Research Article 3855 pRb, Myc and are critically involved in SV40 large T antigen repression of PDGF β-receptor transcription

Hidetaka Uramoto, Anders Hackzell*, Daniel Wetterskog*, Andrea Ballági‡, Hiroto Izumi§ and Keiko Funa¶ Department of Cell Biology, Institute of Anatomy and Cell Biology, Göteborg University, Box 420, SE-405 30 Gothenburg, Sweden *These authors contributed equally to this work ‡Present address: IDEXX Scandinavia AB, Storrymningsvägen 5, SE-748 30 Österbybruk, Sweden §Present address: Department of Molecular Biology, University of Occupational and Environmental Health, School of Medicine, Kitakyushu, 807-8555, Japan ¶Author for correspondence (e-mail: [email protected])

Accepted 17 March 2004 Journal of Cell Science 117, 3855-3865 Published by The Company of Biologists 2004 doi:10.1242/jcs.01228

Summary The expression of the PDGF β-receptor is tightly regulated Furthermore, p53 was found to increase the promoter during a normal cell cycle. c-Myc and p73α repress activity mainly via the upstream Sp1 binding sites together transcription of the receptor through interaction with NF- with the CCAAT motif in the NIH 3T3 cells. This was Y. In ST15A cells which stably express the temperature- confirmed by Schneider’s Drosophila line (SL2) cells sensitive SV40 large T antigen (LT) the receptor expression deficient in both endogenous NF-Y and Sp1. Chromatin and ligand binding decreased under the permissive immunoprecipitation using ST15A cells revealed that both condition. Transient expression of the LT, but not small t, LT and p53 bound the PDGF β-receptor promoter and the decreased the endogenous receptor expression at mRNA binding of p53 diminished when LT was expressed in the and protein levels in NIH3T3 cells but not in the myc-null permissive condition. However, LT binds the promoter in HO15.19 cells. The wild-type LT, but not the various pRb the absence of pRb and p53 in Saos-2 cells stably expressing or p53 binding defective LT mutants, represses the PDGF LT. These results suggest that LT binds the promoter and β-receptor promoter activity. Moreover, the inability of the interferes with NF-Y and Sp1 to repress it in the presence LT-mediated repression in the myc-null cells, the Rb-null of Myc, pRb and p53. 3T3 cells, and the Saos-2 cells lacking pRb and p53, indicates that Myc, pRb and p53 are all necessary elements. PDGF β-receptor promoter-luciferase assays revealed Key words: Platelet-derived growth factor, β-receptor, SV40LT, pRb, that the CCAAT motif is important for the repression. c-Myc, p53

