This Article Abstract Full Text (PDF) All Versions of this Article: 147/6/3093    most recent Author Manuscript (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chang, J. H. Articles by Chen, S. L. Search for Related Content PubMed PubMed Citation Articles by Chang, J. H. Articles by Chen, S. L.

Myogenic Basic Helix-Loop-Helix Proteins Regulate the Expression of Peroxisomal Proliferator Activated Receptor- Coactivator-1

Department of Life Sciences, National Central University (J.H.C., C.H.S., Y.J.C., H.C.C., S.L.C.), Jhongli 32054, Taiwan; and Department of Biochemistry, Chang Gung University (K.H.L.), Kweisan 333, Taiwan

Address all correspondence and requests for reprints to: Dr. Shen Liang Chen, Department of Life Sciences, National Central University, 300 Jhongda Road, Jhongli 32054, Taiwan, Republic of China. E-mail: slchen{at}cc.ncu.edu.tw.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Peroxisomal proliferator-activated receptor- coactivator-1 (PGC-1), a transcriptional coactivator, is selectively expressed in slow-twitch fibers in skeletal muscle. Ectopic expression of the PGC-1 gene in either a cell or an animal has been shown to promote fast to slow fiber-type switch. The expression of PGC-1 in muscle is regulated by myocyte enhancer factor 2 and Forkhead in rhabdomyosarcoma, two transcription factors implicated in terminal muscle differentiation. In this study we found that PGC-1 expression was activated during terminal muscle differentiation in both C2C12 and Sol8 myoblasts. Using retrovirus-mediated MyoD overexpression in C3H10T1/2 cells, we also demonstrated that MyoD, the master regulator of terminal differentiation, could activate PGC-1 expression in vivo. Our transient transfection results also show that myogenic basic helix-loop-helix (bHLH) proteins, especially MyoD, can activate PGC-1 expression by targeting its promoter. Myogenic bHLH protein target sites on PGC-1 promoter were localized to a short region (–49 to +2) adjacent to the transcription start site, which contains two putative E boxes. Mutation of either site significantly reduced MyoD-mediated transactivation in the cells, suggesting that both sites are required for myogenic bHLH protein-mediated activation. However, only one site, the E2 box, was directly bound by glutathione-S-transferase-MyoD protein in EMSAs. Our results indicate that myogenic bHLH proteins not only are involved in lineage determination and terminal differentiation, but also are directly implicated in activation of the key fiber-type and metabolic switch gene, PGC-1.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
PEROXISOMAL PROLIFERATOR-activated receptor- (PPAR) coactivator-1 (PGC1) is a transcriptional coactivator identified in the screening for proteins associating with PPAR. It is highly expressed in tissues requiring high-energy metabolism, such as heart, skeletal muscle, and brown fat (1, 2). Its expression in brown fat is highly induced when animals are exposed to a cold environment, implying its role in metabolic thermogenesis adaptation. Later studies have proven its function in the biogenesis of mitochondria, gluconeogenesis, and fatty acid oxidation (3, 4). Since its identification, PGC-1 has been found to coactivate the function of many nuclear hormone receptors, including thyroid hormone, estrogen, and glucocorticoid receptors, and many more (1, 5, 6, 7, 8). The wide variety of its partners suggests that PGC-1 is not only implicated in the energy metabolism, but is also involved in many other physiological processes.

In muscle, PGC-1 is preferentially expressed in slow-twitch fibers, which are much higher in mitochondria content and are more dependent on oxidative metabolism than fast-twitch fibers (9). In transgenic mice, as the expression of PGC-1 is driven by muscle creatine kinase promoter, the putative fast-twitch fibers are converted into slow-twitch fibers, which are characterized by the expression of troponin I slow, myoglobin, and genes of mitochondria oxidative metabolism (9). The activation of slow-twitch muscle-specific genes by PGC-1 is mediated through MEF2 transcription factors binding to the upstream enhancer sites. MEF2 proteins also activate the transcription of PGC-1 gene and thus establish a positive feedback loop (10). Hormones both activating, such as thyroid hormone, and inhibiting, such as insulin, oxidation metabolism can also regulate the expression of PGC-1 either positively or negatively (11, 12, 13). The signals of insulin in the nucleus are mediated by Forkhead in rhabdomyosarcoma (FKHR), a forkhead transcription factor that binds to the insulin response elements on the PGC-1 promoter to activate its expression (13). FKHR is a transcription factor implicated in the glucose metabolism and longevity of an individual.

Skeletal muscle cells are derived from axial somitic cells during early embryogenesis (14, 15). These myogenic precursor cells induced by axial signals from notochord and neural tube are characterized by their expression of a basic helix-loop-helix (bHLH) transcription factor, either MyoD or Myf5 (16). Once the somitic cells start expressing MyoD or Myf5, they are committed to myogenic lineage only and ultimately become myocytes that can terminally differentiate into myotube, which are the basic building elements of skeletal muscle (17).

Terminal differentiation of myogenic lineage is dictated by the function of MyoD, Myf5, and their family members, myogenin and MRF4 [collectively called myogenic bHLH proteins or, alternatively, myogenic regulatory factors (MRFs)]. Myogenic bHLH proteins can activate the expression of another family of transcription factors, myocyte enhancer factor 2 (MEF2), to commit myoblasts to terminal differentiation (18). Together, myogenic bHLH and MEF2 proteins regulate the differentiation of myoblasts into multinucleated myotubes by activating muscle-specific contractile genes and genes that promote cell cycle exit, such as p21^cip1 and p27^kip1 (19, 20). Both myogenic bHLH protein and MEF2 family members are expressed in slow- and fast-twitch muscle fibers.

