c-Myc Mediates Activation of the cad Promoter via a Post-RNA Polymerase II Recruitment Mechanism
2001; Elsevier BV; Volume: 276; Issue: 51 Linguagem: Inglês
10.1074/jbc.m109014200
ISSN1083-351X
AutoresScott R. Eberhardy, Peggy Farnham,
Tópico(s)Epigenetics and DNA Methylation
ResumoThe c-Myc protein is a site-specific DNA-binding transcription factor that is up-regulated in a number of different cancers. We have previously shown that binding of Myc correlates with increased transcription of the cad promoter. We have now further investigated the mechanism by which Myc mediates transcriptional activation of the cad gene. Using a chromatin immunoprecipitation assay, we found high levels of RNA polymerase II bound to the cad promoter in quiescent NIH 3T3 cells and in differentiated U937 cells, even though the promoter is inactive. However, chromatin immunoprecipitation with an antibody that recognizes the hyperphosphorylated form of the RNA polymerase II carboxyl-terminal domain (CTD) revealed that phosphorylation of the CTD does correlate with c-Myc binding and cad transcription. We have also found that the c-Myc transactivation domain interacts with cdk9 and cyclin T1, components of the CTD kinase P-TEFb. Furthermore, activator bypass experiments have shown that direct recruitment of cyclin T1 to the cad promoter can substitute for c-Myc to activate the promoter. In summary, our results suggest that c-Myc activates transcription ofcad by stimulating promoter clearance and elongation, perhaps via recruitment of P-TEFb. The c-Myc protein is a site-specific DNA-binding transcription factor that is up-regulated in a number of different cancers. We have previously shown that binding of Myc correlates with increased transcription of the cad promoter. We have now further investigated the mechanism by which Myc mediates transcriptional activation of the cad gene. Using a chromatin immunoprecipitation assay, we found high levels of RNA polymerase II bound to the cad promoter in quiescent NIH 3T3 cells and in differentiated U937 cells, even though the promoter is inactive. However, chromatin immunoprecipitation with an antibody that recognizes the hyperphosphorylated form of the RNA polymerase II carboxyl-terminal domain (CTD) revealed that phosphorylation of the CTD does correlate with c-Myc binding and cad transcription. We have also found that the c-Myc transactivation domain interacts with cdk9 and cyclin T1, components of the CTD kinase P-TEFb. Furthermore, activator bypass experiments have shown that direct recruitment of cyclin T1 to the cad promoter can substitute for c-Myc to activate the promoter. In summary, our results suggest that c-Myc activates transcription ofcad by stimulating promoter clearance and elongation, perhaps via recruitment of P-TEFb. transactivation domain carbamoyl-phosphate synthase/aspartate transcarbamoylase/dihydroorotase chromatin immunoprecipitation RNA polymerase II carboxyl-terminal domain positive elongation factor b histone acetyltransferase telomerase reverse transcriptase cyclin-dependent kinase cytomegalovirus glutathioneS-transferase human immunodeficiency virus reverse transcriptase immunoprecipitation c-Myc is a site-specific DNA-binding transcription factor that is conserved throughout evolution. Deregulated expression of c-Myc occurs in many cancers, including lymphomas, breast, colon, and liver cancer. c-Myc is overexpressed in neoplasia by a number of different mechanisms, including gene amplification, translocation, retroviral insertion, and activation of pathways upstream of c-Myc expression (for review, see Ref. 1Henriksson M. Luscher B. Adv. Cancer Res. 1996; 68: 109-182Crossref PubMed Google Scholar). A number of studies have established the importance of