MRG1 Expression in Fibroblasts is Regulated by Sp1/Sp3 and an Ets Transcription Factor*
Baoqin Han,Naili Liu,Xiaoming Yang,Hui Sun,Yu-Chung Yang
DOI: https://doi.org/10.1074/jbc.m007470200
2001-01-01
Abstract:MRG1 (melanocyte-specific gene 1 (MSG1)-related gene), a ubiquitously expressed transcription factor that interacts with p300/CBP, TATA-binding protein and Lhx2, is the founding member of a new family of transcription factors. Initial characterization of this newly discovered transcription factor has underscored its potential involvement in many important cellular processes through transcriptional modulation. We previously demonstrated that MRG1 can be induced by various biological stimuli (Sun, H. B., Zhu, Y. X., Yin, T., Sledge, G., and Yang, Y. C. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 13555–13560). As a first step in understanding its role in different biological processes, we investigated mechanisms that regulate transcription of the mouse MRG1 gene in fibroblasts. Transient transfection of Rat1 fibroblast cells with sequential 5′-deletions of mouse MRG1 promoter-luciferase fusion constructs indicated that the −104 to +121 region contains the full promoter activity. Deletion and site-directed mutations within this region revealed that the Ets-1 site at −97 to −94 and the Sp1 site at −51 to −46 are critical for MRG1 expression in fibroblasts. Gel mobility shift and supershift assays performed with Rat1 nuclear extracts identified nucleoprotein complexes binding to the Ets-1 site and the Sp1 site. In Drosophila SL2 cells, which lack the Sp and Ets family of transcription factors, expression of Sp1, Sp3, and Ets-1 or Elf-1 functionally stimulated MRG1 promoter activity in a synergistic manner. These results suggest that multiple transcription factors acting in synergy are responsible for MRG1 expression and the responsiveness of cells to different biological stimuli. MRG1 (melanocyte-specific gene 1 (MSG1)-related gene), a ubiquitously expressed transcription factor that interacts with p300/CBP, TATA-binding protein and Lhx2, is the founding member of a new family of transcription factors. Initial characterization of this newly discovered transcription factor has underscored its potential involvement in many important cellular processes through transcriptional modulation. We previously demonstrated that MRG1 can be induced by various biological stimuli (Sun, H. B., Zhu, Y. X., Yin, T., Sledge, G., and Yang, Y. C. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 13555–13560). As a first step in understanding its role in different biological processes, we investigated mechanisms that regulate transcription of the mouse MRG1 gene in fibroblasts. Transient transfection of Rat1 fibroblast cells with sequential 5′-deletions of mouse MRG1 promoter-luciferase fusion constructs indicated that the −104 to +121 region contains the full promoter activity. Deletion and site-directed mutations within this region revealed that the Ets-1 site at −97 to −94 and the Sp1 site at −51 to −46 are critical for MRG1 expression in fibroblasts. Gel mobility shift and supershift assays performed with Rat1 nuclear extracts identified nucleoprotein complexes binding to the Ets-1 site and the Sp1 site. In Drosophila SL2 cells, which lack the Sp and Ets family of transcription factors, expression of Sp1, Sp3, and Ets-1 or Elf-1 functionally stimulated MRG1 promoter activity in a synergistic manner. These results suggest that multiple transcription factors acting in synergy are responsible for MRG1 expression and the responsiveness of cells to different biological stimuli. MSG1-related gene melanocyte-specific gene 1 signal transducers and activators of transcription 3 cAMP response element-binding protein