Mechanism for the reduction of telomerase expression during muscle cell differentiation.

Telomerase, the reverse transcriptase that maintains telomere DNA, is usually undetectable in adult human tissues, but is positive in embryonic tissues and in cancers. However, in rodents, several organs of normal adult animals express substantial amounts of telomerase activity. To elucidate relevant control mechanisms operating on the tissue-specific expression of telomerase in rodents, we examined the transcriptional regulation of telomerase reverse transcriptase (mTERT) gene in muscle cell differentiation. Reverse transcriptase-polymerase chain reaction analysis showed that the reduction of telomerase activity was caused by the decrease of mTERT mRNA level during myogenesis. Transfections of mTERT promoter showed that the proximal 225-base pair region is the core promoter responsible for basal transcriptional activity and also participates in the reduced transcription after muscle differentiation. Electrophoretic mobility shift assays showed that this region contained the GC-boxes, which bind to Sp1 family proteins, and the E-box, which binds to c-Myc. Furthermore, DNA binding activities of Sp1, Sp3, and c-Myc were down-regulated during myogenesis. These data suggest that Sp1, Sp3, and c-Myc have critical roles of TERT transactivation in mouse, and the lack of these transcription factors cause down-regulation of mTERT gene expression in muscle cells differentiation.

Telomerase, the reverse transcriptase that maintains telomere DNA, is usually undetectable in adult human tissues, but is positive in embryonic tissues and in cancers. However, in rodents, several organs of normal adult animals express substantial amounts of telomerase activity. To elucidate relevant control mechanisms operating on the tissue-specific expression of telomerase in rodents, we examined the transcriptional regulation of telomerase reverse transcriptase (mTERT) gene in muscle cell differentiation. Reverse transcriptase-polymerase chain reaction analysis showed that the reduction of telomerase activity was caused by the decrease of mTERT mRNA level during myogenesis. Transfections of mTERT promoter showed that the proximal 225-base pair region is the core promoter responsible for basal transcriptional activity and also participates in the reduced transcription after muscle differentiation. Electrophoretic mobility shift assays showed that this region contained the GC-boxes, which bind to Sp1 family proteins, and the E-box, which binds to c-Myc. Furthermore, DNA binding activities of Sp1, Sp3, and c-Myc were down-regulated during myogenesis. These data suggest that Sp1, Sp3, and c-Myc have critical roles of TERT transactivation in mouse, and the lack of these transcription factors cause down-regulation of mTERT gene expression in muscle cells differentiation.
Telomeres are specialized structures at the ends of chromosomes composed of DNA and proteins that are essential for maintaining the stability of the eukaryote genome (1). In vertebrates, they consist of tandem hexanucleotide repeats, (TTAGGG) n , maintained by a specialized ribonucleoprotein enzyme, called telomerase, which adds motif-specific nucleotides using its RNA subunit as a template. Recently, three major telomerase subunits have been identified. The telomerase RNA component provides the template for telomere repeat synthesis (2,3), the telomerase-associated protein (TEP1) binds to telomerase RNA and coordinates assembly of telomerase holoenzyme (4,5), and the most important component responsible for the catalytic activity of telomerase is telomerase reverse transcriptase (TERT) 1 (6,7). Previous studies have shown that TERT is expressed in most malignant tumors but not in normal human tissues and that expression of TERT is closely associated with telomerase activity, whereas two other components are constitutively expressed in both tumors and normal tissues (8,9). These findings indicate that TERT is a rate-limiting determinant of catalytic activity of telomerase. Analysis of the human and mouse TERT promoters reveals that they are regulated by a number of inducible transcription factors, including c-Myc and NF-B (10 -14). These transcription factors are likely to contribute to the observed instances of TERT gene activation.
