Dynamic Transcriptional Regulatory Complexes, Including E2F4, p107, p130, and Sp1, Control Fibroblast Growth Factor Receptor 1 Gene Expression during Myogenesis*

Developmentally controlled transcriptional regulation of myogenic cell proliferation and differentiation via expression of the fibroblast growth factor receptor 1 (FGFR1) gene is positively regulated by Sp1 and negatively regulated by E2F4-based transcriptional complexes. We report that p107 and p130 formed transcriptional complexes with E2F4 on the FGFR1 promoter and repressed FGFR1 gene transcription in myogenic cells. However, in Drosophila melanogaster SL2 cells, only p107 was able to repress Sp1-mediated transactivation of the FGFR1 promoter. Gel shift assays using transfected myoblast nuclear extracts showed that ectopic p107 reduced Sp1 occupancy of the proximal Sp binding site of the FGFR1 promoter, and coimmunoprecipitation studies indicated that Sp1 interacts with p107 but not with p130. Gel shift assays also demonstrated that Sp1 interacted with p107 in E2F4-p107 transcriptional complexes in myoblasts. The nature of the repressor transcriptional complex was altered in differentiated muscle fibers by the relative loss of the E2F4-p107-Sp1 transcription complex and replacement by the repressor E2F4-p130 complex. These findings demonstrate that activation and repression of FGFR1 gene transcription is governed by interplay between Sp1, p107, p130, and E2F4 in distinct transcriptional complexes during skeletal muscle development.


From the Department of Cell Biology and Anatomy, Chicago Medical School, North Chicago, Illinois 60064
Developmentally controlled transcriptional regulation of myogenic cell proliferation and differentiation via expression of the fibroblast growth factor receptor 1 (FGFR1) gene is positively regulated by Sp1 and negatively regulated by E2F4-based transcriptional complexes. We report that p107 and p130 formed transcriptional complexes with E2F4 on the FGFR1 promoter and repressed FGFR1 gene transcription in myogenic cells. However, in Drosophila melanogaster SL2 cells, only p107 was able to repress Sp1-mediated transactivation of the FGFR1 promoter. Gel shift assays using transfected myoblast nuclear extracts showed that ectopic p107 reduced Sp1 occupancy of the proximal Sp binding site of the FGFR1 promoter, and coimmunoprecipitation studies indicated that Sp1 interacts with p107 but not with p130. Gel shift assays also demonstrated that Sp1 interacted with p107 in E2F4-p107 transcriptional complexes in myoblasts. The nature of the repressor transcriptional complex was altered in differentiated muscle fibers by the relative loss of the E2F4-p107-Sp1 transcription complex and replacement by the repressor E2F4-p130 complex. These findings demonstrate that activation and repression of FGFR1 gene transcription is governed by interplay between Sp1, p107, p130, and E2F4 in distinct transcriptional complexes during skeletal muscle development.
Fibroblast growth factor receptors (FGFRs) 1 have diverse functional roles in mitogenesis, angiogenesis, cell migration, differentiation, mesoderm induction, bone growth and limb development (1). In skeletal muscle, FGFR1 mediates the mitogenic activity initiated by FGF1 and FGF2. During skeletal myogenesis, FGFR1 gene expression is positively regulated in proliferating myoblasts and negatively regulated in differentiated muscle fibers. The functional significance of regulated expression of the FGFR1 gene during myogenesis is demonstrated by overexpression in vivo. Chick embryos overexpressing wild-type FGFR1 displayed delayed myoblast differentiation and muscle fiber formation. On the contrary, chick embryos overexpressing a dominant-negative form of FGFR1 displayed premature muscle fiber formation with decreased muscle mass (2,3).
Although regulation of FGFR1 gene expression is important for normal growth and development of skeletal muscles, the molecular mechanism governing its transcription is poorly understood. Positive regulation of FGFR1 gene expression in proliferating myoblasts is governed by the Sp1 transcription factor. The chicken FGFR1 promoter contains two functional, distal Sp1 binding sites, and the proximal region contains three Sp1 binding sites, all of which are essential for full promoter activity in proliferating myoblasts (4,5). Negative transcriptional regulators control FGFR1 promoter activity as FGFR1 gene expression declines during myogenic differentiation. We recently identified E2F4 as a negative regulator of FGFR1 gene expression in skeletal muscle cells (6). Its repressor activity was mediated by E2F4 binding to a proximal cis-element at Ϫ65 bp. However, E2F4 was present in both myoblast and muscle fiber nuclear extracts and bound to the E2F site. Therefore, the mechanism of E2F4-mediated repression of FGFR1 promoter activity in a cell and developmentally regulated manner could not be explained by E2F4-mediated repression alone.
Members of the E2F family of transcriptional regulators functionally interact with the pocket protein transcription factors, p107, p130, and pRb. The nature of these interactions defines the transcriptional regulatory complexes as activators or repressors. These complexes regulate expression of a variety of genes, many of which are associated with cell cycle regulation (7). E2F1, E2F2, and E2F3 specifically interact with pRb and not p107 or p130 in vivo (8,9). E2F5 binds specifically to p130 in vivo (10,11). In contrast, E2F4 associates with pRb, p107, and p130 and comprises the majority of E2F-pocket protein complexes in vivo (12,13). Primary embryonic mouse fibroblasts that lack E2F4 and -5 or their associated pocket proteins are defective in their ability to exit the cell cycle (14). Repressive E2F-pocket protein complexes are also required for normal development in vivo. Mice lacking E2F4 or -5 display developmental defects resulting in neonatal lethality (15,16).
p107 and p130 expression levels and E2F binding activities are differentially regulated during cell cycle progression and withdrawal from the cell cycle (17)(18)(19)(20). Proliferating human fibroblasts have high levels of p107 and E2F4-p107 complexes and very low levels of p130 (18,19). Levels of p107 decline as cells exit the cell cycle, whereas p130 increases until p130-E2F4 complexes predominate (21). p107 and p130 also interact with other cell cycle-associated proteins. In muscle fibers, p107 and p130 associate with cyclin D3, levels of which increase during myogenesis (20). Formation of these multimeric complexes has been suggested to play a role in control of cyclinE-cdk2 activity in differentiated cells and prevent cell cycle reentry (7).
We demonstrated in this study that pocket protein family members p107 and p130 regulate E2F4-mediated repression of FGFR1 gene expression in skeletal muscle cells. Transient transfections in proliferating myoblasts and electromobility shift assays indicated that these proteins form functional complexes with E2F4 and repress FGFR1 gene promoter activity. We further demonstrate that in proliferating myoblasts, Sp1 interacts with p107 and masks the transcriptional repressor activity of the E2F4-p107 complex at the E2F site located in the proximal promoter region of the FGFR1 gene. These findings provide a novel mechanism for developmentally regulated expression of the FGFR1 gene during skeletal myogenesis.

