Role of Intronic E- and N-box Motifs in the Transcriptional Induction of the Acetylcholinesterase Gene during Myogenic Differentiation*

In this study, we examined whether an intronic N-box motif is involved in the expression of acetylcholinesterase (AChE) during myogenesis. We determined that AChE transcripts are barely detectable in cultured myoblasts and that their levels increase dramatically in myotubes. Nuclear run-on assays revealed that this increase was accompanied by a parallel induction in the transcriptional activity of the AChE gene. These changes in transcription were also observed in transfection experiments using AChE promoter-reporter gene constructs. Mutation of the intronic N-box at position 1 755 base pairs (bp) reduced by more than 70% expression of the reporter gene in myotubes. Disruption of an adjacent E-box, at position 1 767 bp, also reduced expression of the reporter gene following myogenic differentiation. Co-transfection experiments using AChE promoter-reporter gene constructs and a myogenin expression vector showed that expression of this regulatory factor increased expression of the reporter gene in myotubes. Although the AChE promoter contains multi-ple E-boxes, mutation of this intronic one was sufficient to prevent the myogenin-induced increase in reporter gene expression. Together, these results indicate that changes in AChE gene transcription occur during myo-genesis and highlight the contribution of the intronic N-and E-box motifs in the developmental regulation of the AChE gene in skeletal muscle.

Acetylcholinesterase (AChE) 1 is widely recognized as an essential component of cholinergic synapses. In both the central and peripheral nervous systems, it is responsible for the hydrolysis of acetylcholine released from nerve terminals thereby ensuring the efficiency of synaptic transmission. Although a single gene encodes AChE, several molecular forms can be generated as a result of alternative splicing and distinct posttranslational processing (for reviews, see Refs. [1][2][3]. It has been previously suggested that this polymorphism allows for the expression of AChE catalytic subunits in different cell types and at various subcellular locations where specific molecular forms may perform site-specific functions (4).
In skeletal muscle, AChE expression is known to be markedly influenced by the state of differentiation and innervation of the muscle fibers. Numerous studies have indeed reported dramatic changes in the expression of the enzyme during myoand synapto-genesis as well as in mature muscles following alterations in their normal levels of neuromuscular activity and/or in their basic supply of trophic factors. Given that AChE is an excellent marker of synaptic differentiation, several groups have recently begun to decipher the molecular basis underlying these changes in AChE expression. For example, increases in AChE mRNA expression have previously been reported during myogenic differentiation (5)(6)(7)(8)(9). Moreover, activity-induced changes in AChE transcript levels have been observed in both cultured myotubes as well as in skeletal muscle in vivo (10 -14).
Despite these recent advances in our understanding of some of the biosynthetic events regulating AChE expression in skeletal muscle, our knowledge of the specific molecular mechanisms that account for these changes in mRNA expression remains largely unknown. Nonetheless, alterations in the stability of pre-synthesized AChE transcripts have been suggested to account for at least a portion of the changes in mRNA levels seen during myogenic differentiation (5,9) and following muscle denervation (13). On the other hand, the contribution of transcriptional regulatory mechanisms has also been documented in several studies that examined expression of the AChE gene by nuclear run-on assays (13,15) and promoter analyses (16 -18). Together, these studies indicate therefore, that both transcriptional control mechanisms as well as posttranscriptional events contribute to the regulation of AChE mRNA levels in skeletal muscle cells.
In this context, we have recently begun to examine the mechanisms underlying the preferential accumulation of AChE transcripts within the postsynaptic sarcoplasm of muscle fibers (see Refs. 11,17,19). In these studies, we showed that the synaptic accumulation of AChE transcripts results, at least partially, from the local transcriptional activation of the AChE gene (17). By mutation/deletion analysis, we further demonstrated the key role of the first intron in regulating both the muscle-specific expression of the AChE gene as well as its preferential synaptic expression. In particular, our studies have shown the contribution of an intronic N-box motif and of the ets-related transcription factor GABP (see Refs. 20 -24) in the synaptic regulation of the AChE gene in muscle fibers (17).
Given the key role of this intronic N-box in the regulation of AChE in adult skeletal muscle, we sought in the present study to determine whether this DNA element also participates in the control of the AChE gene during myogenic differentiation. In addition, we also examined whether other sites, located within the first intron, are also important in controlling AChE expression by focusing on an adjacent E-box motif.