c-Myc in neoplastic transformation. For example, it has long been known that co-transfection of c-Myc and Ras in rat embryo fibroblasts causes transformation in these cells (2Stone J. deLange T. Ramsay G. Jakobovits F. Bishop J.M. Varmus H. Lee W. Mol. Cell. Biol. 1987; 7: 1697-1709Crossref PubMed Scopus (329) Google Scholar). Additionally, targeted overexpression of c-Myc to various tissues of transgenic mice can cause increased tumor incidence (3Sinn E. Muller W. Pattengale P. Tepler I. Wallace R. Leder P. Cell. 1987; 49: 465-475Abstract Full Text PDF PubMed Scopus (591) Google Scholar, 4Adams J.M. Harris A.W. Pinkert C.A. Corcoran L.M. Alexander W.S. Cory S. Palmiter R.D. Brinster R.L. Nature. 1985; 318: 533-538Crossref PubMed Scopus (1308) Google Scholar, 5Harris A.W. Pinkert C.A. Crawford M. Langdon W.Y. Brinster R.L. Adams J.M. J. Exp. Med. 1988; 167: 353-371Crossref PubMed Scopus (314) Google Scholar, 6Sandgren E.P. Quaife C.J. Pinkert C.A. Palmiter R.D. Brinster R.L. Oncogene. 1989; 4: 715-724PubMed Google Scholar). c-Myc is a member of the helix-loop-helix leucine zipper family of transcription factors. Helix-loop-helix leucine zipper proteins bind to a DNA motif called an E box, which consists of the core consensus sequence CACGTG (7Blackwell T.K. Kretzner L. Blackwood E.M. Eisenman R.N. Weintraub H. Science. 1990; 250: 1149-1151Crossref PubMed Scopus (692) Google Scholar). Efficient binding of c-Myc to an E box can only occur when c-Myc forms heterodimers with the protein Max, another helix-loop-helix leucine zipper protein (8Blackwood E.M. Eisenman R.N. Science. 1991; 251: 1211-1217Crossref PubMed Scopus (1453) Google Scholar). When Myc/Max heterodimers bind to DNA, c-Myc can activate transcription through a transcriptional activation domain (TAD)1 in the amino-terminal region of the protein (9Kato G.J. Barrett J. Villa-Garcia M. Dang C.V. Mol. Cell. Biol. 1990; 10: 5914-5920Crossref PubMed Scopus (309) Google Scholar). c-Myc function is antagonized by the Mad proteins, which can also dimerize with Max and bind to E boxes (10Ayer D.E. Kretzner L. Eisenman R.N. Cell. 1993; 72: 211-222Abstract Full Text PDF PubMed Scopus (621) Google Scholar). Mad proteins repress transcription through their ability to recruit the mSin3 corepressor complex (11Ayer D.E. Lawrence Q.A. Eisenman R.N. Cell. 1995; 80: 767-776Abstract Full Text PDF PubMed Scopus (522) Google Scholar). Although c-Myc is highly expressed in proliferating cells, Mad proteins are found mainly in quiescent and differentiated cells. Thus, the relative abundance of c-Myc and Mad proteins in a cell is thought to dictate gene expression. Recent advances in microarray technology has allowed genome-wide studies of mRNA transcripts responsive to transcription factors, and a number of such experiments have been done to examine which genes are responsive to c-Myc (12Coller H.A. Grandori C. Tamayo P. Colbert T. Lander E.S. Eisenman R.N. Golub T.R. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 3260-3265Crossref PubMed Scopus (700) Google Scholar, 13O'Hagan R.C. Schreiber-Agus N. Chen K. David G. Engelman J.A. Schwab R. Alland L. Thomson C. Ronning D.R. Sacchettini J.C. Meltzer P. DePinho R.A. Nat. Genet. 2000; 24: 113-119Crossref PubMed Scopus (120) Google Scholar, 14Schuhmacher M. Kohlhuber F. Holzel M. Kaiser C. Burtscher H. Jarsch M. Bornkamm G.W. Laux G. Polack A. Weidle U.H. Eick D. Nucleic Acids Res. 2001; 29: 397-406Crossref PubMed Scopus (262) Google Scholar). These studies have confirmed the Myc responsiveness of a number of proposed target genes, such asornithine decarboxylase (odc) (15Bello-Fernandez C. Packham G. Cleveland J.L. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 7804-7808Crossref PubMed Scopus (656) Google Scholar),nucleolin (16Greasley P.J. Bonnard C. Amati B. Nucleic Acids Res. 2000; 28: 446-453Crossref PubMed Scopus (104) Google Scholar), cyclin D2 (ccnd2) (17Bouchard C. Thieke K. Maier A. Saffrich R. Hanley-Hyde J. Ansorge W. Reed S. Sicinski P. Bartek J. Eilers M. EMBO J. 1999; 18: 5321-5333Crossref PubMed Scopus (405) Google Scholar), and cdk4 (18Hermeking H. Rago C. Schuhmacher M. Li Q. Barrett J.F. Obaya A.J. O'Connell B.C. Mateyak M.K. Tam W. Kohlhuber F. Dang C.V. Sedivy J.M. Eick D. Vogelstein B. Kinzler K.W. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 2229-2234Crossref PubMed Scopus (375) Google Scholar). Additionally, genes involved in general responses such as glycolysis and protein synthesis appear to be influenced by c-Myc (19Osthus R.C. Shim H. Kim S. Li Q. Reddy R. Mukherjee M. Xu Y. Wonsey D. Lee L.A. Dang C.V. J. Biol. Chem. 2000; 275: 21797-21800Abstract Full Text Full Text PDF PubMed Scopus (595) Google Scholar, 20Kim S. Li Q. Dang C.V. Lee L.A. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 11198-11202Crossref PubMed Scopus (146) Google Scholar). However, mRNA studies do not conclusively establish the identity of a c-Myc target promoter. It is likely that many mRNAs that respond to c-Myc are indirectly affected by Myc-mediated changes in signaling pathways. To establish a promoter as a c-Myc target, it is necessary to show that binding of c-Myc causes changes in promoter activity. However, a difficulty in studying c-Myc target genes is that overexpression of c-Myc in cells results in only modest increases in promoter activity, typically around 2–5-fold. This may be due to several factors, including the presence of highly abundant proteins, such as USF, which also bind to target gene E boxes and may interfere with c-Myc binding, or a requirement for a coactivator which is at limiting concentrations in the cell. We have previously shown that the cad promoter meets all the criteria of a true c-Myc target. The cad gene encodes the trifunctional enzyme carbamoyl-phosphate synthase/aspartate transcarbamoylase/dihydroorotase, which is required for the first three rate-limiting steps of pyrimidine biosynthesis. Expression ofcad is high in proliferating cells and low in quiescent and differentiated cells, which closely correlates with c-Myc expression. The cad promoter contains a conserved E box downstream of the transcription start site which we have shown to be essential for growth regulation (21Miltenberger R.J. Sukow K. Farnham P.J. Mol. Cell. Biol. 1995; 15: 2527-2535Crossref PubMed Scopus (155) Google Scholar). Importantly, we have used chromatin immunoprecipitation (ChIP) studies to show that c-Myc binds to thecad promoter during peak levels of transcription and is not bound when cad is not expressed (22Boyd K.E. Wells J. Gutman J. Bartley S.M. Farnham P.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13887-13892Crossref PubMed Scopus (244) Google Scholar). Finally,cad is one of the few proposed c-Myc target genes whose expression is decreased in c-Myc null cells (23Bush A. Mateyka M. Dugan K. Obaya A. Adachi S. Sedivy J. Cole M. Genes Dev. 1998; 12: 3797-3802Crossref PubMed Scopus (161) Google Scholar). Transcription of protein-coding genes by RNA polymerase II (RNAP II) is believed to involve three main events (24Lee T.I. Young R.A. Annu. Rev. Genet. 