EWS-ATF-1 Chimeric Protein in Soft Tissue Clear Cell Sarcoma Associates with CREB-Binding Protein and Interferes with P53-Mediated Trans-Activation Function

Oncogene (2001) 20, 6653 ± 6659 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc ORIGINAL PAPERS EWS-ATF-1 chimeric protein in soft tissue clear cell sarcoma associates with CREB-binding protein and interferes with p53-mediated trans-activation function

Yasuo Fujimura1, Habibur Siddique1, Leo Lee2, Veena N Rao1 and E Shyam P Reddy*,1

1Program of Cancer Genetics, Cancer center, MS #481, Department of Biochemistry, School of Medicine, MCP Hahnemann University, Broad and Vine, Philadelphia, Pennsylvania, PA 19102, USA; 2SAIC, Frederick, Maryland, MD 21702, USA

The recurrent t(12;22) (q13;q12) chromosomal transloca- fused to various transcriptional factors like ETS family tion associated with soft tissue clear cell sarcoma results members Fli-1 (Ben-David et al., 1991; Prasad et al., in a chimeric protein EWS-ATF-1 that acts as a 1992; Delattre et al., 1992), erg (Rao et al., 1987; constitutive transcriptional activator. The CBP/p300 Reddy et al., 1987; Sorensen et al., 1994), ETV1 (Jeon transcriptional coactivator, which links various transcrip- et al., 1995), E1A-F (Kaneko et al., 1996; Urano et al., tional factors to basal transcription apparatus, partici- 1996), FEV (Peter et al., 1997), Tumor suppressor gene pates in transcriptional activation, growth and cell cycle WT-1 (Gerald et al., 1995), orphan nuclear receptor control and dierentiation. In this study, we show that TEC (Labelle et al., 1995), and C/EBP family CHOP EWS-ATF-1 associates constitutively with CBP both in (Panagopoulos et al., 1996). In addition, TLS/FUS, a vitro and in vivo. Both EWS and ATF-1 fusion domains gene closely related to EWS, is also shown to be are needed for this interaction. Here, we demonstrate involved in chromosomal translocations with erg in that EWS-ATF-1 represses p53/CBP-mediated trans- human myeloid leukemia (Ichikawa et al., 1994; activation function. Overexpression of CBP can counter- Panagopoulos et al., 1994), and with CHOP in myxoid act this repressive eect of EWS-ATF-1. Taken liposarcoma (Crozat et al., 1993; Rabbitts et al., 1993). together, these ®ndings suggest that one of the EWS has several conserved RNA binding motifs mechanisms by which EWS-ATF-1 may cause tumors called RNP-CS and RGG boxes (Burd and Dreyfuss, is through targeting CBP/p300 resulting in the loss of 1994). We have shown that EWS preferentially binds function of p53. This novel mechanism may be to poly G and poly U RNA (Ohno et al., 1994). The responsible for the development of these and other RNA binding activity of EWS is located in the RGG related solid tumors. Oncogene (2001) 20, 6653 ± 6659. box (Ohno et al., 1994) which is present in the carboxy terminal region. In these aberrant chimeric proteins, Keywords: EWS-ATF-1; CBP; p53; Ewing's sarcoma; this carboxyl terminal RNA binding domain of EWS is clear cell sarcoma replaced by DNA-binding domain of several transcriptional factors. EWS-Fli-1, EWS-erg and TLS/FUS-erg Introduction have been shown to function as sequence speci®c transcriptional activators (Ohno et al., 1993, 1994; May Soft tissue clear cell sarcoma or malignant melanoma et al., 1993; Bailly et al., 1994; Prasad et al., 1994). of soft parts is rare and aggressive neuroectodermal Antagonizing EWS-fusion gene expression in the tumor that occurs in the deep soft tissue usually near tumors showed reduced tumorigenicity and clonogeni- tendons and aponeuroses (Epstein et al., 1984). Recent city, suggesting that a certain threshold level of molecular characterization of this tumor reveals expression of these chimeric products is needed for recurrent chromosomal translocation t(12;22) maintaining oncogenicity (Ouchida et al., 1995). We (q13;q12) resulting in the fusion of EWS gene to have found that over expression of aberrant fusion Activating Transcription Factor 1 (ATF-1) gene (Zuc- proteins (EWS-erg, EWS-Fli-1 and TLS/FUS-erg) and man et al., 1993). The EWS gene on chromosome 22 is its normal counterpart (Fli-1 and erg) in mouse a member of RNA binding protein family which ®broblasts inhibit apoptosis induced by either serum includes TLS/FUS, hTAFII68, SARFH/Cabeza (Ber- deprivation or calcium ionophore treatment, indicating tolotti et al., 1996; Immanuel et al., 1995; Stolow and the anti-apoptotic role of these fusion proteins in these Haynes, 1995). Chromosomal translocations associated tumors (Yi et al., 1997). with EWS gene are found in Ewing family tumors and In the EWS-ATF-1 protein, the amino-terminal other solid tumors. In these tumors, the EWS gene is domain of EWS is juxtaposed with DNA binding domain of ATF-1 (Figure 1). ATF-1 is a member of CREB/ATF basic leucine-zipper type of transcriptional factor family, which includes CREB and CREM. *Correspondence: ESP Reddy Received 23 January 2001; revised 23 May 2001; accepted 31 May CREB and ATF-1 share signi®cant homology in 2001 Kinase inducible domain (KID), basic and leucine- EWS-ATF-1 binds to CBP and inhibits p53 transactivation Y Fujimura et al 6654