EXPERIMENTAL PROCEDURES
Cell Culture-Mouse C2C12 cells (ATCC, Manassas, VA) were cultured on Matrigel (Collaborative Biomedical Products, Bedford, MA)coated 60-mm culture dishes in Dulbecco's modified Eagle's medium (Life Sciences/Life Technologies, Inc., Burlington, Ontario) supplemented with 10% fetal bovine serum, 292 ng/ml L-glutamine, and 100 units/ml penicillin-streptomycin, in a humidified chamber at 37°C containing 5% CO 2 . Confluent myoblasts were induced to differentiate and fuse into myotubes by replacing the growth media with differentiation media that contained low serum (5% horse serum). Culture media were changed every 48 h.
RNA Extraction and Reverse-transcription and Polymerase Chain Reaction (RT-PCR)-Total RNA was isolated from myoblasts and myotubes by using 1.0 ml of the TriPure Isolation Reagent (Life Sciences/ Life Technologies, Inc.) per 60-mm culture dish according to the manufacturer's instructions. Briefly, following cell lysis with the TriPure reagent, chloroform was added and the solution was mixed vigorously prior to centrifugation at 12,000 ϫ g for 15 min at 4°C. The RNA contained in the resulting aqueous layer was precipitated with isopropanol, and the pellet was washed several times with 70% ethanol. The RNA was then resuspended in RNase-free water and stored at Ϫ80°C.
Quantitation of the amount of total RNA in each sample was performed using the Amersham Pharmacia Biotech Gene Quant II RNA/ DNA spectrophotometer. Each sample was adjusted to a final concentration of 80 ng/l. Semi-quantitative RT-PCR was carried out as previously described in detail elsewhere (11,13,19,25). Reverse transcription, using random hexamers, was performed for 45 min at 42°C, followed by a 5-min incubation at 99°C. PCR was then used to amplify cDNAs corresponding to AChE and 12 S rRNA. Primers that amplify AChE T transcripts were synthesized according to a previous report (8). These were located in exons 4 (5Ј-3Ј: CTGGGGTGCGGATCGGTGTAC-CCC) and 6 (5Ј-3Ј: TCACAGGTCTGAGCAGCGTTCCTG) in the AChE gene and gave rise to a 670-bp PCR product. The PCR cycling parameters included 1-min denaturation at 94°C followed by a 3-min primer annealing and extension at 70°C (11,13,25). The rRNA primers were synthesized based on available sequences (26). The 5Ј (5Ј-3Ј: GGAAG-GCATAGCTGCTGG) and 3Ј (5Ј-3Ј: CCTCGATGACATCCTTGG) primers amplified a 368-bp target. The cycling parameters for rRNA were 1-min denaturation at 94°C, 1-min primer annealing at 54°C, and 2-min extension at 72°C. For each reaction, a final 10-min elongation step was carried out at 72°C following the last cycle of amplification. Quantitation of the relative abundance of the PCR products was performed following agarose gel electrophoresis and visualization of the amplified cDNAs using the fluorescent dye VistraGreen (Amersham Pharmacia Biotech, Arlington Heights, IL). The analysis was performed using a Storm PhosphorImager and the accompanying ImageQuaNT software (Molecular Dynamics, Inc., Sunnyvale, CA). The values obtained for AChE were standardized relative to the corresponding level of rRNA in the same sample. In these experiments, negative controls consisted of reactions in which total RNA was replaced by RNase-free water as well as reactions in which the reverse transcriptase was omitted.
All RT-PCR measurements aimed at determining the relative abundance of AChE mRNA and rRNA were performed during the linear range of amplification (see Refs. 11,13,19,25). Typically, the cycle numbers were 32 for AChE and 28 for rRNA. RT-PCR conditions (primer concentrations, input RNA, choice of RT primer, cycling conditions) were initially optimized, and these were identical for all samples. Appropriate precautions were taken to avoid contamination and RNA degradation. All samples as well as negative controls were prepared using common master mixes containing the same RT and PCR reagents, and they were always run in parallel. In all experiments, PCR products were never detected for the negative controls.
