![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Ara Parlakian,7,
Julien Giordani,1,4,5,6
Charlotte Lahoute,1,4,5,6
Anne Bertrand,1,4,5,6
Athanassia Sotiropoulos,8
Laure Renou,1,4,5,6
Alain Schmitt,3,4,5,6
Judith Melki,9
Zhenlin Li,7
Dominique Daegelen,1,4,5,6* and
David Tuil1,4,5,6*
Département de Génétique et Développement,1 Plate-Forme de Recombinaison Homologue,2 Plate-Forme de Microscopie Electronique, Institut Cochin, Paris F-75014, France,3 INSERM, U567, Paris F-75014, France,4 CNRS, UMR 8104, Paris F-75014, France,5 Université Paris 5, Faculté de Médecine René Descartes, UM 3, Paris F-75014, France,6 CNRS, UMR 7079, BP256, Université Paris 6, 7 quai St.-Bernard, 75005 Paris, France,7 INSERM, U344, Université René Descartes Paris 5, Faculté Necker, 156 rue de Vaugirard, 75015 Paris, France,8 Laboratoire de Neurogénétique Moléculaire, INSERM, Université d'Evry, E0223, Genopole, 2 rue Gaston Crémieux, 91057 Evry, France9
Received 24 January 2006/ Returned for modification 17 March 2006/ Accepted 7 June 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several families of transcription factors play crucial roles in the differentiation and specification of skeletal muscle cells. They include the myogenic regulatory factor (MRF) family, including MyoD, Myf5, MRF4, and the MEF-2 factors, which belong to the MADS box protein family. Serum response factor (SRF) is a distinct and widely expressed MADS box transcription factor produced in particularly large amounts in all skeletal muscles (3). SRF binds as a homodimer to the critical 10-bp consensus CArG box regulatory element, thereby activating genes involved in diverse biological processes, including cell proliferation, migration, survival, and muscle differentiation (for a review, see reference 19). Indeed, functional CArG boxes have been found in the cis-regulatory regions of various muscle-specific genes, such as the skeletal alpha-actin (SKA), muscle creatine kinase (MCK), dystrophin, tropomyosin, and myosin light chain 1/3 genes. SRF was first shown to be essential for both skeletal muscle cell growth and differentiation in experiments performed with C2C12 myogenic cells. In this model, SRF inactivation abolished MyoD and myogenin expression, preventing cell fusion in differentiated myotubes (36). Further experiments demonstrated that MyoD expression was modulated by a RhoA/SRF signaling cascade (6). However, the mechanisms resulting in the SRF-dependent activation of muscle-specific genes through CArG boxes are not entirely understood. Complexes of SRF and other muscle-specific partners, such as myogenin-E12 and MyoD-E12 heterodimers, may act as the target of the muscle differentiation signal (10). Myocardin-related transcription factors may also intervene as partners of SRF, activating SRF in response to a muscle-specific Rho signaling and actin polymerization pathway (15, 21, 33). The possible involvement of SRF in the physiology of adult skeletal muscle was highlighted by the observation of modulated SRF expression in association with mechanical overload-induced muscle hypertrophy (9). Collectively, these findings suggest that SRF is necessary for early myogenesis and may also regulate skeletal muscle growth.
Early embryonic lethality of SRF knockout mice made it impossible to use this model for studies of the role of SRF during in vivo myogenesis (2). We addressed this issue by developing a conditional SRF gene inactivation strategy in the mouse, based on the Cre-LoxP system that has been used successfully to demonstrate that SRF is crucial for cardiomyogenesis (26) and the maintenance of adult cardiac function (25). We investigated the role of SRF in the postnatal development of skeletal muscles, using an HSA-Cre transgenic line (20) in which Cre-mediated recombination occurs in postmitotic myofibers but not in satellite cells (23). Despite the death of 30% of the mutant mice lacking SRF in skeletal muscle fibers during the perinatal period, we were able to obtain and further analyze surviving mutant mice. These mutant mice soon displayed growth retardation and a major decrease in muscle mass due to severe myofiber hypotrophy resulting from impaired postnatal growth. Satellite cells were unaffected by the mutation, but SRF-depleted myofibers did not regenerate following injury. Moreover, myofibers lacking SRF displayed a reduced myonuclear number. Several superimposed mechanisms may account for this phenotype. The observed loss of SKA expression in mutant muscles may be responsible for a large proportion of the myofiber growth defects. This observation is consistent with the results of another very recent study using Cre-expressing mice to generate an earlier, muscle-specific SRF gene disruption, which was lethal during the perinatal period (17). We show here that the loss of SRF also led to a postnatal downregulation of transcription for both the IL-4 and IGF-1 genes. Thus, alterations in the corresponding pathways may also contribute to the phenotype via the impairment of satellite cell activation and/or recruitment by preexisting mutant myofibers. We identified SRF as a possible direct transcriptional regulator of both the IL-4 and IGF-1 genes, suggesting that SRF plays a key role in pathways involved in skeletal muscle growth and regeneration.