Introduction NF-Y transcription factor. NF-Y binds the consensus CCAAT Platelet-derived growth factor (PDGF) is a serum mitogen motif in the PDGF β-receptor promoter, and controls its basal consisting of four isoforms: A, B, C and D. These isoforms activity together with Sp1 that binds in close proximity exist as dimers: AA, BB, AB, CC and DD, and bind two (Ballagi et al., 1995; Ishisaki et al., 1997; Molander et al., distinct PDGF receptors, α and β, with different affinities, 2001). NF-Y consists of NF-YA, NF-YB and NF-YC subunits thereby exerting their effects on target cells (Bergsten et al., that are all necessary for DNA binding (Kim et al., 1996). c- 2001; Gilbertson et al., 2001; Heldin and Westermark, 1990). Myc and p73α bind the C-terminal HAP domain of NF-YB Upon PDGF stimulation of cells, several intracellular enzymes and NF-YC and inactivate the transcription. become activated to propagate cascades of different signalling Previously it has been shown that the large T antigen (LT) pathways, controlling various cellular functions such as and small t antigen of simian virus 40 (SV40) down-regulated growth, differentiation, chemotaxis and survival (Heldin and the expression of both types of PDGF receptor in various kinds Westermark, 1999). In normal cells, the expression of PDGF of fibroblasts (Wang et al., 1996). They mainly studied the β-receptor changes during the cell cycle, i.e. decreasing PDGF α-receptor downregulation at the mRNA level, which following the G0/G1 exit after growth factor stimulation to was shown to occur independently of p53 and Rb. However, prevent excessive growth signals. This cell-cycle dependent the mechanism behind the PDGF β-receptor repression by LT change seems to be regulated by several key molecules and whether c-Myc, p73α, or as yet undiscovered molecules encoded by proto- and suppressor genes. are involved in the mechanism, still remains unresolved. It has previously been reported that c-Myc represses PDGF SV40 was first isolated in Rhesus monkey cells used to grow β-receptor expression at the transcriptional level through its the active polio vaccine developed in the late 1950s (Hilleman, interaction with NF-Y (Izumi et al., 2001; Oster et al., 2000). 1998). The virus belongs to the polyoma virus family and has We have recently demonstrated that c-Myc interacts with p73 been extensively studied as a model to investigate the control (Uramoto et al., 2002), which, independently of c-Myc, also mechanism of cell growth (for a review, see Weiss et al., 1998). represses transcription of PDGF β-receptor (Hackzell et al., In addition, the involvement of SV40 in human tumours was 2002). Similar mechanisms seem to prevail in the repression recently verified by the identification of the viral sequence in exerted by c-Myc and p73α through their interaction with the certain tumour tissues (Klein et al., 2002), emphasising the 3856 Journal of Cell Science 117 (17) importance of understanding the molecular mechanism of pRb-binding motif (C105G), the mutant ∆434-444, lacking the p53- SV40 in tumour pathogenesis. The LT, a 90-kDa binding motif and the H42Q with an amino acid substitution at phosphoprotein, is the only essential for SV40 position 42, inactivating the J domain function of LT in pSG5 replication. expression vector, were provided by Dr DeCaprio (Chao et al., 2000). The LT was shown to be a molecular chaperone, promoting The J domain is thought to assist pRb binding. The expression levels the proper folding of proteins and preventing protein of all the constructs were examined and the expression of LT in two different vectors was judged to be similar by immunoblotting 3T3 aggregation during cellular stress (Hartl and Martin, 1995). cell lysates after transfection. For functional promoter assays, the The chaperone activity is necessary for viral replication, promoter region of mouse PDGF β-receptor was inserted in a transcriptional control, virion assembly, and transformation luciferase expression vector, pGL3 (Promega) (Ishisaki et al., 1997). (Sullivan and Pipas, 2002). The LT has several functions, The dominant negative mouse NF-YA (DNNFYA) was provided by including control of the activities of ATPase, DNA-binding, Dr Mantovani (Mantovani et al., 1994), and Sp1 cDNA was provided oligomerisation and DNA helicase. Following infection, LT by Dr Tjian (Howard Hugh Medical Institute, University of affects host gene expression and growth control by binding to California, CA). The pPacSP1 and pPacNF-Y expression vectors were a wide variety of transcription factors that are important for provided by Dr Suske (University of Marburg, Germany) and Dr both replication and cell cycle regulation including tumour Zetterberg (Göteborg University, Sweden). suppressors, p53 and family proteins (Moens et al., 1997). These interactions are thought to play crucial roles Receptor binding assay in the pathogenesis and progression of tumours. To estimate the relative amounts of functional PDGF receptor in the The (pRb) prevents cell cycle presence or absence of LT expression, we used the ST15A cell line. progression by binding and sequestering E2F (Classon and The amount of PDGF receptor binding was determined by analysing Harlow, 2002). It has been shown that the E2F–Rb complex serial dilutions of cold PDGF ligand with regard to its ability to can also directly repress the promoter of certain cell cycle compete with [125I]PDGF ligand for binding to the cells. Cells were genes by recruiting histone deacetylase (Magnaghi-Jaulin et grown on 24-well plates (Becton Dickinson) at 33°C, or first at 33°C al., 1998). The N-terminal LT binds the E2F-Rb complex and and then at 39°C for at least 5 days before being used for the assay. For the β-receptor assay the α-receptor was depleted by a 60-minute dissociates it and thereby preventing the repression (Laufen et ° al., 1999). One of the activated downstream target genes is c- preincubation with 50 ng/ml PDGF-AA at 37 C. Cell cultures were washed once in binding buffer (phosphate buffered saline, PBS, myc (Batsche et al., 1994), which is necessary to induce and containing 1 mg/ml bovine serum albumin, BSA, 0.9 mM CaCl2, and maintain proliferative cell states (for a review, see Facchini and 0.5 mM MgCl2) and then incubated at 0°C for 2 hours in 200 ml Penn, 1998). In this manner, SV40 keeps host cells in a binding buffer containing various dilutions of PDGF-BB. The cells proliferating state to use them for its own replication. Another were washed in binding buffer before addition of the labelled ligand critical control mechanism of SV40 on host replication is (0.5-2 ng containing 15,000-30,000 cpm). After incubation at 0°C for through the interaction of the C-terminal LT with the p53 1 hour, the cells were washed with binding buffer, then lysed in 200 tumour suppressor (Kierstead and Tevethia, 1993). The aim of ml of 20 mM Tris-HCl, pH 7.5, 1% Triton X-100 and 10% glycerol this study is to clarify the mechanisms that LT uses to repress at room temperature for 20 minutes. The amount of solubilised [125I] γ PDGF β-receptor expression by using normal 3T3 fibroblasts radioactivity was measured in a -counter. and cell lines deficient in c-Myc, pRb, p53 or NF-Y and Sp1. Immunohistochemistry For immunohistochemical staining of PDGF β-receptor, ST15A cells Materials and Methods cultured on chamber slides (Nunc) at 33°C or 39°C, as described Cell culture above, were fixed in cold 4% paraformaldehyde in PBS. They were Murine NIH3T3 fibroblasts, mouse Rb null 3T3 fibroblasts (Classon stained with the anti-rat PDGF β-receptor (Ab-1; Science) and Harlow, 2002), human Saos-2, pRb and p53 null cell line using an ABC immunoperoxidase method essentially as described (Chandar et al., 1992), and rat HO15.19, c-myc null cell line (Mateyak (Funa et al., 1996). The fixed cells were incubated with 20% normal et al., 1997), as well as rat ST15A cell line (derived from primary goat serum, 2% normal rat serum and 2% BSA in PBS for 10 minutes. rat cerebellar neurons immortalised by stable incorporation of Incubation with the primary antibody diluted 1:500 in PBS was temperature sensitive LT) (Frederiksen et al., 1988) were used in this performed for 1 hour at room temperature. Following incubation with study. The Saos-2-LT cell line was established by stable transfection the biotinylated anti-rabbit Ig, the immunocomplex was coupled to of Saos-2 cells with pcDNA3 containing LT by selection with G418 ABC Elite complex (Vector) which was visualised with 3-amino-9- (Calbiochem). They were maintained in Dulbecco’s modified Eagle’s ethylcarbazole as the chromogen and 0.02% H2O2 as substrate. The medium (DMEM) containing 10% fetal calf serum (FCS), 