Most telomerase-positive cells are highly regenerative or immortal. Among normal tissues of adult humans, telomerase activity was almost undetectable except germ line cells, although very small amounts of activity are detectable in normal bone marrow, peripheral blood leukocytes, lymphoid cells, and skin epidermis (15)(16)(17)(18). In contrast, highly regenerative tissues of normal rodents express modest levels of telomerase even in adult animals (19,20). In the normal mouse, telomerase activity exists in colon, liver, ovary, and testis but not in brain, heart, stomach, and muscle (21,22). Moreover, cell differentiation causes reduced telomerase activity in some kinds of cell types such as murine F9 teratocarcinoma and C2C12 myoblast cells (23,24). These results suggest that this remarkable difference among various tissues in mouse may reflect the different regulation mechanisms of telomerase expression, which may cause the tissue-specific features of differentiation and proliferation. However, the detailed mechanisms that contribute to the telomerase expression during development and cell differentiation in mouse are largely unknown. In this study, to understand the mechanisms that reduce mTERT gene expression during muscle cell differentiation, we constructed a series of reporter plasmids containing the 5Ј-franking sequence of mTERT gene and transfected these constructs into mouse cell lines, including C2C12 myoblasts. Transcriptional activity of mTERT was dependent on the proximal 225-bp region of promoter, and this promoter region contained E-boxes and GCboxes, which bind to bHLH proteins and Sp1 family proteins, respectively. We attempted to identify the transcription factors directing mTERT expression and found that Sp1, Sp3, and c-Myc, but not MyoD, play crucial roles in the regulation of mTERT transcription during muscle cell differentiation.
Cell Culture-Mouse NIH3T3 fibroblasts, C3H10T1/2 fibroblasts, and C2C12 myoblasts were obtained from Riken Cell Bank (Tukuba, Japan) and maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 units/ml penicillin, and 100 g/ml streptomycin (standard medium). In some experiments, subconfluent C2C12 cells were induced to differentiate by lowering fetal bovine serum to final concentration of 0.5% (differentiation medium).
RNA Isolation and Analysis-Total cytoplasmic RNA was isolated using the guanidine isothiocyanate method from the cells that were cultured in standard medium or differentiation medium for 1, 4, 7, or 10 days. Samples of 20 g of total RNA were denatured, separated by electrophoresis in a 1% agarose gel containing formaldehyde, and transferred to GeneScreen membranes (PerkinElmer Life Sciences). The membranes were prehybridized and then hybridized with cDNA probes labeled with [␣-32 P]dCTP using a random primer labeling system (Amersham Pharmacia Biotech, Buckinghamshire, UK). After hybridization, the membranes were washed and exposed to x-ray film. All blots were rehybridized with a glyceraldehyde-3-phosphate dehydrogenase cDNA probe to normalize for mRNA loading differences. To quantify the contents of mRNA in the cells, the membranes were exposed to imaging plates and radioactivities were measured with a bioimage analyzer, Fijix BAS 1500 (Fuji Film, Tokyo, Japan).
RT-PCR Amplification of a Mouse TERT cDNA Fragment-The RT-PCR primers, 5Ј-AGACTSCGCTTCATCCCCAAG-3Ј (sense) and 5Ј-GT-CTGGAGGCTGTTCACCTGC-3Ј (antisense), were constructed according to the mouse TERT cDNA (GenBank accession number AF073311) (7). Total RNA was isolated from C2C12 cells as described above. First-strand cDNA was synthesized with 1 g of total RNA using a First Strand cDNA synthesis kit (Life Technologies, Inc., Gaithersburg, MD) in the presence of 1.6 g of oligo-(dT) 2 primer in a final volume of 25 l. After denaturation for 5 min at 94°C, 4 l of reaction product was amplified by PCR for 29 cycles (94°C for 30 s; 55°C for 30 s; 72°C for 1 min). The amplified products were separated by electrophoresis on either a 2% agarose gel or 8% polyacrylamide gel electrophoresis (PAGE) and visualized with ethidium bromide. The DNA sequences of RT-PCR products were confirmed at least once by DNA sequencing and were found to be identical to the corresponding sequence of mouse TERT cDNA (nucleotides 1857-2975, data not shown). Each RT-PCR was performed three times with independent preparations of RNA, and typical results are shown in Fig. 1. As an internal control, RT-PCR of glyceraldehyde-3-phosphate dehydrogenase was performed for all RNA samples, using the PCR primers, 5Ј-ACCACAGTCCATGCCATCAC-3Ј (sense) and 5Ј-TCCACCACCCTGTTGCTGTA-3Ј (antisense). The linearity of RT-PCR with respect to RNA amount was determined, and the estimation of mRNA for mTERT was done within a linear range.