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Muscle histology, immunohistochemistry, and morphometric measurements. All experiments involved comparisons of control and mutant littermates. Hind limb muscles from newborn to 8-week-old mice were removed and embedded in Cryomatrix, frozen in isopentane cooled in liquid nitrogen, and sectioned in a microtome cryostat (Leica). For assessment of tissue morphology, 5-µm-thick transverse sections were stained with hematoxylin and eosin (H&E) and examined under a light microscope. For fiber type analysis, serial sections were processed with a set of antibodies against the various myosin heavy chain (MyHC) isoforms, as previously described (4). We analyzed fiber size and determined the number of nuclei per myofiber by incubating muscle sections with mouse anti-dystrophin Dys2 antibody (Novocastra) and staining them with Hoechst stain (24). Hoechst-stained nuclei within the dystrophin-positive sarcolemma were counted in each myofiber of the entire muscle section. The fiber cross-sectional area (CSA) and the number of nuclei per myofiber were determined for three consecutive sections from five animals in each group, using Metamorph, version 2.56, software. For SRF immunodetection, paraffin-embedded longitudinal hind limb muscle sections were incubated with a 1:100 dilution of a rabbit polyclonal antibody directed against the carboxy-terminal domain of SRF (Santa Cruz Biotechnology), as described elsewhere (25).
Single myofiber isolation and primary muscle cell culture. Single myofibers with associated satellite cells were isolated from the tibialis anterior (TA) muscles of 2-month-old control and mutant mice, as described by Rosenblatt et al. (28). Primary cultures were derived from TA muscles, as described by Ohanna et al. (24).
Regeneration. Regeneration experiments were performed as described elsewhere (23). Briefly, four control, four heterozygous HSA-Cre:Sf/+, and four mutant mice were anesthetized with isoflurane (0.75 to 1.0% in oxygen) and received a single injection of cardiotoxin (0.25 µg/g of body weight) into the TA muscle. Mice were killed 9 days after injection, and the TA muscle was removed and processed for histological analysis.
Electron microscopy. Electron microscopy was performed on TA muscles from mutant and control adult mice, as previously described (30).
Western blot analysis. Western blotting was performed as previously described (26), using anti-SRF (1:200) and anti-glyceraldehyde-3-phosphate dehydrogenase (anti-GAPDH; 1:400) antibodies (Santa Cruz Biotechnology).
Northern blot analysis.
Northern blots were carried out with 10 µg of total RNA from hind limb muscles, as described by Bertrand et al. (4). Membrane blots were successively hybridized at 65°C with murine cDNA probes for SRF, skeletal alpha-actin, MCK, desmin, MyoD1, myogenin, MRF4, the nuclear factor of activated T cells c2 isoform (NFATc2), follistatin, myostatin, ß1-integrin, and IGF-1 labeled with [
-32P]dCTP using the Megaprime DNA labeling system (Amersham). These double-stranded probes were synthesized by reverse transcription-PCR (RT-PCR), using previously described primers (4, 26), and were inserted into the pCR2.1 TOPO vector (Invitrogen). A human NFATc2 probe spanning 927 bp (PstI/PstI) was also used (11). Sample normalization was assessed by hybridization with an 18S ribosomal probe.
Quantitative and semiquantitative RT-PCR analysis. Total RNAs were extracted from proliferating myoblasts from control and mutant TA muscles by using Trizol reagent (Invitrogen) and were reverse transcribed with Moloney murine leukemia virus reverse transcriptase (Invitrogen) and random hexamers (Promega) to generate cDNAs. Semiquantitative PCRs were then performed, using primers for SRF, IL-4, and GAPDH. Quantitative PCR analysis was performed as already described (25).
ChIP assays. C2C12 myoblasts were cultured in Dulbecco's modified Eagle's medium supplemented with 20% fetal bovine serum and were allowed to differentiate into myotubes by decreasing the concentration of serum to 2% for 48 h. Chromatin immunoprecipitation (ChIP) assays were performed by a modified version of the protocol described by Shang et al. (34). Myotubes were cross-linked by incubation in 1% formaldehyde for 10 min at 37°C, rinsed in phosphate-buffered saline, and lysed. Nuclei were collected by centrifugation and lysed in nuclear lysis buffer. Chromatin was subjected to sonication, and the mixture was then centrifuged to remove cellular debris. Supernatants were precleared with protein G-agarose and then immunoprecipitated with 2 µg of rabbit anti-SRF antibody or 2 µg of preimmune immunoglobulin G (Santa Cruz Biotechnology) overnight at 4°C. Antibody-protein-DNA complexes were isolated by immunoprecipitation with blocked protein G-agarose. Bound DNA fragments were eluted and analyzed by PCR. As a positive PCR control, we amplified 10 ng of genomic DNA. The H4 promoter was used as a negative control for ChIP assays. PCR primer sequences are available upon request.