100 slides were counterstained with haematoxylin and mounted in units/ml penicillin, and 60 µg/ml streptomycin in a 5% CO2 glycerol-gelatin. A negative control for each slide was performed by atmosphere at 33°C. To inactivate LT in ST15A cells, the temperature substituting the primary antibody with 1% BSA in PBS. Cells in the incubator was raised to 39°C for 5-7 days until cells were used. transfected with a mouse PDGF β-receptor expression vector or the The SL-2 cell line (Börestrom et al., 2003) was cultured at 25-28°C empty vector was used as positive or negative controls, respectively. in Schneider’s Drosophila medium (Gibco, Life Technologies). Transient plasmid transfection and immunoblotting Plasmid constructs NIH3T3 cells and HO15.19 cells in 10 cm dishes were transiently The cDNAs for SV40LT wild type (WT) and the mutant K1 (E107K) transfected with 12 µg of expression plasmids using 60 µl Fugene containing an amino acid substitution at position 107 in the pRb- (Roche). Cells were transfected with LT, small t or expression vector, binding motif in pCDNA3 were provided by Dr Livingston (Dana- and cultured with 10% serum. After 48 hours cells were collected and Faber Cancer Institute, Boston, MA). The cDNAs for the mutant lysed with 50 mM Tris-HCl, pH 8.0, 5 mM EDTA, 150 mM NaCl, C105G, containing an amino acid substitution at position 105 in the 1% Triton-X100, 0.05% sodium dodecyl sulfate (SDS), and 1 mM Mechanism for SV40LT downregulation of PDGF β-receptor 3857 phenylmethylsulfonyl fluoride (PMSF). Cell lysate, 200 µg, was X, once with high salt Buffer X (500 mM NaCl), once with LiCl buffer separated by SDS-polyacrylamide gel electrophoresis (PAGE) (10 mM Tris, 1 mM EDTA, 0.25 M LiCl, 1% NP-40, 1% sodium and transferred onto immobilon-P membrane filters. After deoxycholate, pH 8.1) and twice with TE (10 mM Tris, 1 mM EDTA, immunoblotting with anti-PDGF β-receptor (958; Santa Cruz), anti- pH 8.0). Immune complexes were eluted twice with 250 µl of elution LT (Pab 108; Santa Cruz), or anti-Sp1 (1C6, Santa Cruz) antibodies, buffer (0.1 M NaHCO3, 1% SDS). To reverse protein-DNA cross- the membranes were developed by the enhanced chemiluminescence linking, eluted samples were incubated with 0.2 M NaCl for 4-5 hours (ECL) protocol (Amersham). Wild type and Rb–/– 3T3 cells were at 65°C. Samples were digested with Proteinase K (0.04 mg/ml) for serum-depleted in DMEM with 0.5% FCS for 48 hours before being 2 hours at 45°C and then with RNase A (0.02 mg/ml) for 30 minutes stimulated with 10% FCS for the indicated time periods. Cells were at 37°C. DNA was purified with phenol:chloroform followed by harvested and immunoblotted with anti-PDGF β-receptor, anti-c-Myc ethanol precipitation. Purified DNA was resuspended in 10 µl H2O. (9E10, Santa Cruz), and anti-actin (AC-40, Sigma) antibodies. Total Aliquots of 2 µl serial dilution were analysed by PCR with the cell lysate from ST15A cells was extracted from cells cultured at 33°C appropriate primer pairs. The proximal PDGF β-receptor promoter and 39°C, respectively, and immunoblotted with anti-LT and p53 (–229/+273) primers were 5′-GGGAGGGAGCAGGAGGGAAAGG- (DO-1, Santa Cruz) antibodies as described above. AG-3′ and 5′-GAATCAGGGGAATGGAGAGGGTGC-3′. The distal PDGF β-receptor promoter (–1764/–1424) primers were 5′-CCTC- AGGTAGTCATGGTCTC-3′ and 5′-TGCCAGACCACAGGATAA- Reverse Transcription (RT)-PCR TG-3′. Amplification was performed for a predetermined optimal 3T3 cells were seeded in 6-cm dishes at a density of 1.1×105 number of cycles. PCR products were separated by electrophoresis on cells/dish. After 24 hours, half of the cells of each cell type were 2% agarose gels, which were stained with ethidium bromide. transfected with 2 µg of LT expression plasmid/dish and the remaining cells were transfected with 2 µg of the vector plasmid alone. The cells were kept in 10% FCS, and RNA was extracted essentially as Results described (Chomczynski and Sacchi, 1987) at 0, 2, 4 and 8 hours after Binding and expression of PDGFβ-receptor on ST15A transfection. 3 µg of total RNA from each time point was transcribed cells into cDNA with random primers and Moloney murine leukaemia virus In order to examine the functionality of the receptors, we (Invitrogen) according to the manufacturer’s 125 protocol. PCR was performed by using Taq polymerase (Fermentas) performed binding assays with [ I]PDGF ligands. At a and a programmable thermal block (LabLine). PDGF β-receptor and permissive temperature, cells continue to proliferate and as the β-actin primers were used as described previously (Ballági-Pordany temperature increases and LT becomes inactivated, cells et al., 1991). PCR products were identified by dot-blot hybridisation, undergo normal programmed differentiation (Frederiksen et and analysed on 1.5% TAE-agarose gel stained with ethidium bromide al., 1988). PDGF-AA binding was detectable only on the for comparison. ST15A cells at the non-permissive temperature of 39°C. However, the binding could be only partially blocked by the Promoter reporter assay increasing amount of unlabelled ligand (data not shown). After blocking the α-receptor, PDGF-BB binds only the PDGF β- Cells were seeded in 12-well plates at a density of 2×104 cells/well receptor with high affinity, which demonstrates that functional in 10% FCS. The following day, the cells were transiently transfected β ° with 0.2 µg reporter plasmid, 0.25 to 0.5 µg of expression plasmid. -receptor expression was induced at 39 C (Fig. 1A). This Each expression plasmid and reporter plasmid was standardised binding was completely blocked by the unlabelled ligand. We individually using a molar ratio of 1:1 and the total amount of DNA could not perform a Scatchard analysis because it was difficult per well was adjusted to 1.0 µg by addition of mock DNA plasmid. to know the exact number of cells after the differentiation Sp1 was used as a positive control. After 48 hours cells were lysed process. There was no difference in the receptor expression of with 100 µl/well of reporter lysis buffer (Promega). Luciferase activity LT-transfected 3T3 cells at 33°C or 39°C, confirming that the was measured according to the vendor’s instruction (Promega). temperature change alone does not affect the receptor β Standardisation by co-transfection with -galactosidase reporter expression (data not shown). Immunohistochemistry on plasmid was avoided because LT influences expression of most ST15A cells cultured at 33°C showed a dense population of promoters of plasmids used as internal controls (Moens et al., 2001). round cells that stained negative for PDGF β-receptor. Results shown were thus normalised to protein concentration and ° were representative of at least three independent experiments. Following the temperature shift to 39 C, cells became sparse and slender, extending processes, and the majority of cells stained positive for the receptor (Fig. 1B). Chromatin immunoprecipitation (Chip) assay Chip assay was performed as previously described (Uramoto et al., β 2002). Briefly, protein and DNA were cross-linked by incubating Downregulation of PDGF -receptor expression by ST15A cells with formaldehyde at a final concentration of 1% for 10 SV40LT transfection minutes at room temperature. Cells were then lysed in Buffer X (50 We examined the level of PDGF β-receptor protein expression mM Tris-HCl at pH 8.0, 1 mM EDTA, 120 mM NaCl, 0.5% NP-40, by immunoblotting 3T3 cells following transfection of LT, 10% glycerol, and 1 mM PMSF) for 15 minutes on ice. The lysate small t, or the vector alone. Cells transfected with LT alone was sonicated and soluble chromatin was pre-cleared by addition of expressed a significantly lower level of PDGF β-receptor, 10 mg Protein A-sepharose. An aliquot of pre-cleared chromatin was whereas cells transfected with small t showed only a slight removed and used in the subsequent PCR analysis. The remainder of decrease in the expression level (Fig. 2A). To see whether c- the chromatin was diluted with Buffer X. Then protein-DNA was incubated with 2 µg LT antibody, p53 antibody (DO-1), NF-YB (FL- Myc is directly involved in the LT-induced repression of PDGF β –/– 207X, Santa Cruz), normal rabbit serum or mouse IgG in a final -receptor expression, we examined the c-myc HO15.19 volume of 800 µl overnight at 4°C. Immune complexes were collected fibroblast cell line. PDGF β-receptor expression was high in by incubation with 15 µl Protein G-agarose (Santa Cruz) for 1 hour this cell line (Oster et al., 2000) and transfection with LT did at 4°C. Protein G-agarose pellets were washed once with 1 ml Buffer not alter the expression level (Fig. 2B). We examined the 3858 Journal of Cell Science 117 (17)