Because MEF2 and FKHR are both implicated in terminal muscle differentiation, it prompts us to study the regulation of PGC-1 in muscle by other muscle-specific transcription factors, such as MyoD and its family members, which are implicated in the same process. In this study we show that myogenic bHLH proteins can activate PGC-1 expression by targeting its promoter. Myogenic bHLH proteins target sites on PGC-1 promoter localized to a short region (–49 to +2) adjacent to the transcription initiation site, which was designated as the PGC-1 core promoter. Analysis of its sequence revealed two putative E boxes localized to this region. Mutation of either site significantly reduced myogenic bHLH protein-mediated transactivation in the cells, suggesting that both sites are required for myogenic bHLH protein-mediated activation. However, only one site, the E2 box, was shown to be directly bound by glutathione-S-transferase (GST)-MyoD protein in EMSAs. Because PGC-1 is suggested to be the principal factor regulating muscle fiber type determination, our results indicate that myogenic bHLH proteins not only are involved in the lineage determination and terminal differentiation, but also are implicated in the fiber-type and metabolic switch of mature muscle cells.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Plasmids
Human (h) PGC-1 promoter (–992 to +90) in pGL3-basic vector was a gift from Dr. Akiyoshi Fukamizu (Institute of Applied Biochemistry, University of Tsukuba, Tsukuba, Japan). Deletion clones of this promoter, –676 to +90 and –630 to +90, were constructed by digesting the full-length clone with PstI or EcoRI, respectively, in conjunction with XhoI, whose cutting was located on the 5' multiple cloning sites. Digestion of the full-length clone with XhoI/BgII or BgII/HindIII created deletion clones –444 to approximately +90 and –992to approximately –444, respectively. Additional deletion of PGC-1 promoter and site-specific mutation were achieved by specific PCR. The primers used for amplification of each fragment are listed below: hPGC-1 promoter –233 to approximately +90: forward, 5'-GCGGGTACCAGCCTCCAAAAGTCTAAGTG-3'; reverse, 5'-GCGGCTAGCAGCTCCTGAATGACGCCAGT-3'; hPGC-1 promoter –95 to approximately +90: forward, 5'-GCGGGTACCAAGGGAGGCTGGGTGAGTGA-3'; reverse, 5'-GCG-GCTAGCAGCTCCTGAATGACGCCAGT-3'; hPGC-1 promoter –95 to approximately +19: forward, 5'-GCGGGTACCAAGGGAGGCTGGGTGAGTGA-3'; reverse, 5'-GCCCTCGAGAGGCAACCAGCCCCT-TACTG-3'; hPGC-1 promoter –49 to approximately +2 (wild type): forward, 5'-CGCGGTACCTTGTCATGTGACTGGGGACT-3'; reverse, 5'-CGGGCTAGCCTGAGAGTGAACTGAAGGCA-3'; hPGC-1 promoter –49 to approximately +2 [mutant 1 Mt1)]: forward, 5'-TTGTAAAAAAACTGGGGACTGTAGTAAGA-3'; reverse, 5'-CTGAGAGTGAACTGAAGGCACCTGTCTTACTACA-3'; hPGC-1 promoter –49 to approximately +2 (Mt2): forward, 5'-TTGTCATGTGACTGGGGACTGTAGTAAGA-3'; reverse, 5'-CTGAGAGTGAACTGAAGGAAAAAATCTTACTACA-3'; and hPGC-1 promoter –49 to approximately +2 (Mt3): forward, 5'-TTGTAAAAAAACTGGGGACTGTA-GTAAGA-3'; reverse, 5'-CTGAGAGTGAACTGAAGGAAAAAATCTTACTACA-3'.

Amplified DNA fragments were digested with specific restriction enzymes to create sticky ends for ligation into KpnI and NheI sites of pGL3-basic vector. For reporter constructs containing multiple copies of PGC-1 core promoter, wild-type core promoter was amplified by forward and reverse primers containing the NheI site. Then PCR product was digested overnight with NheI before self-ligation to created concatenations of PCR product. This ligation product was then inserted into NheI site of pGL3-basic vector. The copy number and orientation of each construct were verified by sequencing. All PCR-created clones were sequenced to confirm their sequence integrity.

The expression vectors for MyoD, myogenin, Myf5, MRF4, and MEF2C were described previously (33). Expression vector for FKHR was a gift from Dr. William Seller (Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA). vascular cell adhesion molecule-1 and 4RE-thymidine kinase-luciferase plasmids were acquired from Drs. George Muscat and Brett Hosking (Institute for Molecular Bioscience, University of Queensland, Brisbane, Australia).

Cell culture and transient transfection
C2C12 and C3H10T1/2 cells were kept in DMEM supplemented with 20% or 10% fetal calf serum, respectively. Cells were split and plated into 12-well culture dish 1 d before transfection. Except for those in myotubes, where cells were 100% confluent, all transfections were performed at 60% cell confluence. Transient transfection assays were performed by mixing aliquots of plasmid DNA together in 1x HEPES buffer [20 mM HEPES (pH 7.0), 187 mM NaCl, 5 mM KCl, 0.7 mM Na₂HPO₄, and 5.5 mM dextrose]. Mixtures of liposome (Lipofectamine, Invitrogen Life Technologies, Inc., Gaithersburg, MD) in 1x HEPES buffer were then added to the DNA mixture (0.67 µg reporter construct and 0.33 µg expression vector/well) and incubated at room temperature for 15–30 min, allowing the DNA/liposome complex to form. Aliquots of culture medium were then added to each tube and mixed gently. Medium containing the DNA/liposome complex was transferred to cells in triplicate. This transfection step was allowed to proceed overnight before the medium was replaced with differentiation medium containing 2% horse serum. Cells were harvested and assayed for luciferase activity 16–24 h after transfection in a Clarity 2 luminometer (BioTEK, Winooski, VT).

Quantitative real-time RT-PCR
Total RNA was extracted from the C2C12 and Sol8 myoblasts and muscle tissue using TRIzol (Invitrogen Life Technologies, Inc.) according to the supplier’s instructions. Then the first strand of cDNA was synthesized using the SuperScript II kit (Invitrogen Life Technologies, Inc.) for RT-PCR. Briefly, total RNA was denatured at 65 C for 10 min in the presence of 0.5 µg oligo(deoxythymidine) and 1 mM deoxy-NTP. After chilling on ice for at least 1 min, RT was allowed to proceed at 25 C for 5 min in the presence of 1x first-strand buffer, 5 mM dithiothreitol, and 40 U ribonuclease inhibitor. The reaction was allowed to proceed at 42 C for another 60 min. The reaction was heat inactivated at 70 C for 10 min.

Real-time quantitative PCR was performed in a 25-µl reaction mixture containing 5 µM forward/reverse primers, 1x SYBER Green reaction mix (Applied Biosystems, Werrington, UK), and various amounts of template. The reaction was performed with preliminary denaturation for 10 min at 95 C to activate Taq DNA polymerase, followed by 40 cycles of denaturation at 95 C for 15 sec and annealing/extension at 63 C for 1 min. Different amounts of template were used in the same reaction to ensure the linear amplification of PCR products. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as internal control amplified in the same PCR assay. The primer sets used for quantification of MyoD, PGC-1, and GAPDH expression are listed below: MyoD: forward, 5'-AACTGCTCTGATGGCATGATG-3'; reverse, 5'-TCGACACAGCCGCACTCTTC-3'; PGC-1: forward, 5'-GAGCGCCGTGTGATTTACGT-3'; reverse, 5'-GCGAAAGCGTCACAGGTGTA-3'; and GAPDH: forward, 5'-CCTCTGGAAAGCTGTGGCGT-3'; reverse, 5'-TTGGCAGGTTTCTCCAGGCG-3'. All reactions were performed in an ABI 7300 sequence detection system (Applied Biosystems).