Figure 1 Schematic representation of the functional domains of EWS-ATF-1 and CBP. (a) Normal EWS, ATF-1, and their chimeric product EWS-ATF-1 are shown. Arrowhead indicates breakpoint of the fusion protein. Carboxy-terminal region of EWS containing RGG box, which functions as an RNA binding domain, is shown. RNP-CS represents the domain conserved in a variety of RNA binding proteins. Basic and leucine zipper domain of ATF-1 responsible for DNA binding and dimerization are shown. The amino-terminal region of ATF-1 containing the Protein Kinase A (PKA) site is replaced by amino-terminal half of EWS in EWS-ATF-1. (b) Schematic representation of co-activator CBP is shown. Three cysteine histidine rich region and Bromodoamin are shown. The region possess Histone Acetyl-Transferase (HAT) is indicated. KIX represents CREB binding site (aa446 ± 684)

zipper domain. Increasing concentrations of cAMP or regulated by CBP/p300 include CREB, ATF-1, c-Myb, calcium induce phosphorylation of KID domain of c-fos, c-Jun, STATs, GATA-1, Ets, MyoD1 and CREB resulting in augmented transactivation proper- nuclear receptors (reviewed in Giles et al., 1998). ties of CREB. The interaction between CBP and Recent studies show that p53 also requires this cofactor CREB is dependent on the phosphorylation status of for its transcriptional activation function (Lill et al., Ser 133 in the KID domain, which represents a protein 1997; Avantaggiati et al., 1997; Gu et al., 1997; kinase A phosphoacceptor site (Chrivia et al., 1993; Somasundaram and El-Deiry, 1997). In these path- Arias et al., 1994). Similarly, when phosphorylated at ways, CBP/p300 is thought to work as a signal Ser 63 (corresponding to Ser 133 of CREB), ATF-1 integrator between multiple transcriptional factors. can bind to CBP and show augmented transcriptional CBP/p300, which interact with other cofactors such activation to certain target promoters (Shimomura et as NcoA-1, pCIP-1 and TFH110, has histone acetyl al., 1996). In EWS-ATF-1, the N-terminal region of transferase activity that makes the holoenzyme com- KID domain including the PKA phosphoacceptor site plex to transmit the activating signals. CBP/p300 and at Ser 63 was replaced by the N-terminal region of other cofactors are thought to participate in chromatin EWS gene (Figure 1), resulting in the loss of the remodeling through its histone acetylation activity to phosphoacceptor site. We and others have demon- open chromatin structure. Both histones and other strated that EWS-ATF-1 functions as a constitutive transcriptional factors including p53 are acetylated by transcriptional activator on certain target promoters CBP/p300 resulting in increased ability to bind target (Fujimura et al., 1996; Brown et al., 1995). DNA sequences suggesting a broad function for CBP/ CBP/p300 are large nuclear phospho-proteins which p300 (Gu and Roeder, 1997). The translocation of CBP function as cofactors to various transcriptional factors has been found in therapy-related myelodysplasia and and seem to have a key role in growth and cell cycle acute myeloid leukemia. In these translocations CBP is control, cellular dierentiation and development. The fused to MLL/ALL-1 and MOZ respectively, although growing numbers of factors that are found to be the functional study of these fusion genes remains to be

