Dystrophin Dp71 expression is down-regulated during myogenesis: role of Sp1 and Sp3 on the Dp71 promoter activity.

Dp71 expression is present in myoblasts but declines during myogenesis to avoid interfering with the function of dystrophin, the predominant Duchenne muscular dystrophy gene product in differentiated muscle fibers. To elucidate the transcriptional regulatory mechanisms operating on the developmentally regulated expression of Dp71, we analyzed the Dp71 expression and promoter activity during myogenesis of the C2C12 cells. We demonstrated that the cellular content of Dp71 transcript and protein decrease in myotubes as a consequence of the negative regulation that the differentiation stimulus exerts on the Dp71 promoter. Promoter deletion analysis showed that the 224-bp 5'-flanking region, which contains several Sp-binding sites (Sp-A to Sp-D), is responsible for the Dp71 promoter basal activity in myoblasts as well as for down-regulation of the promoter in differentiated cells. Electrophoretic mobility shift and chromatin immunoprecipitation assays indicated that Sp1 and Sp3 transcription factors specifically bind to the Sp-binding sites in the minimal Dp71 promoter region. Site-directed mutagenesis assay revealed that Sp-A is the most important binding site for the proximal Dp71 promoter activity. Additionally, cotransfection of the promoter construct with Sp1- and Sp3-expressing vectors into Drosophila SL2 cells, which lack endogenous Sp family, confirmed that these proteins activate specifically the minimal Dp71 promoter. Endogenous Sp1 and Sp3 proteins were detected only in myoblasts and not in myotubes, which indicates that the lack of these factors causes down-regulation of the Dp71 promoter activity in differentiated cells. In corroboration, efficient promoter activity was restored in differentiated muscle cells by exogenous expression of Sp1 and Sp3.

Dp71 expression is present in myoblasts but declines during myogenesis to avoid interfering with the function of dystrophin, the predominant Duchenne muscular dystrophy gene product in differentiated muscle fibers. To elucidate the transcriptional regulatory mechanisms operating on the developmentally regulated expression of Dp71, we analyzed the Dp71 expression and promoter activity during myogenesis of the C2C12 cells. We demonstrated that the cellular content of Dp71 transcript and protein decrease in myotubes as a consequence of the negative regulation that the differentiation stimulus exerts on the Dp71 promoter. Promoter deletion analysis showed that the 224-bp 5-flanking region, which contains several Sp-binding sites (Sp-A to Sp-D), is responsible for the Dp71 promoter basal activity in myoblasts as well as for down-regulation of the promoter in differentiated cells. Electrophoretic mobility shift and chromatin immunoprecipitation assays indicated that Sp1 and Sp3 transcription factors specifically bind to the Sp-binding sites in the minimal Dp71 promoter region. Site-directed mutagenesis assay revealed that Sp-A is the most important binding site for the proximal Dp71 promoter activity. Additionally, cotransfection of the promoter construct with Sp1-and Sp3-expressing vectors into Drosophila SL2 cells, which lack endogenous Sp family, confirmed that these proteins activate specifically the minimal Dp71 promoter. Endogenous Sp1 and Sp3 proteins were detected only in myoblasts and not in myotubes, which indicates that the lack of these factors causes downregulation of the Dp71 promoter activity in differentiated cells. In corroboration, efficient promoter activity was restored in differentiated muscle cells by exogenous expression of Sp1 and Sp3.
Duchenne muscular dystrophy (DMD) 1 is an inherited disorder characterized by progressive muscle degeneration due to the absence of dystrophin (1). Dystrophin is a 427-kDa protein consisting of four major domains as follows: an N-terminal actin-binding domain, a central spectrin-like rod domain consisting of 24 triple helix structures, a cysteine-rich domain, and a unique C-terminal domain (2). In skeletal muscle, dystrophin is associated with a group of sarcolemmal proteins and glycoproteins known collectively as the dystrophin-associated proteins (DAP) (3). One of the proposed functions of dystrophin is to provide a structural link between the actin-based cytoskeleton and the extracellular matrix (4). The DMD gene exhibits a complex transcriptional regulation due to the presence of at least seven independent promoters that generate three fulllength dystrophins (Dp427) and N-terminally truncated gene products (Dp260, Dp140, Dp116, and Dp71) (5-11). Dp71 contains a unique N-terminal sequence of seven amino acids and the cysteine-rich and C-terminal domains of dystrophin (12)(13)(14). Dp71 is expressed in all tissues tested so far, except for skeletal muscle, where its expression is exclusively confined to myoblasts (15)(16)(17). Conversely, Dp427 is not expressed until after the cells begin myogenic differentiation and is the major isoform expressed in mature fibers (18,19).
Based on obvious structural differences between Dp71 and Dp427, it is unlikely that these proteins are functionally interchangeable. In fact, ectopic expression of Dp71 in skeletal muscle of mdx mice, which lack dystrophin, restored the normal levels of DAP but did not alleviate muscle damage (20,21), and more surprisingly, ectopic expression of Dp71 in skeletal muscle of transgenic wild-type mice results in muscle damage, similar to that observed in mdx mice (22).