electrophoretic mobility shift assay Schneider'sDrosophila cell line 2 kilobase(s) polymerase chain reactin sis-interacting element cytomegalovirus mitogen-activated protein kinase GA-binding protein MRG11 and MSG1 are the members of a new family of transcription factors, which share a conserved C-terminal acidic domain (the CR2 domain) that accounts for their transcriptional activity. Through CR2 (1Shioda T. Fenner M.H. Isselbacher K.J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 12298-12303Crossref PubMed Scopus (80) Google Scholar, 2Dunwoodie S.L. Rodriguez T.A. Beddington R.S.P. Mech. Dev. 1998; 72: 27-40Crossref PubMed Scopus (141) Google Scholar, 3Bhattacharya S. Michels C.L. Leung M.K. Arany Z.P. Kung A.L. Livingston D.M. Genes Dev. 1999; 13: 64-75Crossref PubMed Scopus (315) Google Scholar, 4Yahata T. Caestecker M.P. Lechleider R.J. Andriole S. Roberts A.B. Isselbacher K.J. Shioda T.J. J. Biol. Chem. 2000; 275: 8825-8834Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar), MRG1 and MSG1 interact with the CH1 domain of the p300/cAMP response element-binding protein (CBP) complex, nuclear proteins that function as coactivators for basal transcription complexes (5Janknecht R. Hunter T. Curr. Biol. 1996; 6: 951-954Abstract Full Text Full Text PDF PubMed Scopus (207) Google Scholar, 6Goldman P.S. Tran V.K. Goodman R.H. Recent Prog. Horm. Res. 1997; 52: 103-119PubMed Google Scholar, 7Shikama N. Lyon J. La Thangue N.B. Trends Cell. Biol. 1997; 7: 230-237Abstract Full Text PDF PubMed Scopus (424) Google Scholar). MRG1 also interacts with a LIM homeodomain transcription factor, Lhx2, to enhance Lhx2-dependent transcription possibly through recruitment of p300/CBP and TATA-binding proteins (8Glenn D.J. Maurer R.A. J. Biol. Chem. 1999; 274: 36159-36167Abstract Full Text Full Text PDF PubMed Scopus (90) Google Scholar). The interaction of MRG1 with Lhx2 is dependent on the N-terminal region of MRG1, a region not conserved in MSG1. MSG1 interacts with Smad4 and enhances Smad-mediated transcription in a p300/CBP-dependent manner (4Yahata T. Caestecker M.P. Lechleider R.J. Andriole S. Roberts A.B. Isselbacher K.J. Shioda T.J. J. Biol. Chem. 2000; 275: 8825-8834Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar, 9Shioda T. Lechleider R.J. Dunwoodie S.L. Li H. Yahata T. de Caestecker M.P. Fenner M.H. Roberts A.B. Isselbacher K.J. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 9785-9790Crossref PubMed Scopus (100) Google Scholar). The Smad interaction domain resides in the N terminus of MSG1 and thus is unique to MSG1. Based upon these findings, it has been speculated that MRG1 family proteins bind p300/CBP through their conserved C-terminal regions (the CR2 region) while interacting with DNA-binding proteins through the unique N-terminal regions, thus regulating the p300/CBP-dependent transcriptional activation of different target genes (4Yahata T. Caestecker M.P. Lechleider R.J. Andriole S. Roberts A.B. Isselbacher K.J. Shioda T.J. J. Biol. Chem. 2000; 275: 8825-8834Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar). Besides their structural differences, MRG1 and MSG1 show a distinct pattern of expression that may also determine their functional specificity (1Shioda T. Fenner M.H. Isselbacher K.J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 12298-12303Crossref PubMed Scopus (80) Google Scholar, 2Dunwoodie S.L. Rodriguez T.A. Beddington R.S.P. Mech. Dev. 1998; 72: 27-40Crossref PubMed Scopus (141) Google Scholar, 3Bhattacharya S. Michels C.L. Leung M.K. Arany Z.P. Kung A.L. Livingston D.M. Genes Dev. 1999; 13: 64-75Crossref PubMed Scopus (315) Google Scholar, 10Sun H.B. Zhu Y.X. Yin T. Sledge G. Yang Y.C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13555-13560Crossref PubMed