EXPERIMENTAL PROCEDURES
Plasmid DNA Constructs-The plasmid m23/m42/m54FGFR1 containing mutations of three Sp1 sites in the proximal promoter region was described previously (5). Mutation of the E2F site in the above plasmid was made by using the QuikChange site directed mutagenesis kit (Stratagene) (6) with the following forward primer and its antisense oligonucleotide as a reverse primer 5Ј-GGTTCCCATGCAGCTATCATA-CAGGGGGTTAACCTGCAC-3Ј. The base substitutions are underlined. Mutagenesis was confirmed by DNA sequencing, and the plasmid DNA was designated as m23/m42/m54/m65FGFR1. The new truncated promoter versions of plasmid m23/m42/m54FGFR1 and m23/m42/m54/ m65FGFR1 were prepared by deletion of a distal 2.2-kb SacI fragment and ligation to generate the plasmids mSp11058FGFR1 and mE2F/ Sp11058FGFR1, respectively. The truncated plasmid 1058FGFR1 containing wild-type Sp1 and E2F promoter sequences was described previously (4).
The plasmid 3284FGFR1CAT containing the full-length wild-type FGFR1 promoter was described previously (35). Myoblasts were plated at a density of 2.5 ϫ 10 6 cells/6-cm dish in FM medium and transfected using Lipofectamine Plus reagent (Invitrogen) as described previously (5). For all transfections, 1.5 or 3 g of 3284FGFR1CAT plasmid and 1 g of pRSV␤GAL plasmid were kept constant. Increasing amounts of p107 or p130 expression plasmids, pCMVp107 and pCMVp130 encoding human cMyc-tagged p107 and hemagglutinin (HA) tagged p130, respectively, and vector DNA alone were added to bring total DNA to 8 g. Some cultures were transfected with 1.5 g of pCMVTAG (Stratagene). Transfected cells were harvested after 24 h and 10 days in culture and suspended in 100 l of 0.25 M Tris-HCl, pH 7.8. Cells were lysed by three rounds of freezing and thawing, and the CAT assays were performed as described previously (35). Resulting thin layer chromatography plates were then cut and quantified by liquid scintillation counting. ␤-Galactosidase activities were used to normalize transfection efficiencies (36). The CAT activities from expression of wild-type 3284FGFR1CAT co-transfected with increasing amounts of pCMVp107 and pCMVp130 were expressed as percentage activities relative to FGFR1 promoter activity in the absence of exogenous p107 or p130.
For transfection of SL2 cells, cells were grown in Drosophila melanogaster SFM medium (Invitrogen) supplemented with 2 mM L-gluta-mine plus 1ϫ antibiotic-antimycotic (Invitrogen) at 25°C without CO 2 . Cells were plated at a density of 2.5 ϫ 10 6 cells/6-cm dish on the day before transfection. Cells were transfected in D. melanogaster SFM medium without antibiotics using Cellfectin reagent (Invitrogen) according to the manufacturer's protocol. Each plate received various amounts of pCMVp107 and pCMVp130 expression plasmids, 1 g of pRSV␤GAL, 750 ng of Sp1 expression plasmid (pPacSp1), 1.5 or 3 g of promoter-CAT reporter plasmid as well as pKS BlueScript (Stratagene) DNA to bring the total DNA up to 8 g per plate. After transfection, cells were maintained in growth medium. Cells were harvested 48 h after transfection, and CAT assays were performed (35). For all transfection assays, four to five independent experiments were performed.
Immunocytochemistry-Rabbit anti-E2F4, p107, p130, Sp1, E47, G␣q, and HA primary antibodies and horseradish peroxidase-conjugated anti-mouse IgG and anti-rabbit IgG were obtained from Santa Cruz Biotechnology Inc. Mouse anti-pRb was obtained from BD Transduction Laboratories. Mouse anti-cMyc and ␤-actin antibodies were obtained from Invitrogen and Abcam, respectively. DAPI was obtained from Molecular Probes, Inc. Fluorescein-conjugated anti-mouse IgG and anti-rabbit IgG were obtained from Vector Laboratories, Inc.
Myogenic cultures were immunostained 24 h or 10 days after plating. Cells were washed three times with phosphate-buffered saline (PBS) and fixed with 100% methanol for 5 min. Cultures were washed as above and incubated for 1 h with blocking solution (BS) containing 0.3% bovine serum albumin and 0.1% Tween in PBS. Polyclonal anti-rabbit E2F4, p130, p107, and Sp1 antibodies and pRb monoclonal antibody were diluted 1:700 in BS. Cultures were incubated in primary antibodies for 1 h at room temperature and washed as above. Fluoresceinconjugated anti-mouse IgG and anti-rabbit IgG were diluted 1:100 in BS and added to the cultures for 1 h. DAPI at a concentration of 1.2 M in PBS was added to the cultures for 10 min at room temperature and washed five times with PBS. Two drops of 2.5% diazabicyclooctane in glycerol/PBS (9:1) and coverslips were applied, and the cells were viewed with fluorescence microscopy.
Subcellular Fractionation and Western Blotting-Fractionation of cultured cells was performed as described previously (6). For Western blots, proteins (50 g) were resolved in 7.5% SDS-polyacrylamide gels and electrotransferred onto nitrocellulose membranes. The blots were then blocked for 1 h at 37°C or overnight at 4°C in BS containing 5% nonfat dry milk in PBS with 0.05% Tween. Thereafter, the blots were incubated with rabbit polyclonal p107, p130, Sp1, E47, cMyc, or HA antibody and mouse monoclonal pRb antibody, diluted 1:1000 in BS for 1 h at room temperature, washed three times with PBS containing 0.05% Tween, and incubated with horseradish peroxidase-conjugated anti-rabbit IgG as a secondary antibody (1:4000 dilution) for 1 h at room temperature. The blots were washed five to six times in PBS containing 0.05% Tween, and the immunocomplexes were detected using the Super Signal chemiluminescent substrate (Pierce) and x-ray film.