Extraction of Telomerase-C2C12 cells were washed twice with icecold phosphate-buffered saline, scraped off, and transferred to 1.5-ml microtubes. After centrifugation for 3 min at 1000 ϫ g at 4°C, pellets were suspended in 500 l of the wash buffer (pH 7.5) containing 10 mM HEPES, 1.5 mM MgCl 2 , 10 mM KCl, 1 mM dithiothreitol, and centrifuged again for 3 min at 1000 ϫ g at 4°C. Whole cell extracts were prepared by suspending cells in 30 l of CHAPS lysis buffer, containing 10 mM Tris-HCl, pH 7.5, 1 mM MgCl 2 , 1 mM EGTA, 0.1 mM phenylmethylsulfonyl fluoride, 5 mM 2-mercaptoethanol, 0.5% CHAPS, and 10% glycerol. Cell suspensions were subjected to three cycles of freezing and thawing using liquid nitrogen. After being placed on ice for 30 min, they were centrifuged for 30 min at 10,000 ϫ g and the supernatant was collected (CHAPS extracts). Three independent experiments were performed to measure telomerase activity of the cells cultured with differentiation medium.
Assay of Telomerase-The TRAP (telomere repeat amplification protocol) assay was employed with minor modifications described by Kim et al. (25). The cell extract (representing 1 ϫ 10 5 cells) was incubated for 30 min at 37°C in a mixture containing 20 mM Tris-HCl (pH 8.3), 1.5 mM MgCl 2 , 63 mM KCl, 0.005% Tween 20, 1 mM EGTA, 50 M each of dATP, dGTP, dTTP, 10 M dCTP, 2 Ci of [␣-32 P]dCTP, 0.1 g/l bovine serum albumin, 2 units of Taq DNA polymerase, 0.1 g of TS primer (5Ј-AATCCGTCGAGCAGAGTT-3Ј), and 0.1 g of ACX primer (5Ј-GCGCGGCTTACCCTTACCCTTACCCTAACC-3Ј). After the mixtures were heated at 94°C for 90 s, they were subjected to 30 cycles of PCR amplification (one cycle consisting of 94°C for 30 s and 58°C for 45 s). After the reaction, PCR products were analyzed by 12.5% non-denatur-ing PAGE (1-mm thick). The gels were autoradiographed with the x-ray film (Hyperfilm MP, Amersham Pharmacia Biotech, Buckinghamshire, UK) at Ϫ80°C for 2 h, and telomerase activity was assessed on incorporated radioactive substrate in ladders of product DNA, multiplies of 6 bases corresponding to the telomere repeat unit.
Plasmid Constructs-The mouse TERT promoter-luciferase reporter plasmids were constructed by pGL3-basic vector (Promega, WI) and the DNA, which was obtained by PCR amplification using mouse genomic DNA according to a published sequence (GenBank accession number AF121949) (11). Various lengths of DNA fragments upstream of the initiating ATG codon of the mTERT gene were amplified and inserted into the pGL3-basic vector. For the construction of reporter plasmids containing substitution mutations in transcription factor binding sites, site-specific mutagenesis was performed by a PCR-based protocol. Expression vector for c-Myc protein (pcDNA3-cMyc) encoding the fulllength of c-Myc was a kind gift from Róná n C. O'Hagan and Ronald A. DePinho, Harvard Medical School (26). Expression vector pCGN-Sp1, encoding the full-length of human Sp1, and its backbone vector (pCGN) were kind gifts from Thomas Shenk, Princeton University (27). pCGN-Sp3, encoding the full-length of Sp3, was described previously (28).
Stable Transfection-One microgram of mTERT promoter-luciferase expression plasmid and 0.5 g of pSV2neo vector were cotransfected into C2C12 cells grown in a 12-well plate using SuperFect reagent (Qiagen, Hilden, Germany) as previously described (29). The transfected colonies were selected by 400 g/ml G418 (Life Technologies, Inc.) for 3 weeks. More than 50 colonies were pooled. Cell extracts were from the transfectants that were grown in the presence of 200 g/ml G418 and differentiated in Dulbecco's modified Eagle's medium containing 0.5% (v/v) fetal bovine serum with antibiotics and G418 for 10 days. Luciferase activity of these transfectants was normalized against the number of cells. A series of culture experiments were repeated at least three times.