The performance of stage-specific tasks by Dp427 and Dp71 in muscle cells indicates that their expression is tightly controlled (23). The transcriptional regulation of Dp427 in muscle cells has been well established. It is known that high levels of YY1 protein down-regulate the Dp427 promoter in undifferentiated muscle cells, but upon the induction of muscular differentiation, YY1 protein levels are negatively controlled by the action of the protease m-calpain, allowing the dystrophin promoter bending factor to regulate positively the promoter activity (24). In contrast, regulation of Dp71 during myogenesis remains to be approached, and the Dp71 promoter region, which exhibits the structure of a typical housekeeping promoter, has been only partially characterized (11). In this study, we analyzed the expression of Dp71 and characterized the activity of its promoter region during muscle cell differentia-tion of the C2C12 cells, a sub-culture derived from the C2 cell line. Deletion analysis showed that the major Dp71 promoter activity in proliferating myoblasts depends on the proximal 224-bp promoter region, which contains several Sp-binding sites. Altogether, gel shift, chromatin immunoprecipitation, and site-directed mutagenesis assays, as well as transient transfection experiments in Drosophila SL2 cells, established that Sp1 and Sp3 transcription factors interact with the Spbinding sites and transactivate the Dp71 promoter. In differentiating muscle cells, the Dp71 promoter activity is downregulated; the 224-bp proximal promoter region seems to be sufficient to exert such control, and a concomitant reduction in the endogenous Sp1 and Sp3 protein levels was observed. Restoration of significant promoter activity in differentiating cells was obtained after exogenous expression of Sp1 or Sp3 proteins. Our results indicate that the developmentally regulated expression of Dp71 in muscle cells during differentiation is based on the differential expression of Sp1 and Sp3 transcription factors.

EXPERIMENTAL PROCEDURES
Cell Cultures-C2C12 cells (ATCC CRL-1772), a sub-culture derived from C2 cell line (25), were cultured in Dulbecco's modified Eagle's medium (Invitrogen) containing 4.5 g of glucose/liter, supplemented with 10% fetal bovine serum (FBS) and 2 mM L-glutamine (proliferation medium), and maintained at 37°C in a humidified incubator with a 5% CO 2 atmosphere. In some experiments, subconfluent C2C12 cells were induced to differentiate by lowering FBS to a final concentration of 1% (differentiation medium). Schneider's Drosophila Line 2 (SL2, ATCC CRL-1963) cells were cultured in Schneider's Drosophila Medium (Invitrogen) containing L-glutamine and were maintained at 25°C without CO 2 . All culture media contained 100 units of penicillin and 100 g/ml streptomycin.
Real Time RT-PCR-Dp71 transcript levels during C2C12 muscle cell differentiation were measured by quantitative real time RT-PCR using the comparative C T method described by Applied Biosystems. Total RNA was extracted using the TRIzol reagent (Invitrogen) from C2C12 cells induced to differentiate for 0, 3, 6, or 9 days. 5 g of the extracted total RNA was primed with random hexanucleotides and reversed-transcribed by the Moloney murine leukemia virus-reverse transcriptase (Invitrogen), according to the manufacturer's instructions. Real time PCRs were set up in a reaction volume of 25 l using the TaqMan Universal PCR Master Mix (Applied Biosystems). The probe and primers specific for Dp71 detection were designed using the Primer express software from PerkinElmer Life Sciences. The Dp71 fluorogenic probe was 5Ј-(FAM TM )-CCCCAAAGGACTCAAAGAACCT-(TAMRA TM )-3Ј, and the Dp71 PCR primers were Dp71F 5Ј-TGTATTG-CATTTAGAGCCCCAA-3Ј and Dp71R 5Ј-CTTCCTCTGCGCTTAAT-TGC-3Ј (Synthetic Genetics, San Diego, CA). As endogenous control, real time RT-PCR analysis of the eukaryotic 18 S ribosomal (r)RNA gene was performed in parallel. The r18 S fluorogenic probe, labeled with VIC TM dye-TAMRA TM dye, and PCR primers were purchased from Applied Biosystems. DNA amplifications were carried out in a 96-well reaction plate format in a PE Applied Biosystems 7700 Sequence Detector (PerkinElmer Life Sciences). Both Dp71 and r18 S PCRs were carried out in triplicate. Multiple negative control water blanks were included in each analysis.
Transient Cell Transfection-Each Dp71 promoter construct was introduced into C2C12 cells by means of the Lipofectamine Plus reagent (Invitrogen), together with the pRSV-␤-gal vector or with both the pRSV-␤-gal and pPac, AP2␥/pcDNA3.1(ϩ), pPacSp1, and/or PacSp3 expression plasmids according to the manufacturer's instructions. De-fined amounts of each expression vector were incubated with 8 l of plus reagent in 250 l of Opti-MEM medium, mixed well, and added to an equal volume of Opti-MEM medium containing 12 l of Lipofectamine reagent. The lipid/DNA mixture was mixed well, incubated for 15 min at room temperature, and added to cell cultures (previously washed with Opti-MEM medium) covered by 2 ml of Opti-MEM medium. After 48 h of incubation at 37°C in a 5% CO 2 -humidified incubator, Opti-MEM medium was replaced with either proliferation or differentiation medium. For transfection of the SL2 cell line, freshly grown cells from 3-to 4-day-old cultures were plated at density of 2.5 ϫ 10 6 cells/60-mm dish and the day after were transfected using Cellfectin reagent (Invitrogen), according to the manufacturer's instructions. Each plate was transfected previously with defined amounts of pRSV-␤-gal, each Dp71 promoter-CAT reporter plasmids, and pPacSp1 and/or pPacUSp3 expression plasmids. After transfection, C2C12 and SL2 cells were maintained in their respective growth medium for 48 h, and the CAT and ␤-galactosidase assays were performed.