Preparation of Nuclear Extracts and Electromobility Shift Assay-Nuclear extracts were prepared as described previously (37). Complementary oligonucleotides were commercially synthesized (Integrated DNA Technology). The oligonucleotides (10 g each) were boiled for 10 min in 10ϫ kinase buffer and allowed to cool gradually to room temperature. Approximately 200 ng of double-stranded oligonucleotide were 5Ј end-labeled using T4 kinase (Promega) and [␥-32 P ]ATP (MP Biomedicals). The sequence of the wild-type E2F4 binding site oligonucleotide was 5Ј-CATGCAGCAGCGGGACAGGGGGCTG-3, and the mutated E2F4 binding site oligonucleotide was 5Ј-CCATGCAGCTATCAT-ACAGGGGGCTG-3Ј. Base substitutions are underlined. The Ϫ23 Sp1 oligonucleotide sequence was 5Ј-GACTCTCTTTCTCCCCTCCACAGC-TC-3Ј, and the consensus Sp1 oligonucleotide sequence was 5Ј-ATTC-GATCGGGGCGGGGCGAGC-3Ј.
Nuclear extract proteins (12 g) were added to a 20-l binding reaction containing 20 mM HEPES, pH 7.9, 1 mM MgCl 2 , 4% Ficoll, 0.5 mM dithiothreitol, 50 mM KCl, 2 g of poly dIdC (Amersham Biosciences), 300 g/ml bovine serum albumin, and 50 g/ml salmon sperm DNA and incubated at 4°C for 30 min. For supershifts, 2 g of the antibody was added to the reaction mixture. Labeled probe (70,000 cpm in 1 l) was added and further incubated for 25 min at 4°C. Protein-DNA complexes were resolved by electrophoresis in 5% non-denaturing polyacrylamide gels with 1ϫ Tris-borate/EDTA buffer. The gels were dried and exposed to x-ray film overnight.
Transient Electromobility Shift Assay-Transfections of D. melanogaster SL2 cells were performed as described above using Cellfectin reagent (Invitrogen). Each plate received various combinations of transcription factor expression constructs. One set of plates received 1 g of Sp1 expression plasmid (pPacSp1), 3 g of E2F4 expression plasmid (pCMVE2F4), and 3 g of p107 expression plasmid (pCMVp107). The other set received only 1 g of pPacSp1 and 3 g of pCMVp107, and the total amount of DNA transfected was adjusted to 7 g with pKS Blue-Script DNA (Stratagene). Forty-eight hours after transfection, nuclear extracts were prepared (38), and the gel shifts were performed as described above with 3 g of nuclear extracts.
Immunoprecipitations-Nuclear extracts were prepared as described for electromobility shift assays (37). Extracts from ϳ4 ϫ 10 7 cells (100 l containing 200 g of protein) were diluted to 1 ml with chilled NET buffer (150 mM NaCl, 0.1% Nonidet P-40, and 50 mM Tris-HCl, pH 7.5) containing protease inhibitors. The nuclear lysate was precleared by incubation with 50 l of protein A/G agarose (Santa Cruz Biotechnology) for 2 h at 4°C and further centrifugation. The nuclear protein was incubated with 2 g of antibody against E2F4, pRb, p107, p130, and Sp1 (Santa Cruz Biotechnology) overnight at 4°C. After addition of 50 l of protein A/G agarose, the suspension was incubated for another 2 h at 4°C. The beads were pelleted by centrifugation, washed three times with cold NET buffer, and resuspended in 50 l of 2ϫ SDS sample buffer. After the suspension was heated to 95°C for 10 min, 20-l samples were resolved in denaturing SDS-polyacrylamide gels, transferred to nitrocellulose membrane, and probed for the presence of specific proteins by immunodetection as described above.
Chromatin Immunoprecipitation-We performed chromatin immunoprecipitations using a modification of previously published methods (39,40). Formaldehyde was added to a final concentration of 1% directly to the cell culture media of proliferating myoblasts and differentiated myotubes (4 ϫ 10 7 cells). Fixation proceeded at room temperature for 10 min and was stopped by the addition of glycine to a final concentration of 0.125 M. Cells were washed with PBS, trypsinized, and harvested in PBS containing 10% horse serum. Cells were collected by centrifugation and rinsed in cold PBS. The cell pellets were resuspended in swelling buffer (10 mM potassium acetate, 15 mM magnesium acetate, 0.1 M Tris, pH 7.6, 0.5 mM phenylmethylsulfonyl fluoride, and 100 ng/ml leupeptin and aprotinin) and incubated on ice for 20 min, and then processed on a Dounce homogenizer. The nuclei were collected by microcentrifugation and then resuspended in sonication buffer (1% SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1, 0.5 mM phenylmethylsulfonyl fluoride, and 100 ng/ml leupeptin and aprotinin) and incubated on ice for 10 min. Before sonication, 0.1 g of glass beads (212-to 300-m diameter; Sigma) was added to each sample. The samples were sonicated on ice with an Ultra sonicator at setting 3 for four 20-s pulses to an average length of 0.5 to 1.0 kb and then microcentrifuged.
Chromatin was pre-cleared with a mixture of protein A and protein G Sepharose (blocked previously with 1 mg/ml salmon sperm DNA and 1 mg/ml bovine serum albumin) at 4°C for 4 h two times. Precleared chromatin was equally divided and was separately incubated with 4 g of anti-E2F4, anti-p107, anti-p130, anti-Sp1, and G␣q antibodies, or no antibodies overnight at 4°C. Immunoprecipitation, washing, and elution of immune complexes were carried out as described previously (41). Before the first wash, 20% of the supernatant from the reaction with no primary antibody for each sample (myoblasts and myotubes) was saved as total input chromatin and was processed with the eluted immunoprecipitates beginning at the cross-link reversal step. Cross-links were reversed by the addition of NaCl to a final concentration of 200 mM, and RNA was removed by the addition of 10 g of RNase A per sample followed by incubation at 65°C for 4 to 5 h. The samples were then ethanol precipitated. The samples were resuspended in 100 l of Tris-EDTA, pH 7.5, 25 l of 5ϫ proteinase K buffer (1.25% SDS, 50 mM Tris, pH 7.5, and 25 mM EDTA), and 10 g of proteinase K (Sigma) and incubated at 45°C for 2 h. Samples were extracted with phenol-chloroform and ethanol-precipitated. The pellets were collected by microcentrifugation, resuspended in 30 l of water, and analyzed by PCR.
PCR mixtures contained 2 l of immunoprecipitate or 2 l of a 1:200 dilution of the total sample, 50 ng of each primer, 25 mM MgCl 2 , 2.5 mM dNTP mix, and 2.5 units of TaqDNA polymerase in a total volume of 100 l. After 25 cycles of amplification, the PCR products were run in 1% agarose gels and analyzed by ethidium bromide staining. The PCR primers used to amplify the Ϫ321 to ϩ9 region were: 5Ј CTGTTT-TCAGTGCCAACT 3Ј (forward primer, Ϫ321 to Ϫ302) and 5Ј CAT-GGGGCCCCGTCGGCCGCTG 3Ј (reverse primer, ϩ9 to Ϫ13). The primers used to amplify the Ϫ1950 to Ϫ1788 region were: 5Ј CCGCGT-GAACAAGTGCTTCTTTG 3Ј (forward primer, Ϫ1950 to Ϫ1927) and 5Ј GGGAAAGGGTTCATTGGGAAAC 3Ј (reverse primer, Ϫ1788 to Ϫ1809).