Luciferase Assays-Transient transfection of luciferase reporter plasmids was carried out using SuperFect reagent (Qiagen) as previously described (29). In general, the day before transfection, cells were plated onto 12-well tissue culture plates at a density of 20,000 cells/ well, supplemented with fresh medium before transfection. A total of 0.825 g of DNA consisting of 0.75 g of the indicated luciferase reporter plasmid and 0.075 g of the pRL-thymidine kinase control vector (pRL-TK) (Promega, WI) per well was used for transfection studies. For cotransfection studies, cells grown in 12-well plates were transfected with 0.75 g of luciferase reporter plasmid, 0.075 g of pRL-TK, and 0.75 or 0.25 g of the indicated expression plasmids. PCGN or pcDNA3 was used to adjust the total amount of expression plasmid DNA. After harvest, the cells were assayed by the Dual-Luciferase Reporter Assay system (Promega, WI), using a luminometer (EG&G Berthold, Germany). Protein concentrations of the cell lysates were determined by the method of Bradford with the Bio-Rad protein assay dye reagent (Bio-Rad, CA). Promoter activities were expressed as relative activities, normalized against the concentration of the protein. All transfection experiments were repeated at least three times.
Nuclear Extract Preparation-Nuclear extracts were prepared from the C2C12 cells that were cultured in differentiation medium for 1, 4, 7, or 10 days using the rapid preparation as previously described (29). The protein concentrations of the nuclear fractions were determined by the Bradford assay, and all extracts were stored at Ϫ70°C.
Western Blot Analysis-Cells cultured in differentiation medium for 1, 4, 7, and 10 days were lysed in a buffer consisting of 20 mM Tris-HCl (pH 8.0), 137 mM NaCl, 2 mM EDTA, 10% (v/v) glycerol, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, 10 g/ml aprotinin, 1 mM sodium vanadate, and 1% (v/v) Triton X-100. Each cell lysate containing 25 g of protein was run under reducing conditions in 8% or 10% sodium dodecyl sulfate-polyacrylamide electrophoresis gels (SDS-PAGE), transferred onto polyvinylidene difluoride membranes (Millipore, MA), and reacted with antibodies as described. The primary antibody was detected by counterstaining with an HRP-linked antibody and visualized by enhanced chemiluminescence. Nuclear extracts (20 g) were subjected to SDS-PAGE and Western blotting.

RESULTS
Decreased Telomerase Activity and mTERT Expression during Muscle Cells Differentiation-C2C12 cells in culture were induced to differentiate by lowering serum concentration. Differentiation of C2C12 myoblasts into myotubes was associated with a diminished activity of telomerase (Fig. 1A), as observed previously (23,24). During the culture in differentiation medium, mTERT mRNA levels were also gradually decreased without changing the expression of glyceraldehyde-3-phosphate dehydrogenase mRNA level (Fig. 1B). However, the culture in differentiation medium resulted in up-regulation of muscle-specific genes, MyoD or myogenin (Fig. 1B), in agreement with previous reports (31)(32)(33)(34). Thus, the decrease of mTERT mRNA level was negatively correlated with the expressions of muscle-specific genes. These results indicate that the reduction of telomerase activity during myogenesis is regulated by the reduction of mTERT gene expression in C2C12.
Identification of cis-Elements in the Core Promoter of the mTERT Gene Essential for Transcriptional Activation-To understand the mechanisms of the mTERT expression in cell differentiation, analysis of the mTERT promoter is indispensable. As the first step, luciferase assays were performed using mouse NIH3T3 and C3H10T1/2 as well as C2C12 cell lines. These cell lines were transiently transfected with a series of 5Ј terminus-truncated mutants of the mTERT promoter linked to the luciferase reporter gene. As shown in Fig. 2, a 1.6-kb region of mTERT promoter (Ϫ1561/ϩ53-Luc) demonstrated the significant transcriptional activities in all these cell lines. NIH3T3 and C3H10T1/2 conferred higher transcriptional activity than C2C12. C2C12 exhibited modest transcriptional activity equivalent to 30 -40% of NIH3T3, but this activity was still significant, because it was 250-fold the activity in promoterless reporter plasmid (pGL3-basic).
Although deletions from Ϫ1561 to Ϫ225 of mTERT gene promoter resulted in no significant alterations in luciferase activity, truncation of a further 166 base pairs (from Ϫ225 to Ϫ59) led to a remarkable reduction of the transcriptional activity in these three cell lines (Fig. 2). These findings suggest that the proximal 225 bp is the core promoter essential for basal transcriptional activation of mTERT. Subsequently, C2C12 cells were stably transfected with reporter plasmids and induced to differentiate by the culture in differentiation medium for 10 days. Transfectants with Ϫ1561/ϩ53-Luc, Ϫ414/ ϩ53-Luc, or Ϫ225/ϩ53-Luc reporter plasmids exhibited the reduction of transcriptional activities (Fig. 3), being consistent with the decrease of mTERT mRNA level in muscle cell differentiation (Fig. 1B). These results suggest that the core promoter region (Ϫ225/ϩ53) is also responsible for reduction of mTERT transcription in muscle cell differentiation.