Oncogene EWS-ATF-1 binds to CBP and inhibits p53 transactivation Y Fujimura et al 6655 determined (Satake et al., 1997; Borrow et al., 1996; Results Gu et al., 1992). Several DNA viral oncoproteins were shown to EWS-ATF-1 interacts with CBP both in vitro and in vivo target transcriptional cofactor CBP/p300 suggesting that this phenomenon may contribute to neoplasia. Both CREB and ATF-1 proteins were shown earlier to We speculated whether a similar mechanism for bind to KIX domain of CBP. To test whether EWS- cellular transformation is used by EWS-ATF-1 fusion ATF-1 binds to CBP/p300, we performed co-immuno- protein. To test this hypothesis, we studied the precipitation assays with in vitro translation products interaction of EWS-ATF-1 and CBP. In this study, of EWS-ATF-1 and CBP (KIX domain) (Figure 2a). we show that EWS-ATF-1 associates constitutively Both EWS-ATF-1 and KIX domain of CBP (aa446 ± with CBP and represses the p53-mediated transcrip- 684) were co-translated and labeled with 35S-methionine tional activation function through regulation of CBP/ (Figure 2a, lane 1), then subjected to immunoprecipita- p300. To our knowledge, this is the ®rst report tion with anti-ATF-1 monoclonal antibody. KIX showing that aberrant chimeric proteins inhibit p53 domain of CBP was co-immunoprecipitated along with signaling pathways by squelching transcriptional EWS-ATF-1 (Figure 2a, lanes 5, 6). Similar results coactivator CBP. were observed when EWS-ATF-1 and KIX domain of

a b

Figure 2 EWS-ATF-1 interacts with CBP in vitro and in vivo.(a) Co-immunoprecipitation studies of in vitro translated CBP and EWS-ATF-1. EWS-ATF-1 or CBP RNAs were translated in vitro either alone or together with CBP RNA. These [35S]methionine- labeled translated products were immunoprecipitated with monoclonal antibody raised against carboxy-terminal region of ATF-1. Lane 1, In vitro co-translation of EWS-ATF-1 and CBP (aa446 ± 684); Lane 2, In vitro translation of EWS-ATF-1; Lane 3, In vitro translation of CBP (aa 446 ± 684) (KIX domain of CBP); Lane 4, EWS-ATF-1 and CBP proteins were synthesized separately and then mixed and subjected to immunoprecipitation with monoclonal antibody raised against carboxy-terminal region of ATF-1; Lanes 5 and 6, EWS-ATF-1 and CBP proteins were synthesized together by co-translation and then this reaction mixture was immunoprecipitated with the above mentioned ATF-1 monoclonal antibody. Lane 7, represents in vitro translation carried out in the absence of RNA (negative control); Lane 8, in vitro translated EWS-ATF-1 was immunoprecipitated with anti-ATF-1 antibody; Lane 9, in vitro translated CBP (aa446 ± 684) was immunoprecipitated with anti-ATF-1 antibody. (b) GST pull down assay. Schematic representation of EWS-ATF-1 deletion mutants used in the in vitro binding studies are shown. GST or GST-CBP (aa 446 ± 684) was incubated with in vitro translated [35S]methionine-labeled proteins and the bound complexes were characterized by polyacrylamide gel electrophoresis and visualized by autoradiography. Lanes 1 ± 5 represents input of in vitro translation products. Lanes 6 ± 15, represent [35S]methionine-labeled products bound to GST and GST-CBP. (c) Puri®ed GST and GST-CBP (aa 446 ± 684) proteins used for GST pull down assays are shown by Coomassie staining. (d) Immunoblot analysis of SU-CCS-1 whole cell extracts with anti-ATF-1 monoclonal antibody after immunoprecipitation with CBP (lane 3), normal rabbit IgG (lane 2). Lane 1, detection of EWS-ATF-1 in whole cell extract without immunoprecipitation