For transient chromatin immunoprecipitation assays (42), fetal chick myoblasts (embryonic day 13) were plated at a density of 2.5 ϫ 10 6 cells/10 cm and were transfected with 1.5 g of the 1058FGFR1 wildtype promoter construct and its mutant derivatives by using Lipofectamine Plus reagent (Invitrogen) as described previously (5). After a 24-h incubation, the cells were cross-linked with the use of formaldehyde and harvested, and chromatin immunoprecipitations were performed as described above. The resulting DNA was analyzed by PCR reactions with a forward primer (Ϫ321 to ϩ9 bp) common to all constructs and a reverse primer to the pCAT3Basic (Promega) plasmid backbone (5Ј-GATATATCAACGGTGGTATATCCAGTG-3Ј) at ϩ351 to ϩ325 bp.

Cellular Localization of Pocket Proteins in Chicken Muscle
Cells-We have recently shown that E2F4, present in the nuclei of both myoblasts and myotubes, acts as a negative regulator of FGFR1 gene expression (6). To identify potential E2F4 binding partners within nuclei, we analyzed the intracellular distribution of pocket proteins in undifferentiated myoblasts and differentiated myotubes by immunocytochemistry. p107, p130, and pRb proteins were localized to nuclei of both myoblasts and myotubes (Fig. 1). Immunofluorescent localization of these proteins in the nuclei of myoblasts and myotubes colocalized with 4,6-diamidino-2-phenylindole fluorescence.
Western blot analysis was carried out using fractionated cell lysates of both myoblasts and myotubes to validate the immunostaining results (Fig. 2). p107, p130, and pRb proteins were detected in the nuclear extracts from both myoblasts and myotubes. These proteins were not detected in cytoplasmic extracts. Whereas p107 was abundant in myoblasts, pRb was abundant in myotubes. Relative mobility differences of pRb proteins in extracts from myoblasts and myotubes may have been caused by differences in phosphorylation states. The relative abundance of p130 was approximately the same in myoblasts and myotubes.
p107 and p130, but Not pRb, Formed Complexes at an E2F4 Binding Site-To determine whether p107, p130, and/or pRb from chicken muscle cells interact with E2F4, we performed gel supershift assays using myoblast and myotube nuclear extracts, the E2F4 binding site of the FGFR1 promoter, and antibodies directed against p107, p130, pRb, and E2F4 (Fig. 3). Nuclear extracts from myoblasts and myotubes formed a protein-DNA complex with the FGFR1 E2F4 binding site, which was supershifted by the E2F4 antibody. Addition of p107 and p130 antibodies in the reactions produced a supershifted protein-DNA complex, indicating the presence of these transcrip-FIG. 1. Intracellular distribution of endogenous p107, p130, and pRb in chicken muscle cells. Embryonic day 13 chicken myoblasts were cultured for either 1 or 10 days. Cultures maintained for 10 days were incubated in medium containing 10 g/ml cytosine arabinoside from days 4 to 10. Cells were immunostained with rabbit polyclonal antibodies against p107 and p130 and mouse monoclonal pRb antibody followed by fluorescein isothiocyanate-conjugated secondary antibodies. Nuclei were visualized by 4,6-diamidino-2-phenylindole (DAPI) staining. Cells were also incubated in fluorescein isothiocyanate-conjugated secondary antibody alone (No 1 o Ab) to assess nonspecific, background immunostaining. p107, p130, and pRb were nuclear localized in myoblasts and myotubes. tion factors in the E2F4-DNA complex. p107 was a prominent component of the E2F4-DNA complex derived from myoblast nuclear extracts and declined in the E2F4-DNA complex derived from myotube nuclear extracts. In contrast, p130 was a prominent component of the E2F4-DNA complex in myotubes relative to the p107 supershifted complex in myotubes. Inclusion of the pRb antibody in the binding reactions did not cause a supershift, indicating its absence in the E2F4-DNA complex formed by both myoblast and myotube nuclear extracts. These results indicate that p107 and p130 bind to the FGFR1 promoter at the E2F4 binding site and suggest that they have an important role in the developmentally regulated transcription of the FGFR1 gene in skeletal muscle cells.
p107 and p130 Repress FGFR1 Promoter Activity in Proliferating Myoblasts-We next assessed the function of p107 and p130 in the transcriptional regulation of the FGFR1 gene in myoblasts. The wild-type FGFR1 promoter-reporter construct 3284FGFR1CAT and the separate expression plasmids pC-MVp107 and pCMVp130, encoding p107 and p130, respectively, were transfected into myoblasts. Exogenously expressed p107 and p130 were detected by Western blot analysis using cMyc and HA antibodies, respectively, directed against the expressed fusion proteins. p107 and p130 proteins repressed FGFR1 promoter activity in a dose-dependent manner (Fig. 4). Transfection of 1.5 g of pCMVp107 reduced FGFR1 promoter activity to basal levels, whereas transfection of 1.5 g of pC-MVp130 reduced promoter activity to ϳ50%. Increasing amounts of pCMVp130 up to 6 g did not further reduce promoter activity (data not shown). Repression of FGFR1 promoter activity by pCMVp107 and pCMVp130 was not caused by the presence of the strong, constitutive CMV promoter because transfection of the cloning vector pCMVTAG did not reduce FGFR1 promoter activity (Fig. 4B). These results indicate that p107 is a strong repressor of FGFR1 promoter activity and that p130 functions as a repressor but is relatively less effective.
p107 Represses Sp1-mediated Activation of the FGFR1 Promoter-To assess the transcriptional effects of p107 and p130 on Sp1-mediated transactivation of FGFR1 promoter activity, the Sp1 expression construct pPacSp1 was co-transfected with pCMVp107 or pCMVp130 along with the wild-type 3284FGFR1CAT construct into D. melanogaster SL2 cells. These cells have no detectable levels of endogenous Sp transcription factors (43). As shown in Fig. 5, expression of Sp1 significantly increased FGFR1 promoter activity, in agreement with our previous observations (5). Co-transfection of pC-MVp107 repressed FGFR1 promoter activity to ϳ48% in a dose-dependent manner (Fig. 5A). In contrast, co-transfection of pCMVp130 had no effect on Sp1-mediated FGFR1 promoter activity (Fig. 5B). These results suggest that a p107-Sp1 interaction may regulate FGFR1 promoter activity in myoblasts.