From these observations, we defined a 225-bp fragment as the mTERT minimal promoter and assumed that myogenesisresponsive elements are located in this region. According to data base analysis, we noticed that three GC clusters and two E-boxes are located in this region, which are known to bind with Sp1 family proteins and bHLH proteins, respectively (Fig.  4). To examine the role of these cis-elements more precisely, luciferase assays were performed with reporter plasmids with site-specific mutants, which were introduced into the mTERT minimal promoter (Fig. 5). One E-box is located at the 5Ј-end of the minimal promoter, and the deletion of this site resulted in a 50% reduction in transcriptional activity in C2C12 cells, as shown in Fig. 2 (Ϫ225/ϩ53-Luc and Ϫ185/ϩ53-Luc). Abrogation of this E-box (E-201) by substitution mutations also led to a 50% reduction of transcriptional activity (Fig. 5, Ebox-mt #1), suggesting that this E-box is essential for transactivation. In contrast, the mutation of another E-box (E-4) that is adjacent to the transcription start site led to no significant alteration of transcriptional activity (Ebox-mt #2). Although a mutation in the GC-box (GC-88) at Ϫ88 bp of minimal promoter slightly increased transcriptional activity (GC-mt #3), the mutations in the other two GC-boxes (GC-143 and GC-105) reduced the transcriptional activity (GC-mt #1, #2). Especially, abrogation FIG. 1. Decrease of telomerase activity during muscle cell differentiation. A, CHAPS extracts were prepared from uninduced C2C12 (C) and differentiated myoblasts after 1 week in differentiation medium (D) and assayed for telomerase activity of CHAPS extracts (representing 1 ϫ 10 5 cells) by the TRAP method as described under "Experimental Procedures." At least three independent experiments were performed, and typical results are shown. B, C2C12 myoblasts were cultured in differentiation medium for the indicated time. The cells were harvested, and total RNA were isolated and subjected to RT-PCR for mouse TERT (top). 20 g of total RNA was subjected to Northern blot analysis and probed with a MyoD and myogenin cDNA, which was followed by a glyceraldehyde-3-phosphate dehydrogenase probe to normalize for loading differences. of the GC-box located at Ϫ143 bp (GC-143) of the minimal promoter led to a remarkable reduction (Ͼ95%) of transcriptional activity (GC-mt #1). Abrogation of the GC-box located at Ϫ105 bp (GC-105) also caused a Ͼ50% reduction of transcriptional activity (GC-mt #5). Mutations of all of two E-boxes and three GC-boxes led to a marked loss of transcriptional activity (99.9 Ϯ 0.2% reduction, GC-E-mt), and mutations of only two GC-boxes (GC-143 and GC-105) except GC-88 also caused a Ͼ98% reduction of transcription activity (GC-mt #4). These results indicate that several cis-elements, including GC-143 and -105, as well as E-201, are important for basal transcriptional activity of mTERT gene. In particular, two GC-boxes (GC-143 and -105) are the most essential cis-elements for basal transactivation of mTERT gene in C2C12 cells.
GC-boxes Are Important for Down-regulation of mTERT Gene during Myogenesis-As shown in Fig. 3, a core promoter region (Ϫ225/ϩ53) is responsible for reduction of mTERT transcription during myogenesis. To identify the role of cis-elements in this core promoter, C2C12 cells were stably transfected with reporter plasmids containing mutations in five protein binding sites and induced to differentiate with differentiation medium for 10 days (Fig. 6). Transfectants with GC-mt #1, GC-mt #2, and Ebox-mt #1 reporter plasmids, which have single mutation among these cis-elements, reduced transcriptional activity during myogenesis (75-85% reduction), as well as transfectants with either Ϫ1561/ϩ53-Luc or Ϫ225/ϩ53-Luc reporter plasmid. Mutation of two E-boxes (Ebox-mt #2) also caused a decrease of luciferase activity (75% reduction). This result indicates that GC-boxes in the mTERT core promoter are responsible for differentiational down-regulation of mTERT gene in C2C12 cells. Abrogation of two GC-boxes (GC-143 and GC-105; GC-mt #4) also caused a slight decrease of transcription activity during myogenesis (15% reduction, Fig.  6). However, we could not clearly ascertain whether the rest of the E-boxes in the mTERT core promoter are indispensable for muscle cell differentiation or not, because there was not enough gene expression activity to be lost further with GC-mt #4 reporter plasmid, as shown in Fig. 5. In establishing these stable transfectants, several clones were shown to have different basal transcriptional activities from those of the transient transfectants (Fig. 5), because these results might represent the difference of the number of reporter gene copies, which were integrated into host chromosomal DNA.