Oncogene EWS-ATF-1 binds to CBP and inhibits p53 transactivation Y Fujimura et al 6656 CBP were translated separately (Figure 2a, lanes 2 and checked these eects in Saos-2 cells which do not 3), then mixed and subjected to the immunoprecipita- express p53 (Figure 3b). We co-transfected p53 tion analysis (Figure 2a, lane 4). expression plasmid along with EWS-ATF-1 expression Since a kinase regulatory domain of EWS-ATF-1 is plasmid into Saos-2 cells. Again, EWS-ATF-1 revealed replaced by the EWS regulatory domain (Fujimura et similar extent of repression (55%) of p53-mediated al., 1996), we speculated that phosphorylation of ATF- transcriptional activation (Figure 3b). We wanted to 1 fusion domain by PKA may not be necessary to bind know whether sequestration of CBP is involved in this to CBP. To investigate the nature of the interaction in repressive eect. To test this hypothesis, we transfected vitro, CBP-KIX domain was expressed as a GST-fusion increasing concentration of CBP along with EWS- protein, and then tested its ability to bind to in vitro ATF-1 (Figure 3c). Interestingly, over expression of translated EWS-ATF-1 using GST pull-down assay CBP relieved EWS-ATF-1 mediated repression in a (Figure 2b,c). Consistent with our previous results, dose dependent manner (Figure 3c). These results EWS-ATF-1 binds to GST ± CBP (446 ± 684) whereas suggest that EWS-ATF-1 interfere with p53-mediated unphosphorylated ATF-1 does not bind under these transactivation by sequestration of CBP/p300 proteins. experimental conditions. In order to further understand mechanisms of interaction, we in vitro translated EWS N-terminal Discussion fusion domain and ATF-1 C-terminal fusion domain separately (Figure 2a) and subjected them to GST pull- CBP/p300 plays a pivotal role in transcriptional down assay (Figure 2b). Neither EWS N-terminal half control of broad aspects of cellular regulation includ- nor ATF-1 C-terminal half showed detectable interac- ing dierentiation, homeostasis, and growth control. In tion with KIX domain of CBP under these experi- this study, we have shown that EWS-ATF-1 associates mental conditions, indicating that fusion partners by with CBP both in vitro and in vivo through the KIX themselves are not sucient to bind to CBP (Figure domain of CBP (Figure 2a ± c). We also demonstrate 2b; lanes 13, 15). It appears that the aberrant fusion of that EWS-ATF-1 fusion protein interferes with the EWS to ATF-1 may cause conformational change to p53-mediated transcriptional activation through com- the fusion protein that facilitates binding to CBP. petition between p53 and EWS-ATF-1 in binding to Next, we asked whether this interaction occurs in CBP (Figure 3). vivo in malignant melanoma of soft part (MMSP). SU- The tumor suppressor gene p53, which is mutated in CCS-1 is one of MMSP cell lines, which expressed several human cancers, plays an important role in the EWS-ATF-1 (Brown et al., 1995). We tested whether regulation of cell cycle and DNA repair (Levine, 1997). immunoprecipitation of CBP brings down EWS-ATF-1 The transcriptional activity of p53 is regulated through in co-immunoprecipitation assays (Figure 2d). Using the binding with CBP/p300 cofactor. Viral oncopro- anti-CBP antibody, we obtained a band with a teins such as adenovirus E1a, HTLV-I Tax, and human molecular mass around 60 kDa (Figure 2d, lane 3). papilloma virus E6 are known to interfere with p53 This band co-migrates with the EWS-ATF-1 band mediated trans-activation function through the binding obtained from Western analysis (Figure 2d, lane 1). We with CBP and was suggested that this may be in part conclude from these results that EWS-ATF-1 associ- responsible for transformation by these viruses. Inhibi- ates with CBP both in vitro and in vivo. tion of histone acetylase activity of CBP by E1a, which is involved in remodeling of chromatin structure, might explain one of the consequences of inhibition. In the EWS-ATF-1 represses p53-mediated trans-activation case of human T-cell leukemia virus type-I (HTLV-I) Recent studies revealed that p53 requires CBP/p300 for viral protein Tax, it's competition with p53 in binding its transcriptional activation function (Lill et al., 1997; to CBP might account for repression of the p53- Avantaggiati et al., 1997; Gu et al., 1997). Since EWS- mediated transcriptional activity (Ariumi et al., 2000). ATF-1 interacts with CBP, we hypothesized that EWS- HTLV-1 TAX binds to KIX domain of CBP. Recent ATF-1 may sequester CBP/p300 (which is needed for studies reveal that p53 also binds to KIX domain of p53 transactivation function) leading to the inhibition CBP besides the carboxy terminal and C/H1 region of p53-mediated trans-activation function. To test this (Van Orden et al., 1999). Both p53 and Tax were hypothesis, we studied the eect of EWS-ATF-1 on shown to bind to KIX domain in a mutually exclusive p53-mediated transcriptional activation function. U2- fashion. It is possible to assume such squelching OS cells, which express wild type p53 (Kastan et al., mechanism is operative in EWS-ATF-1 since EWS- 1992), were transfected with various amounts of EWS- ATF-1 also binds to KIX domain. ATF-1 expression plasmid along with p53 reporter and There is an increasing body of evidence indicating CAT assays were performed (Figure 3a). EWS-ATF-1 that the gene dosage of CBP/p300 is critical to keep repressed p53-mediated trans-activation in a dose proper cellular function. The patients of hereditary dependent manner (Figure 3a). Nearly half of the disease Rubinstein-Taybi syndrome are heterozygous at activity (49%) compared to control was observed when CBP locus resulting in developmental abnormality. In 1 mg of EWS-ATF-1 expression plasmid was added to some of these cases, one allele is inactivated by large the cells. To eliminate the possibility that EWS-ATF-1 deletion suggesting haplo-insuciency model of this aects endogenous p53 expression levels, we also disease. These patients are prone to develop neural and