CAT and ␤-Galactosidase Assays-To prepare cell extracts for the CAT and ␤-galactosidase expression assays, cells were scraped into 1ϫ phosphate-buffered saline and centrifuged at 2600 ϫ g at 4°C for 5 min. The cell pellet was resuspended in 150 l of 0.25 M Tris and 1 mM EDTA, pH 8.0, and then subjected to six freeze-thaw cycles with dry ice. Cell debris was removed by centrifugation at 4°C for 5 min at 3000 ϫ g, and the resulting supernatant was removed and its protein content determined by the Bradford assay (Bio-Rad). These clarified extracts were used directly in enzymatic assays. ␤-Galactosidase activity was measured using 2.5 mM -nitrophenyl ␤-D-galactopyranoside (Sigma) as substrate and 30 l of cell extract in a reaction mixture consisting of 60 mM Na 2 HPO 4 , 40 mM NaH 2 PO 4 , 2 mM MgCl 2 , and 50 mM ␤-mercaptoethanol. After incubation for 30 min at 37°C, the reaction was stopped by adding 1 M Na 2 CO 3 , and the absorbance of the reaction product was read at 405 nm. CAT activity was determined by using 20 g of cell extract, 80 mM acetyl-CoA (Sigma), and 0.15 Ci of [ 14 C]chloramphenicol (Amersham Biosciences) in a total volume of 180 l. The reaction mixture was incubated at 37°C for 60 min, extracted with 800 l ethyl acetate, and dried in a vacuum centrifugal evaporator. The dry reaction products were resuspended in 20 ml of ethyl acetate, and the acetylated and nonacetylated forms of [ 14 C]chloramphenicol were separated by thin layer silica gel chromatography for 60 min at room temperature with chloroform/methanol (19:1, v/v) as mobile phase. Percentage conversion of chloramphenicol to its acetylated forms was determined using a radioactive image analyzer (AMBIS 4000).
Isolation of Nuclear Extracts-Nuclear extracts were prepared from undifferentiated and differentiated C2C12 cells, as described previously (26). Briefly, cells were washed with ice-cold phosphate-buffered saline and resuspended in 400 l of lysis buffer (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride). Cell lysates were incubated on ice for 15 min and then centrifuged for 3 min at 3,000 ϫ g at 4°C. The pellet was resuspended in 50 l of extraction buffer (20 mM HEPES, pH 7.9, 400 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride). After that, cells were vigorously shaken at 4°C for 15 min and pelleted by centrifuging for 5 min at 12,000 ϫ g at 4°C. The supernatant was recovered as the nuclear extract, and its protein concentration was determined by the Bradford assay. Nuclear extracts were stored at Ϫ70°C until use.
Antibodies and Western Blot Analysis-Antibodies directed to Sp1, Sp3, and myogenin were purchased from Santa Cruz Biotechnology. The anti-Dp71 polyclonal antibody 2166 was a donation of Dr. D. J. Blake (27). The anti-actin monoclonal antibody was generously provided by Dr. J. M. Hernandez. C2C12 cells cultured in 35-mm culture dishes were scraped, lysed as described previously (28), and centrifuged at 12,000 ϫ g for 10 min at 4°C. Protein samples were quantified by the Bradford assay and denatured at 100°C for 5 min in a protein sample buffer containing 1% SDS and 1% dithiothreitol. One hundred g of total protein extracts were loaded in each lane and subjected to 10% PAGE under denaturing conditions (SDS-PAGE) and transferred to a nitrocellulose membrane for immunoblotting. Immunoblots were probed with the corresponding primary antibodies and developed by using the ECL Western blotting analysis system (Amersham Biosciences).
Chromatin Immunoprecipitation Assay-C2C12 cells (2 ϫ 10 7 ) were treated with 1% formaldehyde to cross-link protein-DNA complexes. Immunoprecipitates of cross-linked complexes were prepared with anti-Sp1 and anti-Sp3 antibodies, treated with proteinase K for 2 h, and then incubated at 65°C to eliminate cross-linking. DNA was purified by phenol/chloroform extraction and ethanol precipitation. DNA samples were quantified by the picogreen assay (Molecular Probes) and then analyzed by PCR amplification of the Dp71 promoter sequence located between Ϫ127 and ϩ78 with 30 cycles of PCR using the following of 32 P-labeled oligonucleotides: Dp71 (upper), 5Ј-CTATCCAGGTTTC-CCCAGGTC-3Ј, and Dp71 (lower), 5Ј-CGGAGGAGTAAGCTTGC-CCAA-3Ј. Different cycle numbers were employed to ensure linearity of the amplification reaction. As negative control, a DMD gene region localized on the junction of the DMD gene intron 63 and exon 64 (from ϩ41,508 to ϩ41,646, relative to the Dp71 transcription start site, Gen-Bank TM accession number AL645848) was amplified with the following 32 P-labeled oligonucleotides: Dys 64/64 (upper), 5Ј-ATAATGTCAGGT-TCTCCGCG-3Ј, and Dys 64/64 (lower), 5Ј-TCAAAAATCCCCAAGC-CCCA-3Ј. As positive control, a DNA region of the human telomerase promoter, which contains several Sp-binding sites, was amplified as positive control with the following 32 P-labeled oligonucleotides Tert (upper), 5Ј-AACACATCCAGCAACCACTGA-3Ј, and Tert (lower), 5Ј-AAGATGAGGAGGGAACGGAGG-3Ј. PCR products were analyzed by 6% PAGE and visualized by autoradiography.
Construction of Mutant Plasmids-A series of mutations was performed in the Dp71 proximal promoter region (p224CAT) by using the QuickChange TM site-directed mutagenesis kit (Stratagene) and each of the following double-stranded oligonucleotides: mutSp-A, 5Ј-C-CCTTCCCAGCCCTGCTCCGTCTGCACGC-3Ј; mutSp-B, 5Ј-CCCCTC-CCTGTCCTGTCCGCCCGCCAGC-3Ј; mutSp-C, 5Ј-CCCCTCCCTGTC-CCGCCTGTCCGCCAGC-3Ј; mutSp-D, 5Ј-GGCTGCGAGCCTGGGCT-TCGTGCGCTTGAC-3Ј; and mutAP2, 5Ј-CCCGCCCGCCAGCCAACC-AGGCAGCGGCGG-3Ј. The underlined letters indicate the mutated bases. Briefly, 50 ng of the p224CAT promoter plasmid was used in a PCR containing 5 l of Pfu polymerase buffer, 2.5 units of Pfu turbo DNA polymerase (Stratagene), 10 mM dNTPs, 125 ng of each oligonucleotide, and water was added to 50 l. Following temperature cycling (PerkinElmer Life Sciences), DpnI treatment was performed to cleave parental DNA and to improve the efficiency of the mutant plasmid screening. The reaction was transferred into XL-1 Blue competent cells, and the transformation mixture was plated on LB ampicillin plates. The authenticity of the mutants was established by DNA sequencing.