Fig. 1. (A) Downregulation of PDGF β-receptor by LT on ST15A cells, containing a temperature-sensitive mutant of LT cultured at 33°C or 39°C. The amount of PDGF receptor binding was analysed by addition of serial dilutions of cold PDGF ligand to compete with [125I]PDGF ligand for cell binding. The α-receptor was depleted by preincubation of cells with PDGF-AA at 37°C. (B) Photomicrographs of ST15A cells immunostained with a PDGF β-receptor antibody after culture at 33°C (left) and after differentiation at 39°C (right).

β kinetics of the mRNA expression in NIH3T3 cell line at 0, 2, Fig. 2. (A) Immunoblotting with PDGF -receptor in 3T3 fibroblasts 4 and 8 hours after transfection with LT or a control vector by transiently transfected with LT, small t (st) or expression vector using Fugene cultured with 10% FCS. Lane 1: non-transfected cells, lane RT-PCR. PDGF β-receptor mRNA, as judged by PCR products µ µ µ β 2: 2 g pcDNA3 vector, lane 3: 1 g LT and 1 g pcDNA3 vector, and in comparison with -actin mRNA, gradually decreased lane 4: 1 µg st and 1 µg pcDNA3 vector. (B) No alteration of PDGF following the transfection of LT but not of a control vector (Fig. β-receptor expression in myc-null HO15.19 cells was shown by 2C). immunoblotting after transfection with LT, or pcDNA3 vector. Immunoblots with anti-LT and anti-actin antibodies are shown. (C) Expression of PDGF β-receptor mRNA at 0, 2, 4 and 8 hours The CCAAT motif is important for SV40LT repression on after transfection of LT or pcDNA3 vector in 3T3 cells. The RT-PCR PDGF β-receptor promoter activity products for PDGF β-receptor, and actin were run on an agarose gel β We examined whether the LT downregulation of PDGF β- and are also shown as relative ratios of PDGF -receptor to actin receptor stems from the effect on the transcriptional activity, obtained by densitometric analysis. and if so, which area of the promoter is responsible for the repression (Fig. 3A). The SacI/SacI promoter reporter construct (pGL3–1481; –1481 to +18) contains a CCAAT transfection of a LT expression vector decreased the luciferase motif located –62 bp upstream of the first initiation site and activity of the SacI/SacI PDGF β-receptor promoter in a also the GC boxes that were shown to bind Sp1 (Molander et concentration-dependent manner when compared with that of al., 2001) and the SacI/SacI mCCAAT with a in the a control vector alone. Small t decreased the activity only CCAAT motif (Ishisaki et al., 1997). The MluI/SacI (–122 to slightly as judged by repeated experiments. LT also decreased +18) also contains the CCAAT motif and the GC boxes, the the activity of the HindIII/SacI as well as the MluI/SacI HindIII/SacI (–69 to +18) contains the CCAAT motif without promoter constructs, indicating that the CCAAT motif may be a GC box (Ballagi et al., 1995). As shown in Fig. 3B, co- needed for the repression to occur (Fig. 3C). Sp1 used as a Mechanism for SV40LT downregulation of PDGF β-receptor 3859

Fig. 3. Effects of PDGF β-receptor promoter activity by LT in 3T3 cells. (A) Schematic illustration of the promoter with relevant restriction enzyme sites. The GC- rich area (white box) and CCAAT motif (black box) are shown. (B) Relative luciferase activity of the SacI/SacI- reporter construct containing CCAAT motif co- transfected with st, LT, and vector alone (Mock). Basic vector without promoter was used as a basal control. Values represent mean promoter activity, and error bars indicate standard deviation of triplicate samples. A representative result of three repeated experiments is shown. The expression levels of LT and st are shown by immunoblotting after transfection. (C) The HindIII/SacI- reporter construct containing CCAAT motif, the MluI/SacI-reporter construct containing CCAAT motif and GC box or the SacI/SacI-construct was co-transfected with Sp1, LT or expression vector alone (mock). The expression of Sp1 is shown by immunoblotting. (D) The SacI/SacI construct was co-transfected with Sp1, LT or vector in the presence of DNNFYA and compared with the effect of LT in the absence of DNNFYA. The expression of NF-YA and DNNFYA is shown by immunoblotting. positive control showed a strong increase when the major Sp1 binding GC-rich sequence located between the MluI and HindIII sites was present in the promoter reporter. In order to determine whether the CCAAT- binding transcription factor NF-Y is required for the LT mediated repression, we co-transfected the DNA- binding defective and thus dominant negative NF- YA (DNNFYA) (Mantovani et al., 1994) together with the LT expression vector and the SacI/SacI promoter luciferase plasmid (Fig. 3D). As expected, co-expression of 0.5 µg DNNFYA decreased the PDGF β-receptor promoter activity to 51% of that obtained with vector alone. Co-expression of LT decreased the promoter activity obtained by trans- fection with control vector alone by 62%, whereas a promoter decrease of only 32% was seen when 0.5 µg DNNFYA construct was additionally transfected together with LT or the control vector. Sp1 used as a positive control increased the promoter activity about twofold even in the presence of DNNF-Y when compared with DNNF-Y without Sp1 co- transfection (Molander et al., 2001). However, the increase was smaller than that obtained when Sp1 was transfected alone without DNNF-Y, which resulted in around a fourfold increase (Fig. 3C).