Sp1 and Sp3 Specifically Bind to the GC-boxes in the mTERT Core Promoter-To identify the proteins that bind to these GC-boxes (GC-143 and GC-105) in mTERT gene core promoter, EMSA was performed with 32 P-labeled oligonucleotides corresponding to each GC-box and nuclear extracts from C2C12 cells cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Two major DNA⅐protein complexes were observed by using either GC-143 or GC-105 for probes (Fig. 7A, lane 1 or 5). The specificity of these complexes was confirmed by competition assay, using an excess amount of unlabeled Sp1 consensus oligonucleotides, which almost eliminated detection of complex formation (Fig. 7B). Given that these complexes appeared similar to those observed in previous reports (35,36), we investigated whether the complexes consisted of Sp1 or Sp3. The slower migrated complex was shifted by the addition of either anti-Sp1 or anti-Sp3 antibodies (Fig.  7A, lanes 2, 3, 6, and 7), whereas the faster migrated complex was shifted by the addition of anti-Sp3 antibody (lanes 3 and 7). Two major bands were almost eliminated by the addition of both antibodies (Fig. 7A, lanes 4 and 8). Based on these results, we conclude that the slower migrated band was derived from the Sp1⅐DNA or Sp3⅐DNA complexes, whereas the faster migrated band was derived from Sp3⅐DNA complex. However, using GC-88 as probe, no shifted bands were observed by the addition of either anti-Sp1 or anti-Sp3 antibodies, although several DNA⅐protein complexes were detected (data not shown). To determine whether Sp1 and Sp3 modulate the tran- scriptional activity of the mTERT promoter, C2C12 myoblasts were cotransfected with one of expression vectors for Sp1 (pCGN-Sp1), Sp3 (pCGN-Sp3), or null vector (pCGN) together with reporter plasmid containing mTERT core promoter (Ϫ225/ ϩ53-Luc). Sp1 and Sp3 induced a large transactivation of mTERT promoter activity (6.5-fold and 4.2-fold, respectively) in C2C12 myoblasts (Fig. 7C). These findings suggest that Sp1 and Sp3 function as potent transactivators of mouse TERT gene.
c-Myc Specifically Bind to the E-box in the mTERT Core Promoter-As shown in Fig. 5, E-box at Ϫ201 to Ϫ196 (E-201) of mTERT promoter was also responsible for the basal transcription. To identify the proteins that bind to this E-box in mTERT gene core promoter, EMSAs were carried out with nuclear extracts prepared from the C2C12 myoblasts (Fig. 8A). Three bands (bands A-C) were observed by using E-201 oligonucleotides as probe, although band C sometimes diminished. These three (or two) bands were competed by homologous competitors (Fig. 8A, lanes 4 -6), but not by MyoD binding oligonucleotides nor mutated oligonucleotides in which the E-box sequences were altered (lanes 7-12). Only band B, but neither band A nor band C, was eliminated by an excess amount of unlabeled c-Myc consensus sequences (Fig. 8A, lanes 1-3).
Therefore, band B may represent the c-Myc⅐DNA complex. The formations of bands A and C were not affected by adding either an excess amounts of c-Myc consensus sequence (lanes 1-3) or the anti-c-Myc antibody (data not shown). Therefore, these bands may not be the complex of Myc⅐Max nor their degradation products and have thus remained to be identified.
Then we tested the modulation of mTERT transcription by c-Myc in a myoblast using the luciferase reporter plasmid. C2C12 myoblasts were cotransfected with c-Myc expression vector pcDNA3-cMyc, together with a reporter plasmid with mTERT core promoter (Ϫ225/ϩ53), or with a longer 1.6-kb upstream region (Ϫ1561/ϩ53). As shown in Fig. 8B, expressed c-Myc increased significantly the promoter activities in both cases. Interestingly, the stimulation (1.9-fold) was higher, with the reporter plasmid harboring a long promoter region of Ϫ1561/ϩ53 than with that contained in the core promoter of Ϫ225/ϩ53 (1.3-fold). These observations suggest that c-Myc, in addition to Sp1 and Sp3, also act on the mTERT promoter to enhance transcription of mTERT gene.