EMSA
Forward and reverse primers encompassing PGC-1 promoter –49 to approximately +2 (as listed above) were heated to 80 C, then allowed to anneal in PCR buffer by the overlapping complementary sequence in the middle region. The recessive ends in the annealed fragments were filled in with Klenow fragment (New England Biolabs, Beverly, MA) before being used as a template for Pfu-mediated DNA amplification to create DNA fragments with blunt ends. These blunted DNA fragments were labeled on both ends by polynucleotide kinase (PNK)-mediated DNA phosphorylation reaction in the presence of [-³²P]ATP. ³²P-Labeled probe was additionally purified by gel purification in 15% acryamide buffered by 1x TBE (Tris, boric acid, EDTA). The ³²P-labeled DNA band was cut off from the gel, then eluted onto diethylaminoethyl-cellulose filter paper before washed out from the paper with high salt solution. DNA washed out from the paper was ethanol precipitated and resuspended in distilled water, then it was ready for the binding assay.

Nuclear extract (10 µg) from Sol8 cells was used to bind PGC-1 promoter probes in 20 µl binding buffer [25 mM HEPES (pH 7.4), 5 mM MgCl₂, 4 mM EDTA, 2 mM dithiothreitol, 110 mM NaCl, 5 µg/ml BSA, and 0.8% Ficoll] at room temperature for 30 min. Protein and DNA complexes were resolved on a 5% acryamide low ionic gel containing 5% glycerol at 4 C for at least 3.5 h. Signals on the gels were viewed by autoradiography. Antibodies (0.5 µg) against MyoD (clone MoAb 5.8A; BD Pharmingen, San Diego, CA), Myo (clone F5D, BD Pharmingen), and MEF2C (polyclonal C-21, Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were included in the binding reaction to verify their presence in the shifted bands on the gel. In some reactions, bacteria-expressed GST-MyoD (2.0 µg), GST-myogenin (1.0 µg), or GST (2.0 µg) proteins were included in the binding reactions to bind the PGC-1 core promoter probes.

Establishment of stable clones expressing MyoD
MyoD cDNA was released from pCDNA3.1 vector by EcoRI and HindIII digestion, then blunted with Klenow fragment reaction before being inserted into the pMSEV-neoEB vector. pMSEV-neoEB vectors carrying sense MyoD cDNA were transfected into GP+E-86, a retrovirus package cell line. Then, cells were incubated with the transfection complex overnight and selected with G418 (400 µg/ml) 48 h after transfection. Stable clones were harvested 2–3 wk after selection for extraction of total RNA. The expression levels of sense and antisense MyoD transcripts were determined by RT-PCR with 25 cycles. After confirmation of the expression of sense and antisense MyoD transcripts, retrovirus was harvested from the culture medium of GP+E-86 cells and directly transferred to C3H10T1/2 cell medium for infection. Infection was allowed to proceed before G418 (400 µg/ml) was added to the medium 2 d after the initial infection, and the selection was continued for 2–3 wk. Then total RNA was isolated from the stable clones to verify the expression levels of MyoD transcripts. The expression level of MyoD was stable during the experimental period (at least 10 passages). After confirmation of MyoD expression, these stable clones were expended, then were ready for transdifferentiation assay and determination of the expression levels of PGC-1. Both GP+E-86 cells and pMSCV-neoEB cells were gifts from Dr. Robert G. Hawley (George Washington University Medical Center, Washington, DC). For transdifferentiation assays, parental and stable clone C3H10T1/2 cells were allowed to grow confluent, then growth medium containing 10% fetal calf serum was replaced by differentiation medium containing 2% horse serum. Cells were allowed to differentiate for at least 72 h before being harvested for isolation of total RNA. The morphology of each clone and parental cell was observed and photographed every 24 h. The PGC-1 expression level was determined in cells differentiated for 72 h.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
PGC-1 expression is significantly activated during terminal differentiation
Although PGC-1 has been reported to be highly expressed in mature skeletal muscle fibers, its expression pattern during terminal differentiation of fast- and slow-twitch fiber types has not been examined in detail. Whether it is differentially expressed during the formation of these two types of muscle fibers is currently unknown. Because it is too difficult to trace the differentiation of myoblasts in vivo, we approached this issue by observing the terminal differentiation of two different fiber type-oriented myoblasts, Sol8 and C2C12, in vitro. Both myoblasts can be induced to undergo terminal differentiation by the withdrawal of serum when they are confluent. However, the myotubes that result are different fiber types; myotubes formed from Sol8 myoblasts are of the slow-twitch phenotype (21), but those formed from C2C12 are of mixed phenotype, in which both fast- and slow-twitch contractile protein genes are expressed at similar levels (22, 23). Therefore, each of these two myoblast cell lines can be used as a model representing myotubes forming either slow-twitch or mixed fibers. The expression level of PGC-1 in both proliferating C2C12 and Sol8 myoblasts is very low and was not detectable by Northern blot assay. Alternatively, we employed real-time PCR to detect its expression in these two myoblasts. The expression level of PGC-1 in proliferating Sol8 myoblasts was about 1/2^17.5 that of GAPDH (Ct = 17.5) in a standard real-time PCR program [Fig. 1A, proliferating (PMB)]. The expression level was not changed significantly when Sol8 myoblasts became confluent [Fig. 1A, confluent (CMB)]. However, PGC-1 expression was significantly activated when Sol8 myoblasts were terminally differentiated into myotubes [Fig. 1A, myotube (MT)]. We failed to detect the expression of PGC-1 in proliferating C2C12 cells, but its expression in CMB became detectable and was 1/2^16.7 that of GAPDH (Ct = 16.7; Fig. 1B, CMB). PGC-1 expression was also significantly activated in C2C12 myotubes (Fig. 1B, CMB). The expression levels of PGC-1 in myotubes of Sol8 and C2C12 were similar (Ct = 14.1 and 14.3, respectively) even though each type of myotube was supposed to be different in metabolism and speed of contraction.

View larger version (18K): [in this window] [in a new window]   FIG. 1. PGC-1 expression is significantly activated during terminal differentiation. Total RNA isolated from PMB, CMB, and MT of Sol8 and C2C12 myoblasts was reverse transcribed, and the expression levels of GAPDH and PGC-1 were determined by real-time PCR. PGC-1 expression in each sample was normalized to that of GAPDH. A, The expression level in PMB was arbitrarily set at 1, and the expression levels in CMB and MT was compared with that of PMB. B, The expression level of PGC-1 in CMB C2C12 myoblasts was set at 1, and that of MT was compared with that of CMB. PGC-1 expression in PMB C2C12 myoblasts was undetectable in our assay. C, PGC-1 promoter is activated during terminal differentiation. PGC-1 and p21CIP1/WAF1 promoters and their empty vector, pGL3-basic (0.67 µg), were transfected into Sol8 confluent myoblasts (CMB) and myotubes (MT) overnight. Luciferase activity was determined 48 h after transfection. Luciferase activity in CMB cells was arbitrarily set at 1-fold activation. Both PGC-1 and p21CIP1/WAF1 promoter activities were significantly increased (*, P < 0.05).