Exogenous p107 Interferes with Sp1 Binding to the Proximal Region of the FGFR1 Promoter-To gain further insight into the mechanism by which p107 represses Sp1-mediated transcriptional activation of the FGFR1 gene, we first determined whether p107 interfered with Sp1 binding to FGFR1 promoter sequences. The proximal E2F4 binding site of the FGFR1 promoter located at Ϫ65 bp is near the three proximal Sp1 binding sites, located between Ϫ65 and Ϫ23 bp from the transcription start site (Fig. 6A). We have shown previously that the most proximal (Ϫ23 bp) Sp1 binding site conferred the strongest transcriptional activation to the FGFR1 promoter (5). Using the Ϫ23 bp Sp1 binding site sequence, we performed gel shift assays with nuclear extracts prepared from myoblasts transfected with 3 g of pCMVp107 and pCMVp130 expression plasmids (Fig. 6B). Overexpression of p107 reduced the formation of the Sp1-DNA complex. Similar results were obtained using the consensus Sp1 binding site oligonucleotide. However, overexpression of p130 did not interfere with Sp1 binding to the Ϫ23 bp Sp1 site of the FGFR1 promoter or to the Sp1 consensus oligonucleotide. These results indicate that p107 antagonizes Sp1-mediated transcriptional activation of FGFR1 gene expression by inhibiting its occupancy of the proximal Sp binding site of the FGFR1 promoter.
Exogenous p107 Physically Interacts with Sp1 in Proliferating Myoblasts-We hypothesized that repression of Sp1-mediated transcription of the FGFR1 promoter by p107 was caused by direct protein-protein interaction. To determine whether exogenous p107 physically interacted with endogenous Sp1 independent of its binding to the FGFR1 promoter, we performed coimmunoprecipitation assays using nuclear extracts prepared from myoblasts transfected with the pCMVp107 expression plasmid. For comparison, we also used nuclear extracts from myoblasts transfected with pCMVp130 expression plasmid. The expression of each construct was examined by Western blot analysis using anti-cMyc and anti-HA antibodies FIG. 2. Western blot analyses of p107, p130, and pRb subcellular distribution in myoblasts (MB) and myotubes (MT). Cell extracts were fractionated and equal amounts (50 g) of cytoplasmic and nuclear extracts were resolved by 7.5% SDS-PAGE. Proteins were transferred to nitrocellulose membranes and detected with specific rabbit polyclonal antibodies for p107 and p130 and a mouse monoclonal antibody for pRb followed by horseradish peroxidase-conjugated secondary antibodies and chemiluminescence. Detection of ␤-actin in cytoplasmic fractions served only as controls for protein loading of cytoplasmic fractions and subcellular fraction of cytoplasmic versus nuclear proteins.
FIG. 3. p107 and p130 but not pRb form complexes with E2F4 at the E2F binding site of the FGFR1 promoter. Gel shift and supershift assays were performed with the FGFR1 E2F binding site, nuclear extracts from myoblasts and myotubes, and E2F4, p107, p130, and pRb antibodies. Lanes 1 and 7 contain probe alone. Lanes 2-6 contain protein-DNA complexes formed with the FGFR1 E2F oligonucleotide and myoblast nuclear extract. Lanes 8 -12 contain myotube nuclear extract. Antibodies added to the reactions are indicated. The arrow indicates the position of the specific E2F4 complexes formed from both myoblast and myotube nuclear extracts, and the arrowhead indicates the position of the supershifted E2F4 complex.
to detect cMyc-p107 and HA-p130 fusion proteins. Detection of the fusion proteins was restricted to extracts from transfected cells (Fig. 7A). cMyc antibody coimmunoprecipitated Sp1 from the nuclear extracts prepared from myoblasts transfected with pCMVp107 (Fig. 7B). In contrast, Sp1 coimmunoprecipitation was not detected with the HA antibody from the nuclear extracts prepared from myoblasts transfected with pCMVp130 (Fig. 7C). Immunoprecipitates with nonspecific antibody (anti-E47) or without any antibody did not yield any specific, detectable interactions. These results indicate that p107 interacts with Sp1 in myoblasts, whereas p130 does not.
Sp1-p107 Interaction Is Developmentally Regulated-The interactions between endogenous Sp1 and pocket proteins in proliferating myoblasts and differentiated myotubes were examined by coimmunoprecipitation of non-transfected cells. Nuclear extracts were prepared from myoblasts and myotubes, and protein complexes were immunoprecipitated with antibodies directed against p107, p130, Sp1, and E2F4. Co-immunoprecipitated proteins were analyzed by immunoblotting. As shown in Fig. 8, Sp1 coimmunoprecipitated with p107 and vice versa in proliferating myoblasts. However, coimmunoprecipitation of Sp1 with p107 from myotube extracts was not detected. This result is consistent with previously published results demonstrating a significant decline in Sp1 abundance in myotubes relative to myoblasts (5). Sp1 coimmunoprecipitation with p130 from myoblasts and myotubes was not detected. p130 did coimmunoprecipitate with E2F4 from myoblast or myotube nuclear extracts. These results verify that Sp1 interacts with p107 and not p130 in proliferating myoblasts and that the Sp1-p107 interaction declines in differentiated myotubes.