2000; 34: 77-137Crossref PubMed Scopus (618) Google Scholar). First, chromatin modification and remodeling of the promoter occurs to facilitate transcription factor binding. Then, general transcription factors bind to the promoter to recruit RNAP II to the promoter to form the preinitiation complex. Finally, RNAP II clears the promoter and begins elongation of mRNA. Stimulation of elongation appears to require phosphorylation of the RNAP II large subunit carboxyl-terminal domain (CTD), which contains multiple tandem repeats of the amino acid sequence YSPTSPS. A number of kinases have the ability to phosphorylate the RNAP II CTD, including the cyclin-cdk complexes cyclin H-cdk7, cyclin C-cdk8, and cyclin T-cdk9. Cyclin H and cdk7 are part of the general transcription factor TFIIH, which has been shown to be important in stimulating elongation (25Dvir A. Conaway R.C. Conaway J.W. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 9006-9010Crossref PubMed Scopus (113) Google Scholar). Cyclin C and cdk8 are part of the mediator complex, and while their role in transcriptional elongation has not been fully elucidated, they have been shown to co-purify with RNAP II (26Cho H. Orphanides G., X., S. Yang X.J. Ogryzko V. Lees E. Nakatani Y. Reinberg D. Mol. Cell. Biol. 1998; 18: 5355-5363Crossref PubMed Scopus (246) Google Scholar). Cdk9 and the cyclin T proteins, which consist of cyclin T1 and cyclins T2a and T2b, are part of a complex known as P-TEFb (for positive transcriptionelongation factor b), which has been shown to be important for HIV transcriptional elongation through its association with Tat (27Price D.H. Mol. Cell. Biol. 2000; 20: 2629-2634Crossref PubMed Scopus (562) Google Scholar). Current models for c-Myc function suggest that transcriptional control of c-Myc target genes may occur through chromatin modification. c-Myc has recently been shown to interact with the protein TRRAP, which associates with the SAGA complex that contains Gcn5, a protein that has histone acetyltransferase (HAT) activity (28McMahon S.B. Wood M.A. Cole M.D. Mol. Cell. Biol. 2000; 20: 556-562Crossref PubMed Scopus (373) Google Scholar). In contrast, Mad proteins interact with histone deacetylases through association with the mSin3 corepressor complex (29Sommer A. Hilfenhaus S. Menkel A. Kremmer E. Seiser C. Loidl P. Luscher B. Curr. Biol. 1997; 7: 357-365Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). These observations have led to a model in which c-Myc target genes are repressed by Mad due to deacetylation of histones. Replacement of Mad/Max by Myc/Max is then proposed to result in recruitment of HAT activity and acetylation of histones, subsequently allowing RNAP II to bind and initiate transcription. In support of this model, recent studies have shown that at some c-Myc target promoters, such as nucleolin, changes in histone acetylation do correlate with binding of c-Myc (30Frank S.R. Schroeder M. Fernandez P. Taubert S. Amati B. Genes Dev. 2001; 15: 2069-2082Crossref PubMed Scopus (418) Google Scholar). Also, a detailed analysis of the cyclin D2 promoter using chromatin immunoprecipitation indicates that c-Myc and TRRAP bind to thecyclin D2 promoter when the gene is transcribed whereas Mad and histone deacetylases 1 bind to the promoter when the gene is turned off (31Bouchard C. Dittrich O. Kiermaier A. Dohmann K. Menkel A. Eilers M. Luscher B. Genes Dev. 2001; 15: 2042-2047Crossref PubMed Scopus (272) Google Scholar). Finally, a study of the telomerase reverse transcriptase (tert) promoter has also shown that histone acetylation correlates with transcription and c-Myc binding (32Xu D. Popov N. Hou M. Wang Q. Bjorkholm M. Gruber A. Menkel A.R. Henriksson M. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 3826-3831Crossref PubMed Scopus (268) Google Scholar). Although Myc clearly regulates the expression of some, and perhaps most, target promoters by recruiting TRRAP and causing an increase in acetylated histones, we have not seen large changes in the levels of histone acetylation at the cad promoter (33Eberhardy S.E. D'Cunha C.A. Farnham P.J. J. Biol. Chem. 2000; 275: 33798-33805Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar). In fact, expression studies using c-Myc constructs containing or lacking the TRRAP-binding domain indicate that c-Myc-mediated activation of thecad promoter does not require recruitment of TRRAP and the associated histone acetyltransferases. 