Statistical analysis. Results are expressed as means ± standard errors of the means (SEM). The significance of differences between means was assessed with Student's t test. P values of <0.05 were considered statistically significant.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Exon 2 of SRF was efficiently excised, as shown by PCR analysis of genomic DNAs extracted from the skeletal muscles of 8-week-old HSA-Cre:Sf/Sf mice (Fig. 1A). In contrast, no excision was observed in other mutant mouse tissues. Western blotting showed that SRF protein levels were much lower in mutant than in control skeletal muscles (Fig. 1B). Cells other than myofibers are present in muscle samples and may account for the low level of nonexcised Sf/Sf alleles and the detection of residual amounts of SRF protein in mutant muscles. Consistent with these findings, SRF-immunoreactive nuclei in the control myofibers were clearly detected, whereas mutant myofibers uniformly lacked SRF-stained nuclei, demonstrating that the recombination efficiency was high and that SRF had been depleted successfully from all mutant skeletal muscle cells (Fig. 1C). We observed only a few cells with SRF-stained nuclei in histological tissue sections, and most of these cells were endothelial and smooth muscle cells surrounding blood vessels (not shown).
|
51% were still alive at 6 weeks of age (Fig. 2D). A comparative study of growth from birth to adulthood revealed that both male and female surviving mutant mice developed severe growth retardation, as demonstrated by total body weight measurements (Fig. 2A and B). Heterozygous HSA-Cre:Sf/+ mice were healthy and displayed normal growth (not shown). In addition, all surviving mutant mice rapidly displayed scoliosis, a sign of muscle weakness (Fig. 2C). As shown in Fig. 2E for various hind limb muscles, adult mutant mice displayed severe and generalized decreases in muscle mass. On average, muscle weights in mutants were only about 25% those in controls. This large decrease in muscle mass in mutants is illustrated in Fig. 2F for the diaphragm and in Fig. 4A for the extensor digitorum longus (EDL), demonstrating that the entire CSA of the muscle was affected. Mutant mice may die prematurely as a result of respiratory insufficiency, due to the particularly thin diaphragms in these animals. In accordance with the results of the growth curves, mutant mice also presented generalized dwarfism affecting all organs, as illustrated for the liver. However, the decrease in muscle mass was much more severe than the decreases in size for other organs.
|
|
|
80% lower for mutant than for control TA muscles, and the fiber size distribution was profoundly altered, with the vast majority of fibers having CSAs of <800 µm2 (Fig. 3C). Moreover, both H&E and dystrophin/Hoechst staining revealed no evidence of regenerating myofibers in adult mutant muscles, as no increase in the number of fibers with central nuclei was observed. Such a low percentage of centrally nucleated myofibers, identical to that observed in controls, was also observed in 3-day-old mutants (see Fig. S1 in the supplemental material). However, while in control muscles the percentage of centrally nucleated fibers increased between days 3 and 12, probably reflecting active recruitment of satellite cells by preexisting fibers for growth, no such increase was observed in mutants. Altogether, these data indicate that the decrease in muscle mass in HSA-Cre:Sf/Sf mutant mice results from severe myofiber hypotrophy, which settles during the postnatal period.
Fiber type composition shift in adult HSA-Cre:Sf/Sf skeletal muscles. The maturation of skeletal muscles is accompanied by the transcriptional activation of an array of muscle-specific genes, the products of which confer unique contractile (fast/slow) and metabolic (glycolytic/oxidative) properties on the various types of myofibers, reflecting the physiological specialization of muscles (29). The CSA of myofibers depends on their contractile and energy metabolism status, with the slow/oxidative fibers that express MyHC 1 having the smallest CSA. Moreover, it has been suggested that SRF regulates MyHC 2B gene expression (1). We therefore investigated whether the distribution of fast/slow myofiber subtypes was modified in mutant muscles, using antibodies against the various MyHC isoforms. Immunostaining of serial sections of mutant and control EDLs showed that mutant EDLs contained a much smaller number of fast/glycolytic type 2B myofibers, whereas the levels of fast/oxidative type 2A myofibers and slow/oxidative type 1 myofibers were much higher in all muscles of the mutants than in those of the controls (Fig. 4A). The control soleus muscle typically contained 50 to 60% MyHC 1 fibers, whereas the mutant soleus displayed an almost complete shift to MyHC 1 fibers, with only a very few MyHC 2A fibers remaining (Fig. 4B). As illustrated with the mutant soleus, even mutant type 1 fibers were much smaller than control type 1 fibers. These results indicate a major fast-to-slow fiber type transition in mutant skeletal muscles which is associated with a switch in oxidative metabolism, as revealed by NADH staining (not shown). Thus, the observed fiber type shift accounts for only a small proportion of the overall decrease in the mutant myofiber CSA, and myofiber hypotrophy affects all mutant fiber types. In addition, embryonic MyHC continued to be expressed strongly in mutant neonate muscles (4-day-old animals) but was not detected in the corresponding control muscles (not shown). Taken together, our data highlight the importance of SRF for the acquisition of a mature fiber type-specific pattern in adult muscle.