Scopus (90) Google Scholar). MSG1 transcripts are predominantly expressed in cultured human and mouse epidermal melanocytes, whereas MRG1 transcripts are detected in all of the human and mouse cell lines and adult tissues examined (1Shioda T. Fenner M.H. Isselbacher K.J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 12298-12303Crossref PubMed Scopus (80) Google Scholar, 2Dunwoodie S.L. Rodriguez T.A. Beddington R.S.P. Mech. Dev. 1998; 72: 27-40Crossref PubMed Scopus (141) Google Scholar, 3Bhattacharya S. Michels C.L. Leung M.K. Arany Z.P. Kung A.L. Livingston D.M. Genes Dev. 1999; 13: 64-75Crossref PubMed Scopus (315) Google Scholar, 10Sun H.B. Zhu Y.X. Yin T. Sledge G. Yang Y.C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13555-13560Crossref PubMed Scopus (90) Google Scholar). Similarly, MRG1 and MSG1 expression profiles during early development are distinct from each other (2Dunwoodie S.L. Rodriguez T.A. Beddington R.S.P. Mech. Dev. 1998; 72: 27-40Crossref PubMed Scopus (141) Google Scholar). MSG1 is predominantly expressed in a subset of mesoderm derivatives, whereas MRG1 transcripts are restricted to anterior visceral endoderm prior to gastrulation. Interestingly, MRG1 is expressed during heart development and its expression can be detected throughout embryogenesis from 8.5 days post-conception, which is consistent with the ubiquitous expression of this gene in adult tissues. These results suggested that MRG1 and MSG1 might play different roles during mouse embryogenesis. MRG1 was cloned in our laboratory as part of an effort to isolate cytokine-inducible genes. Overexpression in Rat1 cells results in loss of cell contact inhibition, anchorage-independent growth in soft agar, and tumor formation in nude mice, suggesting that MRG1 is a transforming gene (10Sun H.B. Zhu Y.X. Yin T. Sledge G. Yang Y.C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13555-13560Crossref PubMed Scopus (90) Google Scholar). Bhattacharya et al. demonstrated that MRG1 is up-regulated by hypoxia and deferoxamine (3Bhattacharya S. Michels C.L. Leung M.K. Arany Z.P. Kung A.L. Livingston D.M. Genes Dev. 1999; 13: 64-75Crossref PubMed Scopus (315) Google Scholar), and represses HIF-1α-mediated transactivation through competitive interaction with the CH1 domain of p300/CBP. Elevated expression of MRG1 in hypoxic cells has been postulated to negatively regulate the cellular response to hypoxia, which is cytokine-regulated. The promoter region of human MRG1 has been reported but not analyzed in detail (11Leung M.-K. Jones T. Michels C.L. Livingston D.M. Bhattacharya S. Genomics. 1999; 61: 307-313Crossref PubMed Scopus (44) Google Scholar), and the transcription factors that regulate it have not been identified. We have now isolated the mouse MRG1 genomic sequence and analyzed the promoter activity by deletion mapping and transient expression in cells. The basal mouse MRG1 promoter maps to region to −104 to +121, which is highly conserved in the human and mouse genes. Deletion analysis in conjunction with site-directed mutagenesis shows that an Ets-1 site at −97 to −94 and an Sp1 site at −51 to −46 are critical for MRG1 promoter activity. Electrophoretic mobility shift assays (EMSA) show that Sp1 and Sp3 bind the Sp1 element, whereas transcription factors binding to the Ets-1 element are yet to be identified. Cotransfection in insect cells that lack Sp and Ets factors demonstrates that Sp1, Sp3, Ets-1, or Elf-1 are important for MRG1 gene expression and therefore for the responses of cells to different biological stimuli. NIH3T3, Rat1, and HepG2 cells were maintained in Dulbecco's modified Eagle's medium with 10% fetal calf serum. Schneider's Drosophila cell line 2 (SL2) was maintained in Schneider's Drosophila medium