2M. A. Nikiforov, S. Chandriani, J. Park, I. Kotenko, D. Matheos, A. Johnsson, S. B. McMahon, and M. D. Cole, submitted for publication. 2M. A. Nikiforov, S. Chandriani, J. Park, I. Kotenko, D. Matheos, A. Johnsson, S. B. McMahon, and M. D. Cole, submitted for publication. These findings, in combination with our previous studies showing that thecad promoter contains high levels of acetylated histones at all times, have led us to investigate other possible mechanisms through which c-Myc may activate transcription. Using the ChIP assay, we have now examined RNAP II binding at the cad gene. We have found that RNAP II is bound to the cad promoter at all times but levels of the hyperphosphorylated form of RNAP II change in correlation with c-Myc binding. We also demonstrate an interaction between the c-Myc TAD and P-TEFb, and activator bypass experiments show that cyclin T1 can substitute for c-Myc to activate the cad promoter. These results suggest that c-Myc functions to stimulate cad transcription by recruitment of the P-TEFb complex. NIH 3T3 cell cultures were maintained and synchronized as previously described (35Boyd K.E. Farnham P.J. Mol. Cell. Biol. 1997; 17: 2529-2537Crossref PubMed Scopus (137) Google Scholar). Briefly, 6–8 × 106 cells were seeded into 500-cm2tissue culture dishes (1–2 dishes were used per antibody per time point in the formaldehyde cross-linking experiments) and incubated in starvation medium for 48–72 h. Cells were then either stimulated to enter the cell cycle by the addition of stimulation medium for 4 h (early G1 phase) or 12 h (G1/S phase) prior to cross-linking. Progression of the cells through the cell cycle was measured by flow cytometric analysis of propidium iodide-stained cells as described previously (36Means A.L. Slansky J.E. McMahon S.L. Knuth M.W. Farnham P.J. Mol. Cell. Biol. 1992; 12: 1054-1063Crossref PubMed Scopus (132) Google Scholar). Data were acquired on a FACScan flow cytometer (Becton Dickinson) using CellQuest acquisition and analysis software (data not shown). For cad-luciferase co-transfections, 1 × 105 cells were co-transfected with cad 3G4luc and the indicated Gal4 fusion protein expression plasmid as described previously (22Boyd K.E. Wells J. Gutman J. Bartley S.M. Farnham P.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13887-13892Crossref PubMed Scopus (244) Google Scholar). U937 cells were maintained in spinner flasks containing RPMI 1640 medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C and 5% CO2. Differentiation of U937 cells was performed by adding all-trans- retinoic acid (Sigma) to a final concentration of 1 μm. Cells were then incubated for an additional 5 days prior to harvesting; cell cycle analysis of U937 cells was performed as described above for 3T3 cells (data not shown). The plasmid GST-Myc-(1–143) contains the sequence encoding the first 143 amino acids of human c-Myc in the plasmid pGEX-4T-1 (Amersham Bioscience, Inc.). The human c-Myc sequence was obtained by PCR from the plasmid Gal4-Myc-(1–262), a gift from C. Dang, using the 5′ primer, 5′-CATAGAATAAGTGCGACATCATCATCGG-3′ and the 3′ primer, 5′-GGATCCTCGAGCTTGGCGGCGGCCGAGAA-3′. PCR products from the GST-Myc-(1–143) primers were isolated and digested withEco RI and Xho I and ligated intoEco RI/Xho I digested pGEX-4T-1. The plasmid GST-USF-(1–181) contains the sequence encoding the first 181 amino acids of human USF1. The USF sequence was obtained from the plasmid Gal4-USF1, a gift from M. Eilers (37Desbarats L. Gaubatz S. Eilers M. Genes Dev. 1996; 10: 447-460Crossref PubMed Scopus (110) Google Scholar). The plasmid Gal4-USF1 was cut with Eco RI, and the insert was gel purified and ligated intoEco RI-digested pGEX-4T-1. Correct orientation of GST-Myc-(1–143) and GST-USF-(1–181) was confirmed by sequencing. The plasmids CMV-cdk9, CMV-cyclin T1, Gal4-cdk9, and Gal4-cyclin T1 were gifts from Luigi Lania (38Majello B. Napolitano G. Giordano A. Lania L. Oncogene. 