MCB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grifone, R.
Right arrow Articles by Maire, P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Grifone, R.
Right arrow Articles by Maire, P.
Molecular and Cellular Biology, July 2004, p. 6253-6267, Vol. 24, No. 14
0270-7306/04/$08.00+0     DOI: 10.1128/MCB.24.14.6253-6267.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Six1 and Eya1 Expression Can Reprogram Adult Muscle from the Slow-Twitch Phenotype into the Fast-Twitch Phenotype

Raphaelle Grifone,1 Christine Laclef,1 François Spitz,1,{dagger} Soledad Lopez,1 Josiane Demignon,1 Jacques-Emmanuel Guidotti,1 Kiyoshi Kawakami,2 Pin-Xian Xu,3 Robert Kelly,4 Basil J. Petrof,1,{ddagger} Dominique Daegelen,1 Jean-Paul Concordet,1 and Pascal Maire1*

Departement Génétique, Développement et Pathologie Moléculaire, Institut Cochin-INSERM 567, CNRS UMR 8104, Université Paris V, 75014 Paris,1 Pasteur Institute, 75015 Paris, France,4 Department of Biology, Jichi Medical School, Minamikawachi, Kawachi, Tochigi 329-0498, Japan,2 McLaughlin Research Institute for Biomedical Sciences, Great Falls, Montana 594053

Received 17 December 2003/ Returned for modification 4 February 2004/ Accepted 26 April 2004


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Muscle fibers show great differences in their contractile and metabolic properties. This diversity enables skeletal muscles to fulfill and adapt to different tasks. In this report, we show that the Six/Eya pathway is implicated in the establishment and maintenance of the fast-twitch skeletal muscle phenotype. We demonstrate that the MEF3/Six DNA binding element present in the aldolase A pM promoter mediates the high level of activation of this promoter in fast-twitch glycolytic (but not in slow-twitch) muscle fibers. We also show that among the Six and Eya gene products expressed in mouse skeletal muscle, Six1 and Eya1 proteins accumulate preferentially in the nuclei of fast-twitch muscles. The forced expression of Six1 and Eya1 together in the slow-twitch soleus muscle induced a fiber-type transition characterized by the replacement of myosin heavy chain I and IIA isoforms by the faster IIB and/or IIX isoforms, the activation of fast-twitch fiber-specific genes, and a switch toward glycolytic metabolism. Collectively, these data identify Six1 and Eya1 as the first transcriptional complex that is able to reprogram adult slow-twitch oxidative fibers toward a fast-twitch glycolytic phenotype.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Muscle fibers exhibit different mechanical and metabolic properties through the expression of different sets of contractile proteins and metabolic enzymes. The relative proportions of the different fiber types within a given muscle are adapted to its function. For example, a muscle will be mainly composed of slow-twitch oxidative fibers if it is designed to sustain prolonged repetitive activity, whereas muscles involved in intermittent bursting efforts are composed primarily of fast-twitch glycolytic fibers. Differentiated adult muscle fibers also show a considerable degree of plasticity, as indicated by their ability to adjust their phenotype in response to altered circumstances, including training, hormonal shifts, or aging (reviewed in references 10 and 63). However, despite recent progress, the molecular mechanisms of muscle fiber diversity and adaptation remain controversial and not fully understood.

Muscle specialization is thought to arise from distinct myoblast populations, which are intrinsically committed to form either fast or slow multinucleated fibers. Such precociously specified myoblasts have been partly identified as adaxial cells in zebra fish (4, 20, 22, 37, 48), and may also exist in other vertebrates. In amniotes, the genetic determination of fast and slow myoblasts that are partly responsible for fiber type diversity has been largely demonstrated during avian myogenesis (2, 21, 44, 47, 69) and has also been proposed to occur in rodents (55). During later periods of myogenesis, an interaction between such genetically determined intrinsic commitment and various environmental cues, such as electrical activity supplied by the nerve in particular, participates in determination of the final phenotype of each fiber (28, 42, 56).

The central role of innervation in the maintenance of the slow phenotype in mature adult fibers has been amply demonstrated through denervation and cross-innervation experiments (8, 17, 27, 29, 52, 53) as well as different electrical stimulation paradigms (39, 61). A number of recent studies began to dissect the molecular events underlying the influence of slow motoneuron activity on muscle fiber phenotypes (65). Thus, several different signaling pathways have been reported to link nerve stimulation to determination of the slow-twitch muscle fiber phenotype, such as those involving activated calcineurin and NFAT (14, 46, 65), calcium-dependent CaM kinase (41, 71, 72), peroxisome proliferator-activated receptor-gamma coactivator 1 (PGC1{alpha}) (38, 73), and Ras (45). However, in contrast to slow fiber type determination, very little is known about the molecular mechanisms which operate to establish and maintain the fast fiber phenotype in adult skeletal muscle.

We have previously characterized the proximal regulatory sequences of the pM promoter of the aldolase A gene which are necessary and sufficient to reproduce its fast twitch type IIB- and IIX-specific activity in transgenic mice (16, 58, 66, 68). In this model, pM is expressed at levels at least 100-fold higher in the fast gastrocnemius (Gas) and tibialis anterior (TA) muscles than in the slow soleus (Sol) muscle. Although mutated versions of this promoter which were tested failed to significantly reduce this fast-twitch glycolytic muscle-specific activity, mutation of the MEF3 sites of the promoter precluded pM-driven transgene activity; therefore, its possible implication could not be tested. However, when the aldolase A MEF3 sites and an adjacent NFI binding site were placed upstream of an unrelated promoter, this was sufficient to drive chloramphenicol acetyltransferase (CAT) transgene expression in the fast Gas muscle but not in the slow Sol muscle (68). We have shown that transcription factors which modulate aldolase A expression through the MEF3 site are members of the Six/sine oculis family of homeoproteins (67). In the present study, we now show that nuclear Six1 and its partner Eya1 are enriched in fast-twitch fibers and that forced expression of these proteins in slow-twitch muscle can activate genes of the glycolytic metabolic pathway as well as sarcomeric genes of the fast contractile apparatus. Our data indicate that Six1 and Eya1 are able to act in a synergistic fashion to drive the transformation of slow-twitch oxidative fibers toward a fast-twitch glycolytic phenotype even in the presence of persistent slow motoneuron innervation. To our knowledge, these findings represent the first evidence of a transcriptional pathway controlling the fast-twitch glycolytic phenotype of adult skeletal muscle.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
In vivo transfections. In vivo transfection experiments were carried out on 7- to 10-week-old C57Bl6 females and 12-month-old pM-alkaline phosphatase (pM-AP) and MLC3f-ßgal (MlC3f-nlacZ-2E) (32) transgenic males. pM-AP transgenic mice contain the AP reporter gene under the control of the aldolase A pM promoter. Briefly, pM-AP transgene was obtained by subcloning both proximal regulatory sequence 1792 to 2250 from the human aldolase A pM (40) and an intronic sequences (from 4098 to 4616) which precedes the first coding exon, upstream of the human placental AP cDNA (19) and transgenic lines were created (G. Tavernier and J.-P. Concordet). A total of 4 µg of plasmid DNA in phosphate-buffered saline (PBS) was injected into the TA and Sol muscles under conditions of ketamine (100 µg/g of body weight) and xylazine (10 µg/g) anesthesia. An electrical field was then applied to both muscles with 70-mm-diameter circular electrodes (BTX tweezertrodes; QBiogen, Illkirch, France) placed on the medial and lateral sides of the limb. A voltage of 130 V/cm was applied six times in 60-ms square wave pulses at 10 Hz with a BTX electro cell manipulator electroporator (QBiogen). In experiments described in this paper (see Fig. 5, 7, 8, 9, and 10), muscles were pretreated 2 h before plasmid injection with 0.4 U of bovine hyaluronidase (Sigma)/µl in 25 µl of PBS (43). These experiments allowed transfection efficiencies from 10 to 50% of Sol and TA fibers 1 week after transfection, as revealed by electroporation of a control ß-galactosidase (ß-Gal) reporter plasmid. The number of independently electroporated muscles is given elsewhere in the text. For FK506 treatment, adult mice were injected daily with FK506 (2.5 µg/g) for a total of 12 weeks. All experiments using animals were conducted in accordance with the European guideline for the care and use of laboratory animals.