PGC-1 expression is regulated by MyoD in vivo
The up-regulation of PGC-1 expression upon terminal differentiation suggests that its expression might be targeted by transcription factors that control the process of terminal differentiation. Terminal differentiation of myogenic cells is under the regulation of myogenic bHLH and MEF2 proteins. It has been shown that MEF2 plays a key role in the calcium-mediated PGC-1 activation during physical training and long-term workload (9, 24). Whether myogenic bHLH proteins play similar roles remains to be determined. To explore the possible roles played by myogenic bHLH proteins in the activation of PGC-1 in vivo, MyoD cDNA was overexpressed in C3H10T1/2 cells by a retrovirus-mediated method to observe its effects on the expression levels of endogenous PGC-1. The expression levels of MyoD in these cell types were examined at both RNA and protein levels. MyoD RNA transcript was not detectable in parental and control retrovirus-transduced polyclonal C3H10T1/2 cells (Fig. 2A, lanes 5 and 7), but was abundantly expressed in MyoD retrovirus-transduced C3H10T1/2 cells (Fig. 2A, lane 9). MyoD protein was not detectable in parental and control retrovirus-transduced polyclonal C3H10T1/2 cells (Fig. 2B, lanes 3 and 4); however, it was highly expressed in polyclonal stable cells infected by MyoD retrovirus (Fig. 2B, lane 5). The expression level of MyoD in polyclonal stable cells was similar to that in C2C12 and Sol8 myoblasts (Fig. 2B, lanes 1 and 2). The expression level of MyoD in the cells was also reflected by their ability to transdifferentiate into myoblasts and myotubes. Only cells infected with MyoD retrovirus had the ability to transdifferentiate into myotubes (Fig. 8C, MyoD). These results demonstrate that MyoD retrovirus-transduced C3H10T1/2 cells express MyoD abundantly, which transforms these fibroblasts into myoblasts.

View larger version (54K): [in this window] [in a new window]   FIG. 2. PGC-1 expression is regulated by MyoD in vivo. A, Expression levels of MyoD transcript in parental C3H10T1/2 cells (10T; lanes 4 and 5) and C3H10T1/2-derived stable clones transduced with control retrovirus (Con; lanes 6 and 7) or retrovirus expressing MyoD (MyoD; lanes 8 and 9) were detected by RT-PCR. The no-template control (N) is shown in lane 2, and DNA markers (M; 100-bp ladder) in base pairs are shown in lane 1. Total RNA (50 ng) was used as the PCR template for reactions shown in lanes 4, 6, and 8 and served as a control for genomic DNA contamination. Equivalent amounts of reverse-transcribed cDNA were used as PCR template for the reactions shown in lanes 5, 7, and 9. C2C12 cDNA was amplified as a positive control (lane 2). Among C3H10T1/2 cells, only cells transduced by MyoD-expressing retrovirus express MyoD (lane 8). The primer set for MyoD amplified a 130-bp sequence spanning exons 2 and 3, so genomic contamination (500 bp) could be easily identified. B, MyoD protein in 30 µg nuclear extract and total lysate isolated from C2C12, Sol8, and Sol3 stable clones was detected by monoclonal antibody (0.2 µg /ml) against MyoD. The expression of GAPDH was used as a loading control for total lysate. Different exposure times were used for total lysate and nuclear extract. The relative mobilities of protein markers are shown to the left (in kilodaltons). C, The transdifferentiation abilities of parental and retrovirus-transduced stable clone cells wee observed when they were confluent in growth medium containing 20% fetal calf serum or 24 and 48 h after serum withdrawal in differentiation medium (DM). Multinucleated myotubes were only observed in MyoD stable clones. All pictures were taken at x200 under phase-contrast light. D, Relative PGC-1 mRNA levels in C3H10T1/2-derived stable clone cells. Total RNA was reverse transcribed, and the expression levels of GAPDH and PGC-1 were determined by real-time PCR. PGC-1 expression in each sample was normalized to that of GAPDH. The expression level in control virus-transduced C3H10T1/2 cells was arbitrarily set at 1-fold (Con, bar no. 1), and used as a standard for comparison with the expression levels in other samples. The expression of PGC-1 was significantly activated in MyoD stable clone cells (*, P < 0.05).

View larger version (24K): [in this window] [in a new window]   FIG. 8. Both E boxes in the core promoter are required for MyoD-mediated activation. A, Diagrams show the sequences of wild-type (Wt) and E box mutant PGC-1 core promoters. The sequences in the mutants that are the same as those in the wild type are represented by a dashed line, and the sequence of E boxes in each mutant is shown. The E1 box (–45 to –40) was mutated to AT pairs in both Mt1 and Mt3. The E2 box (–22 to –17) was similarly mutated in Mt2 and Mt3. B, MyoD failed to activate mutants carrying any E box mutation. Wild-type and E box mutant core promoters were cotransfected with either GFP or MyoD expression vectors into C3H10T1/2 cells to examine their regulation by MyoD. Wild-type core promoter was significantly activated by MyoD, but none of the E box mutants was significantly activated by MyoD. *, P < 0.05.

PGC-1 promoter is activated by MyoD expression
It has been reported that both MEF2C and FKHR can activate PGC-1 expression by targeting enhancer sites located on the PGC-1 promoter. Thus, both can serve as positive controls for testing the regulation of PGC-1 by putative factors. To clarify whether MyoD can regulate the expression of PGC-1 by targeting its promoter, hPGC-1 promoter (–992 to +90) was transfected into C3H10T1/2 pluripotent fibroblasts in the presence or absence of MyoD. In the presence of MyoD, PGC-1 promoter activity was increased 8.3-fold compared with its basal activity in the absence of MyoD, but with vectors expressing green fluorescence protein (GFP) cotransfected (Fig. 3). MEF2C and FKHR activated PGC-1 promoter 3.7- and 2.3-fold, respectively, significantly less than that induced by MyoD (Fig. 3). This result clearly suggests that MyoD can target PGC-1 promoter either directly or indirectly to enhance its activity.