1999; 18: 4598-4605Crossref PubMed Scopus (59) Google Scholar). The plasmid Gal4-CBP is a gift from John Chrivia (39Swope D.L. Mueller C.L. Chrivia J.C. J. Biol. Chem. 1996; 271: 28138-28145Abstract Full Text Full Text PDF PubMed Scopus (149) Google Scholar). The plasmid cad 3G4luc was constructed by the same method as cadG4luc (22Boyd K.E. Wells J. Gutman J. Bartley S.M. Farnham P.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13887-13892Crossref PubMed Scopus (244) Google Scholar), but contains three Gal4 sites in tandem. RNA was prepared from NIH 3T3 and U937 cells as described previously (33Eberhardy S.E. D'Cunha C.A. Farnham P.J. J. Biol. Chem. 2000; 275: 33798-33805Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar). RNase protections forcad mRNA levels in NIH 3T3 cells was performed as described previously (22Boyd K.E. Wells J. Gutman J. Bartley S.M. Farnham P.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13887-13892Crossref PubMed Scopus (244) Google Scholar, 40Slansky J.E. Li Y. Kaelin W.G. Farnham P.J. Mol. Cell. Biol. 1993; 13 (Correction (1993) Mol. Cell. Biol. 13, 7201): 1610-1618Crossref PubMed Scopus (257) Google Scholar). RT-PCR was performed using the 5′ primer, 5′-CTCACTGATCCCTCCTACAA-3′, and the 3′ primer, 5′-GTGGATACGACACTGGGATA-3′, to amplify human cad mRNA. The mRNA from the human gapdh gene was detected using the 5′ primer, 5′-GAGCCAAAAGGGTCATC-3′, and the 3′ primer, 5′-GTGGTCATGAGTCCTTC-3′. RT-PCR reactions contained 10 μl of EZ buffer (PerkinElmer Life Sciences), 2 μm of each primer, 1 m betaine (Sigma), 200 μm each dNTP, 5 units of rTth polymerase (PerkinElmer Life Scineces), 5 μl of 25 mm Mn(OAc)2, and 300 ng of RNA in 50 μl total volume. The formaldehyde cross-linking and chromatin immunoprecipitation assays of tissue culture cells were performed as described previously with the following modifications (22Boyd K.E. Wells J. Gutman J. Bartley S.M. Farnham P.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13887-13892Crossref PubMed Scopus (244) Google Scholar, 41Boyd K.E. Farnham P.J. Mol. Cell. Biol. 1999; 19: 8389-8399Crossref Scopus (147) Google Scholar). Immunoprecipitations were performed overnight at 4 °C using 1 μg of anti-c-Myc sc-764-X (Santa Cruz), 1 μl of anti-USF1 serum (a gift from Emery Bresnick), anti-pol II sc-899 (Santa Cruz), or 3–5 μl of anti-pol II phospho-CTD serum (a gift from David Bentley). For the ChIP experiments with the anti-phospho-CTD antibody, 10 mmsodium pyrophosphate (pH 8) was added to all solutions to inhibit phosphatase activity. Before the first wash, 20% of the supernatant from the no antibody IP for each condition was saved as total input chromatin and was processed with the eluted IPs beginning with the cross-linking reversal step. After the final ethanol precipitation, each IP sample was resuspended in 30 μl of PCR-grade TE (10 mm Tris-HCl, pH 7.6, 1 mm EDTA, pH 8.0). Total input chromatin samples were resuspended in 30 μl of TE and further diluted 1/100 (total dilution 1/500). PCR reactions contained 2 μl of IP sample or 2 μl of diluted total input (0.2% total input chromatin), 1.2 mmMgCl2, 50 ng of each primer, 200 μm each dATP, dGTP, dCTP, and dTTP, 4 μl of 5 m betaine (Sigma), 1 × Thermophilic buffer (Promega), and 1.25 units ofTaq DNA