supplemented with 10% fetal bovine serum at room temperature with atmospheric CO2. Cell culture reagents were obtained from Life Technologies Inc. (Gaithersburg, MD). Antibodies to nuclear proteins Sp1, Sp3, and STAT3 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). A 129/SvJ mouse liver genomic library in Lambda FIXII (Stratagene) was screened using the 2.1-kb MRG1 cDNA as the probe. The probe was random-labeled with [α-32P]dCTP using the Prime-It kit (Stratagene). A 5-kbHindIII fragment from one of the positive clones was subcloned into the HindIII site of a cloning vector, pUC19, resulting in plasmid pUCHi. Sequence analysis showed that the fragment contains a 2.7-kb 5′-flanking region, the first and the second exons, the introns, and most of the third exon. To identify putativecis-regulatory elements, 1 kb of the mouse MRG1 gene 5′-flanking region was analyzed with the Mat Inspector program (12Quandt K. Frech K. Karas H. Wingender E. Werner T. Nucleic Acids Res. 1995; 23: 4878-4884Crossref PubMed Scopus (2414) Google Scholar). The transcription start site was predicted by the neural network promoter prediction (NNPP) program (13Reese M.G. Harris N.L. Eckman F.H. Large Scale Sequencing Specific Neural Networks for Promoter and Splice Site Recognition.Biocomputing: Proc..in: Lawrence Hunter Terri E.Klein 1996 Pacific Symposium. World Scientific Publishing Co., Singapore01-1996: 2-7Google Scholar). The −1036 to +121 region relative to the transcription start site was PCR-amplified using primers: P-1036, 5′-CCCAAGCTTCCCAACCCGTCAGGCAAGA; and P+121, 5′-GGGGTACCCGAGGTTGGCGCCGAGGTCTT. The PCR product was digested withHindIII and KpnI and inserted in a promoterless luciferase reporter vector pXP2 (14Nordeen S.K. BioTechniques. 1988; 6: 454-457PubMed Google Scholar), resulting in plasmid pXP-1036/+121. To construct pXP-2700/+121, a 2.5 kbHindIII/BamHI fragment digested from genomic subclone pUCHi and 0.21kb BamHI/KpnI fragment from −1036/+121 PCR product was ligated and inserted betweenHindIII and KpnI site of pXP2. pXP-762/+121, pXP-338/+121, pXP-92/+121 were generated by means of restriction enzyme digestion. PXP-200/+121, pXP-104/+121, pXP-72/+121, pXP-56/+121, and pXP-44/+121 were generated by PCR using P+121 as 3′-end primer and the following 5′-end primers: P-200, 5′-CCAAGCTTCGGTCCGGAGACCTGCT; P-104, 5′-CCAAGCTTGCTGATGTTCCGGGATC; P-72, 5′-TTAAGCTTGCCGCCGGGGAGGC; P-56, 5′-TTAAGCTTCGCTCCGCCCTTCC; P-44, 5′-CCAGCTTCCTGAGATCCTTATAT, respectively. Constructs generated by PCR were verified by sequencing. pXP-56mSp1 and pXP-104mEts were generated by PCR using P+121 as 3′-end primer and the following primers as 5′-end primers: P-56mSp1, 5′-TTAAGCTTCGCTTTGCCCTTCC; P-104mEts, 5′-CTAAGCTTGCTGATTAAGGGATCCGTGT. Specific mutation for the Sp1 site in the −104/+121 region was generated by PCR. Briefly, two simultaneous PCR reactions, using pXP-2700/+121 as template were performed. The first one used primer P-104 or P-104mEts as a 5′-end primer and PmSp1B (5′-GATCTCAGGAAGGGCAAAGCGACAGCCTCC) as a 3′-end primer; the second one used PmSp1A (5′-GGGAGGCTGTCGCTTTGCCCTTCCTGAGAT) as a 5′-end primer and P+121 as a 3′-end primer. Amplified fragments from each PCR reaction were purified, mixed, and subjected to a second round of PCR using two external primers (P-104 or P-104mEts with P+121). The amplified PCR product was inserted into pGEM-T vector (Promega) and verified by sequencing. The MRG1 promoter fragment was released by digestion withHindIII and KpnI and inserted into theHindIII and KpnI sites of pXP2. One day prior to transfection, NIH3T3 or Rat1 cells were seeded onto 6-well plates at 2 × 105 cells/well. Cells were transfected with 1 μg of reporter plasmid and 0.25 μg of pCMVβ (CLONTECH) using Superfect Transfection Reagents (Qiagen). 