Oncogene EWS-ATF-1 binds to CBP and inhibits p53 transactivation Y Fujimura et al 6657 a developmental tumors (Miller and Rubinstein, 1995). Truncated mutations of p300 were found in epithelial cancers which was accompanied inactivation of the second allele (Gayther et al., 2000). In a mouse model, CBP nullizygous mutants die in the mid-gestation. The hemizygous mutants of CBP develop leukemia support- ing the suggestion that certain level of expression of CBP is required to prevent tumor development (Kung et al., 2000). Taken together, these ®ndings support the view that CBP/p300 has a tumor suppressor function. It appears from our results that aberrant fusion proteins and viral proteins follow similar mechanism in cellular transformation. This includes targeting transcriptional cofactors (such as CBP/p300 etc.) that play an essential role to transmit and combine several signal transduction cascades. Therefore, this novel molecular mechanism may be responsible for the initiation and/or progression of solid tumors involving EWS-ATF-1. It is possible other EWS/TLS chimeric proteins may use similar pathway in transformation. b

Materials and methods

Plasmids

PG13CAT and MG15CAT were kindly provided by B Vogelstein (Kern et al., 1992). pRc/CMV mCBP-HA-1 was a gift from R Goodman. Construction of EWS-ATF-1 and deletion mutants were described elsewhere (Fujimura et al., 1996). Wild type p53 was cloned into pSG5 vector. CREB interaction domain (aa446 ± 684) of CBP was generated by PCR using pcDNA3/CBP as a template and cloned into pGex2TK (Amersham Pharmacia Biotech).