Nuclear Run-on Assay-Nuclear run-on assays were performed as previously described in detail (13,25,27). Briefly, nuclei from myoblasts or myotubes were first isolated and resuspended in a transcription buffer that contained GTP, ATP, CTP, and [␣-32 P]UTP. RNA was transcribed for 30 min at 30°C in the presence of an RNase inhibitor (Promega, Madison, WI). Following RQ1 DNase (Promega) treatment, the nascent radiolabeled RNA was isolated using the TriPure method (see above) and hybridized for 48 h to 10 g each of linearized AChE and myogenin cDNAs immobilized on a GeneScreen Plus nylon membrane. Following hybridization, the membranes were thoroughly washed at 42°C with a 1ϫ saline-sodium citrate, 0.1% sodium dodecyl sulfate (SSC) solution and subjected to autoradiography. The intensity of the resulting signals was quantified using the Storm PhosphorImager and the accompanying ImageQuaNT software (Molecular Dynamics). The signals corresponding to AChE and myogenin were standardized relative to the signals obtained for genomic DNA.
In Vivo and In Vitro Analyses of Rat AChE Promoter-reporter Gene Expression-The 1.9-kb rat AChE promoter fragment termed NRAP (N-box-containing rat AChE promoter) was used in the present studies. This promoter fragment has been described in detail recently (see Ref. 17). Briefly, it is located ϳ600 bp from the translation start site and contains 807 bp upstream of the initiator element and 1075 bp of intron 1. For the current studies, NRAP was inserted into a luciferase reporter construct (pGL3-Basic vector from Promega). Mutagenesis of putative regulatory regions within NRAP was performed using the Altered Sites II in vitro mutagenesis system (Promega). Briefly, the first intronic N-box (positioned in a reverse orientation), located at position ϩ755 bp (17), was mutated from the wild-type CCGGAA to CTTGAA, and the resulting mutant promoter fragment was termed mN-NRAP. In addition, the intronic E-box element adjacent to this N-box motif and located at position ϩ767 bp, was mutated from the wild-type CAGCTG to CCTAGG, and designated as mE-NRAP. Each mutant promoter fragment was inserted into the luciferase reporter vector. Finally, a third mutant, in which the E-and N-box motifs were both mutated, was also used in our studies (mNmE-NRAP).
Plasmid DNA was prepared using the Qiagen (Chatsworth, CA) mega-or midi-prep procedure. Pellets were resuspended in sterile phosphate-buffered saline (PBS). Transfection of cultured myogenic cells was performed using the LipofectAMINE reagent kit (Life Sciences/Life Technologies, Inc.). Myoblasts, at 50 -60% confluence, were transfected with the appropriate promoter-reporter gene construct. They were subsequently either maintained as myoblasts for 2 days, or induced to differentiate into myotubes. Myoblasts or myotubes were then washed with ice-cold PBS and lysed in Reporter Lysis buffer (Promega) following two cycles of freeze-thaw. The solution was then centrifuged at 15,000 ϫ g, and the resulting supernatant was assayed for luciferase activity as described elsewhere (28).
Direct gene transfer into mouse skeletal muscle was performed as previously described (13,17,29). 25 l of the appropriate promoterreporter constructs, diluted at a final concentration of 2 g/l, were injected into mouse tibialis anterior (TA) muscle. Seven days later, muscles were excised, quickly frozen in liquid nitrogen, and stored at Ϫ80°C for subsequent analysis. For detection of luciferase activity, muscles were homogenized in Reporter lysis buffer (Promega) using a Polytron set at maximum speed. Following centrifugation at 15,000 ϫ g, the supernatant was assayed for luciferase activity.
For both direct injection and transfection experiments, a chloramphenicol acetyltransferase (CAT) plasmid driven by the SV40 promoter, was used to control for the efficiency of transduction/transfection. The luciferase activity obtained in each sample was therefore normalized to CAT activity.