Muscle-restricted SRF depletion results in an impaired gene expression program. We further investigated the phenotype of mutant muscles by studying the expression of well-known SRF target genes and genes whose expression varies according to muscle maturity by Northern blotting of RNAs from adult mutant and control limb muscles. Consistent with the efficient excision of the SRF gene and loss of the SRF protein, only very low levels of SRF transcripts were detected in mutant muscles (Fig. 5A). The loss of SRF in mutant muscles led to a drastic downregulation of the transcription of SKA and MCK, both of which are well-characterized CArG-dependent genes, thereby demonstrating that SRF plays a crucial role in the transcriptional regulation of these two genes in vivo in the context of the adult myofiber (Fig. 5A and C). In contrast, the expression of other described SRF target genes was either moderately increased by the loss of SRF, as for desmin (see Fig. 6C), or strongly increased, as for MyoD (Fig. 5A). Consistently, as a consequence of MyoD gene reexpression, expression of the p21 gene, which is MyoD dependent, was also upregulated in mutant myofibers. Interestingly, myogenin expression, which was barely detectable in adult control limb muscles, was also strongly upregulated in SRF-depleted muscles, whereas MRF4 expression increased only moderately (not shown). These changes to the muscle gene expression program evoke defects in the maturation of adult mutant muscles. Moreover, the decrease in SKA expression may itself contribute to the mutant myofiber growth defect by impairing the construction of new sarcomere units.
|
|
Reduced number of myonuclei and impairment of the postnatal IL-4 pathway in myofibers lacking SRF. Postnatal myofiber growth involves cytoplasmic growth, through the addition of new contractile filaments to the preexisting sarcomere, and the incorporation of new nuclei through myoblast recruitment and fusion with nascent myotubes. We wondered whether the postnatal myofiber growth defect in mutants could also result from the defective activation and/or recruitment of satellite cells for fusion into the preexisting SRF-depleted fibers. For this purpose, we counted the Hoechst-stained myonuclei inside the dystrophin-stained sarcolemma in transverse TA and EDL sections (Fig. 6A and B). A major defect in the cytoplasmic growth of mutant myofibers was reflected by an approximately 60% decrease in the cytoplasmic domain regulated by a single myonucleus. However, the number of myonuclei was significantly lower (about 40%) in both SRF mutant muscles than in control muscles, suggesting that defective cytoplasmic growth is not the only mechanism accounting for mutant myofiber hypotrophy. Since satellite cells were unaffected by the SRF mutation in mutant muscles (see below and Fig. 7A and B), alterations in the ability of SRF-depleted myofibers to activate and/or recruit satellite cells for postnatal growth may also play a significant role in decreasing their CSA.
|
Regeneration and IGF-1 expression are impaired in adult HSA-Cre:Sf/Sf muscle. Surprisingly, although they clearly developed myopathy and muscle weakness, adult HSA-Cre:Sf/Sf mice presented no signs of muscle regeneration, such as centrally located myonuclei. In several ways, muscle regeneration recapitulates certain steps in developmental myofiber growth. We assessed the ability of myofibers lacking SRF to regenerate by subjecting the TA muscles of adult control, HSA-Cre:Sf/+ heterozygous, and SRF-depleted mice to cardiotoxin injury. Nine days after cardiotoxin injection, new myofibers had formed in the injured control TA muscles and in heterozygous mice (not shown), as demonstrated by the presence of centrally located myonuclei, whereas the injured mutant TA muscles were clearly atrophied and reproducibly failed to undergo normal regeneration (Fig. 7C). In injured mutant muscles, only very small numbers of regeneration foci and newly formed myotubes were observed, and muscle tissue was replaced by interstitial tissue composed of inflammatory and fibroblastic cells. This failure to regenerate was not due to a cell-autonomous defect of satellite cells, as the HSA-Cre transgenic line used in this study does not allow Cre-mediated recombination in satellite cells (23), and myoblasts isolated from mutant muscles contained normal amounts of SRF (Fig. 7B). In addition, similar numbers of myogenic progenitor cells migrated from single myofibers isolated from mutant muscles and from control myofibers (Fig. 7A), indicating that the poor regeneration in mutant muscles was not linked to alterations in the satellite cell pool. In differentiation medium, myoblasts from mutant muscle underwent normal differentiation into myotubes. This result is consistent with the observation that in muscle cell cultures, the Cre recombinase, under control of the HSA promoter, becomes active only in late myotubes (23). Indeed, we failed to detect efficient SRF exon 2 excision in cultured mutant myotubes at day 14 after satellite cell plating (see Fig. S2 in the supplemental material).
Downregulation of the IL-4 pathway is not sufficient in itself to account for the lack of mutant myofiber regeneration, as IL-4–/– myofibers do regenerate (11). IGF-1, a growth hormone known to play a key role in muscle growth and regeneration (13), has been implicated in almost all stages of muscle regeneration in adults (35). Northern blotting demonstrated that the 7.5-kb IGF-1 transcript level was strongly and reproducibly downregulated in mutant muscles lacking SRF (Fig. 7D). This downregulation of IGF-1 expression may account for the impaired regeneration observed in muscles lacking SRF.
SRF binds the promoter regions of both IL-4 and IGF-1 in vivo. We investigated whether the downregulation of IL-4 and IGF-1 expression observed in myofibers lacking SRF could result from direct transcriptional regulation via the binding of SRF to the corresponding promoter regions. Examination of 10-kb promoter regions for the mouse IL-4 and IGF-1 genes led to the identification of four putative SRF-binding sites in the IL-4 promoter region and two putative SRF-binding sites in the IGF-1 promoter region (Fig. 8A). Although these motifs were not strictly conserved at the same place in rat and human genes, several CArG-like sequences were also found in these regions. CArG-like sequences often constitute functional SRF-binding sites (19), and the most proximal IL-4 CArG-like sequence, with a 1-bp deviation (underlined) (CCATTCTTGG), was considered, as a similar sequence has been shown to be essential for MCK expression in vivo in muscle tissues (38). We investigated whether SRF bound these putative CArG motifs in the IL-4 and IGF-1 promoters in intact mouse myofibers by carrying out ChIP assays on differentiated mouse C2C12 myotubes. As expected, SRF bound specifically to one of the known consensus CArG boxes of the SKA promoter sequence, whereas no PCR signal was obtained if immunoprecipitation was performed with an irrelevant antibody (Fig. 8B). The CArG box-deficient histone H4 promoter was used as a negative control. Interestingly, three of the four putative CArG boxes within the IL-4 mouse promoter region were clearly enriched in SRF immunoprecipitates, including the nonconsensus proximal CArG-like sequence CCATTCTTGG, which is conserved at an identical location in the rat IL-4 promoter (Fig. 8B). The anti-SRF antibody gave specific enrichment of the most proximal CArG box 1 element of IGF-1 but not of the distal motif. Our ChIP assays indicate that SRF can bind the promoters of IL-4 and IGF-1 in differentiated myotubes, suggesting that these two genes are probably directly regulated by SRF.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Development of a myopathy characterized by myofiber hypotrophy and impaired maturation in skeletal muscles lacking SRF. All adult mutant mice developed generalized growth retardation and scoliosis linked to muscle weakness and myopathy, characterized by a decrease in total skeletal muscle mass and severe myofiber hypotrophy. However, mutant muscles did not display the characteristic pathological features of muscle necrosis/regeneration processes, such as macrophage infiltration and centrally nucleated fibers. Unexpectedly, the MyoD and myogenin genes, which are generally reexpressed in regenerating fibers and following denervation injury, were strongly expressed in adult mutant muscles. In the absence of signs of regeneration, this may reflect mutant muscle immaturity and/or changes in motor neuron transmission at the neuromuscular junction, suggesting a role for SRF in regulation of the genetic program for adult skeletal muscle differentiation. Indeed, adult skeletal muscles lacking SRF also displayed an overall increase in the number of slow/oxidative fibers, indicating a disruption of the early postnatal maturation of the various types of fiber, which is also partly dependent upon innervation (29). It was recently shown that SRF activates the fast/glycolytic fiber-specific MyHC 2B gene promoter, whereas a CArG-like element in the MyHC 2A promoter may act as a repressor (1). Activation of the MyHC 2B and MCK genes during muscle differentiation has been shown to involve chromatin remodeling and SRF recruitment (5). The absence of SRF in mutant muscles led to lower levels of MyHC 2B and MCK expression and higher levels of MyHC 2A expression. SRF may therefore play an essential role in the terminal differentiation of myofibers into the fast/glycolytic type 2B myofiber phenotype. Another, nonexclusive hypothesis is that the slow/oxidative phenotype of mutants could also be related to a physiological adaptation of muscles to the overload endogenously generated by a disproportionate body mass. NEM muscles also display such an increase in the number of slow/oxidative fibers generated by endogenous overload (14).
Different roles for SRF depending on skeletal muscle developmental stage. The absence of SRF protein in adult mutant skeletal muscles was associated with the strong downregulation of known target genes, such as SKA and MCK. In contrast, in experiments using a disruption of SRF occurring earlier in muscle development and resulting in the death of mutant neonates, Li et al. observed only mild effects on SKA transcription and no effect on MCK transcription at birth (17). These apparent discrepancies probably reflect differences in SRF function according to the skeletal muscle developmental stage. SRF seems to be particularly important for the transcription of SKA and MCK during postnatal muscle development. These observations are consistent with our previous observations concerning the cardiac alpha-actin gene. Expression of the cardiac alpha-actin gene is unaffected in embryonic hearts lacking SRF (26) but strongly decreased if SRF inactivation is triggered in the adult heart (25).
The absence of SRF in adult muscles led to an increase in mRNA levels for the desmin, troponin T, and MyoD genes, which have also been described as SRF-regulated target genes (16, 18, 39). These results reveal an unexpected complexity with regard to in vitro studies and show that the behavior of CArG box-dependent genes may depend on the cellular context and the developmental stage of the skeletal muscle tissue. Thus, the CArG box present in the distal regulatory region of the mouse MyoD gene, which is required for expression in myoblasts, is no longer necessary in the context of the differentiated myofiber.
Crucial role for SRF in various muscle growth mechanisms. Why do mice lacking SRF in their skeletal muscles display such severe postnatal myofiber growth defects? Around birth, mutant skeletal muscle mass and myofiber CSA seemed only very mildly affected, indicating that the mutant phenotype results primarily from the defective postnatal growth of skeletal myofibers. Using an earlier expressing Cre mouse model, Li et al. observed that SRF is necessary for muscle growth during late embryogenesis (17). In this study, the phenotype observed for HSA-Cre:Sf/Sf mice reflects the timing of SRF gene deletion, which mostly occurs during the postnatal period due to the use of the HSA promoter to direct Cre expression. Collectively, our results suggest that several superimposed defects may contribute to the postnatal muscle growth defect of HSA-Cre:Sf/Sf mice, as follows.
(i) SRF loss led to an important decrease in SKA expression and, to a lesser extent, Tpm2 expression, which could account for a large proportion of the postnatal cytoplasmic growth defect and the smaller size of the myonuclear domain. SKA expression is upregulated at birth, and this isoform predominates in adult skeletal muscle, in which it accounts for up to 90% of the total muscle actin present in the adult sarcomere structure. The profound disorganization of the sarcomere ultrastructure observed in SRF mutant myofibers is also consistent with the impaired stoichiometry of contractile thin filament gene expression (SKA, Tpm2, and TnnT1). Thus, it is not surprising that these ultrastructural alterations resemble those observed in human NEM, which are caused by mutations affecting the same genes (31, 37). Moreover, like our SRF mutants, SKA-null mice appear normal at birth but within 3 days display marked growth retardation, reduced muscle strength, and scoliosis. However, all SKA-null mice die early in the neonatal period (7).
(ii) SRF controls the expression of IL-4 and IGF-1, two locally secreted factors essential for muscle growth and regeneration. Skeletal myofiber atrophy due to defective cytoplasmic growth, as observed in mice deficient for the mTOR substrate S6K1, is not systematically accompanied by alterations in the number of myofiber nuclei (24). In contrast, myofibers lacking SRF displayed a decrease in the number of myonuclei similar to that observed in mice homozygous for an IL-4 null mutation (11). Indeed, IL-4 expression was strongly downregulated in mutant muscles during the period in which postnatal growth and satellite cell recruitment normally occur. The very low percentage of centrally nucleated myofibers observed in 12-day-old mutants compared to that observed in controls also argues for such defective satellite cell recruitment during postnatal growth. Our ChIP assays with myotube chromatin also revealed that SRF bound three CArG box sequences present in the IL-4 promoter in vivo, strongly suggesting that SRF regulates IL-4 transcription directly. Other transcription factors, such as NFATc2, have been shown to be involved in the control of IL-4 production (11). Unexpectedly, we found that NFATc2 expression was clearly upregulated in mutant muscle. As recently described for regulation of the smooth muscle -actin gene by an NFATc3/SRF complex (8), we cannot exclude the possibility that NFATc2 and SRF interact cooperatively to regulate IL-4 gene expression. The increased level of NFATc2 transcripts may reflect a compensatory attempt to counterbalance mutant myofiber hypotrophy. Similarly, follistatin mRNA levels were very high in mutant muscle, whereas myostatin mRNA levels were low. This balance in the pattern of gene expression would favor myoblast recruitment and fusion (12). Similar abortive compensatory hypertrophy mechanisms have also been observed in adult SRF-depleted cardiomyocytes (25). These data identify SRF as a crucial regulator of IL-4 transcription in growing myofibers.
The failure of myofibers lacking SRF to regenerate suggests that these cells are unable to promote the efficient activation and/or recruitment of satellite cells for their growth and regeneration. Although IL-4 expression is upregulated in damaged adult muscle, downregulation of the expression of this gene in myofibers lacking SRF is not sufficient in itself to explain this phenotype because IL-4–/– myofibers do regenerate, but with a subsequent defect in size (11). In contrast, the downregulation of the predominant IGF-1 mRNA variant expressed in skeletal muscle (35) observed in mutant myofibers may contribute to impairments in both growth and regeneration. Indeed, IGF-1 acts as an autocrine/paracrine mediator, promoting myofiber hypertrophy and regeneration through myoblast proliferation and myogenic differentiation. Targeted expression of IGF-1 in skeletal muscle has been shown to promote myofiber hypertrophy and to accelerate muscle regeneration (22, 27). We also found that SRF bound to a CArG element of the endogenous IGF-1 gene promoter in intact chromatin. Very little is known about the sequences controlling IGF-1 expression in skeletal muscle, but our data suggest an important role for SRF. Interestingly, liver-specific downregulation of IGF-1 expression has also been observed in a model of conditional disruption of SRF in the liver (M. U. Latasa, D. Couton, C. Mitchell, C. Charvet, A. Lafanechère, J. E. Guidotti, Z. Li, D. Tuil, D. Daegelen, and H. Gilgenkrantz, submitted for publication), consistent with SRF playing a direct role in the transcriptional regulation of IGF-1. Another, nonexclusive explanation is that the lack of IL-4 expression in mutant muscles also contributes to IGF-1 downregulation. Indeed, IL-4 has been shown to stimulate IGF-1 expression in macrophages, and a similar pathway may operate in muscle cells (40).
We show here that SRF may play a new role in muscle growth and regeneration by controlling local IL-4 and IGF-1 expression. Further promoter dissection experiments should be carried out to confirm our initial hypothesis that IL-4 and IGF-1 are target genes for SRF.
Our data demonstrate a central role for SRF in postnatal myofiber hypertrophy, maturation, and regeneration through control both of the building of new sarcomere units and of two crucial pathways for skeletal muscle physiology. These findings open up several lines of future research. The generation of mice with inducible skeletal muscle-restricted inactivation of the SRF gene would make it possible to elucidate further the relationships between the SRF, IL-4, and IGF-1 pathways.
|
ACKNOWLEDGMENTS |
|---|
The NFATc2 cDNA probe was kindly provided by G. Pavlath (Emory University, Atlanta, Georgia). We thank E. Souil from the Tissue Morphology Technology Facility and F. Letourneur from the Genome and Sequencing Facility (Cochin Institute Paris 5 University) for their assistance. We also thank D. Couton and D. Bellanger for valuable technical help, H. Gilgenkrantz for critically reading the manuscript, and Region Ile de France for contributing to the Cochin Institute animal care facility.
|
FOOTNOTES |
|---|
Supplemental material for this article may be found at http://mcb.asm.org/.
C.H. and A.P. contributed equally to this work.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Arsenian, S., B. Weinhold, M. Oelgeschlager, U. Ruther, and A. Nordheim. 1998. Serum response factor is essential for mesoderm formation during mouse embryogenesis. EMBO J. 17:6289-6299.[CrossRef][Medline]
3. Belaguli, N. S., L. A. Schildmeyer, and R. J. Schwartz. 1997. Organization and myogenic restricted expression of the murine serum response factor gene. A role for autoregulation. J. Biol. Chem. 272:18222-18231.
4. 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.
5. Caretti, G., M. Di Padova, B. Micales, G. E. Lyons, and V. Sartorelli. 2004. The polycomb Ezh2 methyltransferase regulates muscle gene expression and skeletal muscle differentiation. Genes Dev. 18:2627-2638.
6. Carnac, G., M. Primig, M. Kitzmann, P. Chafey, D. Tuil, N. Lamb, and A. Fernandez. 1998. RhoA GTPase and serum response factor control selectively the expression of MyoD without affecting Myf5 in mouse myoblasts. Mol. Biol. Cell 9:1891-1902.
7. Crawford, K., R. Flick, L. Close, D. Shelly, R. Paul, K. Bove, A. Kumar, and J. Lessard. 2002. Mice lacking skeletal muscle actin show reduced muscle strength and growth deficits and die during the neonatal period. Mol. Cell. Biol. 22:5887-5896.
8. Gonzalez Bosc, L. V., J. J. Layne, M. T. Nelson, and D. C. Hill-Eubanks. 2005. Nuclear factor of activated T cells and serum response factor cooperatively regulate the activity of an alpha-actin intronic enhancer. J. Biol. Chem. 280:26113-26120.
9. Gordon, S. E., M. Fluck, and F. W. Booth. 2001. Selected contribution: skeletal muscle focal adhesion kinase, paxillin, and serum response factor are loading dependent. J. Appl. Physiol. 90:1174-1183.
10. Groisman, R., H. Masutani, M. P. Leibovitch, P. Robin, I. Soudant, D. Trouche, and A. Harel-Bellan. 1996. Physical interaction between the mitogen-responsive serum response factor and myogenic basic-helix-loop-helix proteins. J. Biol. Chem. 271:5258-5264.
11. Horsley, V., K. M. Jansen, S. T. Mills, and G. K. Pavlath. 2003. IL-4 acts as a myoblast recruitment factor during mammalian muscle growth. Cell 113:483-494.[CrossRef][Medline]
12. Iezzi, S., M. Di Padova, C. Serra, G. Caretti, C. Simone, E. Maklan, G. Minetti, P. Zhao, E. P. Hoffman, P. L. Puri, and V. Sartorelli. 2004. Deacetylase inhibitors increase muscle cell size by promoting myoblast recruitment and fusion through induction of follistatin. Dev. Cell 6:673-684.[CrossRef][Medline]
13. Jacquemin, V., D. Furling, A. Bigot, G. S. Butler-Browne, and V. Mouly. 2004. IGF-1 induces human myotube hypertrophy by increasing cell recruitment. Exp. Cell Res. 299:148-158.[CrossRef][Medline]
14. Joya, J. E., A. J. Kee, V. Nair-Shalliker, M. Ghoddusi, M. A. Nguyen, P. Luther, and E. C. Hardeman. 2004. Muscle weakness in a mouse model of nemaline myopathy can be reversed with exercise and reveals a novel myofiber repair mechanism. Hum. Mol. Genet. 13:2633-2645.
15. Kuwahara, K., T. Barrientos, G. C. Pipes, S. Li, and E. N. Olson. 2005. Muscle-specific signaling mechanism that links actin dynamics to serum response factor. Mol. Cell. Biol. 25:3173-3181.
16. L'Honore, A., N. J. Lamb, M. Vandromme, P. Turowski, G. Carnac, and A. Fernandez. 2003. MyoD distal regulatory region contains an SRF binding CArG element required for MyoD expression in skeletal myoblasts and during muscle regeneration. Mol. Biol. Cell 14:2151-2162.
17. Li, S., M. P. Czubryt, J. McAnally, R. Bassel-Duby, J. A. Richardson, F. F. Wiebel, A. Nordheim, and E. N. Olson. 2005. Requirement for serum response factor for skeletal muscle growth and maturation revealed by tissue-specific gene deletion in mice. Proc. Natl. Acad. Sci. USA 102:1082-1087.
18. Mericskay, M., A. Parlakian, A. Porteu, F. Dandre, J. Bonnet, D. Paulin, and Z. Li. 2000. An overlapping CArG/octamer element is required for regulation of desmin gene transcription in arterial smooth muscle cells. Dev. Biol. 226:192-208.[CrossRef][Medline]
19. Miano, J. M. 2003. Serum response factor: toggling between disparate programs of gene expression. J. Mol. Cell Cardiol. 35:577-593.[CrossRef][Medline]
20. Miniou, P., D. Tiziano, T. Frugier, N. Roblot, M. Le Meur, and J. Melki. 1999. Gene targeting restricted to mouse striated muscle lineage. Nucleic Acids Res. 27:e27.
21. Miralles, F., G. Posern, A. I. Zaromytidou, and R. Treisman. 2003. Actin dynamics control SRF activity by regulation of its coactivator MAL. Cell 113:329-342.[CrossRef][Medline]
22. Musaro, A., K. McCullagh, A. Paul, L. Houghton, G. Dobrowolny, M. Molinaro, E. R. Barton, H. L. Sweeney, and N. Rosenthal. 2001. Localized Igf-1 transgene expression sustains hypertrophy and regeneration in senescent skeletal muscle. Nat. Genet. 27:195-200.[CrossRef][Medline]
23. Nicole, S., B. Desforges, G. Millet, J. Lesbordes, C. Cifuentes-Diaz, D. Vertes, M. L. Cao, F. De Backer, L. Languille, N. Roblot, V. Joshi, J. M. Gillis, J. Melki, T. Frugier, F. D. Tiziano, E. Lacene, M. H. Moreau, P. Miniou, D. Tiziano, and M. Le Meur. 2003. Intact satellite cells lead to remarkable protection against Smn gene defect in differentiated skeletal muscle. J. Cell Biol. 161:571-582.
24. Ohanna, M., A. K. Sobering, T. Lapointe, L. Lorenzo, C. Praud, E. Petroulakis, N. Sonenberg, P. A. Kelly, A. Sotiropoulos, and M. Pende. 2005. Atrophy of S6K1(–/–) skeletal muscle cells reveals distinct mTOR effectors for cell cycle and size control. Nat. Cell Biol. 7:286-294.[CrossRef][Medline]
25. