4
:
556
-64,  
2002
.
41
Roovers K, Assoian RK. Integrating the MAP kinase signal into the G1 phase cell cycle machinery.
Bioessays
,
22
:
818
-26,  
2000
.
42
Fukuma M, Okita H, Hata J, Umezawa A. Upregulation of Id2, an oncogenic helix-loop-helix protein, is mediated by the chimeric EWS/ets protein in Ewing sarcoma.
Oncogene
,
22
:
1
-9,  
2003
.
43
Nishimori H, Sasaki Y, Yoshida K, et al The Id2 gene is a novel target of transcriptional activation by EWS-ETS fusion proteins in Ewing family tumors.
Oncogene
,
21
:
8302
-9,  
2002
.
44
Lasorella A, Uo T, Iavarone A. Id proteins at the cross-road of development and cancer.
Oncogene
,
20
:
8326
-33,  
2001
.
45
Norton JD. ID helix-loop-helix proteins in cell growth, differentiation and tumorigenesis.
J Cell Sci
,
113
:
3897
-905,  
2000
.
46
Yates PR, Atherton GT, Deed RW, Norton JD, Sharrocks AD. Id helix-loop-helix proteins inhibit nucleoprotein complex formation by the TCF ETS-domain transcription factors.
EMBO J
,
18
:
968
-76,  
1999
.
47
Shao J, Sheng H, DuBois RN, Beauchamp RD. Oncogenic Ras-mediated cell growth arrest and apoptosis are associated with increased ubiquitin-dependent cyclin D1 degradation.
J Biol Chem
,
275
:
22916
-24,  
2000
.
48
Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B.
Nature (Lond)
,
378
:
785
-9,  
1995
.
49
van Weeren PC, de Bruyn KM, de Vries-Smits AM, van Lint J, Burgering BM. Essential role for protein kinase B (PKB) in insulin-induced glycogen synthase kinase 3 inactivation. Characterization of dominant-negative mutant of PKB.
J Biol Chem
,
273
:
13150
-6,  
1998
.
50
Matsumoto Y, Tanaka K, Nakatani F, et al Downregulation and forced expression of EWS-Fli1 fusion gene results in changes in the expression of G1 regulatory genes.
Br J Cancer
,
84
:
768
-75,  
2001
.
51
Eliazer S, Spencer J, Ye D, Olson E, Ilaria RL, Jr. Alteration of mesodermal cell differentiation by EWS/FLI-1, the oncogene implicated in Ewing’s sarcoma.
Mol Cell Biol
,
23
:
482
-92,  
2003
.
52
Mao X, Miesfeldt S, Yang H, Leiden JM, Thompson CB. The FLI-1 and chimeric EWS-FLI-1 oncoproteins display similar DNA binding specificities.
J Biol Chem
,
269
:
18216
-22,  
1994
.
53
Silvany RE, Eliazer S, Wolff NC, Ilaria RL, Jr. Interference with the constitutive activation of ERK1 and ERK2 impairs EWS/FLI-1-dependent transformation.
Oncogene
,
19
:
4523
-30,  
2000
.
54
Lessnick SL, Dacwag CS, Golub TR. The Ewing’s sarcoma oncoprotein EWS/FLI induces a p53-dependent growth arrest in primary human fibroblasts.
Cancer Cell
,
1
:
393
-401,  
2002
.
55
Deneen B, Denny CT. Loss of p16 pathways stabilizes EWS/FLI1 expression and complements EWS/FLI1 mediated transformation.
Oncogene
,
20
:
6731
-41,  
2001
.
56
Urashima M, Teoh G, Akiyama M, et al Restoration of p16INK4A protein induces myogenic differentiation in RD rhabdomyosarcoma cells.
Br J Cancer
,
79
:
1032
-6,  
1999
.