View larger version (15K): [in this window] [in a new window]   FIG. 3. PGC-1 promoter is activated by MyoD expression. Vectors (0.33 µg) expressing GFP, MyoD, MEF2C, and FKHR were cotransfected individually with PGC-1 promoter in the pGL3-basic vector (0.67 µg) into C3H10T1/2 cells overnight. Luciferase activity was determined 24 h after transfection. Luciferase activity in cells transfected with GFP and PGC-1 promoter was arbitrarily set at 1-fold activation. All myogenic factors activated PGC-1 promoter significantly (*, P < 0.05; **, P < 0.01).

Dose-dependent activation of PGC-1 promoter by MyoD

View larger version (11K): [in this window] [in a new window]   FIG. 7. Core promoter can function in a dose-dependent manner. Reporter constructs (0.67 µg) carrying different copies of PGC-1 core promoter were transfected into C3H10T1/2 cells in the presence and absence of MyoD-expressing vector (0.33 µg). Luciferase activity in cells transfected with GFP and reporter construct containing one copy of PGC-1 core promoter was arbitrarily set at 1-fold activation. The total DNA amount was made equal by including GFP-expressing vector in the liposome mixture.

View larger version (19K): [in this window] [in a new window]   FIG. 4. Dose-dependent activation of PGC-1 promoter by MyoD. A, Increasing amounts of MyoD-expressing vector were cotransfected with PGC-1 promoter (0.67 µg) into C3H10T1/2 cells to examine the effect of MyoD expression. Luciferase activity in cells transfected with GFP and PGC-1 promoter was arbitrarily set at 1-fold activation. All cells that received MyoD-expressing vector showed significantly higher luciferase activity than cells that received GFP-expressing vector. Cells transfected with FKHR-expressing vector were used as a positive control. B, MyoD-mediated PGC-1 promoter activation was promoter specific. No significant activation of pGL3-basic vector by MyoD was observed.

View larger version (23K): [in this window] [in a new window]   FIG. 5. MyoD target the PGC-1 core promoter region. Wild-type and deletion mutant promoters (0.67 µg) were cotransfected with vectors (0.33 µg) expressing GFP or MyoD into C3H10T1/2 fibroblasts (A and C) or RD rhabdomyosarcoma cells (B) to examine their regulation by MyoD. All promoter constructs, except –992 to –444, showed significant activation by MyoD in C3H10T1/2 cells (*, P < 0.05; **, P < 0.01; ***, P < 0.005). D, A schematic representation of the wild-type PGC-1 promoter (–992 to +90) and its deletion mutants. Their activation by MyoD is shown to the right. Key regulatory elements, such as cAMP response element (CRE), insulin response sequence (IRS1–3), and putative E boxes on the wild-type promoter are also shown.

The region from –444 to +90 was further deleted from the 5' end to make a series of deletion mutants. All 5' deletion mutants could be specifically activated by the expression of MyoD, and no significant difference was observed between these mutants with respect to their activation by MyoD (Fig. 5C). The MyoD-mediated activation was still very robust even when the reporter construct carried only –49 to approximately +2 region of the PGC-1 promoter. Because this region of promoter is 1) very short, 2) covers the transcriptional Initiator, and 3) supports MyoD-mediated activation, we thus named it the core promoter of the PGC-1 gene. The basal activity of the PGC-1 core promoter in the absence of MyoD is the same as that of the pGL3-basic vector (Fig. 5C).

PGC-1 core promoter is targeted by all myogenic bHLH proteins
The DNA-binding domain bHLH motif is highly conserved among MyoD gene family members. Therefore, it is of interest to know whether other members of this family can target the same region on the PGC-1 promoter. PGC-1 core promoter was cotransfected into C3H10T1/2 cells together with MyoD family members, and we found that all MRFs could activate this core promoter (Fig. 6A). No significant difference was observed among activations mediated by MyoD, myogenin, and MRF4. However, the activation mediated by Myf5 was consistently and significantly lower than those mediated by other MRFs. Myf5 is a factor mainly implicated in the lineage determination of myogenesis. Its function in the later stages of myogenesis is not well documented. Our result suggests that Myf5 is a poorer activator of PGC-1 expression than other members of the same family.

View larger version (18K): [in this window] [in a new window]   FIG. 6. PGC-1 core promoter is targeted by all myogenic bHLH proteins. PGC-1 promoters or pGL3-basic vector (0.67 µg) was cotransfected with vectors (0.33 µg) expressing either GFP or myogenic bHLH protein genes into C3H10T1/2 (A), HeLa (B), and CV-1 (C) cells. Luciferase activity in cells transfected with GFP and PGC-1 promoter was arbitrarily set at 1-fold activation. All myogenic bHLH proteins significantly activated PGC-1 promoter. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Core promoter can function in a dose-dependent manner
The core promoter targeted by MyoD lies just adjacent to the transcriptional initiation site. It is of interest to know whether this core promoter can function as an enhancer. Therefore, reporter constructs driven by different orientations and copies of this core promoter were created, and their activation by MyoD was tested in transient transfection assays. We found that the function of this core promoter was very orientation sensitive. It could not be activated by MyoD if it was presented in antisense orientation in the reporter constructs (data not shown). However, we also found that multiple copies of this core promoter could enhance MyoD-mediated activation and function in a copy number-dependent manner (Fig. 7). These results suggest that this core promoter can be targeted by MyoD only when it is presented in the right context. In this situation, E box-bound MyoD can cooperate to activate transcription synergistically.

Both E boxes in the core promoter are required for MyoD-mediated activation
The PGC-1 core promoter is only 51 bp long and can be activated by all myogenic bHLH proteins; this prompts us to consider that MyoD may activate this core promoter by targeting its response elements located on this promoter. Response elements mediating MyoD’s binding to DNA are called E boxes (consensus sequence, CANNTG), because MyoD needs to form heterodimers with ubiquitous E2A proteins (E12, E47, and E2–5) to achieve a stable binding complex on DNA. Two putative E box elements were identified after scrutinizing the core promoter sequence by the MATInspector function of the GENOMATIX suit (www.genomatix.de). The E box located between –45 and –40 was arbitrarily designated the E1 box, and the other located between –22 and –17 was designated the E2 box. The E1 box was predicted to be the target of Dec1/Stra13/Sharp2, a bHLH transcription factor involved in chondrocyte differentiation, but the E2 box was predicted to be 0.995 similar to the MyoD target consensus sequence (0.995 matrix similarity; a perfect match is 1.00). To confirm this bioinformatic prediction, the sequences of both E boxes were mutated into AT pairs independently or together to make mutant core promoter carrying either one (Mt1 and Mt2) or two (Mt3) mutated E boxes (Fig. 8A). Mutation of either E box on the core promoter ablated MyoD-mediated activation (Fig. 8B). Simultaneous mutation of both E boxes on the core promoter did not further reduce its activation by MyoD. Although E box mutation on the core promoter ablated its activation by MyoD, its basal activity was still similar to that of the wild-type promoter (data not shown) and was not affected by E box mutation. Therefore, this artificial mutation affects MyoD-mediated activation without reducing the basal activity of the promoter. This result of site-specific mutation strongly suggests that both E1 and E2 boxes are required for MyoD to activate PGC-1 core promoter. Whether both E boxes are specifically bound by MyoD-E protein heterodimer needs to be clarified by other assays.