View larger version (70K):
[in this window]
[in a new window]
  		FIG. 5. Eya1 is also enriched in the nuclei of fast-twitch muscle fibers. (A) Northern blot experiment performed with 4 µg of poly(A)+ mRNA, showing the accumulation of Eya1 mRNAs in adult mouse muscles (upper panel) together with the relative ß-actin mRNA content (lower panel). Lane 1, Gas; lane 2, Sol. (B to G) Immunohistochemistry performed on adult Gas and Sol muscle sections with Eya1 (B and C, respectively), Eya2 (D and E, respectively), and Eya4 (F and G, respectively) antibodies. (H) Gel mobility shift assay performed with in vitro-translated Six1 and Eya1 proteins. Lane 1, 1 µl of in vitro-translated Six1; lane 2, 4 µl of in vitro-translated Flag-Eya1; lane 3, 1 µl of in vitro-translated Six1 plus 3 µl of in vitro-translated Eya1; lane 4, 1 µl of in vitro-translated Six1 plus 3 µl of in vitro-translated Eya1 plus 1 µl of Six1 antibodies; lane 5, 1 µl of in vitro-translated Six1 plus 3 µl of in vitro-translated Eya1 plus 1 µl of Flag antibodies. The results of fluorescent imaging of Sol muscle transfected 4 days earlier with 50 ng of Yfp-Eya1 alone (I) or 50 ng of Yfp-Eya1 in the presence of 500 ng of pCMV-Six1 expression vector (J). Magnification in panels I and J, x200.

 


View larger version (73K):
[in this window]
[in a new window]
  		FIG. 7. Forced expression of Six1/Eya1 in the adult slow Sol muscle activates pM-AP and MLC3f-ßgal transgenes. (A) Sol muscle cross sections obtained from pM-AP transgenic mice transfected with Six1/Eya1 expression vectors. At 2 weeks after transfection, AP activity was detected within the transfected Sol muscles of pM-AP transgenic mice. (B to E) Gfp and ß-Gal activity in Sol muscles of MLC3f-nlsßgal transgenic mice transfected with a control green fluorescent protein (Gfp) expression plasmid to allow identification of transfected fibers in the presence of Six1 and Eya1 expression vectors (C and E) or of pCR3 control plasmid (B and D). In panel C, transfected Gfp-positive fibers within the Sol corresponded to fibers demonstrating nuclear ß-Gal staining (E). Note that the ß-Gal blue nuclei are at the periphery of the fibers, indicating that these fibers have not undergone regeneration.

 


View larger version (122K):
[in this window]
[in a new window]
  		FIG. 8. Forced expression of Six1 and Eya1 in the adult slow Sol muscle drives a slow to fast transition 2 weeks after transfection. Immunohistochemical analysis at the Sol-Gas muscle level of expression with serial cross sections from pM-AP transgenic mice (B to F) or wild-type mice (G to N) transfected with Six1/Eya1 expression vectors (except in panels G and H, which show expression of MyHCI and MyHCIIX for a wild-type, not transfected, Sol section). Note that the antibody used to reveal MyHCIIX recognizes all MyHCs except MyHCIIX; thus, pure MyHCIIXs in panel H are unstained or are only weakly stained, as exemplified by those fibers marked by an asterisk in the Gas muscle image (D). (A) Schematic representation of the Gas-Sol muscle portions as presented in panels B to F; (B) AP detection; (C) MyHCIIB expression; (D) MyHCIIX expression; (E) MyHCIIA expression; (F) MyHCI expression. In panels B to F, the three long arrows indicate the same three fibers on serial cross sections, revealing that pM-AP-positive fibers also express MyHCHIIB, and perhaps MyHCIIX, but do not express MyHCIIA or MyHCI. The short arrow shows the same fiber on serial cross sections expressing MyHCIIB but not pM-AP; also note that this fiber no longer expresses either MyHCI or MyHCIIA. (G) MyHCI expression in wild-type Sol muscles at the periphery near the fibula. (H) MyHCIIX expression in Gas (left side of panel) and Sol (right side of panel) muscles of wild-type mice. (I to N) Six1/Eya1 expression vectors have been transfected into the Sol muscles together with a pDsRed expression plasmid (to visualize transfected fibers), and serial cross sections at the periphery of the muscle near the fibula are shown. (I) SERCA-1 expression, (J) DsRed detection, (K to L) MyHCIIB expression, (M) NADH-TR detection, (N) SDH detection. Note that both SDH and NADH-TR activities are downregulated in MyHCIIB-positive fibers.

 


View larger version (56K):
[in this window]
[in a new window]
  		FIG. 9. Nuclear accumulation of both Six1 and Eya1 is required to observe a slow to fast fiber type switch in Sol fibers. The results of immunohistochemical analysis of MyHCIIB expression on serial cross sections of Sol muscles transfected with 500 ng of pCMV-Six1 plus 500 ng of pCMV-Eya1 (A to C), with 500 ng of pCMV-Six1 only (D to E), or with 500 ng of pCMV-Eya1 only (F to G) are shown. MyHCIIB expression (A, D, and F) is only found within Sol fibers in which both Six1 and Eya1 nuclear accumulation are observed. Six1 nuclear expression as revealed by immunocytochemistry with Six1 antibodies (B and E). Eya1 nuclear expression as revealed by immunocytochemistry with Eya1 antibodies (C and G). Note that MyHCIIB is not detected in Sol fibers expressing only Six1 (D) or Eya1 (F). The stars in panels D and F point to the Sol fibers in which Six1 and Eya1 nuclear expression, respectively, was detected.

 


View larger version (79K):
[in this window]
[in a new window]
  		FIG. 10. Slow to fast fiber type switching occurs through Six1 targets. Wild-type (A and B) or pM-AP (C to E) Sol muscles were transfected with 100 ng of pDsRed plus 500 ng of pCMV-Six1 plus 500 ng of pCMV-Eya1mutant (A and B) or with 100 ng of pDsRed plus 500 ng of pCMV-Six1-VP16 chimeric protein only (C to E). At 15 days later, transfected Sol fibers were analyzed. (A) DsRed-positive transfected fibers. (B) Absence of MyHCIIB-positive fibers in an adjacent section. (C) Immunohistochemical analysis of MyHCIIB expression. (D) AP detection. (E) MyHCemb expression on serial sections.

 
CAT assays. pM164CAT and pM125CAT (60) activities in TA and Sol muscles were analyzed by in vivo transfections with 3.5 µg of pM164CAT and pM125CAT plasmids as described above. Transactivation by Six1 was performed by cotransfecting 100 ng of full-length mouse Six1 cDNA under the control of the CMV promoter present in pCR3-expressing vectors (Invitrogen) in a final volume of 50 µl. Empty pCR3 was used as a negative control. Transfection efficiency was monitored by cotransfection with pRSV-luciferase plasmid (0.5 µg). At 1 week after injection, muscles were dissected and homogenized in 500 µl of 0.1 M (pH 7.8) potassium phosphate buffer containing 1 mM dithiothreitol, and CAT as well as luciferase activities were assayed according to standard methods (58, 60).

Histological analysis. Muscles were dissected, embedded in cryomatrix, and quickly frozen in isopentane cooled with liquid nitrogen. Cryostat sections (15 µm) were fixed with 2% paraformaldehyde for 30 min, left for 1 h in blocking solution (1x PBS, 2% bovine serum albumin, 1% fetal calf serum, 0.1% Triton X-100). Rabbit polyclonal antibodies directed against Six1 (used at 1:800 dilution), Six4 (1:1,000), Six5 (1:1,000), Eya1 (1:3,000), Eya2 (1:3,000) or Eya4 (1:3,000) have been characterized previously (24, 67) or will be described elsewhere (Eya4 [H. Ozaki and K. Kawakami]). These antibodies and antibodies directed against SERCA1 (Calbiochem) (1:50) were then applied to the treated sections. Monoclonal antibodies against myosin heavy chain I (MyHCI) (Sigma) (1:5,000), MyHCIIA, MyHCIIB, and all MHC types except type IIX (SC-71, BF-F3, and BF-35, respectively) were used on unfixed 15-µm sections. Bound primary antibodies were detected with biotinylated secondary antibodies followed by horseradish peroxidase-conjugated streptavidin (Vectastain ABC kit; Vector laboratories) and DAB (Zymed laboratories) reaction. X-Gal (5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside) staining or AP staining was carried out on 2% paraformaldehyde-fixed 15-µm cryosections incubated at 37°C for 1 h for MlC3f-ßgal detection (see Fig. 7) or overnight in X-Gal staining solution (1 mM X-Gal, 5 mM potassium ferrocyanide, 5 mM ferricyanide, and 2 mM MgCl2 in 1x PBS) or AP staining solution (100 mM Tris-HCl [pH 9.5], 100 mM NaCl, 50 mM MgCl2, 337.5 µg of nitroblue tetrazolium [NBT]/ml, 175 µg of 5-bromo-4-chloro-3-indolyl phosphate [BCIP]/ml), at room temperature for 2 h, after inactivation of endogenous activity for 1 h at 65°C in PBS. NADH tetrazolium reductase (NADH-TR) and succinate dehydrogenase (SDH) histochemistry were carried out on 15-µm unfixed sections and incubated at 37°C in NADH staining solution (0.2 M Tris [pH 7.4], 1 mg of NBT chloride/ml, 0.4 mg of ß-NAD/ml) for 30 min, or in SDH staining solution (0.2 M sodium phosphate buffer [pH 7.6], 0.2 M sodium succinate, 1 mg of NBT chloride/ml) for 2 h.

Northern blots. RNA from Gas and Sol muscles of adult mice were prepared by the guanidinium thiocyanate procedure (15). Poly(A+) mRNAs were purified on an oligo(dT) column essentially as described previously (62). For Northern blot experiments, 20 µg of total RNA or 4 µg of poly(A+) mRNA was denatured with a formaldehyde-formamide mix before MOPS (morpholinepropanesulfonic acid)-agarose gel electrophoresis (62). mRNAs were transferred to a nylon membrane and hybridized with Six1, Eya1, R45 (complementary to 18S rRNA), or ß-actin probes.

Plasmids and protein synthesis. Comparison of mouse Six1 cDNA (50) with the human cDNA (6) revealed that the first 14 amino acids (aa) were lacking in the published mouse sequence. Full-length mouse Six1 cDNA was obtained by addition of a PCR fragment obtained from mouse genomic DNA corresponding to the first 14 aa of Six1 to the Six1-pCR3 expression vector already obtained (67). Six1-Vp16 expression vector was obtained by inserting the VP16 activation domain of the herpes simplex virus in frame with the last amino acids of Six1. Wild-type and mutant Eya1 cDNA expression vectors have been cloned in pCDNA3 (9). The yellow fluorescent protein (Yfp)-Eya1 expression plasmid has been obtained by ligation of the Eya1 EcoRI-ApaI fragment in EcoRI-ApaI-opened Yfp-C1 plasmid (Clontech). The polyMEF3-nlsßGal plasmid was obtained by ligation of six repeats of the MEF3 aldolase A sequence of 6x MEF3-tk (67) in front of a TATA-nlsßGal plasmid containing the –35 to + 45 aldolase A minimal promoter. Total protein extracts from Sol and Gas muscles were obtained by crushing frozen muscles in liquid nitrogen. Powder of muscle was then directly resuspended in Laemmli's buffer and ultrasonicated. Insoluble material was eliminated by centrifugation. A total of 100 µg of protein was then subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Transfer of the proteins as well as immunochemical detection were performed as described already (67). In vitro synthesis of Eya1 and Six1 was obtained with a transcription/translation T7 TNT quick-coupled transcription/translation kit (Promega). Gel mobility shift assays were performed as previously described (60) with aldolase A MEF3 double stranded DNA probes (67). Supershift experiments were performed with Six1 or Flag M2 (Sigma) antibodies.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The MEF3 element present in aldolase A pM promoter is required for its specific expression in fast-twitch adult skeletal muscles. To investigate the role of the MEF3 sites present in the aldolase A pM in fiber-type-specific transcriptional activation, we performed in vivo transient transfections of the fast TA and the slow Sol muscles of adult mice by injection-electroporation of different plasmid constructs. At 1 week after transfection, we compared the activity of a reporter gene driven by the –164 to +45 pM promoter fragment (pM164) (58), which includes the MEF3 sites, to that of the same promoter lacking the MEF3 sites (pM125) (60). In these in vivo transient transfection experiments, pM164 was about fivefold more active in the TA than in the Sol muscles. Deletion of the MEF3 sites abolished this preferential fast-twitch muscle activity (Fig. 1); the activity of pM125 in the TA was decreased to the levels observed in the Sol muscles with pM164. In contrast, deletion of the MEF3 sites did not significantly modify the low level of pM activity observed in the Sol muscles (Fig. 1). These findings suggest that the transcription factors binding to the MEF3 sites activate transcription in a fiber-type-specific manner (in fast TA but not in slow Sol muscles). However, it was previously shown that in a transgenic model (i.e., in a chromatin rather than an episomal context) mutation of the MEF3 sites also impaired the weak expression of pM in the Sol muscles while having a more dramatic effect on its level of expression in fast-twitch muscles (58, 68). Therefore, besides activating transcription in fast-twitch fibers, the MEF3-bound protein complexes may be involved in opening the chromatin structure at the pM promoter site in both fast and slow skeletal muscles.



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 1. The MEF3 element is required for specific expression of pM in adult fast-twitch skeletal muscle. Fast TA and slow Sol muscles were transfected with 3.5 µg of pM164CAT and pM125CAT plasmids as well as with 0.5 µg of pRSV-luciferase plasmid (the latter to allow normalization of transfection efficiencies). At 7 days after transfection, muscles were isolated and levels of activity of pM164CAT and pM125CAT in TA and Sol muscles (means ± standard errors of the means) were determined. Values for CAT activity are expressed relative to the level obtained in fast TA after transfection of pM164CAT (arbitrarily set as 100%). Differences in pM164CAT activity levels in fast TA muscle are represented. Each point represents the mean of six experiments.

 
MEF3 activity is higher in MyHCIIB fibers. It has previously been shown that a 70-bp fragment of the aldolase A promoter containing MEF3 and NFI binding sites was able to drive transcription specifically in fast-twitch muscles of the hindlimb (68), but our group was unable to characterize more precisely the DNA binding motif responsible for this specificity. To test the hypothesis that aldolase A MEF3 sites on their own can direct gene expression within a specific subset of fast-twitch fibers, we generated a mouse transgenic line which carries an artificial promoter composed of aldolase A MEF3 sites driving expression of the nlsßgal transgene. Although this transgene is completely silent in myogenic precursors migrating into the limb and, at later stages, in embryonic and adults limb muscles (R. Grifone et al., unpublished data), it is highly active in trunk and head muscles of both embryo and adults, indicating that MEF3 binding proteins are sufficient to drive muscle-specific gene transcription in these muscles. Serial sectioning of axial muscles was performed to test the fiber-type-specific expression of the polyMEF3-driven transgene. Whereas the nlsßgal transgene was expressed in MyHCIIB fibers, it remained silent in all other fiber types, thus indicating that the protein complex recruited to MEF3 sites in MyHCIIB-positive fibers is more active or more abundant than those in other fiber types (Fig. 2 and data not shown). It also appears that in MyHCIIB fibers, the protein complex recruited to the polyMEF3 promoter can activate transcription without the need for additional myogenic DNA binding motifs. Therefore, these data support the notion that MEF3 promoter elements are preferentially active in MyHCIIB fibers and that this activity is sufficient (even in the absence of other muscle-specific binding sites such as the E box, MEF2 sites, and NFI sites originally present in the aldolase A pM promoter) to drive transcription in a subset of adult muscle fast-twitch fibers.



View larger version (89K):
[in this window]
[in a new window]
  		FIG. 2. Detection of polyMEF3-nlsßgal activity in MyHCIIB fibers of transgenic animals. (A) Schematic representation of the polyMEF3-nlsßgal transgene. (B) nlsßgal activity on a transverse section of the longissimus dorsi muscle of a polyMEF3-nlsßgal transgenic adult animal. (C) Immunohistochemical analysis of MyHCIIB expression on a serial cross section. Note that only MyHCIIB positive fibers express the transgene.

 
The MEF3 binding homeoprotein Six1 is enriched in fast muscle nuclei. We have previously identified the proteins that bind to the MEF3 sites as members of the Six family, which are orthologs of the sine oculis protein originally identified in Drosophila melanogaster eye development (67). Among the six different Six genes identified in vertebrates, only Six1, Six4, and Six5 are expressed in adult skeletal muscles (67). To characterize the pattern of expression of Six1/4/5 in fast and slow muscles of the adult mouse, we performed immunohistochemical detection of Six1, Six4 and Six5 on cross-sections of the fast Gas and slow Sol muscles. These experiments revealed a preferential accumulation of Six1 in the nuclei of the fast-twitch Gas (Fig. 3A), whereas the level of accumulation of Six4 and Six5 in the nuclei of the Sol and Gas fibers did not appear to be different (Fig. 3B to C). Surprisingly, we also found that the levels of mouse Six1 mRNA, as measured by Northern blot analyses, were higher in Sol than in Gas or other fast-twitch muscles (Fig. 3D), suggesting that the specific enrichment of Six1 protein in fast-type muscles is posttranscriptionally controlled. To test this hypothesis, we performed Western blot experiments with total proteins from Sol and Gas muscles (Fig. 3E). These experiments revealed that Six1 levels in the Gas are no higher than those in the Sol muscles. Thus, absence of Six1 nuclear accumulation in slow Sol fibers compared to that of the fast Gas is posttranslationally controlled.



View larger version (74K):
[in this window]
[in a new window]
  		FIG. 3. Six1 is enriched in fast-twitch muscle nuclei. (A to C) Immunohistochemistry assays of adult mouse muscle sections at the Sol-Gas level performed using polyclonal antibodies directed against Six1 (A), Six4 (B), and Six5 (C). (D) Northern blot experiment performed with 20 µg of total RNA, showing the accumulation of Six1 mRNAs (upper panel) in adult muscles together with the relative levels of 18S rRNA content (lower panel) present in the different samples analyzed. Lane 1, Gas; lane 2, extensor digitorum longus; lane 3, Sol. (E) Western blot experiment performed with 100 µg of total protein, showing the presence of Six1 (upper) and {alpha}-tubulin (below) in Gas (lane 1) and Sol (lane 2) muscles.

 
Forced expression of Six1 alone in the Sol muscles is unable to change its phenotype. We hypothesized that the enrichment of Six1 in muscle nuclei of fast-twitch muscles could make this protein a strong candidate to account for the fiber-type specificity of pM activity. Accordingly, we tested whether its forced expression in the Sol muscles could ectopically activate pM. The Sol muscle of adult mice was cotransfected with pM164 CAT and a full-length Six1 cDNA expression vector (driven by a CMV enhancer-promoter), and analyzed for CAT activity after one week. In keeping with the above hypothesis, pM164 CAT activity was significantly higher in the Sol muscle transfected with 100 ng of Six1 expression vector (Fig. 4A), showing that increasing Six1 protein concentration in the Sol muscle allowed the protein to accumulate in the nuclei (see below) and efficiently activate transcription. Six1 trans activation also required the MEF3 sites, since no trans activation was observed with the pM125CAT reporter gene (Fig. 4A). Moreover, forced higher nuclear levels of Six1 in the slow-twitch Sol muscle was sufficient to activate pM transcription to levels close to those observed in the fast-twitch TA, thus suggesting that higher levels of Six proteins could be the basis for the specific activity of pM164 in fast muscles.



View larger version (67K):
[in this window]
[in a new window]
  		FIG. 4. Forced expression of Six1 in adult slow Sol muscles. (A) fast TA and slow Sol muscles were transfected with 3.5 µg of pM164CAT or pM125CAT plasmids, together with 0.5 µg of pRSV-luciferase to control for transfection efficiency, and either pCMV-Six1 expression plasmid (100 ng) or empty pCR3 vector (100 ng). Values for CAT activity at 7 days after transfection of pM164CAT into Sol muscles and pM125CAT into TA or Sol muscles (means ± standard errors of the means) are expressed relative to the level obtained with 100 ng of pCR3-pM164CAT in TA muscles (arbitrarily set as 100%). Each point represents the mean of six experiments. (B) Serial section at the distal hindlimb level of untransfected pM-AP transgenic line, showing that AP activity is not detectable in Sol muscles. The inset in the right bottom corner shows a macroscopic view of the Sol, with a border of positive-testing Gas fibers. (C) Serial sections of Sol muscles transfected with 500 ng of pCMV-Six1 expression vector in pM-AP transgenic mice. At 7 days after transfection, AP activity was revealed histochemically in the Gas-Sol muscles; while most Gas fibers express the AP transgene, none of the transfected Sol fibers express AP (the position of the Sol muscle adjacent to the plantaris-gas muscle masses is demarcated by a solid line). (D) Six1-transfected fibers of the Sol show a detectable amount of Six1 protein accumulation in their nuclei, as revealed by immunochemistry with Six1 antibodies. Note that panel D is an enlargement of a Sol serial section of panel C which has been revealed with Six1 antibodies.

 
Whereas the above experiments confirmed the ability of Six1 expression to trans activate pM within an episomal context, we wished to ascertain whether Six1 could also activate fast-twitch specific muscle genes which were integrated into a chromatin environment. This was done either by evaluating effects on previously characterized aldolase A pM transgenes that exhibit preferential expression in fast-twitch muscles, or by determining the effects of forced Six1 expression on endogenous fast-twitch muscle gene expression. We first transfected the Six1 expressing vector into the Sol muscles of adult transgenic mice carrying the pM-AP transgene (aldolase A pM promoter driving an AP reporter gene). This transgene is silent in Sol muscles, while it is very active in the adjacent Gas muscle (Fig. 4B), as expected from previous studies of pM activity in transgenic mice (59, 66). One week (n = 5) or two weeks (n = 2) after in vivo transfection of Six1 expression vector in the Sol muscles, we did not detect any AP positive fibers within this transfected muscle (Fig. 4C), despite the fact that immunohistochemistry revealed nuclei of many Sol transfected fibers expressing Six1 protein (Fig. 4D) at a level which was comparable to adjacent untransfected fast-twitch Gas muscle. In keeping with these findings, we were also unable to detect any modification of MyHC expression after one (n = 8) or two (n = 8) weeks of forced production of Six1 in the Sol muscles (data not shown, but see Fig. 9). Therefore, accumulation of Six1 protein in Sol muscle nuclei is not sufficient to activate the aldolase A pM promoter in a chromatin environment, nor is it able to modify the endogenous slow-oxidative phenotype of the Sol muscle. Taken together, these data suggest that endogenous Six1 protein levels in Sol muscle nuclei are not high enough to be detected by classical immunohistochemical techniques or activate episomal MEF3 targets, whereas forced overexpression of the Six1 gene in the Sol muscles permits myonuclear accumulation of the protein and transcriptional activation of episomal, but not integrated, MEF3 targets.

Eya1, a Six1 cofactor, is enriched in the nuclei of fast-twitch muscles. Differences in the ability of Six1 to activate fast-twitch genes in the episomal and chromatin environments suggested the possibility that Six1 cofactors might be required to activate the same genes in a chromatin environment, and that such cofactors could be absent from slow fibers. In this regard, eyes absent (Eya) proteins were initially identified in Drosophila as nuclear proteins able to interact with sine oculis (So) (54). Four Eya genes have been identified in mammals (Eya1-4). Similarly to So and Eya which act synergistically in eye development of Drosophila, Six and Eya proteins can interact physically through evolutionarily conserved protein domains and synergistically activate transcription of MEF3 dependent genes in mammals (23, 49) (C. Laclef and P. Maire, unpublished data). To determine whether Eya genes participate in fast-twitch muscle gene expression, we first monitored the presence of their mRNAs by Northern blot. Eya1, Eya2 and Eya4 mRNAs were found to accumulate in all muscles analyzed: vastus lateralis, masseter, extensor digitorum longus, diaphragm, Gas, and Sol (Fig. 5 and data not shown). However, immunohistochemistry performed on adult muscle cryosections revealed a preferential enrichment of Eya1 protein in the nuclei of fast-twitch Gas compared to the slow-twitch Sol muscle (Fig. 5B to C), while Eya2 and Eya4 nuclear accumulation levels were comparable in the two muscles (Figs. 5D to G). As the overall levels of Eya1 mRNA were essentially equal in the Sol and Gas muscles (Fig. 5A), this specific enrichment of Eya1 protein in the nuclei of fast-twitch muscles must result from differential posttranscriptional mechanisms in different fiber types, as is the case for Six1. We also tested whether Six1 and Eya1 proteins, which are both enriched in the nuclei of fast-twitch muscles, could interact. These proteins were in fact able to interact physically on the aldolase A MEF3 DNA element, forming a ternary Eya1-Six1-MEF3 complex, as shown by electrophoretic mobility shift assay performed with in vitro-synthesized Six1 and Flag-Eya1 proteins (Fig. 5H). These results suggest that in vivo, within the adult fast-twitch muscle nuclei where these two proteins accumulate, such an interaction could take place and synergistically drive transcription of fast-type genes. Results showing that Six1 and Eya1 can also interact in vivo (Fig. 5I to J) strengthen this hypothesis. Hence, expression of a chimeric Yfp-Eya1 protein in Sol fibers led to a diffuse localization of the protein. However, when the same chimeric Yfp-Eya1 protein was coexpressed with Six1 within Sol muscle fibers, this led to increased Eya1 accumulation in the nuclei. Altogether, these findings suggest that Six1 can drive Eya1 into the nuclear compartment of the living adult Sol fiber, possibly through direct physical interaction of the two proteins.

Six1 and Eya1 do not accumulate in Sol muscle myonuclei of mice treated with FK506. It has been shown that prolonged periods of treatment with the calcineurin inhibitor FK506 results in a transition of the slow-twitch muscle fibers of the mouse Sol toward a fast-twitch phenotype (14). To test whether Six1 and Eya1 proteins accumulate in Sol nuclei during this fiber type transition, we treated mice with FK506 daily for 12 weeks. Results of the FK506 treatment are summarized in Fig. 6A. After such treatment, 100% of Sol fibers were MyHCIIA positive (with 17% coexpressing the MyHCI isoform). In contrast, in untreated control mice only 46% of Sol muscle fibers were MyHCIIA positive (Fig. 6A). However, no specific enrichment of Six1 and Eya1 proteins was observed in the Sol muscles of FK506-treated animals (Fig. 6B). Interestingly, after this treatment we did not detect a fiber type transition in the adjacent plantaris muscle, in which in both control and treated animals we observed 22 and 24% of MyHCIIA (untreated and treated, respectively), 25 and 26% of MyHCIIX (untreated and treated, respectively), and 57 and 53% of MyHCIIB fibers (untreated and treated, respectively). These results suggest that blocking calcineurin activity only permits a shift from MyHCI to MyHCIIA and that nuclear accumulation of Six1 and Eya1 in Sol fibers is not linked to the calcineurin activation pathway present in slow fibers. Accordingly, Six1 and Eya1 may not play a role in establishing the fast-twitch oxidative (MyHCIIA) phenotype but rather play a role in establishing the fast-twitch glycolytic (MyHCIIB) phenotype.



View larger version (77K):
[in this window]
[in a new window]
  		FIG. 6. The slow to fast MyHCIIA transition observed in Sol fibers of mice treated with FK506 does not lead to Six1/Eya1 nuclear enrichment. (A) The percentages of fibers expressing different MyHC isoforms in Sol and plantaris muscles of mice treated with FK506, as well as in those of untreated controls, were determined by immunohistochemistry assays performed on hindlimb muscle cross sections at the Gas-Sol-plantaris level (n = 2 mice per group). (B) Immunohistochemical analysis of Six1 and Eya1 expression on cross-sections of FK506-treated or control Gas and Sol muscles, revealing the absence of Six1 or Eya1 nuclear enrichment in Sol fibers of FK506-treated animals.

 
Forced expression of Six1 and Eya1 together in the adult slow Sol muscles activates transgenes expressed in fast glycolytic fibers. To test the hypothesis that Six1 and Eya1 nuclear accumulation control the fast-twitch glycolytic type IIX and IIB phenotype, we cotransfected Six1 and Eya1 into Sol fibers of transgenic mice carrying fast-twitch-specific transgenes. In contrast to what was observed with transfection of Six1 alone (n = 2) (Fig. 4C), coexpression of Six1 and Eya1 in the Sol muscles of adult mice was able to activate expression of the pM-AP transgene (n = 2) (Fig. 7A). Furthermore, a similar observation was made using transgenic mice carrying the MLC3f-nlsßgal transgene (myosin light chain-3f promoter driving a nuclear-localized ß-Gal reporter gene), for which the activity level is generally far higher in fast Gas muscles than in the slow Sol muscles (32). Thus, coexpression of Six1, Eya1 and green fluorescent protein (Gfp) (the latter to visualize transfected fibers) revealed activation of the MLC3f-nlsßgal transgene in Gfp-positive (i.e., transfected) Sol fibers (Fig. 7C and E), whereas adjacent Gfp-negative (i.e., nontransfected) fibers and Sol fibers transfected by Gfp alone (Fig. 7 B and D) failed to demonstrate activation of this fast-type myosin promoter (n = 2). We also noted that accumulation of the ß-Gal reporter in Six1/Eya1-transfected fibers was always in peripherally located myonuclei, as opposed to the centrally located nuclei characteristic of regenerated muscle fibers in the mouse. As MyHCemb has been shown to be reexpressed during muscle regeneration, we checked for its expression but failed to detect it, which is consistent with a lack of muscle regeneration in our model (see Fig. 10). Therefore, these findings indicate that in our model, the switch in muscle fiber phenotype from slow to fast is achieved without the need for an associated muscle regenerative process.

Forced expression of Six1 and Eya1 together in the adult slow Sol muscle drives a slow to fast phenotype transition of endogenous sarcomeric and metabolic genes. We next tested the ability of Six1/Eya1 production to activate different endogenous fast-type muscle genes in slow Sol fibers. MyHCIIB was never detected in the Sol muscles of the 7- to 10-week old C57BL females used in this study (n > 7 for control animals). At 15 days after Six1 and Eya1 transfection, endogenous MyHCIIB and MyHCIIX gene activation was revealed by specific antibodies on serial cross-sections. Expression of MyHCIIB and MyHCIIX was observed in transfected Sol muscles of wild-type mice as well as in those of mice carrying either the pM-AP or MLC3f-ßgal fast-twitch-specific transgenes. Immunohistochemical analysis revealed that the accumulation of MyHCIIB and MyHCIIX proteins in pM-AP-positive fibers was associated with a complete extinction of MyHCI and MyHCIIA expression (Fig. 8B to F). Interestingly, some of the transfected Sol fibers failed to activate the aldolase A pM promoter while their MyHC expression profile nonetheless shifted, appearing as type IIB and IIX mixed hybrid fibers with no MyHCI or MyHCIIA expression (Fig. 8B to F).

In serial cross-sections of wild-type Sol muscles, we determined that less than 2% of the fibers expressed MyHCIIX (Fig. 8H and 6A) and that 60% of the fibers near the fibula were of the slow type I phenotype, the others being of type IIA phenotype (Fig. 8G and data not shown). The fibers expressing Six1/Eya1 in this location in animals whose Sol muscles had been transfected were MyHCIIB positive (n = 9) (Fig. 8K and L), arguing that type I as well as type IIA fibers are converted to a type IIB phenotype after forced expression of Six1 and Eya1. All MyHCIIB-positive fibers within the Sol muscles expressing Six1/Eya1 also expressed the fast sarco(endo)plasmic reticulum Ca2+ -ATPase (SERCA1), whereas SERCA1 was found in only 50% of normal control Sol fibers (Fig. 8I and data not shown). In addition, we noted that MyHCIIB-positive Sol fibers (Fig. 8L) strongly expressed enolase ß (data not shown), while both SDH and NADH-TR (as indicators of the oxidative potential of the muscle fiber) (Fig. 8M and N) activities were decreased to levels as low as those observed in IIB Gas fibers. A detailed analysis of Sol fibers expressing Six1/Eya1 revealed that more than 75% of these fibers appeared to have turned off their oxidative program and/or MyHCI/IIA expression while at the same time turning on MyHCIIB, MLC3f, aldolase A, enolase ß, and SERCA1 expression.

Six1 and Eya1 nuclear accumulation is required to reprogram the adult Sol fiber. We next sought to establish whether a correlation existed between Six1 and Eya1 nuclear accumulation in transfected fibers of the Sol and MyHCIIB expression within these same fibers. As shown in Fig. 9A, MyHCIIB accumulated only within those Sol fibers in which both Six1 and Eya1 could be detected in peripheral nuclei of the fibers (Fig. 9B to C). Indeed, we never observed MyHCIIB or MyHCIIX expression by immunohistochemistry in Sol fibers when the Sol muscle was transfected with empty vector (n = 3) or Six1 (n = 8) or Eya1 (n = 5) expression vectors delivered alone under the same experimental conditions (Fig. 9D to G).

Taken together, these results demonstrate that supplying Sol slow-twitch oxidative fibers with Six1 and its Eya1 cofactor leads to the activation of endogenous glycolytic enzymes and fast-twitch sarcomeric proteins as well as to the concomitant down-regulation of slow oxidative markers. Therefore, the Six1/Eya1 protein complex is able to convert adult slow-twitch oxidative fibers to a fast-twitch glycolytic phenotype, thus overriding (at least in part) the influence of the slow motoneuron firing pattern on mature muscle fibers of the Sol.

The Six1/Eya1 synergy driving slow to fast phenotype transitions in the Sol muscle is based on activation of Six1 target genes. To determine whether the synergistic effects of Six1 and Eya1 in producing the slow to fast phenotype transition within the Sol muscles are due to a direct interaction between these two proteins, we next coexpressed Six1 with a mutated Eya1 protein. This mutated Eya1 protein bears a single point mutation (replacing Leu 472 with Arg) which interferes with Six-Eya1 interactions (9, 51) and also causes the pathological branchio-oto-renal syndrome in humans (1). Coexpression of Six1 with this mutated Eya1 protein in Sol fibers failed to induce a slow to fast transition, and no MyHCIIB expression was detected in any of the transfected fibers (n = 7) (Fig. 10A and B). Thus, in our model it appears that the main function of Eya1 is to activate transcription of fast-type genes via its interaction with Six1 and not through some other transcription factors. Additional evidence for the central role of Six1 target genes in the slow to fast transition is provided by experiments in which we replaced Eya1 with another strong activating cofactor (VP16) linked to Six1. At 15 days after a chimeric Six1 protein linked to the strong activating domain of VP16 was transfected into Sol fibers, MyHCIIB- as well pM-AP-positive fibers were detected in Six1/VP16-expressing fibers (n = 3) (Fig. 10C and E). Therefore, the data are consistent with a mechanism whereby the Six1/Eya1 synergy which leads to reprogramming of the slow oxidative Sol muscles to a fast glycolytic phenotype is achieved mainly through activation of Six1 target genes, with Eya1 playing a critically important transactivating role in this process.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Members of the Six protein family were originally identified as homologues of the sine oculis protein, which is essential for development of the ocular system in Drosophila (12, 50, 64). Other Drosophila proteins which act together with sine oculis within a highly conserved transcription factor network include eyes absent (Eya), eyeless, and dachshund (Dac) (11, 54). Eya-deficient Drosophila results argue for a role for Eya in the specification of a subset of muscle fibers (7), and dSIX4, one of the three So-like proteins present in Drosophila, is coexpressed with Eya in the mesoderm and required for the fusion process of myoblasts (33). The mammalian homologues of these Drosophila proteins include the six Six proteins (31, 50), four Eya proteins (5, 75), nine Pax proteins (13), and four Dac-related proteins (25). These proteins are expressed in different cell types during embryogenesis (35, 76), suggesting that they could participate in distinct genetic programs triggered by various environmental cues. It was also found that vertebrate Eya2 is able to synergize with Dach2 or with Six1 to drive chick muscle differentiation (26). These observations are consistent with the fact that Six and Eya proteins can interact physically through domains which are conserved from Drosophila to mammals (26, 49).

Although the Six1/Eya2 complex was previously implicated in early myogenesis (26), in the present report we show that the Six1/Eya1 complex is involved in another crucial aspect of myogenesis, i.e., the metabolic and contractile specialization of mature muscle fibers. Hence, we provide the first evidence that a Six1/Eya1 transcriptional complex exerts regulatory control over the expression of fast isoforms of the contractile apparatus (MyHCIIB, MyHCIIX, MLC3f, and SERCA1) as well as the glycolytic enzymes aldolase A and enolase-ß. Six1/Eya1 thus appears to represent a novel regulatory transcriptional complex involved in the contractile-metabolic specialization of adult myofibers. Whether activation of all the aforementioned genes is directly under the control of the Six1/Eya1 complex through binding to MEF3 control elements remains to be determined. However, at least for the aldolase A pM promoter, this activation appears to be direct through MEF3 sites which are responsible for both muscle-specific (68) and fast phenotype-specific (this study) expression from this promoter.

As our transfection-electroporation method has been shown to target plasmids in all fibers independently of their fast-slow phenotype (3), the Six1 and Eya1 complex appears to be able to convert adult type IIA as well as type I myofibers of the Sol to a fast-glycolytic IIB phenotype. While it is still unclear how the accumulation of Six1 and Eya1 proteins is controlled within muscle fibers, we also provide preliminary evidence that differential enrichment within myonuclei of fast and slow fibers involves posttranscriptional regulation. The specific enrichment of Six1 and Eya1 in the nuclei of the fast Gas compared to the slow Sol muscles is reminiscent of the impaired nuclear targeting of NFATc1 found in isolated muscle fibers converted to a fast phenotype by electrostimulation (34). The mechanisms which prevent Six1 and Eya1 accumulation in Sol nuclei remain to be characterized. These events, however, can be bypassed by forced expression of Six1 and Eya1, suggesting that proteins present in Sol fibers that sequester Six1 and Eya1 outside of the nucleus can be titrated by this forced expression. Interestingly, Six1 has been found previously in both the cytoplasm and nucleus of myogenic cells during embryogenesis (24), and nuclear accumulation of Eya2, which, like Eya1, lacks a nuclear localization signal, has been shown to depend upon interactions with G{alpha} and Six proteins (23, 49). Direct interactions between Eya2 and G{alpha}z and G{alpha}i proteins which lead to the sequestration of Eya2 outside of the nuclei have been characterized (23). Moreover, it has been shown recently that activation of the acetylcholine receptor-G{alpha}q pathway in innervated avian muscle fibers in cell culture could provoke a slow- to fast-fiber type transition mediated by PKC activity (30). Whether such pathways are differentially activated in fast and slow adult muscles and control the nucleocytoplasmic shuttling of the Six1/Eya1 complex will have to be tested. The assessment of the distribution of Six1 and Eya1 in the nuclei of cross-innervated muscles and in fast- and slow-type muscle fibers during postnatal development will help us to understand how these proteins, in relation with specific innervation patterns, control the fast glycolytic phenotype.

The ability of Six1/Eya1 to regulate the contractile apparatus, calcium-regulatory proteins, and metabolic genes is analogous to the calcineurin pathway, which regulates slow-specific contractile proteins (14, 18, 72) as well as myoglobin, a protein related to oxidative metabolism (14). In response to slow motoneuron impulses, activated calcineurin directly controls the phosphorylation state of NFAT and allows its translocation to the nucleus, thereby leading to activation of target slow-type muscle proteins in cooperation with MEF2 and other regulatory proteins (14, 46, 65). Calcium-dependent CaM kinase activity is also upregulated by slow motoneuron activity. Furthermore, CaM kinase IV has been shown to amplify the slow-type calcineurin-generated responses by promoting MEF2 transactivator functions and enhancing the oxidative capacity of slow myofibers through stimulation of mitochondrial biogenesis (41, 71, 72). PGC1-{alpha} has been shown to interact directly with MEF2 to synergistically activate slow-twitch muscle genes. PGC1{alpha} is also capable of activating mitochondrial genes involved in oxidative metabolism independently of any known muscle regulatory protein (38, 73). Finally, a Ras-mitogen-activated protein kinase signaling pathway has also been shown to participate in the nerve-dependent induction of the slow program in regenerating muscle (45). However, while the Ras-mitogen-activated protein kinase and calcineurin transduction pathways appear to regulate the type I and IIA fiber phenotypes (45, 65), our results point to Six1/Eya1 participation in a separate, distinct pathway which controls the initiation and or maintenance of a MyHCIIX/MyHCIIB glycolytic muscle program. It appears unlikely that myonuclear accumulation of Six1/Eya1 in mammals is controlled by the calcineurin pathway, since no nuclear accumulation of Six1/Eya1 was observed in Sol fibers treated with an inhibitor of the calcineurin pathway, a treatment which in any case leads only to a partial change (from type I to IIA only and not to types IIX and IIB) of fiber phenotype toward the fast end of the spectrum. Moreover, absence of Eya1 protein enrichment in myonuclei of adult slow-twitch muscles in the mouse is reminiscent of the fact that Eya1 mRNA is not expressed in adaxial slow pioneer cells in zebrafish (57). These observations raise the possibility that during the course of evolution, different strategies have been selected to deprive slow type muscle cells of Eya1 activity.

Both Eya1 (74) and Six1 (35, 36) knockout mice show neonatal lethality, precluding analysis of postnatal muscle maturation in the absence of these proteins. Conditional Six1 and Eya1 knockout models will help to characterize the network of genes under the control of these proteins in adult muscle and their physiological relevance. In addition, it is conceivable that altered Six1/Eya1 function could play a role in certain neuromuscular diseases and aging, which show a preferential loss of MyHCIIX/MyHCIIB fibers (70).


   ACKNOWLEDGMENTS
 
We thank D. Tuil for helpful discussions concerning electroporation, Sophie Gautron for critical reading of the manuscript, I. Hanson and G. Borsani for the gift of mouse Eya4 cDNA, C. Thornton for the gift of Six5 antibodies, A. Keller for the gift of enolase ß antibodies, and Florence Bertin, Evelyne Souil, and Arlette Porteu for helpful technical assistance. We thank Fujisawa GmbH, Munich, Germany, for generously providing FK506.

R.G. and C.L. have been supported by a fellowship from the Ministere de la Recherche et de l'Education Nationale, and F.S. has been supported by a grant from the Société de Secours des Amis des Sciences. B.J.P. was supported by an Institut National pour la Santé et la Recherche Médicale (INSERM) travel fellowship. Financial support to this work has been provided by the INSERM, by an Action Concertée Incitative (ACI 0220514), and by the Association Française contre les Myopathies.


   FOOTNOTES
 
* Corresponding author. Mailing address: Departement Génétique, Développement et Pathologie Moléculaire, Institut Cochin-INSERM 567, CNRS UMR 8104, Université Paris V, 24 Rue du Faubourg Saint Jacques, 75014 Paris, France. Phone: 33 1 44 41 24 16. Fax: 33 1 44 41 24 21. E-mail: maire{at}mail.cochin.inserm.fr. Back

{dagger} Present address: Department of Zoology and Animal Biology, University of Geneva-Sciences III, 1211 Geneva 4, Switzerland. Back

{ddagger} Present address: Respiratory Division and Meakins-Christie Laboratories, McGill University Health Center, Montreal, Quebec, Canada. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 
1. Abdelhak, S., V. Kalatzis, R. Heilig, S. Compain, D. Samson, C. Vincent, F. Levi-Acobas, C. Cruaud, M. Le Merrer, M. Mathieu, R. Konig, J. Vigneron, J. Weissenbach, C. Petit, and D. Weil. 1997. Clustering of mutations responsible for branchio-oto-renal (BOR) syndrome in the eyes absent homologous region (eyaHR) of EYA1. Hum. Mol. Genet. 6:2247-2255.[Abstract/Free Full Text]

2. Anakwe, K., L. Robson, J. Hadley, P. Buxton, S. Allen, C. Hartmann, B. Harfe, T. Nohno, D. Evans, and P. Francis-West. 2003. Wnt signalling regulates myogenic differentiation in the developing avian wing. Development 130:3503-3515.[Abstract/Free Full Text]

3. Bertrand, A., V. Ngo-Muller, D. Hentzen, J.-P. Concordet, D. Daegelen, and D. Tuil. 2003. Muscle electrotransfer as a tool for studying muscle fiber-specific and nerve-dependent activity of promoters. Am. J. Physiol. Cell Physiol. 285:C1071-C1081.[Abstract/Free Full Text]

4. Blagden, C., P. Currie, P. Ingham, and S. Hughes. 1997. Notochord induction of zebrafish slow muscle mediated by Sonic hedgehog. Genes Dev. 11:2163-2175.[Abstract/Free Full Text]

5. Borsani, G., A. DeGrandi, A. Ballabio, A. B. Bulfone, L. Bernard, S. Banfi, C. Gattuso, M. Mariani, M. Dixon, D. Donnai, K. Metcalfe, R. Winter, M. Robertson, R. Axton, A. Brown, V. van Heyningen, and I. M. Hanson. 1999. EYA4, a novel vertebrate gene related to Drosophila eyes absent. Hum. Mol. Genet. 8:11-23.[Medline]

6. Boucher, C., N. Carey, Y. Edwards, M. Siciliano, and K. Johnson. 1996. Cloning of the human SIX1 gene and its assignment to chromosome 14. Genomics 33:140-142.[CrossRef][Medline]

7. Boyle, M., N. Bonini, and S. DiNardo. 1997. Expression and function of clift in the development of somatic gonadal precursors within the Drosophila mesoderm. Development 124:971-982.[Abstract]

8. Buller, A., J. Eccles, and A. Eccles. 1960. Interactions between motoneurones and muscles in respect of the characteristic speeds of their responses. J. Physiol. 150:417-439.[Free Full Text]

9. Buller, C., X. Xu, V. Marquis, R. Schwanke, and P. Xu. 2001. Molecular effects of Eya1 domain mutations causing organ defects in BOR syndrome. Hum. Mol. Genet. 10:2775-2781.[Abstract/Free Full Text]

10. Buonanno, A., and R. Fields. 1999. Gene regulation by patterned electrical activity during neural and skeletal muscle development. Curr. Opin. Neurobiol. 9:110-120.[CrossRef][Medline]

11. Chen, R., M. Amoui, Z. Zhang, and G. Mardon. 1997. Dachshund and eyes absent proteins form a complex and function synergistically to induce ectopic eye development in Drosophila. Cell 91:893-903.[CrossRef][Medline]

12. Cheyette, B., P. Green, K. Martin, H. Garren, V. Hartenstein, and S. Zipursky. 1994. The Drosophila sine oculis locus encodes a homeodomain-containing protein required for the development of the entire visual system. Neuron 12:977-996.[CrossRef][Medline]

13. Chi, N., and J. A. Epstein. 2002. Getting your Pax straight: Pax proteins in development and disease. Trends Genet. 18:41-47.[CrossRef][Medline]

14. Chin, E., E. Olson, J. Richardson, Q. Yang, C. Humphries, J. Shelton, H. Wu, W. Zhu, R. Bassel-Duby, and R. Williams. 1998. A calcineurin-dependent transcriptional pathway controls skeletal muscle fiber type. Genes Dev. 12:2499-2509.[Abstract/Free Full Text]

15. Chomczynski, P., and N. Sacchi. 1987. Single-step method of RNA isolation by acid guanidium thyocyanate-phenol-chloroform extraction. Anal. Biochem. 162:156-159.[Medline]

16. Concordet, J. P., M. Salminen, J. Demignon, C. Moch, P. Maire, A. Kahn, and D. Daegelen. 1993. An opportunistic promoter sharing regulatory sequences with either a muscle-specific or a ubiquitous promoter in the human aldolase A gene. Mol. Cell. Biol. 13:9-17.[Abstract/Free Full Text]

17. d'Albis, A., F. Goubel, R. Couteaux, C. Janmot, and J. Mira. 1994. The effect of denervation on myosin isoform synthesis in rabbit slow-type and fast-type muscles during terminal differentiation. Denervation induces differentiation into slow-type muscles. Eur. J. Biochem. 223:249-258.[Medline]

18. Delling, U., J. Tureckova, H. W. Lim, L. J. De Windt, P. Rotwein, and J. D. Molkentin. 2000. A calcineurin-NFATc3-dependent pathway regulates skeletal muscle differentiation and slow myosin heavy-chain expression. Mol. Cell. Biol. 20:6600-6611.[Abstract/Free Full Text]

19. DePrimo, S., P. Stambrook, and J. Stringer. 1996. Human placental alkaline phosphatase as a histochemical marker of gene expression in transgenic mice. Transgenic Res. 5:459-466.[CrossRef][Medline]

20. Devoto, S., E. Melancon, J. Eisen, and M. Westerfield. 1996. Identification of separate slow and fast muscle precursor cells in vivo, prior to somite formation. Development 122:3371-3380.[Abstract]

21. DiMario, J., S. Fernyak, and F. Stockdale. 1993. Myoblasts transferred to the limbs of embryos are committed to specific fibre fates. Nature 362:165-167.[CrossRef][Medline]

22. Du, S., S. Devoto, M. Westerfield, and R. Moon. 1997. Positive and negative regulation of muscle cell identity by members of the hedgehog and TGF-beta gene families. J. Cell Biol. 139:145-156.[Abstract/Free Full Text]

23. Fan, X., L. Brass, M. Poncz, F. Spitz, P. Maire, and D. Manning. 2000. The alpha subunits of Gz and Gi interact with the eyes absent transcription cofactor Eya2, preventing its interaction with the six class of homeodomain-containing proteins. J. Biol. Chem. 275:32129-32134.[Abstract/Free Full Text]

24. Fougerousse, F., M. Durand, S. Lopez, L. Suel, J. Demignon, C. Thornton, H. Ozaki, K. Kawakami, P. Barbet, J. Beckmann, and P. Maire. 2002. Six and Eya expression during human somitogenesis and Myod gene family activation. J. Muscle Res. Cell Motil. 23:255-264.[CrossRef][Medline]

25. Hammond, K. L., I. M. Hanson, A. G. Brown, L. A. Lettice, and R. E. Hill. 1998. Mammalian and Drosophila dachshund genes are related to the Ski proto-oncogene and are expressed in eye and limb. Mech. Dev. 74:121-131.[CrossRef][Medline]

26. Heanue, T., R. Reshef, R. Davis, G. Mardon, G. Oliver, S. Tomarev, A. Lassar, and C. Tabin. 1999. Synergistic regulation of vertebrate muscle development by Dach2, Eya2, and Six1, homologs of genes required for Drosophila eye formation. Genes Dev. 13:3231-3243.[Abstract/Free Full Text]

27. Huey, K., and S. Bodine. 1998. Changes in myosin mRNA and protein expression in denervated rat soleus and tibialis anterior. Eur. J. Biochem. 256:45-50.[Medline]

28. Hughes, S., and P. Salinas. 1999. Control of muscle fibre and motoneuron diversification. Curr. Opin. Neurobiol. 9:54-64.[CrossRef][Medline]

29. Jakubiec-Puka, A., I. Ciechomska, J. Morga, and A. Matusiak. 1999. Contents of myosin heavy chains in denervated slow and fast rat leg muscles. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 122:355-362.

30. Jordan, T., J. Li, H. Jiang, and J. DiMario. 2003. Repression of slow myosin heavy chain 2 gene expression in fast skeletal muscle fibers by muscarinic acetylcholine receptor and G{alpha}q signaling. J. Cell Biol. 162:843-850.[Abstract/Free Full Text]

31. Kawakami, K., S. Sato, H. Ozaki, and K. Ikeda. 2000. Six family genes—structure and function as transcription factors and their roles in development. Bioessays 22:616-626.[CrossRef][Medline]

32. Kelly, R., S. Alonso, S. Tajbakhsh, G. Cossu, and M. Buckingham. 1995. Myosin light chain 3F regulatory sequences confer regionalized cardiac and skeletal muscle expression in transgenic mice. J. Cell Biol. 129:383-396.[Abstract/Free Full Text]

33. Kirby, R., G. Hamilton, D. Finnegan, K. Johnson, and A. Jarman. 2001. Drosophila homolog of the myotonic dystrophy-associated gene, SIX5, is required for muscle and gonad development. Curr. Biol. 11:1044-1049.[CrossRef][Medline]

34. Kubis, H., R. Scheibe, J. Meissner, G. Hornung, and G. Gros. 2002. Fast-to-slow transformation and nuclear import/export kinetics of the transcription factor NFATc1 during electrostimulation of rabbit muscle cells in culture. J. Physiol. 541:835-847.[Abstract/Free Full Text]

35. Laclef, C., G. Hamard, J. Demignon, E. Souil, C. Houbron, and P. Maire. 2003. Altered myogenesis in Six1 deficient mice. Development 130:2239-2252.[Abstract/Free Full Text]

36. Laclef, C., E. Souil, J. Demignon, and P. Maire. 2003. Thymus, kidney and craniofacial abnormalities in Six1 deficient mice Mech. Dev. 120:669-679.

37. Lewis, K., P. Currie, S. Roy, H. Schauerte, P. Haffter, and P. Ingham. 1999. Control of muscle cell-type specification in the zebrafish embryo by Hedgehog signalling. Dev. Biol. 216:469-480.[CrossRef][Medline]

38. Lin, J., H. Wu, P. Tarr, C. Zhang, Z. Wu, O. Boss, L. Michael, P. Puigserver, E. Isotani, E. Olson, B. Lowell, R. Bassel-Duby, and B. Spiegelman. 2002. Transcriptional co-activator PGC-1 alpha drives the formation of slow-twitch muscle fibres. Nature 418:797-801.[CrossRef][Medline]

39. Lomo, T., R. Westgaard, and H. Dahl. 1974. Contractile properties of muscle: control by pattern of muscle activity in the rat. Proc. R. Soc. Lond. B Biol. Sci. 187:99-103.[Medline]

40. Maire, P., S. Gautron, V. Hakim, C. Gregori, F. Mennecier, and A. Kahn. 1987. Characterization of three optional promoters in the 5' region of the human aldolase A gene. J. Mol. Biol. 197:425-438.[CrossRef][Medline]

41. McKinsey, T., C. Zhang, J. Lu, and E. Olson. 2000. Signal-dependent nuclear export of a histone deacetylase regulates muscle differentiation. Nature 408:106-111.[CrossRef][Medline]

42. McLennan, I. 1994. Neurogenic and myogenic regulation of skeletal muscle formation: a critical re-evaluation. Prog. Neurobiol. 44:119-140.[CrossRef][Medline]

43. McMahon, J., E. Signori, K. Wells, V. Fazio, and D. Wells. 2001. Optimisation of electrotransfer of plasmid into skeletal muscle by pretreatment with hyaluronidase-increased expression with reduced muscle damage. Gene Ther. 8:1264-1270.[CrossRef][Medline]

44. Miller, J., and F. Stockdale. 1986. Developmental origins of skeletal muscle fibers: clonal analysis of myogenic cell lineages based on expression of fast and slow myosin heavy chains. Proc. Natl. Acad. Sci. USA 83:3860-3864.[Abstract/Free Full Text]

45. Murgia, M., A. Serrano, E. Calabria, G. Pallafacchina, T. Lomo, and S. Schiaffino. 2000. Ras is involved in nerve-activity-dependent regulation of muscle genes. Nat. Cell Biol. 2:142-147.[CrossRef][Medline]

46. Naya, F., B. Mercer, J. Shelton, J. Richardson, R. Williams, and E. Olson. 2000. Stimulation of slow skeletal muscle fiber gene expression by calcineurin in vivo. J. Biol. Chem. 275:4545-4548.[Abstract/Free Full Text]

47. Nikovits, W. J., G. Cann, R. Huang, B. Christ, and F. E. Stockdale. 2001. Patterning of fast and slow fibers within embryonic muscles is established independently of signals from the surrounding mesenchyme. Development 128:2537-2544.[Abstract/Free Full Text]

48. Norris, W., C. Neyt, P. Ingham, and P. Currie. 2000. Slow muscle induction by Hedgehog signalling in vitro. J. Cell Sci. 113:2695-2703.[Abstract]

49. Ohto, H., S. Kamada, K. Tago, S.-H. Tominaga, H. Ozaki, S. Sato, and K. Kawakami. 1999. Cooperation of Six and Eya in activation of their target genes through nuclear translocation of Eya. Mol. Cell. Biol. 19:6815-6824.[Abstract/Free Full Text]

50. Oliver, G., R. Wehr, N. Jenkins, N. Copeland, B. Cheyette, V. Hartenstein, S. Zipursky, and P. Gruss. 1995. Homeobox genes and connective tissue patterning. Development 121:693-705.[Abstract]

51. Ozaki, H., Y. Watanabe, K. Ikeda, and K. Kawakami. 2002. Impaired interactions between mouse Eyal harboring mutations found in patients with branchio-oto-renal syndrome and Six, Dach, and G proteins. J. Hum. Genet. 47:107-116.[CrossRef][Medline]

52. Pette, D., and R. S. Staron. 1997. Mammalian skeletal muscle fiber type transitions. Int. Rev. Cytol. 170:143-223.[Medline]

53. Pette, D., and G. Vrbova. 1999. What does chronic electrical stimulation teach us about muscle plasticity? Muscle Nerve 22:666-677.[Medline]

54. Pignoni, F., B. Hu, K. Zavitz, J. Xiao, P. Garrity, and S. Zipursky. 1997. The eye-specification proteins So and Eya form a complex and regulate multiple steps in Drosophila eye development. Cell 91:881-891.[CrossRef][Medline]

55. Pin, C., A. Hrycyshyn, K. Rogers, W. Rushlow, and P. Merrifield. 2002. Embryonic and fetal rat myoblasts form different muscle fiber types in an ectopic in vivo environment. Dev. Dyn. 224:253-266.[CrossRef][Medline]

56. Rafuse, V., and L. Landmesser. 2000. The pattern of avian intramuscular nerve branching is determined by the innervating motoneuron and its level of polysialic acid. J. Neurosci. 20:1056-1065.[Abstract/Free Full Text]

57. Sahly, I., P. Andermann, and C. Petit. 1999. The zebrafish eya1 gene and its expression pattern during embryogenesis. Dev. Genes Evol. 209:399-410.[CrossRef][Medline]

58. Salminen, M., S. López, P. Maire, A. Kahn, and D. Daegelen. 1996. Fast-muscle-specific DNA-protein interactions occurring in vivo at the human aldolase A M promoter are necessary for correct promoter activity in transgenic mice. Mol. Cell. Biol. 16:76-85.[Abstract]

59. Salminen, M., P. Maire, J. P. Concordet, C. Moch, A. Porteu, A. Kahn, and D. Daegelen. 1994. Fast-muscle-specific expression of human aldolase A transgenes. Mol. Cell. Biol. 14:6797-6808.[Abstract/Free Full Text]

60. Salminen, M., F. Spitz, M. Y. Fiszman, J. Demignon, A. Kahn, D. Daegelen, and P. Maire. 1995. Myotube-specific activity of the human aldolase A M-promoter requires an overlapping binding site for NF1 and MEF2 factors in addition to a binding site (M1) for unknown proteins. J. Mol. Biol. 253:17-31.[CrossRef][Medline]

61. Salmons, S., and G. Vrbova. 1969. The influence of activity on some contractile characteristics of mammalian fast and slow muscles. J. Physiol. 201:535-549.[Abstract/Free Full Text]

62. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.

63. Schiaffino, S., and C. Reggiani. 1996. Molecular diversity of myofibrillar proteins: gene regulation and functional significance. Physiol. Rev. 76:371-423.[Abstract/Free Full Text]

64. Serikaku, M., and J. O'Tousa. 1994. sine oculis is a homeobox gene required for Drosophila visual system development. Genetics 138:1137-1150.