48 h after transfection, cell extracts were prepared, and luciferase activity was determined with Luciferase assay systems (Promega). β-Galactosidase assays were performed as described previously by Sambrook et al. (15Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: a Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989: 16.66-16.67Google Scholar). For transfection of HepG2 cells, cells were plated onto 24-well plates at 4.0 × 104 cells/well and transfected with 0.6 μg of the reporter plasmid and 0.2 μg of pCMVβ using the calcium phosphate method as described previously (15Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: a Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989: 16.66-16.67Google Scholar). The medium was changed 24 h after transfection. After another 24 h, cell extracts were prepared and luciferase and β-galactosidase assays were performed as described above. For transfection of SL2 cells, 1 day prior to transfection, cells were plated onto 6-well plates at 2.0 ×106 cells/well and transfected by the calcium phosphate method as described (16Dennig J. Beato M. Suske G. EMBO J. 1996; 15: 5659-5667Crossref PubMed Scopus (203) Google Scholar). Each well received 10 μg of DNA, including 5 μg of indicated luciferase reporter construct and varying amounts of expression plasmid such as pPacSp1 (17Coury A.J. Tjian R. Cell. 1988; 55: 887-898Abstract Full Text PDF PubMed Scopus (1069) Google Scholar), pPacUSp3 (16Dennig J. Beato M. Suske G. EMBO J. 1996; 15: 5659-5667Crossref PubMed Scopus (203) Google Scholar), pPacUEts-1, or pPacUElf-1 (18Karantzoulis-Fegara F. Antoniou H. Lai S.-L.M. Kulkarni G. D'Abreo C. Wong G.K.T. Miller L. Chan Y. Atkins J. Wang Y. Marsden P.A. J. Biol. Chem. 1999; 274: 3076-3093Abstract Full Text Full Text PDF PubMed Scopus (183) Google Scholar). Variable amounts of the expression plasmids were adjusted with the control plasmid pPacO. The medium was changed 24 h after addition of DNA, and the cells were harvested for luciferase assays 48 h after transfection. Luciferase values were normalized against total protein concentrations determined by protein assay (Bio-Rad). Nuclear extracts from Rat1 cells were prepared as described (19Dignam J.D. Lebovitz R.M. Roeder R.G. Nucleic Acids Res. 1983; 11: 1475-1489Crossref PubMed Scopus (9119) Google Scholar). Synthesized double-stranded oligonucleotides were end-labeled with [γ-32P]ATP and T4 polynucleotide kinase. Approximately 0.2 ng of the labeled oligonucleotide (20,000 cpm) was added to 10 μg of nuclear extracts in a final volume of 20 μl containing 1 μg of poly(dI-dC), 10 mm Tris-HCl (pH 7.5), 50 mm NaCl, 0.5 mm EDTA, 1 mm MgCl2, 4% glycerol, and 1 mm dithiothreitol and incubated for 15 min at room temperature. Subsequently, the DNA·protein complexes were separated from the free probe by electrophoresis through a 5% nondenaturing polyacrylamide gel. Gels were dried and subjected to autoradiography in the presence of intensifying screens (DuPont) at −80 °C. For supershift analysis, 1 μl of antibody was incubated with nuclear extracts at 4 °C for 1 h in binding buffer, followed by an additional incubation for 30 min at room temperature with labeled oligonucleotides. For competition analysis, unlabeled DNA was incubated with nuclear extracts at 4 °C for 20 min before the addition of the labeled probe. The oligonucleotides used as EMSA probes were annealed prior to labeling. The sequences of the upper strands of the oligonucleotides used were as follows: MRG1 Sp1 (−56/−42), 5′-TTAAGCTTCGCTCCGCCCTTCC; MRG1 Ets-1 (−101/− 85), 5′-GATGTTCCGGGATCCGT. The following oligonucleotides were used for competition analysis: mutated MRG1 Sp1 (−56/−42), 5′-TTAAGCTTCGCTTTGCCCTTCC; mutated MRG 1 Ets-1 (−101/−85), 