Data Analysis-All data are depicted as mean Ϯ S.E. Differences between two groups were validated by Student's t test.

Dp71 Expression Is Down-regulated during C2C12
Myogenesis-To evaluate the expression of Dp71 during C2C12 muscle cell differentiation, its mRNA and protein levels were determined. Dp71 mRNA expression was measured by real time RT-PCR using Dp71-specific primers, whereas Dp71 protein was assessed by Western blotting using the polyclonal antibody 2166, directed against its C-terminal domain. Myoblast cultures were induced to differentiate for 3, 6, and 9 days by lowering serum concentration, and cell-differentiated morphology was monitored by light microscopy analysis. As shown in Fig. 1A, Dp71 mRNA expression was detected in myoblasts but decreased to minimum levels in myotube cultures since day 3 of induced differentiation, at which point myotube formation became apparent under microscopy visualization (data not shown). Dp71 protein expression pattern during C2C12 myogenesis was similar to that of Dp71 transcript; it was immunodetected in undifferentiated cells but disappeared totally from differentiated cultures of 3, 6, and 9 days (Fig. 1B, upper panel). In contrast, C2C12 maintained in differentiation medium resulted in up-regulation of myogenin, a myogenic gene marker; this protein was undetectable in myoblast cells but appeared by day 3 and remained steady throughout the rest of the differentiation treatment (Fig. 1B, lower panel), in agreement with previous reports (29 -32). Thus, the concomitant decreases of Dp71 mRNA and protein levels were negatively correlated with the differentiation process of C2C12 cells and evaluated by the acquisition of a differentiated morphology and induction of myogenin expression. Furthermore, these results indicated that the reduction of Dp71 expression during skeletal muscle differentiation was regulated, at least in part, at a transcriptional level.
Dp71 Promoter Activity during Muscle Cell Differentiation-To characterize the mechanisms controlling the negative transcriptional regulation of Dp71 that occurred during myogenic differentiation, C2C12 myoblasts were transiently Asterisks denote significant differences (p Ͻ 0.05) from undifferentiated cell culture (day 0). B, total protein extracts were prepared from cell cultures induced to differentiate for the indicated days. Protein extracts were subjected to immunoblot analysis using anti-Dp71 (antibody 2166) and anti-myogenin antibodies. Membranes were stripped and reprobed with a monoclonal anti-actin antibody.
cotransfected with pRSV-␤-gal and p1.8CAT vectors; the latter contains 1.8 kb from the mouse Dp71 promoter region fused to the reporter gene CAT (11). Myoblast cultures were induced to differentiate for 2 days, and the Dp71 promoter function was evaluated by CAT assays using ␤-galactosidase activity to normalize transfection efficiency. The 1.8-kb Dp71 promoter fragment (p1.8CAT) drove the efficient expression of CAT in C2C12 myoblast compared with promoterless pSV0CAT (Fig. 3). As expected from the mRNA results described above, reporter activity of the Dp71 promoter decreased by 70% in differentiated cells (Fig. 2). Because the induction of muscle cell differentiation was also modulated by the addition of cAMP (33,34), we decided to evaluate the effect of this alternative inductor of differentiation on the Dp71 promoter activity. For that purpose, myoblasts, transiently transfected with the p1.8CAT vector, were cultured in growth medium for 48 h with or without cAMP. Fig. 2 shows that cAMP treatment resulted in a 40% inhibition of CAT activity, whereas a negative control experiment, in which the nerve growth factor was added to the proliferation medium, provoked no changes in CAT activity. Altogether, these results indicate that Dp71 promoter responds negatively in a specific way to inducers of skeletal muscle differentiation.
Identification of cis-Elements Essentials for the Regulation of the Dp71 Promoter during Myogenesis-To define the cis-acting elements required for Dp71 promoter regulation in muscle cells, a series of mutant promoter constructs containing progressive 5Ј-end or internal deletions of the Dp71 promoter sequence, linked to the CAT reporter gene, were constructed and transfected into proliferating myoblasts. The cell cultures were then grown for 48 h under proliferation or differentiation conditions, and the promoter activity of each mutant plasmid was evaluated. In myoblast cells (Fig. 3, black bars), deletion of the promoter sequence encompassing positions Ϫ1834 to Ϫ1156 caused a near 70% reduction of the reporter activity, whereas further deletion beyond position Ϫ937 virtually recovered the transcriptional activity observed with the wild-type promoter. This suggest that positive and negative regulatory elements may be contained in these two regions, respectively. Further deletions to positions Ϫ709 (p709CAT) and Ϫ586 (p586CAT) did not significantly affect the transcriptional ac-tivity of the promoter constructs compared with the wild-type promoter. Vector p224CAT, which contains only 224-bp from the proximal promoter region, maintained ϳ50% of the wildtype promoter activity. Most interestingly, the removal of the 224-bp proximal promoter region abolished the reporter transcriptional activity of p1500CAT and p900CAT vectors, despite conserving most of the 5Ј-end distal promoter sequences (Fig.  3). Finally, as compared with wild-type promoter, plasmids carrying internal deletion between positions Ϫ1603 and Ϫ711 (pXbaIdelCAT), Ϫ1778 and Ϫ711 (p700CAT), and Ϫ1576 and Ϫ839 (pstuIdelCAT) did not modify substantially the reporter activity (Fig. 3). These findings suggest that the proximal 224-bp region is the core promoter essential for basal transcriptional activation of Dp71 in proliferating myoblast cells. When we analyzed the behavior of the different promoter constructs in differentiating C2C12 cells, we observed that all mutant constructs that already showed appreciable CAT activity in myoblast cells displayed substantial reductions in their reporter activity, ranging from 50 to 70% (Fig. 3, open bars). Because the minimal proximal promoter region (p224CAT vector) maintained the negative regulation displayed in response to the differentiation stimulus, we assumed that this sequence was also responsible for the negative modulation displayed by the Dp71 promoter during myogenesis.