Nuclear Protein Extraction-Nuclear proteins were extracted from myoblasts and myotubes as previously described (30). Briefly, myoblasts and myotubes were scraped into ice-cold PBS and collected by centrifugation at 200 ϫ g for 5 min. The cells were resuspended and washed once in 1 ml of ice-cold PBS and centrifuged as above. The cells were resuspended in 1 ml of buffer A (10 mM Hepes, pH 7.9; 10 mM KCl; 1.5 mM MgCl 2 ; 1.5 mM DTT; and 0.5 mM phenylmethylsulfonyl fluoride (PMSF)) and centrifuged at 200 ϫ g for 5 min at 4°C. The cells were then lysed in 300 l of buffer A containing 0.1% Nonidet P-40 for 20 min on ice. The homogenate was spun at 15,000 ϫ g for 10 min at 4°C. The supernatant was discarded, and the pellet containing the nuclei was resuspended in 35 l of buffer B (20 mM Hepes, pH 7.9, 420 mM NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 25% glycerol, 0.5 mM DTT, 0.5 mM PMSF, and the protease inhibitors spermidine, spermine, aprotinin, leupeptin, and pepstatin) for 45 min on ice to extract nuclear proteins. The nuclear extract was then obtained following centrifugation at 15,000 ϫ g for 15 min at 4°C. An equal volume of buffer C (20 mM Hepes, pH 7.9, 50 mM KCl, 1.0 mM EDTA, 0.1 mM EGTA, 20% glycerol, 0.5 mM DTT, and 0.5 mM PMSF) was subsequently added to the nuclear extract, which was then stored at Ϫ80°C. Protein concentration was determined using the Bradford Assay (Bio-Rad, Hercules, CA).
Electrophoretic Mobility Shift Assay (EMSA)-Equal amounts of nuclear proteins (5 g) were incubated in a reaction mixture containing 17 mM Hepes, 0.2 mM EDTA, 0.04 mM EGTA, 0.7 mM DTT, 140 mM NaCl, 0.5 mM MgCl 2 , 36 mM KCl, 16% glycerol, and 5.0 g of poly(dI-dC) for 10 min at room temperature. The reaction mixtures were then incubated with 0.2 ng of 32 P-labeled double-stranded oligonucleotides corresponding to either the intronic N-or E-box motifs for 20 min at room temperature. 5 l of loading buffer containing 20 mM Hepes, 100 mM KCl, 60% glycerol, 0.5 mM EDTA, 0.5 mM EGTA, and 0.125% bromphenol blue, was added to the reaction mixture prior to electrophoresis on a 5% native polyacrylamide gel. The gels were run in Tris-glycine solution for 2.5 h at 200 V. They were then dried between filter paper and cellophane for 1 h at 80°C under a vacuum, exposed to x-ray films for up to 2 days at Ϫ80°C. For competition assays, the reaction mixture was incubated with a 250-fold excess of unlabeled probe for 20 min at room temperature prior to the addition of the labeled oligonucleotides. The sequence of the synthetic oligonucleotides used for EMSAs were (all in 5Ј33Ј orientation) CTGGAGAAGCCGGAACTACAGCAG (sense) and CTGCTGTAGTTCCGGCTTCTCCAG (antisense) for the intronic N-box; CTGGAGAAGCTTGAACTACAGCAG (sense) and CTGCTGTA-GTTCAAGCTTCTCCAG (antisense) for the mutant intronic N-box (mutations are underlined); AACTACAGCAGCTGTTGCCCCCAA (sense) and TTGATGTCGTCGACAACGGGGGTT (antisense) for the E-box; and AACTACAGCCTAGGTTGCCCCCAA (sense) and TTGAT-GTCGGATCCAACGGGGGTT (antisense) for the mutant E-box (mutations are underlined). For labeling, double-stranded oligonucleotides (20 ng), T4 Polynucleotide kinase and [␣-32 P]ATP (60 Ci) were mixed in a kinase buffer (50 mM Tris-HCl, pH 7.6, 10 mM MgCl 2 , 5% glycerol, and 5 mM DTT) and incubated at 37°C for 1 h. Labeled oligonucleotides were isolated by centrifugation through a Sephadex G-50 column (Amersham Pharmacia Biotech) at 3000 ϫ g for 5 min.

RESULTS
Expression of the AChE Gene during Myogenic Differentiation-In a first series of experiments, we examined expression of the AChE gene during myogenic differentiation of mouse C2C12 cells grown in culture. Specifically, we first compared the relative abundance of AChE mRNAs in undifferentiated myoblasts (50% confluent), confluent myoblasts, and differentiated myotubes. As observed previously (see Refs. 5, 7), levels of AChE transcripts increased dramatically during myogenic differentiation. Fig. 1A shows that AChE mRNAs were barely detectable in undifferentiated myoblasts (50% confluent) and became more expressed, albeit still at low levels, in 100% confluent myoblasts. Upon differentiation and fusion of confluent myoblasts into multinucleated myotubes, AChE mRNA levels increased significantly (p Ͻ 0.05) and by ϳ3to 4-fold in myotubes exposed to differentiation media for 2 days (Fig. 1B). In 4-day-old myotubes, AChE mRNA levels were further increased reaching, in this case, more than 10-fold as compared with the levels seen in 100% confluent myoblasts.