5′-GATGTTAAGGGATCC; consensus ETS/PEA3 oligonucleotide, 5′-CTGAACTTCCTGCTCGAGATC (Santa Cruz Biotechnology); MRG1 −56/−34, 5′-TCGTCCCGCCTTCCTGAGATCC; and SIE (sis-interacting element), 5′-GTCGACATTTCCGTAA ATCTTGT (20Wagner B.J. Hayes T.E. Hoban C.J. Cochran B.H. EMBO J. 1990; 9: 4477-4484Crossref PubMed Scopus (550) Google Scholar). To understand the molecular basis for the ubiquitous expression of MRG1, we analyzed its promoter activity. A 5-kbHindIII fragment from a mouse MRG1 genomic clone, which contains 2.7 kb upstream sequence from the transcription start site and all of the coding regions, was subcloned into the pUC19 vector. Similar to the human MRG1 gene (11Leung M.-K. Jones T. Michels C.L. Livingston D.M. Bhattacharya S. Genomics. 1999; 61: 307-313Crossref PubMed Scopus (44) Google Scholar), the mouse gene consists of three exons, separated by two small introns (data not shown). The transcription start site was predicted according to the neural network promoter prediction (NNPP) program (13Reese M.G. Harris N.L. Eckman F.H. Large Scale Sequencing Specific Neural Networks for Promoter and Splice Site Recognition.Biocomputing: Proc..in: Lawrence Hunter Terri E.Klein 1996 Pacific Symposium. World Scientific Publishing Co., Singapore01-1996: 2-7Google Scholar). Potential regulatory elements have been identified through the Mat Inspector program (12Quandt K. Frech K. Karas H. Wingender E. Werner T. Nucleic Acids Res. 1995; 23: 4878-4884Crossref PubMed Scopus (2414) Google Scholar). These include C/EBP sites at −998 and −758; Sp1 sites at −748, −698, −648, −348, −178, and −51; AP4 site at 718; AP2 site at −218; Ets binding sites at −278, −97, and −44; MZF-1 site at −65; and an E-box at −148 (Fig. 1 A) (the corresponding GenBank™ accession number for mouse MRG1 gene is AF295547). Extensive identity exists between mouse and human proximal promoter regions (Fig. 1 B), introns and exons (data not shown), indicating that the MRG1 gene is well conserved during evolution. As shown in Fig. 1 B, there is an overall 80% identity between mouse and human MRG1 genes in the proximal promoter region between −190 and +121. Sequences further upstream are more diverse, suggesting −190 to +121 may play an important role in MRG1 expression. To assess the promoter activity of the MRG1 5′-flanking region, deletion mutants were generated using the firefly luciferase reporter vector, pXP2 (14Nordeen S.K. BioTechniques. 1988; 6: 454-457PubMed Google Scholar). Each resulting recombinant plasmid DNA was then transiently transfected into Rat1 fibroblasts with pCMVβ to monitor transfection efficiency. Among the constructs tested, pXP-104/+121 still had the full promoter activity, whereas pXP-44/+121 lost most of the promoter activity (Fig.2 A), indicating that the region between −104 and −44 contains critical elements for MRG1 expression. This region contains an Sp1 site (CCGCCC) at −51, an Ets-1 site (TTCC) at −97, and an MZF-1 site (GGGA) at −65. Rat1 cells were chosen, because we previously utilized this cell line to demonstrate that MRG1 expression can be induced by serum, insulin, and platelet-derived growth factor (10Sun H.B. Zhu Y.X. Yin T. Sledge G. Yang Y.C. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13555-13560Crossref PubMed Scopus (90) Google Scholar). The same experiments were also performed in NIH3T3, and HepG2 cells, and the results were similar, consistent with the ubiquitous expression of MRG1 in different cell types (data not shown). We then constructed more detailed deletions within this region (Fig.2 B). As compared with pXP-104/+121, construct pXP-92/+121 with only 12-bp sequence deleted at the 5′-region had a significantly lower promoter activity. This 12-bp region contains an Ets-1 binding