Tissue culture and transfection Saos-2 cells and U-2 OS cells were purchased from the American Type Culture Collection (Rockville, MD, USA). SU-CCS-1 cells were obtained from AL Epstein. Saos-2 cells were grown in McCoy's 5A medium supplemented with 15% c 120 were transiently transfected with 2 mg of p53 reporter plasmid PG -CAT, 2 mg of pCMVb (b-galactosidase) along with various 100 13 amounts of expression vector encoding EWS-ATF-1. All CAT values were normalized to each other based on the respective b- 80 galactosidase activity. Percentage of the control was calculated based on CAT value of the equimolar empty vector transfected cells. Error bars are standard deviations of three independent 60 experiments. (b) Saos-2 cells were co-transfected with 2 mgof PG13CAT, 10 mg of pCH110 (b-galactosidase as internal control), 40 0.1 mg of p53 expression plasmid along with various concentrations (1, 3 or 10 mg) of EWS-ATF-1 expression plasmid or pcDNA3 empty vector. Percentage of the control was calculated 20 based upon fold activation of CAT value of the empty vector transfected cells. Error bars are standard deviations of three 0 independent experiments. (c) Overexpression of CBP relieves the repressive eect of EWS-ATF-1 on p53-mediated trans-activation. U-2 OS cells were transfected with 1 mg of EWS-ATF-1 expression vector, 2 mgofPG13-CAT, 2 mg of pCMVb (as an internal control) along with indicated amount of full-length CBP expression vector. The value obtained with the reporter alone was Figure 3 EWS-ATF-1 represses p53-mediated trans-activation in arbitrarily set as 100. Results shown are average of three a dose-dependent manner. (a) U-2 OS human osteosarcoma cells independent experiments with standard deviations

Oncogene EWS-ATF-1 binds to CBP and inhibits p53 transactivation Y Fujimura et al 6658 fetal bovine serum, 50 units/ml penicillin, 50 mg/ml strepto- 10% SDS ± PAGE gels. After the transfer, membrane was mycin. Both U-2 OS cells and SU-CCS-1 cells were grown in incubated in blocking solution (5% blocking materials (Bio- RPMI-1640 medium supplemented with 10% fetal bovine Rad) in 16TBS) for overnight. Western blot assay were serum, 50 units/ml penicillin, 50 mg/ml streptomycin. The performed using anti-ATF-1 mouse monoclonal antibody cells were transfected by calcium phosphate precipitation (25C10G) (Santa Cruz) as primary antibody. Immunocom- method on a 10-cm plate. Forty-eight hours after transfec- plex were detected using chemiluminescence-based system tion, the cells were harvested and chloramphenicol acetyl (Amersham) as described by the manufacturer. transferase (CAT) activity and b-galactosidase activity were measured. CAT activity was measured using Fuji Bioimage GST Pull-down assay analyser and transfection eciency was normalized using b- galactosidase activity as previously described (Fujimura et al., GST-CBP (KIX) and GST were expressed and puri®ed as 1996). described previously. For in vitro binding assay, 10 mlof35S- methionine labeled in vitro translated proteins was diluted with binding buer (20 m Tris-HCl [pH 7.8], 150 mM Co-immunoprecipitation and Western blot assay M NaCl, 5 mM MgCl2, 0.1 mM EDTA, 0.05% Nonidet P-40, Con¯uent 10-cm plates of SU-CCS-1 cells were washed twice 1mM dithiothreitol) and incubated with 2 ml of GST ± CBP with cold phosphate-buered saline and lysed in lysis buer KIX beads or GST beads for 1 h at 48C. The beads were with freshly added protease inhibitors (50 mM Tris-HCl washed ®ve times with binding buer, boiled in SDS sample [pH 8.0], 150 mM NaCl, 5 mM EDTA, 0.5% IGEPAL CA- buer and loaded on 12% SDS-polyacrylamide gel. The gel 630 (SIGMA), 0.1 mM PMSF and complete protease was stained with Coomassie brilliant blue, dried and scanned inhibitor cocktail (Boehringer Mannheim)). After incubation using a Fuji BioImaging analyser. on ice for 20 min, the lysate was centrifuged at 20 000 g for 10 min at 48C. Supernatants were diluted with one volume of lysis buer without IGEPAL CA-630 and subjected to immunoprecipitation using 1 mg anti-CBP antibody (A-22) Acknowledgments or control sera (Santa Cruz). Immunocomplex were cross- We thank other colleagues of Reddy and Rao's labora- linked to 15 ml protein A/G PLUS Agarose (Santa Cruz) for tories for their kind cooperation. This work is supported in overnight at 48C. The beads were washed three times with part by NIH grants RO1 CA 85343, RO1 CA 58642 and lysis buer containing 0.25% of IGEPAL CA-630. Proteins US Army medical research and command grant DAMD- were solubilized with SDS sample buer and separated on 17-99-1-9060 to ESP Reddy and CA 57322 to VN Rao.