Sp1 and Sp3 Specifically Bind in Vitro to the GC Boxes in the Dp71 Core Promoter-Sequence analysis of the proximal promoter region revealed the presence of four binding motifs for the Sp family of transcription factors located between positions Ϫ76/Ϫ71, Ϫ42/-37, Ϫ38/Ϫ33, and ϩ26/ϩ31 (termed Sp-A, Sp-B, Sp-C, and Sp-D boxes, respectively), and a single AP2 DNA element located in Ϫ29 to Ϫ23 (Fig. 4A). It should be noted that the Sp-B and Sp-C sequences overlap each other. To determine whether the consensus-binding sites present in the core promoter region interact with nuclear protein components, we performed EMSA with double-stranded 32 P-labeled oligonucleotides spanning Sp-A, Sp-BC, Sp-D, and AP2 DNA elements and nuclear extracts from C2C12 myoblasts. Fig. 4B shows that both Sp-A and Sp-BC probes gave rise to several major DNA-protein complexes (C1, C2, C3, and C4) and additional diffuse retarded bands. To define the nature of the DNA-protein complexes formed, competitive binding experiments were performed (Fig. 4B). The four slower migrating complexes formed by Sp-A probe (C1 to C4) as well as the first, third, and fourth slower migrating complexes obtained with the Sp-BC oligonucleotide (C1, C3, and C4) were competed specifically by adding a 100-fold molar excess of either the respective unlabeled probe or a consensus Sp oligonucleotide. In contrast, the DNA-protein complexes remained unaltered after the addition of 100-fold molar excess of either a mutated oligonucleotide or an unrelated oligonucleotide Oct1 (complexes indicated in Fig.  4B). The remaining Sp-binding site, the Sp-D box, formed only a weakly diffused retarded band that did not show changes when challenged with a 100-fold molar excess of unlabeled oligonucleotides Sp-D, consensus Sp, and mutated Sp or Oct1 (Fig. 4B). Finally, the AP2 probe produced two major retarded bands (C1 and C2), and the formation was significantly reduced by adding a 100-fold molar excess of either an unlabeled AP2 probe or a consensus AP2 oligonucleotide; contrary, a 100-fold molar excess of either an unrelated Oct1 oligonucleotide or a mutated AP2 probe did not compete in binding (Fig. 4B).
Because functional analysis of the proximal Dp71 promoter region showed that the Sp-A site is the most important transcription factor binding site for activity of the minimal Dp71 promoter (see below), we decided to investigate whether complexes formed between the Sp-A probe and myoblast nuclear extracts consists of Sp1 and/or Sp3 proteins. Therefore, an FIG. 2. Effect of myogenesis induction on the Dp71 promoter activity. C2C12 cell cultures were cotransfected with vector p1.8CAT that contains the 1.8-kb Dp71 promoter region and pRSV-␤-gal. Transfected cells were either maintained under proliferating conditions or induced to differentiate for 2 days by culturing in differentiation medium (1% FBS) or by adding 1 mM dibutyryl-cAMP (db-cAMP) to the proliferation medium. The resulting CAT activities were normalized against ␤-galactosidase activities, and the reporter activity of undifferentiated control cells (cells cultured in medium supplements with 10% FBS) was set at 100%. In negative control experiments, nerve growth factor (NGF) at 2 nM was added to cells cultured in proliferation medium. Data are expressed as the mean Ϯ S.D. of at least three independent experiments, each performed in duplicate. Asterisks denote significant differences (p Ͻ 0.05).
EMSA for the Sp-A probe in the presence of anti-Sp1 or anti-Sp3 antibodies was performed. Fig. 4C shows that the C1 and C2 complexes were shifted when Sp1 or Sp3 antibodies were included in the assay, respectively. Incubation of the binding reactions with an unrelated IgG did not affect the EMSA pattern confirming the specificity of the assay. Our results indicate that the Sp1 and Sp3 transcription factors bind in vitro to the Sp DNA elements in the Dp71 proximal promoter region.
The Dp71 Minimal Promoter Region Interacts with Sp1 and Sp3 in Vivo-In order to evaluate whether Sp1 and Sp3 proteins are indeed recruited in vivo to the Dp71 promoter, we employed a ChIP assay to cross-link the DNA with bound proteins in situ in C2C12 cells; the protein-DNA complexes were precipitated with antibody specific for Sp1 or Sp3. The DNA fragments containing the Dp71 minimal promoter region were then amplified by radioactive PCR. As shown in Fig. 5, PCR with primers flanking the Dp71 minimal promoter region produced a band from DNA coprecipitated with Sp1 or Sp3, and such a band migrated on an agarose gel to a position identical to the genomic DNA control (Fig. 5B, Input). In the positive control, we obtained an intense PCR product when the Sp1 and Sp3 immunoprecipitates were subjected to a PCR with primers flanking the telomerase promoter region (Fig. 5B, Tert), which contains multiple Sp-binding sites (35,36). In the negative controls, Dp71 promoter primers did not generate any PCR product when the immunoprecipitation reaction was carried out with an unrelated IgG antibody. Likewise, PCR with primers flanking an irrelevant DNA region (a DMD gene region with no Sp boxes) did not produce any PCR products from the Sp1 and Sp3 immunoprecipitates. Our results indicate that the Sp1 and Sp3 transcription factors bind in vivo to the Dp71 proximal promoter region.