either one of the LXCXE motif mutants, K1 or C105G, caused Effects of various SV40LT mutants on the PDGF β- more than a twofold increase of PDGF β-receptor promoter receptor promoter activity in 3T3 cells activity compared with control vector. The J domain- Various LT mutants including the three pRb-binding defective inactivating mutant H42Q neither increased nor decreased the mutants were examined in order to determine the domains of activity. The p53 binding mutant, ∆434-444, did not decrease LT necessary to repress PDGF β-receptor promoter activity in the promoter activity, indicating that the p53-binding domain 3T3 cells. LT contains a four-helix bundle, residues from is also needed for the repression of the promoter. These helices 2 and 4 called the J domain and a loop containing the findings suggest that the binding of LT to both pRb and p53 LXCXE motif. The N-terminal J domain and this loop interact plays an important role in the repression. with the highly conserved A and B domains of pRb (Sullivan In contrast to the wild-type LT, co-transfection of the K1 and Pipas, 2002). As shown in Fig. 4A, co-transfection of mutant increased the activity, in a concentration-dependent 3860 Journal of Cell Science 117 (17) manner, of both the SacI/SacI and the shorter HindIII/SacI expected, LT could not alter the PDGF β-receptor promoter constructs, as shown in Fig. 4B. Furthermore, a mutation in the activity in Saos-2 cells, whereas Sp1 used as a positive control CCAAT motif completely abolished the activation of the activated the promoter. We also used the c-myc–/– HO15.19 cell SacI/SacI construct by K1 (Fig. 4C). Co-transfection of line to see whether c-Myc is directly involved in the LT- DNNF-YA also abolished the activation of the wild type induced repression of the PDGF β-receptor promoter activity SacI/SacI by K1. These results indicate that the K1 activates (Fig. 5B). Transfection of LT did not repress, but almost the promoter through the NF-Y-binding site. doubled the activity of PDGF β-receptor promoter. Sp1 strongly activated the promoter activity in this cell line. Furthermore, we examined the promoter activity in the Rb–/– pRb, Myc, and p53 are required for repressing PDGFβ- 3T3 cell line, which also failed to respond to LT. The receptor promoter activity by SV40LT expression level of PDGF β-receptor was examined in this cell In order to see whether the major targets of LT, pRb, and p53 line at 0, 4, 8, 12, 24, 48 hours after serum stimulation are directly involved in the LT induced repression of PDGF β- following 48 hours’ serum starvation and was compared with receptor promoter, we examined the promoter activity in the that of normal 3T3 cells. Expression of the receptor was higher Saos-2 cell line lacking both pRb and p53 (Fig. 5A). As in Rb–/– cells, and no change was seen during the observed time period as judged by immunoblotting. c-Myc expression increased at 4 hours after serum stimulation even in the absence of pRb as reported elsewhere (Herrera et al., 1996), which was followed by a much slower decrease compared to the rapid decrease seen in normal 3T3 cells. This suggests that downregulation of the PDGF β-receptor, as well as that of c-Myc, is impaired in the absence of pRb.

p53 increases the PDGF β-receptor promoter activity through NF-Y and Sp1 binding motifs We showed that the p53-binding defective LT was unable to repress the PDGF β- receptor promoter activity. We therefore tested whether p53 has any activity at all on the promoter. In a previous study, we showed that p73α bound NF-Y and repressed the promoter activity, but p53 did not (Hackzell et al., 2002). Careful evaluation of the effect of p53 on various PDGF β-receptor promoter constructs in 3T3 cells showed a small but significant increase in the activity (Fig. 6A). The 3′SacI/ApaI construct, consisting of a 142-bp sequence located downstream of the initiation site, did not respond to p53 co-

Fig. 4. Binding to both p53 and pRb is necessary for LT repression on PDGF β-promoter activity and the LXCXE Rb-binding mutants enhance the promoter activity via CCAAT motif. (A) 3T3 cells were co-transfected with the SacI/SacI construct together with H42Q, ∆434-444, C105G, LTK1 or vector. The expression of all the LT vectors used in the assays is shown by immunoblotting. (B) 3T3 cells were co- transfected with the HindIII/SacI, SacI/SacI or basic reporter construct together with LT, K1, or vector alone. (C) 3T3 cells were co-transfected with the SacI/SacI construct containing wild type or mutated CCAAT motif together with vector alone, LTK1, DNNFYA or DNNFYA and LTK1. Luciferase activity was measured as described above. Mechanism for SV40LT downregulation of PDGF β-receptor 3861 transfection (data not shown). The HindIII/SacI construct containing the CCAAT motif showed a 50% increase of basic activity when p53 was co- transfected. The SacI/SacI constructs containing both the CCAAT- and Sp1-binding motifs showed a 70% increase of the activity by p53. However, the mutation in the CCAAT motif in the SacI/SacI still maintained 50% of the activation by p53 expression. These results indicate that the promoter activation by p53 is likely to be dependent on binding of both Sp1 and NF-Y. In order to analyse more exactly the effect of p53, we used the SL-2 cell line lacking both endogenous Sp1 and NF-Y (Courey and Tjian, 1998; Magana et al., 2000; Börestam et al., 2003). In this cell line, the SacI/SacI PDGF β-promoter showed very low activity (Fig. 6B). Co-transfection of Sp1 alone yielded a stronger activation than NF-Y alone, and Sp1 and NF-Y together yielded a synergistic activation of the promoter. Addition of p53 yielded an additive effect on NF-Y transfection but a small synergism was observed with Sp1 on the promoter activation. All these factors together brought about a significant activation due to the synergisms between Sp1 and NF-Y as well as of p53 and Sp1. It is thus likely that p53 activates the promoter, chiefly through the Sp1-binding site.

SV40LT and p53 bind the proximal PDGF β- receptor promoter in vivo In order to see whether LT binds the PDGF β-receptor promoter in vivo, Chip assays were performed. ST15A cells cultured at both permissive and non-permissive temperatures were used. PCR amplification of the proximal PDGF β-receptor promoter was carried out with DNA extracted from the immunocomplex precipitated by an anti-SV40LT antibody or anti-p53 antibody. This promoter region contains the Sp1-binding sites, the CCAAT motif, and the initiation site. As a control, we included about 1.5 kbp of upstream sequence of the same promoter, lacking these consensus motifs. Fig. 7A shows that the PDGF β-receptor promoter sequence was significantly enriched in the complex obtained with the anti-LT antibody in ST15A cells at 33°C when LT is expressed in the cells. No PDGF β- receptor promoter sequence was detected in the