site. pXP-56/+121 still had about 26% promoter activity of pXP-104/+121. However, further deletion of 12 base pairs from −56/+121 (pXP-44/+121), which contains the Sp1 binding site, greatly reduced the promoter activity. Similarly, mutations of the Sp1 site in pXP-56/+121 from CCGCCC to TTGCCC decreased the promoter activity to the level of pXP-44/+121 (Fig. 2 B). It should also be noted that deletion of the region between −92 and −57 also resulted in a significant decrease in the promoter activity. This region contains an MZF-1 binding site (GGGA sequence at −66 to −62). These results demonstrated that the Ets-1 site, Sp1 site, and the region between −92 and −56 are important for the mouse MRG1 promoter activity. To further define the specific elements in the proximal −104 region that contribute to the proximal promoter activity, we generated a series of MRG1 promoter constructs with mutations in Ets-1 and Sp1 elements from pXP-104/+121. Ets-1 motif TTCC was mutated to TTAA, whereas Sp1 motif CCGCCC was mutated to TTGCCC. Upon transfection into Rat1 cells, constructs containing a specific mutation of Sp1 or Ets-1 element resulted in about 90 and 50% decreases in the promoter activity, respectively (Fig.3). The results were similar in the NIH3T3 fibroblast cell line. In HepG2, disruption of the Sp1 site decreased 75% of the promoter activity. In all cases, mutations of both sites resulted in a 90–98% loss of the proximal promoter activity (Fig. 3), demonstrating that both Ets-1 and Sp1 elements are important for the MRG1 proximal promoter activity. To determine the biochemical composition of protein complexes binding to the proximal Sp1 or Ets-1 element of the MRG1 promoter, nuclear extracts from Rat1 fibroblast cells were analyzed by EMSA using a labeled Sp1 or Ets-1 oligonucleotide probe. As illustrated in Fig.4 A, three DNA·protein complexes (A, B, and C witharrows) were detected using labeled Sp1 probe (lane 2). These DNA·protein complexes were specific, because they disappeared in the presence of excess amounts of unlabeled Sp1 competitors (lanes 3 and 4), and inclusion of an excess amount of unlabeled mutated Sp1 oligonucleotide did not affect the formation of these complexes (lane 5). Because members of the Sp1 multigene family share the same binding consensus sequence (GC or GT box) (21Suske G. Gene. 1999; 238: 291-300Crossref PubMed Scopus (973) Google Scholar), we performed supershift experiments to determine the identity of the three complexes. As shown in Fig. 4 B, anti-Sp1 antibody inhibited complex A formation (lane 2), whereas complexes B and C were supershifted in the presence of anti-Sp3 antibody (lane 3). In the presence of both antibodies, complexes B and C were supershifted and the intensity of complex A decreased (lane 4). Anti-STAT3 was included as a negative control and was found not to affect the mobility of these complexes (lane 5). The gel shift pattern was similar when performed using nuclear extracts from NIH3T3 cells (data not shown). These results indicated that Sp1 is a component of complex A and complexesB and C contain Sp3, a typical gel shift pattern as previously described for the Sp1 site (21Suske G. Gene. 1999; 238: 291-300Crossref PubMed Scopus (973) Google Scholar). The potential interaction of the Ets-1 element with nuclear proteins was also evaluated by EMSA (Fig. 4 C). Transcription factors in the Ets family recognize the sequence (A/G)(T/A)(G/A)(A/T)TCC(G/T)(G/C)Y with a similar structure in the DNA binding domain. The TCC sequence is absolutely essential for Ets family members to bind DNA, with the specificity between different members determined by surrounding nucleotides (22Wasylyk B. Hahn S.L. Giovane A. Eur. J. Biochem. 