Sp1 and Sp3 Transcription Factors Transactivate the Dp71 Promoter-The EMSA and ChIP analyses described above indicate that Sp1 and Sp3 bind to the GC boxes present in the Dp71 proximal promoter. Therefore, to determine whether these transcription factors can functionally transactivate the Dp71 promoter, we employed a transfection system consisting of Drosophila SL2 cells, a cell line that naturally lacks the expression of endogenous Sp transcription factors, and vectors pPacSp1 (encoding Sp1), pPacUSp3 (encoding Sp3), and pPac (empty vector). The Dp71 minimal promoter vector (p224CAT) was transfected into SL2 cells with either pPacSp1 or pPacUSp3, and 48 h after transfection CAT activity of each promoter construct was evaluated. As shown in Fig. 6, cells transfected with p224CAT and 0.1 g of pPacSp1 displayed a 150-fold augment in CAT activity; however, increasing amounts of Sp1 expression vector did not modify substantially such transactivation. On the other hand, relative to pPacUSp3 vector, a 2-fold molar excess of Sp3 expressing vector was required to obtain a significant increase in CAT activity controlled by the minimal promoter. Nevertheless, transfection with higher amounts of pPacSp3 (0.5 g) caused a drastic reduction of 70% in the former promoter transactivation, which suggests that elevated amounts make Sp3 change from a positive to a negative transcriptional regulator. These findings strongly suggest that Sp1 and Sp3 play an important role in activating the Dp71 proximal promoter. In addition, it appears that precise amounts of each Sp protein are required to modulate positively the minimal promoter.
Functional Analysis of the Transcription Factor Binding Sites Present in the Proximal Dp71 Promoter-From the experiments described above, it seems that the Sp1/Sp3 proteins are key factors in modulating the Dp71 transcriptional activity in muscle cells. Hence, it is expected that disruption of the Sp1/ Sp3-binding sites located in the Dp71 minimal promoter region would impair the activity of this promoter. To approach this hypothesis, we introduced point mutations that were shown to abolish nuclear protein binding in the EMSA competition experiments into Sp-A, Sp-B, Sp-C, Sp-D, and AP2 DNA ele- ments. As shown in Fig. 7, the point mutation in the Sp-A site resulted in ϳ40% reduction of the CAT activity controlled by the minimal promoter region, whereas the mutation of Sp-B or Sp-D sites showed reductions of ϳ18 and ϳ27% of CAT activity, respectively. Finally, the mutation of the Sp-C site resulted in a slight increase of ϳ14% in the minimal promoter activity (Fig. 7). When point mutations were introduced into all of the potential Sp-binding sites located in the minimal promoter region, the reporter activity was reduced by ϳ54% compared with the wild-type construct; on the other hand, mutation of the AP2 site resulted in no significant alteration in the promoter activity (Fig. 7). These results suggest that the Sp-A site is the most crucial transcription factor-binding site in terms of basal Dp71 expression.
Endogenous Expression Levels of the Sp1 and Sp3 Transcription Factors-As Sp1/Sp3 transactivate significantly the Dp71 proximal promoter in Drosophila SL2 cells, it is likely that they are also involved in the molecular mechanism controlling the down-regulation of this promoter during myogenesis. Therefore, as a first step in clarifying this matter, we analyzed by Western blotting the endogenous expression of these transcription factors in nuclear extracts obtained from proliferating myoblasts and differentiated myotubes. In myoblast nuclear extracts, the expected protein bands for Sp1 (ϳ100 kDa) and The double-stranded oligonucleotides used as probes and wild-type competitors are indicated by the brackets above and below the sequence. The underlined sequences indicate the potential AP2 and Sp-binding sequences, and the transcription start site are indicated by an arrow. B, EMSAs were performed by incubating nuclear extracts from undifferentiated C2C12 cells with the indicated double-stranded 32 P-labeled probes in the absence or presence of a 100-fold molar excess of unlabeled competitors. Mutated competitors for Sp and AP2 are described under "Experimental Procedures." DNA-protein complexes were resolved on nondenaturing 6% polyacrylamide gels and analyzed as described under "Experimental Procedures." Major DNA-protein complexes are indicated as C1-C4. C, specific DNA-binding complexes formed with Sp-A probe were supershifted (open arrows) by adding antibodies for Sp1 or Sp3 proteins to the reaction mixture. Super-retarded complexes (S1 and S2) are indicated by open arrows. An unrelated IgG was used as negative control.
Sp3 (ϳ132 and ϳ70 kDa) were obtained after using their respective specific antibodies. Most interestingly, Sp1 and Sp3 protein bands disappeared from myotubes since the 6th and 3rd day of induced differentiation, respectively (Fig. 8A). These findings indicate that Sp1 and Sp3 protein expression is present but is extinguished as muscular differentiation proceeds. This pattern of protein expression resembled that of Dp71 (Fig.  1). In support of these findings, we observed that DNA-protein complexes formed in vitro between the Sp-A probe and myoblast extracts (complexes C1, C2, and C3) disappeared completely when protein extracts from differentiated cells of 6 days were employed in the EMSA (Fig. 8B).

Overexpression of the Sp3 Transcription Factor Restores the Activity of the Proximal Dp71 Promoter in Differentiated
Cells-Because all of the experimental evidence shown before indicated that down-regulation of the Dp71 expression during myogenesis is caused by the scarcity of Sp1 and Sp3 found in differentiated myotubes, it is expected that the Dp71 promoter activity could be recovered by restoring the expression level of these transcription factors in differentiated muscle cells. To test this idea, the minimal promoter construct was introduced into C2C12 cells together with either pPac, pPacSp1, or pPacSp3 expression vectors, and the transfected cultures were induced to differentiate for 48 h. Fig. 9 shows that the expression of the empty vector (pPac) did not prevent the drop of the promoter activity that occurred in response to the differentiation stimulus. In contrast, increasing amounts of the Sp3expressing vector recovered the reporter activity controlled by the minimal promoter at levels even higher than that observed in proliferating myoblasts. On the other hand, Sp1 overexpression provoked only modest increases in the promoter CAT levels that did not restore completely the activity shown by this promoter before differentiation. Simultaneous overexpression of Sp1 and Sp3 transcription factors did not have an additional effect on CAT activity obtained by the single overexpression of Sp3, indicating that the Sp1 and Sp3 proteins did not act synergistically over the Dp71 minimal promoter region. The participation of Sp1 and Sp3 in the recovery of the promoter activity appears to be specific because the overexpression of AP2 failed to rescue the reporter activity in the differentiated cells (data not shown).