We next examined whether this increase in AChE mRNA expression during myogenic differentiation could be attributed to changes in the transcriptional activity of the AChE gene. To this end, we performed nuclear run-on assays using myonuclei isolated from cultured C2C12 cells at various stages of myogenic differentiation ( Fig. 2A). As expected (see for example Ref. 5), we observed an increase in the transcriptional activity of myogenin during myogenesis. Similarly, we also detected a gradual increase in the transcriptional rate of the AChE gene in muscle cells undergoing myogenic differentiation. This increase was transient, because in 4-day-old myotubes, the rate of transcription appeared to decrease toward levels seen in 100% confluent myoblasts. These results show, therefore, that the initial increase in AChE mRNA levels seen during myogenic differentiation was accompanied by an induction in the transcriptional activity of the AChE gene, thereby indicating that changes in transcription account for some of the increase in AChE expression.
The Transcriptional Induction of the AChE Gene during Myogenic Differentiation Depends on Intronic E-and N-boxes-In a subsequent series of experiments, we assessed the contribution of cis-and trans-acting elements in the transcrip-tional regulation of AChE during myogenic differentiation. Rat AChE promoter-reporter gene constructs were transfected into C2C12 myoblasts at ϳ50 -60% confluence (see "Experimental Procedures"). The cells were subsequently harvested when myoblasts became confluent or following myogenic differentiation. For these experiments, a rat AChE promoter fragment, termed NRAP (17), was linked to a luciferase reporter gene (Fig. 3). Additionally, several mutated NRAP fragments were also used in these studies (see Fig. 3, A and B), including one in which both the intronic E-and N-boxes had been mutated.
In myoblasts, NRAP appeared active since we were able to detect some luciferase activity (Fig. 3C). Mutations of the intronic E-(mE-NRAP) and N-(mN-NRAP) boxes at position ϩ767 and ϩ755 bp (see Ref. 17), respectively, had no effect on expression of the reporter gene in these confluent myoblasts. By contrast, NRAP appeared very active in differentiated myotubes, because the activity of luciferase was significantly (p Ͻ 0.05) higher (ϳ6-fold). Furthermore, mutations of the intronic E-and N-boxes markedly reduced expression of the reporter gene. Specifically, mutation of the N-box at position ϩ755 bp led to a decrease of ϳ75% (p Ͻ 0.05) in luciferase activity.

FIG. 1. Increased expression of AChE mRNAs during myogenic differentiation.
Semi-quantitative RT-PCR analysis was performed using total RNA isolated from myoblasts (MB) at 50% or 100% confluence, and from myotubes (MT) following exposure to differentiation media for 2 or 4 days. A, representative ethidium bromide-stained agarose gels containing RT-PCR products corresponding to AChE mRNA and 12 S rRNA. B, the relative abundance of AChE transcripts in muscle cells at different stages of myogenic differentiation, expressed as a percentage of AChE transcripts in myoblasts (100%). In this analysis, levels of AChE mRNA were normalized according to the levels of 12 S rRNA seen in the same samples. Note the progressive increase in AChE transcript levels as differentiation proceeds. Mean Ϯ S.E. is shown; a minimum of three independent experiments were performed.
Mutation of the adjacent intronic E-box essentially abolished the transcriptional activation of NRAP in differentiated myotubes, because the levels of luciferase were reduced by ϳ90%. Due to this pronounced effect however, we were not able to demonstrate cooperativity between the adjacent E-and N-box motifs, because disruption of both DNA elements (mNmE-NRAP) reduced expression of the reporter gene to a comparable extent in differentiated myotubes.
Given that both the intronic E-and N-box motifs appeared necessary for the transcriptional induction of the AChE gene during myogenic differentiation, we next investigated the contribution of trans-acting elements. To this end, we performed a series of EMSAs using radioactive oligonucleotides encompassing the AChE intronic N-box motif at position ϩ755 bp (17) and the intronic E-box motif at position ϩ767 bp. As expected (31, 32), we detected a relative increase in the binding activity to the E-box with nuclear proteins extracted from myotubes versus myoblasts (Fig. 4).