Sp1 Is Present in the E2F4-p107 Complex Bound to the E2F Site in the FGFR1 Promoter-Having identified a specific protein-protein interaction between Sp1 and p107, we next determined whether Sp1 was present in the E2F4-p107 complex bound to the E2F binding site of the FGFR1 promoter. Electromobility supershift assays were performed using myoblast and myotube nuclear extracts, the FGFR1 E2F4 binding site as a probe, and antibodies directed against E2F4, p130, p107, and FIG. 4. Repression of FGFR1 promoter activity by p107 and p130. Embryonic day 13 chick myoblasts were cotransfected with 1.5 g of wild-type 3284FGFR1CAT reporter plasmid along with variable amounts of pCMVp107 (A), pCMVTAG (B), or pCMVp130 (C) expression plasmids. CAT activities were determined 24 h after transfection and compared with the activity obtained from wild-type 3284FGFR1CAT. The empty expression plasmid vector pCM-VTAG was used as a control to assess nonspecific repression of the FGFR1 promoter by the presence of CMV promoters. Exogenously expressed p107 and p130 were detected in Western blots (WB) using cMyc and HA antibodies, respectively. Detection of E47 served as a loading control of nuclear extracts in Western blot analyses of p107 and p130. Transfection efficiencies were normalized to ␤-galactosidase activities derived from cotransfection of pRSV␤GAL. Data represent the mean CAT activity Ϯ S.E. of four independent experiments. Increased doses of pCMVp107 and pC-MVp130 expression plasmids significantly reduced FGFR1 promoter activity (p Ͻ 0.001 and p Ͻ 0.002) in a dosedependent manner. Sp1 (Fig. 9). Incubation of the protein-DNA complexes with E2F4, p107, and p130 antibodies resulted in a supershifted protein-DNA complex indicating the presence of these tran-scription factors in the E2F4 complex in the nuclear extracts of both myoblasts and myotubes. The protein-DNA complex formed by myoblast nuclear extract and the E2F4 binding site FIG. 5. Transactivation assays of the FGFR1 promoter in D. melanogaster SL2 cells transfected with pC-MVp107, pCMVp130, and pPacSp1 expression plasmids. D. melanogaster SL2 cells were co-transfected with 1.5 g of wild-type 3284FGFR1CAT along with a constant amount of pPacSp1 (750 ng) and variable amounts (1.5 to 6 g) of pC-MVp107 (A) or pCMVp130 (B) expression plasmids. Exogenously expressed Sp1 was detected using an Sp1 antibody in Western blots (WB). Exogenously expressed p107 and p130 were detected with cMyc and HA antibodies, respectively. Immunodetection of E47 served as a loading control of nuclear extracts for Western blot analyses. Differences in transfection efficiencies were normalized relative to ␤-galactosidase activity from co-transfection of pRSV␤GAL. Reported CAT activities are relative to the CAT activity from cultures transfected with full-length wild-type promoter and pPacSp1, but no pCMVp107 or pCMVp130. Data represent the mean CAT activity Ϯ S.E. of five independent assays. Increasing amounts of pCMVp107 expression plasmid significantly (p Ͻ 0.001) repressed Sp1-dependent FGFR1 promoter activity. In contrast, co-transfection with pCMVp130 expression plasmid had no effect (p Ͻ 0.15) on Sp1-mediated FGFR1 promoter activity.
was supershifted by addition of Sp1 antibody, indicating the presence of Sp1 in the E2F4-p107 complex. Moreover, incubation of myoblast nuclear extract with a mutated E2F binding site oligonucleotide did not form any protein-DNA complexes confirming that Sp1 association with the E2F4 binding site oligonucleotide is specific to the E2F4 binding site itself and is not caused by any unknown Sp1 binding sites within the oligonucleotide. Addition of the Sp1 antibody to the protein-DNA complex formed by myotube nuclear extracts did not result in a supershifted complex. These results indicate that Sp1 contributes to the transcriptional complex in myoblasts consisting of E2F4 and p107 at the FGFR1 E2F4 binding site.
Sp1-specific Association with the E2F4-p107 Complex at the FGFR1 E2F4 Binding Site was Maintained when Transiently Expressed in D. melanogaster SL2 Cells-To further prove that Sp1 was present in the E2F4-p107 complex at the FGFR1 E2F4 binding site, we performed transient shift assays with nuclear extracts prepared from D. melanogaster SL2 cells transfected with different combinations of expression plasmids. As shown in Fig. 10, expression of E2F4, p107, and Sp1 proteins formed a protein-DNA complex with the FGFR1 E2F4 binding site oligonucleotide. Incubation of this protein-DNA complex with E2F4, p107, and Sp1 antibodies resulted in a supershift, indicating that these transcription factors formed a tripartite protein complex with the FGFR1 E2F4 binding site oligonucleotide. The p130 antibody did not supershift the protein-DNA complex because pCMVp130 was not expressed in these cells. The lack of a supershifted complex by the p130 antibody demonstrated specificity of the E2F4, p107, and Sp1 antibodies for the protein-DNA complex. However, in the absence of E2F4 protein, this protein-DNA complex was not detectable, suggest-FIG. 6. p107 inhibits Sp1 occupancy of the proximal region of the FGFR1 promoter. A, sequence of the proximal promoter of the FGFR1 gene containing the identified E2F4 and Sp1 binding sites. B, ED13 chick myoblasts were transfected with 3 g of pCMVp107 or pCMVp130 expression plasmid. Twenty-four hours after transfection, myoblast nuclear extracts were prepared and used to generate complexes with the FGFR1 Sp1 binding site oligonucleotide (Ϫ23) (lanes 2-4) and a consensus Sp1 binding site oligonucleotide (Sp1) (lanes 6 -9) as probes. Lanes 1 and 5 contain probe alone. Lane 6 contains antibody against Sp1. Asterisks indicate complexes that do not contain Sp1 and that were previously detected (5). Formation of the Sp1-DNA complex (arrow) was confirmed by addition of Sp1 antibody and appearance of a supershifted protein-DNA complex (arrowhead). Overexpression of p107, but not p130, interfered with Sp1 protein binding to the Ϫ23 Sp1 site of the FGFR1 promoter and the consensus Sp1 binding site.

FIG. 7.
Exogenously expressed p107 interacts with endogenous Sp1. A, nuclear extracts (50 g) of myoblasts transfected with the expression plasmids encoding cMyc-tagged p107 and HA-tagged p130 were probed with anti-cMyc and anti-HA antibodies to detect their exogenous expression levels. B, protein complexes in nuclear extracts from myoblasts transfected with the pCMVp107 expression construct were immunoprecipitated with the cMyc antibody and probed with an anti-Sp1 antibody. C, protein complexes in nuclear extracts from myoblasts transfected with the pCMVp130 expression construct were immunoprecipitated (IP) with an anti-HA antibody and Western blots were probed with an anti-Sp1 antibody. A nonspecific anti-E47 antibody was used as a control and did not immunoprecipitate Sp1. Input protein was diluted 1:3 before loading onto the gel. Sp1 coimmunoprecipitated with cMyc-p107 but not HA-p130.

FIG. 8. Endogenous interactions between p107 and Sp1.