polymerase (Promega) in 20 μl total volume. After 34–36 cycles of amplification, 8 μl of the PCR products was electrophoresed on a 1.5% agarose gel, and DNA was stained with ethidium bromide and visualized under UV light. Alternatively, reactions were performed by incorporating ∼2 μCi of [α-32P]dCTP into the reactions and amplifying for 25 cycles. These products were separated on a 6% polyacrylamide gel and visualized with a Storm PhosphorImager (Molecular Dynamics). Signals were quantitated using ImageQuant for Macintosh v2.1 (Molecular Dynamics). Fold changes in factor binding were calculated by first dividing the IP signal intensity by the 0.2% input signal intensity (to control for varying levels of chromatin in each sample) for both the G1/S (or log) sample and the G0 (or differentiated) sample. Then, the G1/S ratio was divided by the G0 ratio. The sequences of primers used for PCR analysis can be found at our website (mcardle.oncology.wisc.edu/farnham). HeLa cell nuclear extract was prepared from frozen cells as described previously (42Means A.L. Farnham P.J. Mol. Cell. Biol. 1990; 10: 653-661Crossref PubMed Scopus (164) Google Scholar), and was dialyzed against affinity chromatography buffer (ACB) (10 mm HEPES, pH 7.9, 1 mm EDTA, 1 mmdithiothreitol, 20% glycerol) containing 0.5 mmphenylmethylsulfonyl fluoride and 0.1 m NaCl. GST fusion proteins were prepared and immobilized on glutathione-Sepharose 4B (Amersham Bioscience, Inc.), incubated with HeLa nuclear extract and columns were washed and proteins eluted as described previously (43Fry C.J. Pearson A. Malinowski E. Bartley S.M. Greenblatt J. Farnham P.J. J. Biol. Chem. 1999; 274: 15883-15891Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Eluates were analyzed by Western blot analysis with anti-cdk7 sc-529 (Santa Cruz), anti-cyclin H sc-855 (Santa Cruz), anti-cdk8 sc-1521 (Santa Cruz), anti-cyclin C sc-1061 (Santa Cruz), anti-cdk9 sc-7331 (Santa Cruz), and anti-cyclin T1 sc-8128 (Santa Cruz) antibodies. Each binding experiment was repeated at least three times. We had previously shown that there are high levels of acetylated histones on the cad promoter in both quiescent and differentiated cells (33Eberhardy S.E. D'Cunha C.A. Farnham P.J. J. Biol. Chem. 2000; 275: 33798-33805Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar) suggesting that the chromatin encompassing the cad promoter is in an open configuration prior to c-Myc binding. If so, components of the transcription machinery may have access to the cad promoter regardless of promoter activity. We therefore wished to examine the levels of RNAP II on the cad promoter in states of both high and low transcriptional activity. To do this we used both NIH 3T3 and U937 cells, so that we could monitor protein binding to cad under conditions where transcription can be regulated (Fig.1). NIH 3T3 cells can be synchronized by serum withdrawal into quiescence, resulting in low levels ofcad transcription, Addition of serum causes re-entry into the cell cycle, and cad transcription is maximal at S phase (Fig. 1 B). U937 cells are a lymphoma-derived cell line that can be induced to differentiate after incubation with all-trans- retinoic acid. In growing U937 cells,cad expression is high, while differentiated cells show almost no cad mRNA (Fig. 1 B). To monitor RNAP II binding to the cad promoter, we used an antibody that recognizes the amino terminus of the large subunit of RNAP II. As a positive control, we also examined binding of c-Myc to the cad promoter. Immunoprecipitated DNA was analyzed by PCR using both ethidium bromide staining of 34 cycle PCR and [α-32P]dCTP incorporation of 25 cycle PCR. Both methods showed similar changes in protein binding (Figs.2 and 3), so we have used