DISCUSSION
Dp71 is expressed in a wide variety of tissues with the exception of skeletal muscle, where it is known that Dp71 protein levels decrease during myogenesis. On the contrary, dystrophin expression increases in differentiated muscle cells (13,(15)(16)(17)(18)(19)37). The transcriptional regulation of dystrophin during myogenesis has been well established (24,38), while the gene regulation of Dp71 during this cellular process remains to be approached. With this understanding, the purpose of the present study was to establish a muscular cell model that mimics the in vivo expression of Dp71 while allowing for definition of the molecular mechanisms controlling Dp71 expression during muscle cell differentiation. Here we demonstrated that expression of Dp71 is down-regulated during differentiation of C2C12 muscle cells as a consequence of decreases in their mRNA and protein levels (Fig. 1). The cellular content of Dp71 protein and mRNA decreased in parallel with the transcriptional activity of the Dp71 promoter in response to myogenesis (Fig. 2), indicating that the alteration in the transcription at the promoter level accounts for the elimination of Dp71 expression during muscle cell differentiation. Therefore, we characterized by deletion analysis the transcriptional activity of the mouse Dp71 promoter in C2C12 cells, and we determined that the Dp71 core promoter region, spanning from Ϫ224 to ϩ65, is sufficient for the basal transcriptional activity of Dp71 in proliferating myoblasts as well as for the negative transcriptional modulation displayed by this gene in response to myogenesis (Fig. 2). The minimal promoter region of Dp71 lacks a TATA box and contains several potential cis-acting elements as follows: four Sp-binding sites (Sp-A, overlapping Sp-B and -C sites, and Sp-D) and a single AP2-binding site. To ascertain whether these transcription factor-binding sites are necessary for Dp71 transcriptional modulation in muscle cells, different molecular approaches were employed. By using EMSA, we detected several specific DNA-protein complexes formed by myoblast nuclear extracts with oligonucleotides corresponding to the Sp-A-, Sp-BC-, Sp-D-, and AP2-binding sites; with the Empty pPac vector was used as negative control. In all cases, control plasmid pRSV-␤-gal was included to normalize transfection efficiency. The CAT activity obtained with pPac was set at 1%, and all other CAT activities were represented relative to this value. Data represent the mean CAT activities Ϯ S.E. of three independent experiments, each performed in duplicate. exception of Sp-D, the rest of these DNA-protein complexes were competed efficiently by their respective consensus oligonucleotides (Fig. 4B). We have also identified the binding of Sp1 and Sp3 to the Sp-A box by performing supershift assays with Sp1-and Sp3-specific antibodies (Fig. 4C). Furthermore, by ChIP assays, we demonstrated that transcription factors Sp1 and Sp3 are indeed recruited in vivo to the Dp71 core promoter (Fig. 5). Consequently, by using the Drosophila SL2 cell line, we demonstrated that these two transcription factors behave as activators of the Dp71 promoter region with Sp1 being more active than Sp3 (Fig. 6). Altogether, these findings indicate that binding of Sp1 and Sp3 to the Dp71 core promoter has a functional role in myoblast cells. The transcription factors Sp1 and Sp3 are ubiquitously expressed proteins that bind their recognition sequence (GC boxes) with similar affinity. Sp1 is thought to play a primary role in the regulation of a large number of genes including constitutive housekeeping genes and inducible genes (39), whereas Sp3 contains a transcriptional repression domain and can act as an activator or as a repressor of Sp1-mediated activation (40 -43). In fact, we observed that elevated levels of Sp3 change its role from activator to repressor of Dp71 expression in Drosophila SL2 cells (Fig. 6), whereas in differentiating C2C12 cells this transcription factor always acts as an activator regardless of its protein levels (Fig.  9). Because repression function of Sp3 depends on additional proteins which act as corepressors (42), it is likely that the differential behavior of Sp3 observed in our study is determined by the cellular context.
Functional reporter gene studies revealed that the Sp-A site is the most important transcription binding site for the activity of the minimal Dp71 promoter, and cancellation of this Sp site caused a marked decreased of 40% in reporter activity, whereas cancellation of all of the Sp sites resulted in only 54% reduction of reporter activity. The apparent redundancy due to the presence of multiple Sp-binding sites in the Dp71 promoter may constitute a mechanism by which expression of Dp71 is guaranteed in myoblasts; hence, if an Sp-binding site is disrupted, alternative Sp-binding sites may participate in nuclear protein binding to maintain sufficient gene expression. The fact that mutation of all Sp1-binding sites reduced the promoter activity in ϳ54% suggests that additional transcription factor-binding sites, other than Sp1/Sp3 DNA elements, maintain a residual Crossed symbols(s) represent those site(s) mutated in the p224CAT vector. Mutant constructs were transiently transfected into proliferating C2C12 cells with pRSV-␤-gal standardization plasmid as described under "Experimental Procedures." The CAT activities were normalized against ␤-galactosidase activities, and the promoter reporter activity of the wild-type promoter was set at 100%. Data are expressed as the mean Ϯ S.D. of at least three independent experiments, each performed in duplicate.
FIG. 8. Sp1 and Sp3 are repressed during myogenesis. A, C2C12 cells were cultured in proliferating medium (day 0) or induced to differentiate for 3, 6, and 9 days by culturing in differentiation medium. Total protein extracts were resolved by 10% SDS-PAGE, electroblotted onto nitrocellulose membrane, and reacted with anti-Sp1-and anti-Sp3-specific antibodies. Membranes were stripped and reprobed with a monoclonal antibody anti-actin. The arrows indicate Sp1 and Sp3 proteins. B, EMSA was performed using double-stranded 32 P-labeled Sp-A probe and nuclear extracts from undifferentiated (MB) or 6 days differentiated cells (MT). DNA-binding complexes were resolved on nondenaturing 6% polyacrylamide gel, and the specific DNA complexes are indicated by arrows.
activity. Because overexpression of the AP2 protein and the cancellation of the AP2-binding site cause no changes in the Dp71 promoter activity (Fig. 7), it seems that the AP2 DNA element is irrelevant for the function of this promoter in muscle cells. Downstream from the transcription start site, from ϩ43 to ϩ52, a previously unnoted NFB element was identified, which might participate in the Dp71 promoter function. Recently, it was demonstrated that Sp1 and Sp3 transactivate several promoters by interacting directly with NFB-like elements (44,45). Therefore, it is possible that, in the absence of intact consensus Sp DNA-binding sites, Sp1 and Sp3 might act on the Dp71 promoter by binding to the NFB element. Additional work is required to test this hypothesis.
The presence of multiple potential Sp transcription factor binding sites proximal to the start of transcription is a feature of a number of promoters lacking a consensus TATA box, including the promoter regions for utrophin, epidermal growth factor receptor, and insulin-like growth factor genes (46 -48). It has been proposed that such multiple Sp-binding sites and associated proteins may stabilize the transcriptional machinery and establish a site of transcription start in TATA-less promoters.
As Sp1/Sp3 factors seem to be crucial for the transcriptional activity of the Dp71 minimal promoter region in myoblasts, we envisaged that the drop in the activity of this promoter during myogenesis could be caused by impairment in the functioning of these transcription factors. In this direction, we observed that Sp-specific DNA-protein complexes obtained with myoblast extracts disappeared when nuclear extracts from differentiated muscle cells were employed in the EMSA (Fig. 8B). These findings indicate that the protein levels and/or binding activities of the Sp1 and Sp3 proteins are altered in myotubes. Further immunoblotting experiments clarified this matter; we revealed that Sp1 and Sp3 proteins are present in proliferating myoblasts but their levels decrease drastically as differentiation proceeds (Fig. 8A). Most interestingly, the expression pat-tern of Dp71 resembles that of Sp proteins; Dp71 protein is present in proliferating myoblasts but disappears in myotubes since the 3rd day of induced differentiation (Fig. 1B). These results are consistent with the conclusion that both Sp1 and Sp3 directly regulate the expression of Dp71 in C2C12 muscle cells by interacting with the promoter region in myoblast cells, and the lack of these factors appears to be sufficient to cause down-regulation of the Dp71 expression in myotubes (Fig. 10). Supporting our conclusion, we observed that exogenous expression of Sp1 and Sp3 restores the transcriptional activity of the proximal Dp71 promoter in differentiated muscle cells (Fig. 9). Although these results could not rule out completely the possibility that additional unrevealed repressor factors participate in Dp71 down-regulation, they do indicate that the action of Sp1 and Sp3 on the Dp71 promoter constitute the main, if not the unique, regulatory mechanism working in differentiating muscle cells to modulate Dp71 expression.
Because the extensive distribution of Sp1 and Sp3 in mammalian tissues is consistent with the expression profile of Dp71 (14,42), it is plausible to propose that the positive regulation exerted by these transcription factors on the Dp71 promoter is not a specific phenomenon restricted to myoblasts but a general regulatory mechanism present in different cell types.
Muscle cell differentiation is mediated by the transcription factor MyoD, which acts as a master regulator leading to the activation of many muscle-specific genes (49,50). Recent studies (51,52) have shown that the overexpression of MyoD leads to the repression of Sp1 and Sp3. In this context, although MyoD may not directly participate in decreasing Dp71 transcription in our experimental system, it might indirectly suppress the Dp71 promoter transcription in vivo by repressing genes for Sp1 and Sp3.
The modulation of Dp71 expression during myogenesis has a noticeable relevance in the physiology of muscle tissue. In early myogenesis, Dp71 is expressed to participate in cytoskeletal remodeling (23), whereas in mature muscle fibers its expression must be extinguished to allow dystrophin to be the predominant DMD gene product. It has been proposed that elevated Dp71 expression provokes the Dp71 protein competition in muscle sarcolemma with dystrophin for the available binding sites in the DAP complex. This event could interfere with the normal formation of the linkage between the actin cytoskeleton and the DAP complex (20,21). In fact, FIG. 9. Exogenous expression of Sp1 and Sp3 transcription factors restore the activity of the proximal Dp71 promoter in differentiated C2C12 cells. C2C12 cell cultures were transfected with 1 g of p224CAT reporter plasmid (containing the proximal Dp71 promoter) with variable amounts of pPacSp1 (0.1 and 0.3 g), pPacSp3 (0.1 and 0.3 g), and pPacSp1 and pPacSp3 together (0.1 g each one). Empty vector pPac was used as negative control. Transfected cells were either maintained in proliferating medium (open bar) or induced to differentiate for 48 h by culturing in differentiation medium (black bars). In all cases control plasmid pRSV-␤-gal was included to normalize CAT activities against ␤-galactosidase activities. The reporter activity obtained for p224CAT in undifferentiated cells transfected with the empty vector pPac was set at 100% (black bar), and the CAT activities of differentiated cultures were represented relative to this value. Data are expressed as the mean Ϯ S.D. of at least three independent experiments, each performed in duplicate. the ectopic expression of Dp71 in skeletal muscle of transgenic mice with normal dystrophin expression causes a muscular dystrophy phenotype (22).