Immunoprecipitations (IP) were performed with nuclear extracts prepared from myoblasts (MB) or myotubes (MT). A, antibodies to p107 and p130 were used to coimmunoprecipitate p107 and p130 protein complexes. The Western blot was incubated with anti-Sp1 antibody. Sp1 coimmunoprecitated only with p107 in myoblast nuclear extracts. B, antibodies to Sp1 and p130 were used to coimmunoprecipitate protein complexes from myoblasts and myotubes. The Western blot was incubated with anti-p107 antibody. p107 coimmunoprecipitated with Sp1 in myoblast nuclear extracts. C, protein complexes from myoblast and myotube nuclear extracts were coimmunoprecipitated with antibodies to p130 and E2F4. The Western blot was incubated with anti-p130 antibody. p130 coimmunoprecipitated with E2F4 in extracts from myoblasts and myotubes. Nonspecific anti-E47 antibody was used as the control. Input protein was diluted 1:3 before loading onto the gel. Integrity of the MT nuclear extracts in A and B was verified by immunodetection of p130 and E2F4 in C.
ing that Sp1 and p107 cannot be recruited to the FGFR1 E2F binding site without E2F4. As a control, Sp1 consensus oligonucleotide was used to show that nuclear extracts prepared from D. melanogaster SL2 cells transfected with pPacSp1 and pCMVp107, but not pCMVE2F4, did contain Sp1.
Analyses of the Protein-DNA complex at the E2F4 Binding Site of the FGFR1 Promoter in Vivo-The composition of the endogenous protein-DNA complexes in the proximal region of the FGFR1 promoter in both myoblasts and myotubes in vivo was analyzed by chromatin immunoprecipitation. Because FGFR1 gene expression is positively regulated in proliferating myoblasts and negatively regulated in differentiated myotubes, the occupancy of the proximal promoter region by E2F4, p107, p130, and Sp1 was determined in these two cell types (Fig. 11). In proliferating myoblasts, antibodies specific for E2F4, p107, p130, and Sp1 immunoprecipitated DNA fragments that were PCR-amplified by primers surrounding the E2F4 and Sp1 binding sites, yielding the expected product size of 330 bp. These results are consistent with the electromobility shift and supershift assays in Fig. 9. Omission of antibody and inclusion of a nonspecific antibody did not yield an immunoprecipitated DNA fragment detectable by PCR amplification. Similar to the protein-DNA complexes in myoblasts, the protein-DNA complexes in myotubes in vivo consisted of E2F4 and p130 at the E2F4 binding site. However, addition of the p107 antibody immunoprecipitated DNA from myotubes that was only weakly amplified. Furthermore, addition of the Sp1 antibody did not immunoprecipitate the FGFR1 proximal promoter DNA. Antibody specificities were demonstrated by lack of FGFR1 promoter immunoprecipitation by a nonspecific antibody. Specificity of PCR amplification of immunoprecipitated DNA was demonstrated by lack of amplified DNA using control primers for amplification of DNA spanning the Ϫ1950 to Ϫ1788-bp region of the FGFR1 promoter, which has no obvious E2F or Sp1 binding sites. These results indicate that protein-DNA transcriptional regulatory complexes consisting of E2F4, p107, p130, and Sp1 occupy the FGFR1 promoter in proliferating myoblasts. In contrast, the predominant protein-DNA complex in differentiated myotubes consists of E2F4 and p130.
To further demonstrate that in proliferating myoblasts, Sp1 interacts with the E2F4-p107 complex at the E2F4 binding site located in the proximal promoter region of the FGFR1 gene, we performed transient chromatin immunoprecipitation assays using wild-type and mutated FGFR1 promoter constructs. We showed previously that in proliferating myoblasts, three Sp1 sites (Ϫ23, Ϫ42, and Ϫ54 bp) located in the proximal promoter region of FGFR1 up-regulated transcription of this gene, and mutation of these sites abrogated promoter activity (5). In contrast, mutation of the E2F site (Ϫ65 bp) adjacent to the Sp1 FIG. 9. Sp1 contributes to the E2F4-p107 complex bound to the FGFR1 E2F site in myoblasts. Gel supershift assays with the labeled FGFR1 E2F binding site (lanes 1-6 and 9 -14) or with the mutated FGFR1 E2F4 binding site oligonucleotide (lanes 7 and 8) and antibodies to E2F4, p107, p130, and Sp1 were performed using myoblast and myotube nuclear extracts . Lanes 1, 8, and 9 contain probe alone. Lanes 2-6 contain protein-DNA complexes formed with the FGFR1 E2F4 binding site oligonucleotide and myoblast nuclear extract. Lane 7 contains no protein-DNA complexes formed with the mutated FGFR1 E2F4 binding site oligonucleotide. Lanes 10 -14 contain myotube nuclear extract. Antibodies added to the reactions are indicated. The arrow indicates the position of the specific E2F4 complexes, and the arrowhead indicates the position of the supershifted E2F4-based complex.
FIG. 10. Transient shift assays to determine Sp1-p107-E2F4 tripartite complex binding to the FGFR1 E2F binding site in D. melanogaster SL2 cells. As described under "Experimental Procedures," D. melanogaster SL2 cells were transfected with different combinations of expression plasmids (pCMVE2F4ϩpCMVp107ϩpPacSp1 or pCMVp107ϩpPacSp1). Forty-eight hours after transfection, nuclear extracts were prepared and used to generate complexes with the FGFR1 E2F4 binding site oligonucleotide (lanes 2-9) and a consensus Sp1 binding site oligonucleotide (lanes 10 -11) as probes. Lane 1 contains probe alone. Lane 2 contains no protein-DNA complexes formed with nuclear extract prepared from non-transfected cells. Lanes 3-7 contain protein-DNA complexes formed with nuclear extracts prepared from the cells transfected with pCMVE2F4, PCMVp107, and pPacSp1 expression plasmids. Lanes 8 -11 contain protein-DNA complexes formed with nuclear extracts prepared from cells transfected with pC-MVp107 and pPacSp1, and not pCMVE2F4, expression plasmids. Antibodies (Ab) added to the reactions are indicated. p130 antibody was used as a control for nonspecific protein-DNA interactions. The left arrow indicates the position of the specific E2F4-DNA complexes, and the left arrowhead indicates the position of the supershifted E2F4based complex. Formation of the Sp1-DNA complex (right arrow) was confirmed by addition of Sp1 antibody and appearance of a supershifted protein-DNA complex (right arrowhead). Expression of E2F4, p107, and Sp1 proteins formed a tripartite complex with the FGFR1 E2F binding site oligonucleotide, and this complex was not detectable in the absence of E2F4. sites increased promoter activity up to 3-fold, indicating that this site mediates promoter repression (6). The 2.2-kb distal promoter region containing two Sp1 sites was deleted from the wild-type and mutated FGFR1 promoter CAT constructs. After transfection of these promoter constructs into proliferating myoblasts, binding of endogenous E2F4, p107, and Sp1 was tested (Fig. 11B). Wild-type FGFR1 promoter was amplified when E2F4, p107, p130, and Sp1 antibodies were used. No PCR product was detected with the nonspecific G␣q antibody and in the absence of antibody. However, mutations of the E2F4 and Sp1 binding sites did not allow binding of E2F4, p107, p130, and Sp1 to the FGFR1 proximal promoter region. As predicted, E2F4, p107, p130, and Sp1 antibodies immunoprecipitated the promoter containing the wild-type E2F4 binding site and the mutated Sp1 sites, indicating that Sp1 antibody immunoprecipitated the FGFR1 promoter via Sp1 binding to the transcriptional complex at the E2F4 binding site. DISCUSSION The regulation of FGFR1 gene expression is a critical component in the development and growth of skeletal muscles. FGFR1 gene transcription is positively regulated in proliferating myoblasts and negatively regulated during muscle differentiation. We have previously shown that Sp1 is an activator of FGFR1 gene transcription (5) and that E2F4 is a transcriptional repressor (6). However, these studies also showed that E2F4 was nuclear localized in both myoblasts and myotubes and that nuclear extracts of both contained E2F4 that bound to the FGFR1 E2F4 binding site. Therefore, we hypothesized that higher order protein-protein interactions at the E2F4 binding site regulated FGFR1 promoter activity. Results of experi-ments reported here indicate that the pocket proteins p107 and p130 regulate FGFR1 promoter activity via E2F4. The results also indicate that Sp1 specifically interacts with p107 at the E2F4 binding site of the FGFR1 promoter.