Analysis of the N-box binding activity revealed a single and specific band following incubation of the appropriate radiolabeled oligonucleotides with nuclear proteins isolated from C2C12 cells (Fig. 5A). This protein complex was effectively competed by the addition of a 250-fold excess of unlabeled wild-type probe, whereas an excess of unlabeled mutant probe, encompassing the same mutation as used to generate the mN-NRAP construct, failed to compete with this protein complex (Fig. 5A). In addition, this band was also supershifted when the extracts and oligonucleotides were incubated together with antibodies directed against GABP␣ and ␤ (kindly provided by Dr. Laurent Schaeffer) thereby confirming the identity of the proteins present in this complex (Fig. 5B). The binding activity of GABP to the N-box motif appeared slightly elevated in nuclear extracts isolated from myotubes as compared with myoblasts (Fig. 5C). The slight increase in GABP expression in myotubes was confirmed by Western blot analysis (data not shown; see also Ref. 53).
Myogenin Increases Expression of AChE via the Intronic Ebox Motif-In a separate series of studies, we further ascertained the role of myogenic factors and the intronic E-box at position ϩ755 bp in the regulation of the AChE gene during myogenic differentiation. To this end, we co-transfected into myoblasts a myogenin expression vector (33) along with a specific AChE promoter-reporter gene construct and harvested the cells after exposure to the differentiation media for 4 days. Fig.  6 shows that overexpression of myogenin increased by almost 2-fold (p Ͻ 0.05) the activity of luciferase in differentiated myotubes. As suggested by the experiments presented above, this increase in luciferase activity is dependent upon a single E-box motif, because co-transfection of the myogenin expression vector with the mE-NRAP construct, in which the single intronic E-box is mutated (at position ϩ767 bp), totally pre-

FIG. 2. Transcription of the AChE gene increases during myogenic differentiation.
Nuclear run-on assays were performed with nuclei isolated from myoblasts (MB) at 50% or 100% confluence, and from myotubes (MT) following exposure to differentiation media for 2 or 4 days. A, shows representative autoradiograms corresponding to the transcriptional activity of the AChE and myogenin genes during myogenic differentiation. Genomic DNA was used in these assays as a control. B, represents the quantitative analysis. Signals for AChE and myogenin were normalized to that observed for genomic DNA.

FIG. 3. Intronic N-and E-box motifs are involved in the transcriptional induction of the AChE gene during myogenic differentiation.
Myoblasts were transfected with the appropriate AChE constructs (wild-type NRAP, mutated mN-NRAP (A), mutated mE-NRAP (B), or double mutant mNmE-NRAP) and with a constitutively expressed CAT plasmid. Cells were harvested when they were still myoblasts or when they became myotubes. C, the activity of luciferase in proteins extracted from the muscle cells following transfection with the different constructs. Luciferase activity was normalized to that seen with CAT. Asterisks indicate significant differences (p Ͻ 0.05 versus NRAP). Mean Ϯ S.E. is shown; a minimum of three independent experiments were performed in triplicate.
vented the increase seen with co-transfection of the wild type NRAP construct (Fig. 6).
The Intronic E-box and N-box Regulate Expression of the AChE Gene in Vivo-In a final series of experiments, we assessed the contribution of these cis-acting elements in vivo by directly injecting into mouse TA muscles, the various promoterreporter gene constructs and by monitoring 7 days later the activity of luciferase normalized to a co-injected constitutively expressed CAT plasmid. As previously observed (17), the NRAP fragment is very active in vivo and mutation of the intronic N-box at position ϩ755 bp, reduced significantly (p Ͻ 0.05) the expression of the reporter gene (Fig. 7). In agreement with our data obtained from cultured myogenic cells, functional disruption of the adjacent E-box (see Fig. 5B) in NRAP eliminated completely expression of the reporter gene further highlighting the crucial role of this DNA element in regulating expression of the AChE gene in muscle cells.

DISCUSSION
Myogenic differentiation is a developmentally regulated process characterized by a series of coordinated biochemical and morphological changes accompanied by the fusion of mononucleated myoblasts into multinucleated myotubes. Together with cytoskeletal and contractile proteins, expression of several synaptic proteins, including AChR, N-CAM, utrophin, and AChE, is also enhanced to varying degrees during myogenesis. In this context, previous studies have shown that, in the case of AChR (34 -37) and utrophin (38), the increased expression of mRNAs can partially be attributed to an increase in the rate of transcription of these genes. Here, we show that transcription of the AChE gene is also increased during the early phases of myogenic differentiation. In fact, it appears that during the initial stages of muscle cell development, the rate of transcription of the AChE gene correlates well with the pattern of mRNA expression (compare undifferentiated myoblasts, confluent myoblasts, and 2-day-old myotubes in Figs. 1 and 2). In agreement with our findings, Rossi et al. (39) and R. L. Ro-tundo 2 have also reported recently an increase in the expression of the AChE gene that parallels the induction in AChE mRNA at early stages of chick muscle cell development. Thus, as originally suggested by Merlie and Sanes (40) as well as by Klarsfeld (41), expression of genes encoding key synaptic proteins in muscle may indeed be coordinately regulated during myogenic differentiation.