Fig. 5. LT is unable to repress PDGF β-receptor promoter activity in (A) Saos-2 cells, (B) HO15.19 cells and (C) Rb null cells. Transfection of Sp1 was used as a positive control. Cells were co-transfected with the SacI/SacI construct together with LT or vector alone examined in 2 different vectors. Expression of LT is shown by immunoblotting. Luciferase activity was measured as described above. (D) Rb null cells express a stable level of PDGF β-receptor during a cell cycle. Cells were harvested at indicated time points after serum stimulation of starved cells and used for immunoblots with antibodies against PDGF β-receptor, c-Myc and actin. The lowest panel of 3T3 cells shows non-specific bands. 3862 Journal of Cell Science 117 (17) immunocomplex obtained at 39°C or when the antibody was substituted with mouse IgG (Fig. 7B). When the p53-antibody was used for immunoprecipitation, the promoter sequence was amplified. Binding of p53 to the promoter was clearly stronger at 39°C than at 33°C (Fig. 7A,B). The input cell lysate as well as the precipitated DNA were also analysed at one- third dilutions. The amount of total protein in ST15A cells at different temperatures was examined by immunoblotting (Fig. 7C). As expected, LT was expressed only at 33°C, and p53 was expressed almost at the same level or slightly stronger at 33°C. In order to see whether LT binds the promoter in the absence of p53 and pRb, we made a Saos-2 cell clone stably expressing LT (Saos-2-LT), and used for Chip assays. In both parental and Saos-2-LT cell lines, the proximal promoter, but not the distal promoter, was precipitated by an NF-YB antibody used as a positive control (Fig. 7D,E). An anti-LT antibody bound the proximal promoter in Saos- 2-LT cells, suggesting that LT can bind the promoter in the absence of pRb and p53. Expression of LT in Saos-2-LT and Saos-2 cells was compared by immunoblotting as shown in Fig. 6. p53 increases PDGF β-receptor promoter activity. (A) 3T3 cells were co- Fig. 7F. transfected with the HindIII/SacI or the SacI/SacI constructs (wild type or CCAAT- mutant) together with p53 or vector alone. The expression of p53 is shown by immunoblotting. (B) SL2 cells were co-transfected with the SacI/SacI construct and Discussion NF-Y, Sp1, NF-Y + Sp1 or vector alone in combination with p53. Luciferase activity We have shown that the effects of LT can be was measured as described above. detected at the protein, mRNA and transcription levels in 3T3 cells. The effects are mainly exerted through the CCAAT motif, but also through 1996), ‘cell growth’ itself might not be a decisive factor for the Sp1 binding area of the PDGF β-receptor promoter. the repression. Moreover, when examined by co-immunoprecipitation, LT It is possible that LT-bound pRb could act as a repressor by does not bind NF-Y, c-Myc, or p73α (Reichelt et al., 1999) recruiting chromatin-remodelling complexes, such as histone (data not shown), suggesting that the repression is not caused deacetylases (HDACs) and the ATP-dependent SWI/SNF by a direct interaction between LT and those transcription (Zhang et al., 2000). This is supported by our results that both factors that have previously been shown to be involved in the of the LXCXE motif-mutated pRb-binding defective LTs repression of the receptor transcription. In 3T3 cells, both markedly increased the promoter activity in NIH3T3 cells. In types of pRb-binding mutants (in the LXCXE motif and the contrast, neither the p53-binding mutant nor the J domain N-terminal J domain) completely abolished the repression of mutant resulted in clear activation. The LXCXE motif and pRb the PDGF β-receptor promoter activity. In addition, Rb–/– cells create the most extensive interaction (Gjoerup et al., 2000; Kim did not respond to LT in the PDGF β-receptor promoter assay, et al., 2001). It should be noted that the activation on the PDGF indicating that pRb is indispensable for the repression. c-Myc β-receptor promoter occurred through the CCAAT motif, the is a key regulator of the PDGF β-receptor expression by site needed for c-Myc- or p73-induced repression. Thus, the repressing the transcriptional activation of NF-Y (Izumi et al., chromatin modification of this promoter area might be the 2001; Oster et al., 2000). In the Rb–/– cell line, the expression key target of LT. However, the clear mechanism of the of c-Myc was shown to increase after serum stimulation but transactivation by the pRb-binding defective LT mutant with a delayed downregulation, as previously reported remains to be elucidated. (Herrera et al., 1996). The Rb–/– cell line expresses a stable Chip assays in ST15A cells with a p53 antibody revealed level of PDGF β-receptor after serum stimulation, whereas in that p53 binds the promoter, and that the binding was stronger the control 3T3 cell line the expression decreases following in the absence of LT at 39°C than at 33°C. Thus, LT may bind the increase of c-Myc. Nevertheless, Myc is also indispensable and sequester p53 from the promoter as the total amount of p53 for the LT-mediated repression, as LT could not repress the in cells rather increased at 33°C, as judged by immunoblotting promoter activity in myc-null cells. Further study is necessary using total cell lysates including nuclear proteins. to determine the relationship between pRb and Myc in the LT Downregulation of p53 mRNA expression at 39°C in this cell repression on the receptor. As Rb–/– cells exhibit a shorter G1 line was also reported previously (Hayes et al., 1991). It is phase and a faster cell cycle than control cells (Herrera et al., notable that p53 binds the promoter, the bound p53 increases Mechanism for SV40LT downregulation of PDGF β-receptor 3863

Fig. 7. LT and p53 bind PDGF β-receptor promoter in vivo in ST15A cells and Saos-2 cells. (A,B) Chromatin was immunoprecipitated from ST15A cells cultured with anti-LT antibody, anti-p53 antibody or mouse IgG at 33°C (A) and 39°C (B). Immunoprecipitated DNA was purified and analysed by PCR together with whole cell lysate using primers specific for the proximal and distal areas of the PDGF β-receptor promoter. (C) Total protein was extracted from ST15A cells cultured under both conditions and immunoblotted with anti-LT, anti-p53, and anti-actin antibodies. (D,E) Chromatin was immunoprecipitated from Saos-2 cells (D), and Saos-2- LT cells (E) stably expressing LT with normal rabbit serum, mouse IgG, anti-NF-YB, and anti-LT antibodies described as above. (F) Expression of LT and actin is shown in Saos- 2 and Saos-2-LT cells by immunoblotting. when LT is absent and the receptor expression is enhanced. In of binding and repressing the promoter, partly through fact, the p53-binding area of LT overlaps with its DNA-binding inhibiting the p53 action. Further characterisation of the LT surface (Li et al., 2003). Although no clear p53 responsive and p53 binding sites on the promoter is underway in our element is found in this promoter area, LT seems to be capable laboratory. 3864 Journal of Cell Science 117 (17)