MRFs bind directly to the core promoter E2 box
Although mutation of E boxes in the transfection assays suggests that they are unreplaceable elements in the core promoter for MRF-mediated activation, it does not necessarily mean that MyoD binds directly to these two E boxes. Apart from direct binding to the E boxes, MRFs can also activate PGC-1 transcription either by interacting with other E box-binding transcription factors through protein-protein interaction or by activating the expression of these E box-binding transcription factors. Therefore, to understand the mechanisms of MRF-mediated activation, we employed EMSA to examine the binding between PGC-1 core promoter DNA and MRF proteins.

PGC-1 core promoter (–49 to +2) DNA was end labeled with ³²P in a PNK-mediated reaction and used as probe for detecting its binding by proteins. Bacterially expressed GST-MyoD bound both wild-type and Mt1 PGC-1 core promoter (Fig. 9A, lanes 3 and 7), and the complex could be supershifted by MyoD antibody (Fig. 9A, lane 4). The binding between GST-MyoD and Mt2 promoter (Fig. 9A, lane 10) was much lower compared with that of wild-type and Mt1. Similar results were observed when GST-myogenin was used to bind these probes (Fig. 10B). Although MyoD and myogenin are highly similar in their sequence contexts, the complexes formed by these two proteins with the PGC-1 core promoter were very different. GST-MyoD formed two or three discrete lower bands along with a diffuse, low-mobility band. On the contrary, GST-myogenin always formed a discrete major band along with a very faint minor band. To date, we are not sure whether these bands represent monomer, dimer, or multimer proteins complexing with DNA. It is of interest to know whether MyoD and myogenin bind this DNA probe with different affinities. Therefore, increasing amount of GST-bHLH, GST-MyoD, and GST-myogenin (supplemental Fig. S1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org) were used to bind the wild-type probe. No DNA binding was observed when GST-bHLH was used in the assay (Fig. 9C). Both GST-MyoD and GST-myogenin bound wild-type probe in a dose-dependent manner (Fig. 9C, lanes 2–9). However, because the amount of GST-myogenin protein used was only half that of GST-MyoD (supplemental Fig. S1), and the complex formed by GST-myogenin was stronger than that of GST-MyoD (Fig. 9C), it suggests that the binding affinity of GST-myogenin to PGC-1 core promoter was slightly higher than that of GST-MyoD.

View larger version (68K): [in this window] [in a new window]   FIG. 9. MRF proteins bind directly to the core promoter E2 boxes. PGC-1 core promoter (–49 to +2) was end labeled with 32P by PNK-mediated reaction and used for a probe in the EMSA after being purified by the tombstone method. A, GST-MyoD binds directly to the wild-type and mutant1 PGC-1 core promoters, but not to the Mt2 promoter. Wild-type (Wt) and mutant (Mt1 and Mt2) PGC-1 core promoter probes were incubated with GST (2.0 µg; lanes 2, 6, and 9), GST-MyoD (2.0 µg; lanes 2, 4, 7, and 10), or alone (lanes 1, 5, and 8) for 30 min before being run on the gel. The specificity of the GST-MyoD complex was confirmed by its supershifting when antibody (0.25 µg) against MyoD was included in the reaction (lane 4). G, GST protein. B, Similar to A, but with the GST-MyoD and its antibody replaced by GST-myogenin (1.0 µg) and myogenin antibody (0.25 µg; lane 4). C, Both GST-MyoD and GST-myogenin bind wild-type PGC-1 core promoter in a dose-dependent manner. Increasing amounts of GST-myoD (0.8–3.2 µg), GST-myogenin (0.4–1.6 µg), and GST-bHLH (0.8–3.2 µg) were incubated with wild-type PCG-1 promoter to test their binding. The bHLH region in GST-bHLH was derived from that of MyoD. No retarded band was observed when GST alone (lane 1) or GST-bHLH (lanes 10–13) were included in the reaction.

View larger version (71K): [in this window] [in a new window]   FIG. 10. Nuclear proteins bind majorly to the core promoter E1 box. A, Nuclear protein (10 µg) from Sol8 MT was used to bind wild-type PGC-1 core promoter. Cold core promoter DNA of wild-type and E box mutants in 100-fold excess to that of the labeled probe were also included in the reactions shown in lanes 3, 4, and 5. Both wild-type (Wt) and mutant2 (Mt2) specifically competed off the E1 box complex formed by nuclear proteins and wild-type probe. NE, Nuclear extract. B, Increasing amounts (25-, 50-, and 100-fold) of cold wild-type DNA was used to compete off the protein-DNA complex formed by GST-myogenin (1.0 µg; lanes 1–5) or nuclear proteins (10 µg; lanes 6–10) with wild-type probe. A 150-bp DNA fragment (G) corresponding to GAPDH cDNA was included in the reaction to serve as a nonspecific competitor on lanes 5 and 10.

Because nuclear proteins and MRFs form discrete complexes on either the E1 or E2 box, it is of interest to know their relative binding affinities to E boxes. Increasing amounts (25-, 50-, and 100-fold) of cold wild-type DNA was used to compete off the protein-DNA complex formed by either GST-myogenin (1.0 µg) or nuclear proteins (5 µg) with wild-type probe. The E1 box complex could be competed off by cold DNA in a dose-dependent manner (Fig. 10B, lanes 6–9). No competition was observed when the same amount of cold DNA was included to compete off the MRF complex (Fig. 10B, lanes 1–4). The binding between wild-type probe and GST-myogenenin was indeed reduced when up to 800-fold cold DNA as included in the reaction (data not shown). This might reflect the fact that GST-myogenin has strong binding affinity to the E2 box, and it takes a lot of cold DNA to compete it off when a large amount of GST-myogenin is used in the binding assay. Neither the E1 nor E2 box complex was reduced when 100-fold GAPDH cDNA was included in the reaction (Fig. 10B, lanes 5 and 10).

One interesting question that remains to be answered is whether MRFs are present in the E1 box complex? To answer this question, wild-type and E box mutant probes were tested for binding by nuclear protein isolated from Sol8 myotubes. A similar binding pattern was observed when wild-type and Mt2 probes were used (Fig. 11A, lanes 2 and 12). However, no shifted band was observed when the E1 box was mutated (Fig. 11A, lanes 7–10). This confirms that the major protein-DNA complex formed by the nuclear proteins is targeting the E1 box. No supershifted band was observed when antibodies against MyoD (Fig. 11A, lanes 3, 8, and 13; MD), myogenin (Fig. 11A, lanes 4, 9, and 14; Mg), and MEF2C (Fig. 11A, lanes 5, 10, and 15; MEF) were included in the reaction to test their presence in the DNA-protein complex. This result suggests that either these three proteins are not included in this E1 box complex or they are very minor components of this complex. However, to our surprise, when the same gel was exposed longer (8 d), about six or seven very faint shifted bands were observed in reactions probed with Mt1 (Fig. 11B). One of these bands showed supershift when antibodies against MyoD and myogenin were included in the reactions (Fig. 11B, lanes 8 and 9, respectively). This strengthens the conclusion that endogenous MRFs target the E2 box on the PGC-1 promoter.

View larger version (51K): [in this window] [in a new window]   FIG. 11. Endogenous MRFs bind to the core promoter E2 box. A, Endogenous MRF proteins are not included in the E1 box complex. Wild-type and E box mutant probes were incubated with nuclear extract (10 µg) isolated from Sol8 myotubes (all lanes except lanes 1, 6, and 11) or alone. Because no shifted band was observed when the E1 box was mutated (lanes 7–10), the protein-DNA complex formed by the nuclear proteins was therefore called the E1 box complex. No supershifted band was observed when antibodies (0.5 µg) against MyoD (MD; lanes 3, 8, and 13), myogenin (Mg; lanes 4, 9, and 14), and MEF2C (MEF; lanes 5, 10, and 15) were included in the reaction to test their presence in the DNA-protein complex. When the same gel as that in A was exposed longer (8 d), about six or seven very faint shifted bands were observed in reactions probed with Mt1. B, One of these bands showed supershift when antibodies against MyoD (lane 8) and myogenin (lane 9) were included in the reactions. C, Increased amounts of nuclear proteins (56 µg) from Sol8 CMB (lanes 2–6), Sol8 MT (lanes 7–11), and C3H10T1/2 (lane 12) as well as Mt1 probe (106 cpm) were used in the binding reaction to test their binding. A supershifted band was observed when antibodies against MyoD (MD; lanes 3 and 8) and myogenin (Mg; lanes 4 and 9) were included in the reaction to test their presence in the DNA-protein complex. The addition of MEF2C antibody (lanes 5 and 10) and control IgG (lanes 6 and 11) to the reaction had no effect on the mobility of the MRF complex. D, The amounts of MyoD and myogenin in the nuclear extract were detected by Western blotting using the same antibodies as those used in the EMSA. The amount of nuclear protein used was the same as that used in each reaction in C. Both antibodies against MyoD and myogenin were used at a 1:500 dilution.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
MyoD is a skeletal muscle-specific transcription factor that is implicated in keeping the myogenic lineage after determination and driving the entry into terminal differentiation. MyoD drives the entry of terminal differentiation mainly by activating the expression of two terminal differentiation-specific transcription factors, MEF2C and myogenin, which subsequently drive the expression of contractile and cell cycle exit genes. Although the expression of various muscle-specific genes has been found to be under the regulation of MyoD, not a single coactivator has been demonstrated to be the downstream target of MyoD. In this study we demonstrated that, apart from acting through activation of MEF2C expression, MyoD and other MRFs could participate in the transcriptional activation of the PGC-1 gene. Although PGC-1 is not a muscle-specific factor, it is the first coactivator found to date to be directly regulated by MyoD and other MRFs.

Our study shows that E1 and E2 boxes adjacent to the transcription start site are critical to the expression of PGC-1 in vitro and in vivo. This prompts us to consider whether these two E boxes are conserved elements during evolution. We found that the proximal region of the PGC-1 promoter (–170 to +60) is highly conserved (92.5% identity) among mammals (Fig. 12). Both E1 and E2 boxes are 100% conserved in the four mammalian species examined (human, pig, mouse, and rat). The conservation of the proximal promoter sequence, especially these of the E boxes, suggests that the mechanisms of regulation of PGC-1 expression are highly conserved during mammalian evolution. This sequence conservation may reflect the importance of PGC-1 in the survival of mammals. As described above, PGC-1 plays important roles in the homeostasis of energy consumption and cell type determination. Its proper expression, both temporally and spatially, will confer upon individuals who carry it selection advantages and thus conserve its regulatory regions. We also found that this conservation of proximal promoter sequence was significantly reduced between mammals and avian (42% identity; data not shown). The E1 box is totally beyond recognition in the chicken PGC-1 promoter. However, the E2 box is still highly conserved between mammals and chicken, except the 3' G is replaced by T in the chicken PGC-1 promoter.

View larger version (72K): [in this window] [in a new window]   FIG. 12. The PGC-1 proximal promoter is highly conserved in mammals. The PGC-1 promoter sequences are aligned and numbered according to the human PGC-1 gene. The accession numbers for the PGC-1 promoters of the various species are AY382577 (rat), AC116763 (mouse), AF18193 (human), and AY484494 (pig). The arrow indicates the transcription start site. Boxed regions represent E1 and E2 boxes. Regions with identical sequence were shaded in orange.

Some myogenic E boxes are specifically regulated by only one or two MRFs, and the binding of other MRFs is excluded. For instance, MyoD and myogenin act on the chicken myosin light chain I (cMLCI) gene as distinct transcription factors; they bind to E boxes 1 and 5, respectively, on the chicken myosin light chain I promoter. Furthermore, these two myogenic E boxes are not responsive to MRF4 (28). In our study we found that E boxes on the core promoter were similarly activated by all MRFs in C3H10T1/2 cells. No significant preference among MyoD, myogenin, and MRF4 was observed, albeit the transactivation mediated by Myf5 was lower than that mediated by other MRFs. This may reflect the fact that Myf5 functions solely during lineage determination and principally targets the expression of early lineage-specific genes.

It has been reported that MyoD and myogenin are differentially associated with fast and slow muscle fibers. More MyoD expression is observed in the fast fibers, and fewer in the slow ones. The opposite phenomenon is observed with the expression of myogenin (29, 30, 31). Down-regulation of MyoD is also observed during C2C12 terminal differentiation (32). Because the PGC-1 gene is preferentially expressed in the slow fibers, this suggests that myogenin has a more important role than MyoD in the regulation of PGC-1 expression. Our EMSA results support this hypothesis, because the binding of myogenin to the E2 box is slightly stronger than that of MyoD. Surprisingly, stronger activation of the PGC-1 promoter by myogenin was not observed in our study. However, because myogenin is five to 10 times more abundantly expressed than MyoD in the adult slow fibers, this suggests that myogenin might play a more important role in the maintenance of PGC-1 expression in the slow fibers after they acquire their slow fiber phenotype. When does MyoD regulate the expression of PGC-1? It has been reported that MyoD is expressed in the myoblasts of both types of fibers in neonates (30). Thus, the strong activation of PGC-1 by MyoD observed in our study may reflect the regulatory mechanisms functioning in the neonates or embryo. In the adult fast fibers, where PGC-1 is weakly expressed, but more MyoD expression is observed, the activation of PGC-1 by MyoD is probably compromised by the sequestration of key ancillary factors that are essential to MyoD function on the E2 box of the PGC-1 promoter.

One question remains to be answered is do MRFs activate PGC-1 in vivo? If the answer is yes, then is the binding of MyoD to E2 box constitutive or facultative? We are currently using chromatin immunoprecipitation assay to demonstrate the binding of MRFs to the E2 box in vivo and to identify the stimuli that promote MRF-mediated PGC-1 activation. We hope that in the near future we will be able to provide answers to these questions.

	Footnotes

 
This work was supported by the grants from National Science Council of Taiwan, Republic of China (NSC-93-2311-B-008-008) and the Brain Research Center of University System of Taiwan.

The authors have no potential conflicts of interest to declare.

First Published Online March 9, 2006

Abbreviations: bHLH, Basic helix-loop-helix; CMB, confluent; Ct, maximum cycle threshold; DTT, dithiothreitol; FKHR, Forkhead in rhabdomyosarcoma; GAPDH, glyceraldehyde-3-phosphate dehydrogenase, GFP, green fluorescence protein; GST, glutathione-S-transferase; h, human; MEF2, myocyte enhancer factor 2; MRF, myogenic regulatory factor; MT, myotube; Mt, mutant; Pfu, DNA polymerase-derived from Pyrococcus furiosus; PGC-1, peroxisome proliferation activation receptor- coactivator 1; PGC1, peroxisomal proliferator activated receptor- coactivator 1; PMB, proliferating; PNK, polynucleotide kinase; SDS, sodium dodecyl sulfate.

Received October 17, 2005.

Accepted for publication February 27, 2006.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Knutti D, Kaul A, Kralli A 2000 A tissue-specific coactivator of steroid receptors, identified in a functional genetic screen. Mol Cell Biol 20:2411–2422[Abstract/Free Full Text]
2. Puigserver P, Wu Z, Park CW, Graves R, Wright M, Spiegelman BM 1998 A cold-inducible coactivator of nuclear receptors linked to adaptive thermogenesis. Cell 92:829–839[CrossRef][Medline]
3. Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant G, Mootha V, Troy A, Cinti S, Lowell B, Scarpulla RC, Spiegelman BM 1999 Mechanisms controlling mitochondrial biogenesis and respiration through the thermogenic coactivator PGC-1. Cell 98:115–124[CrossRef][Medline]
4. Yoon JC, Puigserver P, Chen G, Donovan J, Wu Z, Rhee J, Adelmant G, Stafford J, Kahn CR, Granner DK, Newgard CB, Spiegelman BM 2001 Control of hepatic gluconeogenesis through the transcriptional coactivator PGC-1. Nature 413:131–138[CrossRef][Medline]
5. Tcherepanova I, Puigserver P, Norris JD, Spiegelman BM, McDonnell DP 2000 Modulation of estrogen receptor- transcriptional activity by the coactivator PGC-1. J Biol Chem 275:16302–16308[Abstract/Free Full Text]
6. Delerive P, Wu Y, Burris TP, Chin WW, Suen CS 2002 PGC-1 functions as a transcriptional coactivator for the retinoid X receptors. J Biol Chem 277:3913–3917[Abstract/Free Full Text]
7. Vega RB, Huss JM, Kelly DP 2000 The coactivator PGC-1 cooperates with peroxisome proliferator-activated receptor in transcriptional control of nuclear genes encoding mitochondrial fatty acid oxidation enzymes. Mol Cell Biol 20:1868–1876[Abstract/Free Full Text]
8. Oberkofler H, Esterbauer H, Linnemayr V, Strosberg AD, Krempler F, Patsch W 2002 Peroxisome proliferator-activated receptor (PPAR) coactivator-1 recruitment regulates PPAR subtype specificity. J Biol Chem 277:16750–16757[Abstract/Free Full Text]
9. Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss O, Michael LF, Puigserver P, Isotani E, Olson EN, Lowell BB, Bassel-Duby R, Spiegelman BM 2002 Transcriptional co-activator PGC-1 drives the formation of slow-twitch muscle fibres. Nature 418:797–801[CrossRef][Medline]
10. Czubryt MP, McAnally J, Fishman GI, Olson EN 2003 Regulation of peroxisome proliferator-activated receptor coactivator 1 (PGC-1) and mitochondrial function by MEF2 and HDAC5. Proc Natl Acad Sci USA 100:1711–1716[Abstract/Free Full Text]
11. Feingold K, Kim MS, Shigenaga J, Moser A, Grunfeld C 2004 Altered expression of nuclear hormone receptors and coactivators in mouse heart during the acute-phase response. Am J Physiol 286:E201–E207
12. Weitzel JM, Radtke C, Seitz HJ 2001 Two thyroid hormone-mediated gene expression patterns in vivo identified by cDNA expression arrays in rat. Nucleic Acids Res 29:5148–5155[Abstract/Free Full Text]
13. Daitoku H, Yamagata K, Matsuzaki H, Hatta M, Fukamizu A 2003 Regulation of PGC-1 promoter activity by protein kinase B and the forkhead transcription factor FKHR. Diabetes 52:642–649[Abstract/Free Full Text]
14. Cossu G, Tajbakhsh S, Buckingham M 1996 How is myogenesis initiated in the embryo? Trends Genet 12:218–223[CrossRef][Medline]
15. Cossu G, Kelly R, Tajbakhsh S, Di Donna S, Vivarelli E, Buckingham M 1996 Activation of different myogenic pathways: myf-5 is induced by the neural tube and MyoD by the dorsal ectoderm in mouse paraxial mesoderm. Development 122:429–437[Abstract]
16. Buckingham M, Bajard L, Chang T, Daubas P, Hadchouel J, Meilhac S, Montarras D, Rocancourt D, Relaix F 2003 The formation of skeletal muscle: from somite to limb. J Anat 202:59–68[CrossRef][Medline]
17. Black BL, Olson EN 1998 Transcriptional control of muscle developme