References

Arias J, Alberts AS, Brindle P, Claret FX, Smeal T, Karin M, Gayther SA, Batley SJ, Linger L, Bannister A, Thorpe K, Feramisco J and Montminy M. (1994). Nature, 370, 226 ± Chin SF, Daigo Y, Russell P, Wilson A, Sowter HM, 229. Delhanty JD, Ponder BA, Kouzarides T and Caldas C. Ariumi Y, Kaida A, Lin JY, Hirota M, Masui O, Yamaoka (2000). Nat. Genet., 24, 300 ± 303. S, Taya Y and Shimotohno K. (2000). Oncogene, 19, Gerald WL, Rosai J and Ladanyi M. (1995). Proc. Natl. 1491 ± 1499. Acad.Sci.USA,92, 1028 ± 1032. Avantaggiati ML, Ogryzko V, Gardner K, Giordano A, Giles RH, Peters DJ and Breuning MH. (1998). Trends Levine AS and Kelly K. (1997). Cell, 89, 1175 ± 1184. Genet., 14, 178 ± 183. Bailly RA, Bosselut R, Zucman J, Cormier F, Delattre O, Gu Y, Nakamura T, Alder H, Prasad R, Canaani O, Cimino Roussel M, Thomas G and Ghysdael J. (1994). Mol. Cell G, Croce CM and Canaani E. (1992). Cell, 71, 701 ± 708. Biol., 14, 3230 ± 3241. Gu W and Roeder RG. (1997). Cell, 90, 595 ± 606. Ben-DavidY,GiddensEB,LetwinKandBernsteinA. Gu W, Shi XL and Roeder RG. (1997). Nature, 387, 819 ± (1991). Genes Dev., 5, 908 ± 918. 823. Bertolotti A, Lutz Y, Heard DJ, Chambon P and Tora L. Ichikawa H, Shimizu K, Hayashi Y and Ohki M. (1994). (1996). EMBO J., 15, 5022 ± 5031. Cancer Res., 54, 2865 ± 2868. Borrow J, Stanton Jr VP, Andresen JM, Becher R, Behm Immanuel D, Zinszner H and Ron D. (1995). Mol. Cell Biol., FG, Chaganti RS, Civin CI, Disteche C, Dube I, Frischauf 15, 4562 ± 4571. AM, Horsman D, Mitelman F, Volinia S, Watmore AE Jeon IS, Davis JN, Braun BS, Sublett JE, Roussel MF, and Housman DE. (1996). Nat. Genet., 14, 33 ± 41. Denny CT and Shapiro DN. (1995). Oncogene, 10, 1229 ± Brown AD, Lopez-Terrada D, Denny C and Lee KA. (1995). 1234. Oncogene, 10, 1749 ± 1756. Kaneko Y, Yoshida K, Handa M, Toyoda Y, Nishihira H, Burd CG and Dreyfuss G. (1994). Science, 265, 615 ± 621. Tanaka Y, Sasaki Y, Ishida S, Higashino F and Fujinaga Chrivia JC, Kwok RP, Lamb N, Hagiwara M, Montminy K. (1996). Genes Chromosomes Cancer, 15, 115 ± 121. MR and Goodman RH. (1993). Nature, 365, 855 ± 859. Kastan MB, Zhan Q, el-Deiry WS, Carrier F, Jacks T, Walsh Crozat A, Aman P, Mandahl N and Ron D. (1993). Nature, WV, Plunkett BS, Vogelstein B and Fornace Jr AJ. (1992). 363, 640 ± 644. Cell, 71, 587 ± 597. DelattreO,ZucmanJ,PlougastelB,DesmazeC,MelotT, Kern SE, Pietenpol JA, Thiagalingam S, Seymour A, Kinzler PeterM,KovarH,JoubertI,deJongP,RouleauG, KW and Vogelstein B. (1992). Science, 256, 827 ± 830. Aurias A and Thomas G. (1992). Nature, 359, 162 ± 165. Kung AL, Rebel VI, Bronson RT, Ch'ng LE, Sie CA, Epstein AL, Martin AO and Kempson R. (1984). Cancer Livingston DM and Yao TP. (2000). Genes Dev., 14, 272 ± Res., 44, 1265 ± 1274. 277. Fujimura Y, Ohno T, Siddique H, Lee L, Rao VN and Reddy ES. (1996). Oncogene, 12, 159 ± 167.