Chip assays in the Saos-2 cells containing LT, confirmed that Ballági-Pordany, A., Pordany, A. and Funa, K. (1991). Quantitative LT-binding of the promoter is not sufficient for the repression determination of mRNA phenotypes by the polymerase chain reaction. Anal. to occur. The presence of intact pRb and p53 is thus crucial. Biochem. 196, 89-94. Batsche, E., Lipp, M. and Cremisi, C. (1994). Transcriptional repression and The proximal primer set was made so that the PCR product activation in the same cell type of the human c-MYC promoter by the would contain the consensus sequences for Sp1 and NF-Y as retinoblastoma gene protein: antagonisation of both effects by SV40 T well as the initiation site. In addition to binding to pRb and antigen. Oncogene 9, 2235-2243. p53, resulting in the repression of the promoter, LT was also Bergsten, E., Uutela, M., Li, X., Pietras, K., Östman, A., Heldin, C. H., Alitalo, K. and Eriksson, U. (2001). PDGF-D is a specific, found to bind Sp1 (result not shown). It is possible that LT protease-activated ligand for the PDGF beta- receptor. Nature Cell Biol. interferes with the activation of Sp1. As Sp1 interacts and 3, 512-516. activates NF-Y (Liang et al., 2001), LT might negatively affect Börestrom, C., Zetterberg, H., Liff, K. and Rymo, L. (2003). Functional the whole complex including both of the factors. In myc–/– interaction of nuclear factor y and sp1 is required for activation of the cells, co-transfection of Sp1 strongly activated the PDGF β- epstein-barr virus C promoter. J. Virol. 77, 821-829. Chandar, N., Billig, B., McMaster, J. and Novak, J. (1992). Inactivation of receptor promoter (Fig. 5B). This suggests that Myc may p53 gene in human and murine osteosarcoma cells. Br. J. Cancer 65, 208- actually repress the promoter activity not only by directly 214. interacting with NF-Y, but also by sequestering Sp1 as reported Chao, H. H., Buchmann, A. M. and DeCaprio, J. A. (2000). Loss of (Gartel et al., 2001). We could not show this activity of Myc p19(ARF) eliminates the requirement for the pRB-binding motif in simian virus 40 large T antigen-mediated transformation. Mol. Cell. Biol. 20, 7624- in 3T3 cells, possibly because endogenous Myc was present 7633. (Molander et al., 2001). Chomczynski, P. and Sacchi, N. (1987). Single-step method of RNA isolation In agreement, the p53-binding defective LT is unable to by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. repress the promoter activity. We previously found that levels Biochem. 162, 156-159. Classon, M. and Harlow, E. (2002). The retinoblastoma tumour suppressor of Myc and p73 increase following serum stimulation when in development and cancer. Nat. Rev. Cancer 2, 910-917. they repress the promoter activity. So far of all cell-cycle Courey, A. J. and Tjian, R. (1988). Analysis of Sp1 in vivo reveals multiple regulating molecules investigated by us, only ∆Np73 has been transcriptional domains, including a novel glutamine-rich activation motif. shown to activate the promoter (Hackzell et al., 2002). The Cell 55, 887-898. p53-mediated activation that we detected here is small yet Facchini, L. M. and Penn, L. Z. (1998). The molecular role of Myc in growth and transformation: recent discoveries lead to new insights. FASEB J. 12, significant and appears to involve the Sp1-binding site of the 633-651. promoter. In fact, in combination with Sp1, the activation can Frederiksen, K., Jat, P. S., Valtz, N., Levy, D. and McKay, R. (1988). be further enhanced. It has been reported that induction of p53 Immortalization of precursor cells from the mammalian CNS. Neuron 1, results in a complex formation with Sp1 through its C- 439-448. Funa, K., Yamada, N., Brodin, G., Pietz, K., Ahgren, A., Wictorin, K., terminus, dissociating HDAC1 from Sp1 on the p21 promoter Lindvall, O. and Odin, P. (1996). Enhanced synthesis of platelet-derived (Lagger et al., 2003). p53 would thus not only collaborate with growth factor following injury induced by 6-hydroxydopamine in rat brain. Sp1 activation, but also prevent deacetylation of histones by Neuroscience 74, 825-833. releasing HDAC1 from Sp1. Whether it happens on this Gartel, A. L., Ye, X., Goufman, E., Shianov, P., Hay, N., Najmabadi, F. and receptor promoter remains to be studied. Tyner, A. L. (2001). Myc represses the p21(WAF1/CIP1) promoter and β interacts with Sp1/Sp3. Proc. Natl. Acad. Sci. USA 98, 4510-4515. In conclusion, LT seems to severely affect the PDGF - Gilbertson, D. G., Duff, M. E., West, J. W., Kelly, J. D., Sheppard, P. O., receptor promoter by interfering with the activation by NF-Y Hofstrand, P. D., Gao, Z., Shoemaker, K., Bukowski, T. R., Moore, M., and Sp1, through a mechanism involving Myc, pRb and p53. et al. (2001). Platelet-derived growth factor C (PDGF-C), a novel growth There is a strong concern as to whether SV40 is involved in factor that binds to PDGF alpha and beta receptor. J. Biol. Chem. 276, 27406-27414. the pathogenesis of human tumours. This possibility exists as Gjoerup, O., Chao, H., DeCaprio, J. A. and Roberts, T. M. (2000). pRB- several spontaneously occurring human tumours, such as dependent, J domain-independent function of simian virus 40 large T mesothelioma, osteosarcoma and other brain tumours contain antigen in override of p53 growth suppression. J. Virol. 74, 864-874. DNA sequences derived from the SV40 (Klein et al., Hackzell, A., Uramoto, H., Izumi, H., Kohno, K. and Funa, K. (2002). p73 2002). However, defective suppressor genes in tumours may independent of c-Myc represses transcription of platelet-derived growth factor beta-receptor through interaction with NF-Y. J. Biol. Chem. 277, render them unresponsive to the repression by LT, thus tumours 39769-39776. could maintain the growth factor receptor expression Hartl, F. U. and Martin, J. (1995). Molecular chaperones in cellular protein uncoupled from the cell cycle. folding. Curr. Opin. Struct. Biol. 5, 92-102. Hayes, T. E., Valtz, N. L. and McKay, R. D. (1991). Downregulation of We thank Drs Livingston, Jat, DeCaprio, Mantovani, Zetterberg, CDC2 upon terminal differentiation of neurons. New Biol. 3, 259-269. Suske, and Tjian for cDNAs, and Drs Sedivy, Classon, Zetterberg, and Heldin, C. H. and Westermark, B. (1990). Growth factors and their receptors. In Handbook of Experimental Pharmacology, and McKay for cell lines. This work was supported by grants from the Mutagenesis, Vol. 94, pt II (ed. C. S. Cooper and P. L. Grover), pp. 353- Swedish Medical Research Council, the Swedish Cancer Foundation, 379. Heidelberg, Germany: Springer Verlag. the Swedish Children’s Cancer Foundation, the Inga and Arne Heldin, C. H. and Westermark, B. (1999). Mechanism of action and in vivo Lundberg Foundation, STINT, and the Swedish Institute. H.U. is role of platelet-derived growth factor. Physiol. Rev. 79, 1283-1316. supported by the Wenner-Gren Society and the Wenner-Gren Herrera, R. E., Sah, V. P., Williams, B. O., Makela, T. P., Weinberg, R. A. Foundation, the Uehara Memorial Life Science Foundation, and the and Jacks, T. (1996). Altered cell cycle kinetics, gene expression, and G1 Kanae Foundation for Life & Socio-Medical Science. restriction point regulation in Rb-deficient fibroblasts. Mol. Cell. Biol. 16, 2402-2407. Hilleman, M. R. (1998). Discovery of simian virus 40 (SV40) and its relationship to poliomyelitis virus vaccines. Dev. Biol. Stand. 94, 183-190. References Ishisaki, A., Murayama, T., Ballagi, A. E. and Funa, K. (1997). Nuclear Ballagi, A. E., Ishizaki, A., Nehlin, J. O. and Funa, K. (1995). Isolation and factor Y controls the basal transcription activity of the mouse platelet- characterization of the mouse PDGF beta-receptor promoter. Biochem. derived-growth-factor beta-receptor gene. Eur. J. Biochem. 246, 142-146. Biophys. Res. Commun. 210, 165-173. Izumi, H., Molander, C., Penn, L. Z., Ishisaki, A., Kohno, K. and Funa, Mechanism for SV40LT downregulation of PDGF β-receptor 3865