1993; 211: 7-18Crossref PubMed Scopus (807) Google Scholar, 23Sharrocks A.D. Brown A.L. Ling Y. Yates P.R. Int. J. Biochem. Cell Biol. 1997; 29: 1371-1387Crossref PubMed Scopus (276) Google Scholar). The −104/+121 region contains two Ets core consensus sequences: one at −97 to − 94 (most closely matched Ets-1 recognition sequence) and the other one at − 45 to −42. With MRG1 Ets-1 element (−101/−85) as a probe, three complexes were formed (lane 1) and could be competed by unlabeled −101/−85 oligonucleotide (lane 2), suggesting they are specific. Conversely, unlabeled oligonucleotide bearing a mutated Ets-1 element (TTCC mutated to TTAA) was unable to compete for binding (lane 3), indicating the core TTCC motif is critical for the formation of DNA·protein complexes. However, the unlabeledsis-interacting element (SIE) oligonucleotide (20Wagner B.J. Hayes T.E. Hoban C.J. Cochran B.H. EMBO J. 1990; 9: 4477-4484Crossref PubMed Scopus (550) Google Scholar), containing a core TTCC motif, could not compete for the complex formation (lane 6), suggesting that flanking sequences are also important for the formation of DNA·protein complexes. Interestingly, the other TTCC-containing oligonucleotide from the MRG1 promoter (−56 to −34) could compete for the formation of complexA but not complexes B and C. Mutations of the TTCC motif in region −56 to −34 did not significantly affect the promoter activity in Rat1 cells (data not shown). Most importantly, the addition of a 100-fold excess of unlabeled Ets/PEA3 binding oligonucleotide (lane 5) inhibited the formation of all three complexes almost as effectively as did MRG1–101/−85 oligonucleotide. These results strongly suggested that −101/−85 contains an Ets-1 binding site. However, addition of monoclonal antibodies directed against various members of Ets factor family (Ets-1, Ets-2, PU.1, Erg-1, Fli-1, and Elk-1) did not change the gel shift pattern (data not shown). The exact identity of proteins in each of the DNA·protein complexes therefore remains to be determined. To directly determine whether Sp1- and Ets-related proteins could functionally modulate MRG1 promoter activity, Drosophila SL2 cells, which are deficient in Sp1-, Sp3-, and Ets-related proteins (16Dennig J. Beato M. Suske G. EMBO J. 1996; 15: 5659-5667Crossref PubMed Scopus (203) Google Scholar, 17Coury A.J. Tjian R. Cell. 1988; 55: 887-898Abstract Full Text PDF PubMed Scopus (1069) Google Scholar) were utilized. The reason for using insect instead of mammalian cells is because Sp1-, Sp3-, and Ets-related factors are expressed in virtually all mammalian cells, which could affect the interpretation of cotransfection experiments. We introduced MRG1 proximal promoter construct pXP-104/+121 along withDrosophila expression vectors pPacSp1 or pPacUSp3 intoDrosophila SL2 cells. As shown in Fig.5 A, both pPacSp1 and pPacUSp3 stimulated pXP-104/+121 promoter in a dose-dependent manner, with pPacSp1 being the more potent activator. Stimulation was about 35-fold with 2.5 μg of pPacSp1. To further investigate whether Sp3 could enhance transcriptional activation by Sp1, pXP-104/+121 (5 μg), pPacSp1 (1 μg), and pPacSp3 (1–4 μg) were cotransfected into Drosophila SL2. As shown in Fig. 5 B, transcriptional activation by Sp1 could clearly be enhanced by Sp3. Moreover, when the Sp1 binding site was mutated, no potentiation was observed with Sp1, Sp3, or the combination of Sp1 and Sp3. These results indicated that the Sp1 element is required for mediating the transactivation effect of both Sp1 and Sp3 on the MRG1 promoter (Fig.5 C). To assay the functional effect of Ets-related proteins on MRG1 expression, pXP-104/+121 or pXP-104mEts-1 (5 μg) was introduced intoDrosophila SL2 cells together with the Drosophilaexpression plasmids for Ets factors, pPacUEts-1 (1 μg), o