Oncogene EWS-ATF-1 binds to CBP and inhibits p53 transactivation Y Fujimura et al 6659 Labelle Y, Zucman J, Stenman G, Kindblom LG, Knight J, Prasad DD, Ouchida M, Lee L, Rao VN and Reddy ES. Turc-Carel C, Dockhorn-Dworniczak B, Mandahl N, (1994). Oncogene, 9, 3717 ± 3729. Desmaze C and Peter M. (1995). Hum. Mol. Genet., 4, Rabbitts TH, Forster A, Larson R and Nathan P. (1993). 2219 ± 2226. Nat, Genet., 4, 175 ± 180. Levine AJ. (1997). Cell, 88, 323 ± 331. Rao VN, Papas TS and Reddy ES. (1987). Science, 237, Lill NL, Grossman SR, Ginsberg D, DeCaprio J and 635 ± 639 Livingston DM. (1997). Nature, 387, 823 ± 827. Reddy ES, Rao VN, Papas TS. (1987). Proc. Natl. Acad. Sci. May WA, Lessnick SL, Braun BS, Klemsz M, Lewis BC, USA, 84, 6131 ± 6135 Lunsford LB, Hromas R and Denny CT. (1993). Mol. Cell Satake N, Ishida Y, Otoh Y, Hinohara S, Kobayashi H, Biol., 13, 7393 ± 7398. Sakashita A, Maseki N and Kaneko Y. (1997). Genes Miller RW and Rubinstein JH. (1995). Am.J.Med.Genet., Chromosomes Cancer, 20, 60 ± 63. 56, 112 ± 115. Shimomura A, Ogawa Y, Kitani T, Fujisawa H and Ohno T, Ouchida M, Lee L, Gatalica Z, Rao VN and Reddy Hagiwara M. (1996). J. Biol. Chem., 271, 17957 ± 17960. ES. (1994). Oncogene, 9, 3087 ± 3097. Somasundaram K and El-Deiry WS. (1997). Oncogene, 14, Ohno T, Rao VN and Reddy ES. (1993). Cancer Res., 53, 1047 ± 1057. 5859 ± 5863. Sorensen PH, Lessnick SL, Lopez-Terrada D, Liu XF, OuchidaM,OhnoT,FujimuraY,RaoVNandReddyES. Triche TJ and Denny CT. (1994). Nat. Genet., 6, 146 ± 151. (1995). Oncogene, 11, 1049 ± 1054. Stolow DT and Haynes SR. (1995). Nucleic Acids Res., 23, Panagopoulos I, Aman P, Fioretos T, Hoglund M, 835 ± 843. Johansson B, Mandahl N, Heim S, Behrendtz M and Urano F, Umezawa A, Hong W, Kikuchi H and Hata J. Mitelman F. (1994). Genes Chromosomes Cancer, 11, (1996). Biochem. Biophys. Res. Commun., 219, 608 ± 612. 256 ± 262. Van Orden K, Giebler HA, Lemasson I, Gonzales M and Panagopoulos I, Hoglund M, Mertens F, Mandahl N, Nyborg JK. (1999). J. Biol. Chem., 274, 26321 ± 26328. Mitelman F and Aman P. (1996). Oncogene, 12, 489 ± 494. Yi H, Fujimura Y, Ouchida M, Prasad DD, Rao VN and Peter M, Couturier J, Pacquement H, Michon J, Thomas G, Reddy ES. (1997). Oncogene, 14, 1259 ± 1268. Magdelenat H and Delattre O. (1997). Oncogene, 14, Zucman J, Delattre O, Desmaze C, Epstein AL, Stenman G, 1159 ± 1164. Speleman F, Fletchers CD, Aurias A and Thomas G. Prasad DD, Rao VN and Reddy ES. (1992). Cancer Res., 52, (1993). Nat. Genet., 4, 341 ± 345. 5833 ± 5837

Oncogene