Sp1, Sp3, and c-Myc Are Repressed during Myogenesis-We found that Sp1, Sp3, and c-Myc bind to the mTERT gene core promoter in myoblasts but not in differentiated myotubes (Figs. 3 and 6). To understand the nature of the mechanisms involved, we determined the levels of each transcriptional factor in nuclear extracts prepared from C2C12 cells cultured in differentiation medium. By the Western blot analysis, the con-tents of all these transcription factors decreased gradually during muscle cell differentiation (Fig. 9A). To confirm the binding activities of these transcription factors to mTERT core promoter, nuclear extracts obtained from myoblasts cultured in differentiation medium were subjected to EMSA using GC-143, GC-105, or E-201 oligonucleotides as probe. Sp1⅐DNA, Sp3⅐DNA, and c-Myc⅐DNA complexes were detected, and each complex was diminished during the culture under serum-starved condition (Fig. 9, B and C). These results suggest that the decrease in DNA binding of Sp1, Sp3, and c-Myc, consistent with the reduction in the concentration of these transcription factors, resulted in the elimination of telomerase activity during myogenesis.

DISCUSSION
In this study, we demonstrated that telomerase activity is down-regulated along differentiation of muscle cells as a consequence of decrease in the mRNA for mTERT. The cellular content of mTERT mRNA decreased in parallel with the transcriptional activity of mTERT promoter in response to myogenesis, indicating that the alterations in the transcription of the mTERT accounts for the elimination of telomerase activity during myogenesis. The putative mTERT core promoter region, ranging from Ϫ225 to ϩ53, contains several GC-boxes and E-boxes. Deletion and mutation of these elements resulted in a significant loss of transcriptional activity in several cell lines, including C2C12 myoblasts. We attempted to identify the ciselements in the mTERT core promoter for transactivation and found that two GC-boxes (GC-143 and GC-105) bound to Sp1 and Sp3, and an E-box (E-201) bound to c-Myc but not to MyoD in C2C12 myoblasts. It was demonstrated that decreased binding of Sp1, Sp3, and c-Myc to three cis-elements in the mTERT core promoter were well correlated with the down-regulation of mTERT gene expression during muscle cell differentiation.
Two GC-boxes, GC-143 and GC-105, were essential for the mTERT core promoter activity, whereas GC-88 was not. Mutation in GC-143 or GC-105 led to the remarkable reduction of transcriptional activity (Ͼ95% or Ͼ60% decrease, respectively; Fig. 5). As expected, double mutations in these two GC-boxes resulted in marked reduction of transcriptional activity (Ͼ98% decrease), suggesting that these two elements may be the most essential cis-elements in the mTERT core promoter. We have also identified the binding of Sp1 and Sp3 to these GC-boxes in the region Ϫ225/ϩ53 of the mTERT gene by EMSAs (Fig. 7, A  and B). Furthermore, overexpression of either Sp1 or Sp3 elevated the level of mTERT transcription in undifferentiated myoblasts (Fig. 7C). Sp1 is thought to be a ubiquitously expressed transcription factor that plays a primary role in the regulation of a large number of genes, including constitutive housekeeping genes and inducible genes (37). Sp3 is considered to be bifunctional such that it represses Sp1-mediated activity of several promoters (38 -41), whereas it acts as an activator of gene expression in mammalian cells (42,43). In this study, we provide evidence that, in mTERT gene transcription, both Sp1 and Sp3 act as positive regulators (Fig. 9). As shown in Fig. 2, NIH3T3 and C3H10T1/2 conferred higher transcriptional activity than C2C12. Consistently, EMSAs showed greater binding activities of Sp1 and Sp3 in NIH3T3 and C3H10T1/2 than in C2C12 (data not shown). These results might be caused by different extents of Sp1 and Sp3 expression among cell types (44). In mouse, high levels of Sp1 expression have been found in spermatids, T cells in thymus, epithelial cells, and hematopoietic cells, whereas Sp1 expressions in heart, skeletal muscle, and smooth muscle cells of stomach have been shown to be low (44). The different expression levels of Sp1 are consistent with tissue-specific expressions of telomerase activity in adult mouse. Telomerase activities are not detected in brain, heart, stomach, and muscle in mouse (22).