Several experimental approaches were used to localize p107, p130, and pRb intracellularly and with respect to the FGFR1 promoter. Immunodetection studies revealed that these proteins were differentially expressed in proliferating myoblasts and differentiated myotubes. Although p107, p130, and pRb were localized to the nuclei of both myoblasts and myotubes, p107 was abundant in myoblast nuclear extracts, and pRb was abundant in myotube nuclear extracts. In contrast, the abundance of p130 remained unchanged in myoblast versus myotube nuclear extracts. Gel mobility shift and supershift assays with the FGFR1 E2F4 binding site oligonucleotide (Fig. 3) also differentially localized p107 and p130 to E2F4-DNA complexes. The E2F4-p107 complex was more prominent in proliferating myoblasts, whereas the E2F4-p130 complex was more prominent in differentiated myotubes. pRb did not form a protein complex with E2F4. These results indicate that a specific transition of E2F4-pocket protein complexes occurs during differentiation of chicken muscle cells and support previous findings that E2F4-p130 complexes replace E2F4-p107 complexes in fully differentiated cells (21).
E2F proteins often activate promoters, and subsequent recruitment of pocket proteins to the E2F site leads to the repression of the E2F-dependent genes such as B-myb, E2F1, cyclin A, and cyclin E (25, 44 -46). Transient transfection assays in proliferating myoblasts indicated that p107 and p130 are dose-dependent corepressors of FGFR1 gene expression. Although both p107 and p130 functioned as repressors in myoblasts, transient transfections in D. melanogaster SL2 cells indicated that p107, and not p130, repressed Sp1-mediated activation of the FGFR1 promoter. To determine the mode of repression of Sp1 activation by p107, we analyzed the protein-DNA complexes at the Sp1 binding site and showed that p107, and not p130, interfered with formation of the Sp1-DNA complex in myoblasts. Because p130 did not interact with Sp1, it is likely that p130 could not interfere with Sp1-mediated activation via direct protein-protein interaction. Furthermore, the inability of p130 to repress Sp1-mediated activation of FGFR1 promoter activity in D. melanogaster SL2 cells is probably caused by the absence of E2F-associated factors and the lack of repressor complex formation.
Repression of Sp1 activation by p107 may occur by physical interference of Sp1 binding at the promoter as a result of the proximity of the binding elements or by protein-protein interactions independent of DNA binding. The possibility of direct protein-protein interactions was assessed by coimmunoprecipitation assays. Exogenously expressed and endogenous p107 interacted with Sp1 in nuclear extracts of myoblasts. No interactions were detected in nuclear extracts of myotubes because Sp1 abundance declines during chicken myogenic differentiation (5). In contrast, p130 did not interact with Sp1 in myoblast nuclear extracts. Electromobility shift and chromatin immunoprecipitation assays in myoblasts and D. melanogaster SL2 cells demonstrated that Sp1 interacts with the transcriptional complex bound to the E2F4 binding site of the FGFR1 promoter. This interaction is mediated by specific Sp1-p107 interaction within the complex. The interactions between p107 and Sp1 in regulation of the FGFR1 promoter activity contribute to other reports of p107-Sp1 interactions in HeLa, COS, and CHOC 400 cells (28,29).
Together, these results suggest a model for the developmentally regulated transcription of the FGFR1 gene during skeletal muscle development (Fig. 12). Sp1-mediated transcriptional activation of FGFR1 gene expression in proliferating myoblasts occurs through at least two distinct mechanisms. Sp1 activates FGFR1 promoter activity in proliferating myoblasts by directly binding to proximal and distal cis-elements (5). In addition, Sp1 interacts with p107 in the repressor p107-E2F4 complex at the E2F4 binding site in myoblasts. Because the p107-E2F4 complex is abundant in proliferating myoblasts, the repressor complex of p130-E2F4 is less abundant. During myogenic differentiation, the relative abundance of p107 declines, as does the abundance of the p107-E2F4-DNA complex relative to the p130-E2F4-DNA complex. The latter protein-DNA complex is not able to interact with Sp1 and exhibits its repressor potential. In addition, the abundance of Sp1 declines during cell differentiation, removing direct FGFR1 promoter activation and allowing repression by remaining p107-E2F4-DNA complexes. These results suggest that Sp1 masks p107-E2F4 repression of FGFR1 promoter activity in proliferating myoblasts. The results provide a model framework in which coordinated interplay between nuclear transcription factors govern the differential regulation of FGFR1 gene expression in proliferating myoblasts and differentiated muscle fibers.
Acknowledgments-We thank Dr. Hui Li for expert technical assistance. We thank Dr. R. Tjian for pPacSp1, Dr. K. Helin for pCMVE2F4, FIG. 12. A dynamic model for developmentally regulated FGFR1 gene transcription. A, in proliferating myoblasts, Sp1 positively regulates FGFR1 gene transcription via binding to multiple proximal and distal Sp binding sites. Sp1 also physically interacts with p107 and masks E2F4-p107-mediated repression of FGFR1 promoter activity. B, in differentiated myotubes, the abundance of Sp1 declines, reducing direct FGFR1 promoter activation by Sp1 and allowing transcriptional repression by remaining E2F4-p107 and E2F4-p130 complexes.
Dr. R. Watson for pCMVp107, and Dr. K. Hansen for pCMVp130 expression constructs.