Together with the recent data obtained by Rossi et al. (39) and Rotundo, 2 our results appear in contrast to the earlier findings of Taylor and colleagues (5,9). In their previous work, these investigators failed to detect an increase in the transcriptional activity of the AChE gene, thereby concluding that posttranscriptional mechanisms operating at the level of mRNA stability accounted for the increased AChE expression seen during myogenic differentiation. However, this discrepancy may be reconciled if we consider that, in their experimental approach, Taylor and co-workers focused on more advanced stages of myogenic differentiation during which maturational events are likely taking place. In support of this, it is important to note that we (present study) and Rossi et al. (39) and Ro-2 R. L. Rotundo, personal communication.

FIG. 4. The E-box binding activity increases during myogenic differentiation.
A, binding activity to the E-box motif in nuclear extracts from myoblasts (MB) and myotubes (MT). The specific bands are indicated by the arrows whereas an arrowhead points to the unbound probe. B, specificity of the signals since the bands were essentially eliminated by competition with a 250ϫ molar excess of wild type unlabeled (WT) probe. Note also that the bands remained unaffected by a competition assay using the mutant oligonucleotide. The oligonucleotide used in this assay contains an E-box binding site but does not encompass an N-box binding site.
FIG. 5. Binding activity to the N-box motif during myogenic differentiation. A, shows that a single complex is seen when nuclear extracts from myoblasts (MB) are incubated with an oligonucleotide encompassing the N-box motif. The complex is competed with a 250ϫ molar excess of wild type unlabeled probe. Note also that the signal is unaffected by competition with the mutant oligonucleotide. B, the result of a supershift experiment using antibodies against GABP ␣ and ␤, which confirms the identity of the trans-acting factors binding to the N-box motif. Supershifted complexes are indicated by the white arrow. C, the relative abundance of the binding activity in myoblasts (MB) versus myotubes (MT). tundo 2 all observed a decrease in the transcriptional activity of the AChE gene in older myotubes, an observation entirely consistent with the results of Taylor and colleagues who also detected a slight reduction in the expression of the AChE gene in fully differentiated myotubes (5). Taken together, these results indicate therefore that transcriptional mechanisms participate in the regulation of AChE mRNA expression during myogenic differentiation but only at the earliest stages.
As stated above, the increase in AChE mRNA expression that occurred in confluent myoblasts and 2-day-old myotubes can be solely explained by an increase in the transcriptional activity of the AChE gene. This increase in transcription appeared transient, because it returned toward control levels in older myotubes. At that stage however, AChE mRNA levels were further increased indicating, therefore, that post-transcriptional mechanisms likely account for this sustained increase. In this context, Taylor and colleagues have also implied an important contribution for post-transcriptional mechanisms during the later stages of muscle differentiation (5,9). In agreement with this view, it is relevant to note that recent studies performed in our laboratory have also illustrated the key role of the AChE 3Ј-untranslated repeat and of distinct RNA-binding proteins in regulating the stability of AChE transcripts in mature myotubes (42). Together, these results indicate, therefore, that during the early stages of myogenic differentiation, transcriptional regulatory mechanisms play a predominant role in the control of AChE expression whereas the contribution of post-transcriptional events becomes more important at later stages. The concerted effects of these various mechanisms are likely important for the rapid increase in AChE expression in developing muscle cells and to ensure that this induction is well maintained in mature myotubes as to prepare them for the arrival of exploratory motor axons.