Other groups have also shown that telomerase activity decreases along muscle cell differentiation (23,24). It was also shown that myogenesis decreases the binding of Sp1 and Sp3 to the promoter region of GLUT1 gene in C2C12 cells, in which Sp1 and Sp3 are down-regulated (35,36). These results are consistent with our conclusion that both Sp1 and Sp3 directly regulate the expression of mTERT in C2C12 cells by the interaction with the promoter region of mTERT gene in myogenesis. Muscle cell differentiation is mediated by a transcription factor MyoD, which acts as a master regulator leading to the activation of many muscle-specific genes (32,45,46). Recent studies showed that the overexpression of MyoD leads to the repression of Sp1 and Sp3 (35,36). In this context, although MyoD may not take part directly in decreasing mTERT gene transcription in our experimental systems, it might suppress mTERT gene transcription in vivo by an indirect mechanism, i.e. via repressing genes for Sp1 and Sp3.
c-Myc, a transcription factor encoded by a proto-oncogene, is suppressed during myogenic differentiation through post-transcriptional mechanisms (47)(48)(49)(50). In humans, c-Myc binds to the hTERT gene promoter and plays a critical role in regulation of hTERT expression (10 -12). Similar relationships between mTERT and c-Myc were observed during differentiation of mouse erythroleukemia cells and mitogen-stimulated lymphocyte proliferation. These findings suggest there is a potential link between increased c-Myc and up-regulation of mTERT in normal proliferating and transformed cells (8,11). Consistent to these observations, we showed that c-Myc enhanced the activity of the mouse TERT promoter and that transcriptional stimulation by c-Myc was higher with the reporter plasmid harboring a longer promoter region than with that contained in the core promoter region (Fig. 8B). These observations indicate that c-Myc, in addition to Sp1 and Sp3, also plays an important role in mTERT gene transcription and that there are other c-Myc-responsive cis-elements outside the core promoter region. In this context, the decrease of c-Myc expression in muscle cell differentiation may be responsible for the reduction of mTERT transcription. It is generally accepted that c-Myc activates transcription as part of a heterodimeric complex with a number of interacting partners, including members of the Max and Mad families (51,52). Kyo et al. (13) showed recently that human TERT gene expression is variable depending on the Myc/Max ratio or cell types used. In our study, although mTERT transactivation by c-Myc was lower than that by Sp1 or Sp3 (Figs. 7C and 8B). the expression of appropriate Myc/ Max ratio might also lead to remarkably increased mTERT promoter activity as well as human TERT.
During myogenesis, the promoter activity of the Ϫ1561/ϩ53 long upstream region decreased to a greater extent than that of the core promoter (Ϫ225/ϩ53) (Fig. 3). In parallel, expression of mTERT mRNA largely decreased to a greater extent than the expression of luciferase conjugated with Ϫ225/ϩ53 during the differentiation. Moreover, as shown in Fig. 8B, other c-Myc responsive cis-elements are probably located in the Ϫ1561/ϩ53 long upstream region. According to data base analysis, we also noticed that a number of putative cis-elements, including GC- The C2C12 cells were cultured in differentiation medium for 1, 4, 7, or 10 days. A, nuclear extracts were analyzed by immunoblotting with antibodies against Sp1, Sp3, and c-Myc. B, 5 g of nuclear extracts was subjected to EMSAs using GC-box region at Ϫ143 (GC-143; lanes 6 -10) or at Ϫ105 (GC-105; lanes 1-5) as probe. C, 5 g of nuclear extracts was subjected to EMSAs using E-box region at Ϫ201 (E-201) as probe. Representative autoradiograms are shown.
boxes or E-boxes, are located 5Ј upstream beyond the core promoter region of mTERT (data not shown). These putative cis-elements may also regulate the transcription of mTERT gene by binding transcription factors such as the bHLH proteins or the Sp1 family. It remains to be solved whether chromatin remodeling factors and/or nucleosomal packaging, as well as transcription factors, may also play a role for mTERT transactivation through the enhancer or silencer region outside the core promoter sequence. To analyze further the tissuespecific expression of telomerase in mouse, experiments using an upstream region of the mTERT core promoter are underway in our laboratory.
In summary, our results showed that Sp1 and Sp3 have critical roles of mTERT transactivation in mouse. In particular, two GC-boxes (GC-143 and GC-105) in the mTERT core promoter are the most essential for transcriptional activity. In addition to Sp1 and Sp3, c-Myc also activated mouse TERT gene expression. Our data also indicated that the repression of these transcription factors causes down-regulation of the mTERT gene in muscle cell differentiation.