ethidium bromide staining of 34 cycle PCR for the remainder of this study. As expected, we saw differences in the amount of c-Myc on the promoter which correlated with transcriptional activity. However, we found that levels of RNAP II did not correlate with transcriptional activity. Although c-Myc binding on thecad promoter increased 9.2-fold in NIH 3T3 cells from quiescence to S phase, RNAP II binding was unchanged (Fig.2 A). We also examined RNAP II binding on intron 1 of thenucleolin gene, which contains five E boxes which regulatenucleolin expression, and shows large changes in histone acetylation correlating with c-Myc binding (30Frank S.R. Schroeder M. Fernandez P. Taubert S. Amati B. Genes Dev. 2001; 15: 2069-2082Crossref PubMed Scopus (418) Google Scholar). In contrast to thecad promoter which showed no difference in RNAP II binding, this region of the nucleolin gene shows a 5.2-fold change in RNAP II binding. Using the second model system, we saw a 20-fold difference in c-Myc binding on the cad promoter between growing and differentiated U937 cells, but RNAP II binding was unchanged in these cells (Fig. 2 B). We also examined another cell cycle-regulated promoter, b-myb, for RNAP II binding in U937 cells. In contrast to the cad promoter, we saw an 8.8-fold difference in RNAP II binding to the b-myb promoter between growing and differentiated cells (Fig. 2 B). The results obtained by monitoring the nucleolin gene and the b-myb promoter indicate that our assay is capable of detecting changes in RNAP II binding. Therefore, we conclude that RNAP II is bound to the cad promoter in quiescent NIH3T3 cells and differentiated U937 cells when cad transcription is low.Figure 3Increased RNA polymerase II is seen at the 3′ end of the cad gene when transcription is activated. A, NIH 3T3 cells synchronized at G0 and S phase were cross-linked and chromatin was immunoprecipitated with a RNAP II antibody. Immunoprecipitated DNA was analyzed by PCR amplification with primers specific for the 3′ end of the cad gene. On the left is a PCR reaction done for 34 cycles and stained with ethidium bromide and on theright is the same PCR done for 25 cycles with [α-32P]dCTP incorporation. PCR signals were quantitated as detailed under “Materials and Methods”. B, asynchronously growing U937 cells and U937 cells differentiated for 5 days with 1 μm all-trans- retinoic acid were cross-linked and chromatin was immunoprecipitated with a RNAP II antibody. Immunoprecipitated DNA was analyzed by PCR as inA.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Although we see high levels of RNAP II on the cad promoter at all times, one would expect to see changes in the amount of RNAP II on the coding regions of the cad gene correlating with changes in transcription. To test this hypothesis, DNA isolated from ChIP experiments was PCR amplified with primers to the 3′ end of thecad gene. We found that RNAP II binding changes about 3–5-fold on the 3′ end of cad in NIH 3T3 cells between G0 and S phase or as growing U937 cells are differentiated (Fig. 3, A and B). We have repeated the ChIP assay several times and have obtained similar results. These results suggest that c-Myc may indeed activate transcription of cad by stimulating the transition between the initiation complex and the elongation complex, thus resulting in an increase in full-length transcripts. Our results clearly indicate that RNAP II is bound to thecad promoter in the absence of c-Myc. These results raise the question as to how RNAP II is recruited to the cad promoter. The cad promoter does not have a TATA box so recruitment of RNAP II by TFIID might not be the mechanism by which RNAP II is brought to the promoter. We have previously shown that USF is boun
Referência(s)