K. (2001). Mechanism for the transcriptional repression by c-Myc on PDGF C. and Mathis, D. (1994). Dominant negative analogs of NF-YA. J. Biol. beta-receptor. J. Cell Sci. 114, 1533-1544. Chem. 269, 20340-20346. Kierstead, T. D. and Tevethia, M. J. (1993). Association of p53 binding and Mateyak, M. K., Obaya, A. J., Adachi, S. and Sedivy, J. M. (1997). immortalization of primary C57BL/6 mouse embryo fibroblasts by using Phenotypes of c-Myc-deficient rat fibroblasts isolated by targeted simian virus 40 T-antigen mutants bearing internal overlapping deletion homologous recombination. Cell Growth Differ. 8, 1039-1048. . J. Virol. 67, 1817-1829. Moens, U., Seternes, O. M., Johansen, B. and Rekvig, O. P. (1997). Kim, H. Y., Ahn, B. Y. and Cho, Y. (2001). Structural basis for the Mechanisms of transcriptional regulation of cellular genes by SV40 large inactivation of retinoblastoma tumor suppressor by SV40 large T antigen. T- and small T-antigens. Virus Genes 15, 135-154. EMBO J. 20, 295-304. Moens, U., van Ghelue, M., Kristoffersen, A. K., Johansen, B., Rekvig, O. Kim, I. S., Sinha, S., de Crombrugghe, B. and Maity, S. N. (1996). P., Degre, M. and Rollag, H. (2001). Simian virus 40 large T-antigen, but Determination of functional domains in the C subunit of the CCAAT- not small T-antigen, trans-activates the human cytomegalovirus major binding factor (CBF) necessary for formation of a CBF-DNA complex: immediate early promoter. Virus Genes 23, 215-226. CBF-B interacts simultaneously with both the CBF-A and CBF-C Molander, C., Hackzell, A., Ohta, M., Izumi, H. and Funa, K. (2001). Sp1 subunits to form a heterotrimeric CBF molecule. Mol. Cell. Biol. 16, 4003- is a key regulator of the PDGF beta-receptor transcription. Mol. Biol. Rep. 4013. 28, 223-233. Klein, G., Powers, A. and Croce, C. (2002). Association of SV40 with human Oster, S. K., Marhin, W. W., Asker, C., Facchini, L. M., Dion, P. A., Funa, tumors. Oncogene 21, 1141-1149. K., Post, M., Sedivy, J. M. and Penn, L. Z. (2000). Myc is an essential Lagger, G., Doetzlhofer, A., Schuettengruber, B., Haidweger, E., negative regulator of platelet-derived growth factor beta receptor expression. Simboeck, E., Tischler, J., Chiocca, S., Suske, G., Rotheneder, H., Mol. Cell. Biol. 20, 6768-6778. Wintersberger, E. et al. (2003). The tumor suppressor p53 and histone Reichelt, M., Zang, K. D., Seifert, M., Welter, C. and Ruffing, T. (1999). deacetylase 1 are antagonistic regulators of the cyclin-dependent kinase The yeast two-hybrid system reveals no interaction between p73 alpha and inhibitor p21/WAF1/CIP1 gene. Mol. Cell. Biol. 23, 2669-2679. SV40 large T-antigen. Arch. Virol. 144, 621-626. Laufen, T., Mayer, M. P., Beisel, C., Klostermeier, D., Mogk, A., Reinstein, Sullivan, C. S. and Pipas, J. M. (2002). T antigens of simian virus 40: J. and Bukau, B. (1999). Mechanism of regulation of hsp70 chaperones by molecular chaperones for viral replication and tumorigenesis. Microbiol. DnaJ cochaperones. Proc. Natl. Acad. Sci. USA 96, 5452-5457. Mol. Biol. Rev. 66, 179-202. Li, D., Zhao, R., Lilyestrom, W., Gai, D., Zhang, R., DeCaprio, J. A., Uramoto, H., Izumi, H., Ise, T., Tada, M., Uchiumi, T., Kuwano, M., Fanning, E., Jochimiak, A., Szakonyi, G. and Chen, X. S. (2003). Yasumoto, K., Funa, K. and Kohno, K. (2002). p73 Interacts with c-Myc Structure of the replicative helicase of the oncoprotein SV40 large tumour to regulate Y-box-binding protein-1 expression. J. Biol. Chem. 277, 31694- antigen. Nature 423, 512-518. 31702. Liang, F., Schaufele, F. and Gardner, D. G. (2001). Functional interaction Wang, J. L., Nister, M., Bongcam-Rudloff, E., Ponten, J. and Westermark, of NF-Y and Sp1 is required for type a natriuretic peptide receptor gene B. (1996). Suppression of platelet-derived growth factor alpha- and beta- transcription. J. Biol. Chem. 276, 1516-1522. receptor mRNA levels in human fibroblasts by SV40 T/t antigen. J. Cell Magana, M. M., Koo, S. H., Towle, H. C. and Osborne, T. F. (2000). Physiol. 166, 12-21. Different sterol regulatory element-binding protein-1 isoforms utilize Weiss, R., Giordano, A., Furth, P., DeCaprio, J., Pipas, J., Ozer, H., distinct co-regulatory factors to activate the promoter for fatty acid synthase. Strickler, H., Procopio, A., Garcea, R. and Carbone, M. (1998). SV40 J. Biol. Chem. 275, 4726-4733. as an oncogenic virus and possible human pathogen. Dev. Biol. Stand. 94, Magnaghi-Jaulin, L., Groisman, R., Naguibneva, I., Robin, P., Lorain, S., 355-360. le Villain, J. P., Troalen, F., Trouche, D. and Harel-Bellan, A. (1998). Zhang, H. S., Gavin, M., Dahiya, A., Postigo, A. A., Ma, D., Luo, R. X., Retinoblastoma protein represses transcription by recruiting a histone Harbour, J. W. and Dean, D. C. (2000). Exit from G1 and S phase of the deacetylase. Nature 391, 601-605. cell cycle is regulated by repressor complexes containing HDAC-Rb- Mantovani, R., Li, X. Y., Pessara, U., Hooft van Huisjduijnen, R., Benoist, hSWI/SNF and Rb-hSWI/SNF. Cell 101, 79-89.