Recently, a DNA element termed an N-box motif was identified (20) and shown to be critical for directing the preferential synaptic expression of genes encoding the acetylcholine receptor ␦- (20,22) and ⑀-(21) subunits as well as the utrophin (24) and AChE (17) genes. Interestingly, in the case of the AChE gene, we noted that, although four N-boxes are located within 800 bp of the initiator element (two in the promoter region and two in the first intron; see Ref. 17), the one located in the first intron at position ϩ755 bp was critical for controlling the synapse-specific expression of the reporter gene. Given our results showing the transcriptional induction of the AChE gene during myogenic differentiation, we therefore became interested in determining whether this N-box also participated in the regulation of AChE expression in developing muscle cells.
Our transfection experiments showed, in agreement with our nuclear run-on data, that indeed expression of the AChE gene is increased during myogenic differentiation. Furthermore, these experiments revealed the key role played by the intronic N-box in controlling expression of the AChE gene during differentiation of muscle cells. To our knowledge, this is the first report demonstrating the importance of the N-box motif during myogenic differentiation.
Several recent studies have shown that the ets-related transcription factors GABP ␣ and ␤ can bind to the N-box motif to transactivate genes encoding synaptic proteins (17,(22)(23)(24)43). Therefore, we sought to determine whether expression of GABP was affected during myogenic differentiation. Electrophoretic mobility shift assays revealed that GABP-binding activity to the N-box was slightly increased in nuclear extracts from myotubes versus myoblasts, and this was confirmed by Western blot analysis. Although the increase appears rather modest, it is important to note that the transactivation potential of GABP is also influenced by its phosphorylation status with stronger transcriptional activation occurring in the ab-FIG. 6. Myogenin increases expression of AChE-promoter-reporter gene constructs. Myoblasts were co-transfected with vectors expressing myogenin and CAT, and with either the wild-type NRAP or the mE-NRAP construct (see Fig. 3B). The cells were harvested at the myotube stage. Expression of luciferase and CAT in the transfected cells was assessed in protein extracts. All luciferase activity data were normalized to CAT. The asterisks indicate a significant difference (p Ͻ 0.05) versus NRAP, and E indicates a difference (p Ͻ 0.05) versus NRAP ϩ myogenin. Mean Ϯ S.E. is shown; a minimum of three independent experiments were performed in triplicate.

FIG. 7.
The intronic E-and N-box motifs are essential for in vivo expression of the AChE gene. Mouse TA muscles were directly injected with the plasmids containing the wild type NRAP promoter fragment, mN-NRAP and mE-NRAP (see Fig. 3, A and B) and with a constitutively active CAT plasmid. Luciferase activity was then assayed and corrected to the activity seen for CAT. The asterisks indicate p Ͻ 0.05 versus NRAP. Mean Ϯ S.E. is shown; a minimum of four muscles were analyzed per construct. sence of an increase in DNA binding activity (22). These results, together with the promoter analysis, further highlight the importance of the trans-acting element GABP and its corresponding cis-acting element, i.e. the N-box motif, in the expression of AChE during myogenic differentiation.
The ets-related transcription factors often cooperate with other transcription factors (for reviews, see Refs. 44,45), including AP-1 (46), or with cofactors such as the CREB-binding protein also known as CBP/p300 (47), to exert their effects. Because ets-related factors, including GABP, may also possess a conserved domain with homology to basic helix-loop-helix transcription factors such as myogenic factors (48), it appeared possible that GABP could in fact interact with myogenic factors to regulate expression of the AChE gene. This view appeared particularly attractive given the presence of an E-box in the immediate vicinity of the N-box motif at position ϩ755 bp in the first intron of the AChE gene. Promoter analysis showed that indeed this E-box is crucial in regulating expression of the AChE gene during myogenic differentiation, because its mutation reduced by more than 90% expression of the reporter gene in transfected myotubes. Consistent with these findings, we further showed by co-transfection of AChE promoter-reporter gene constructs and a myogenin expression vector, that myogenin increases the expression of luciferase in myotubes. Remarkably, this effect was dependent upon a single E-box, because mutation of the intronic E-box prevented the myogenin-induced increase in reporter gene expression despite the presence of numerous E-boxes throughout the promoter region of the AChE gene (see Ref. 17). Our findings are therefore in agreement with the well-known effects of myogenic factors on expression of genes in differentiating muscle cells (for reviews, see Refs. 49 -51). In addition, our direct plasmid injection in mouse TA muscles also showed that this E-box plays a critical role in regulating the basal level of expression of the AChE gene in vivo. Taken together with our previous study (17), these data clearly illustrate the importance of elements located within the first intron of the AChE gene in the control of its expression in skeletal muscle (see also Ref. 52).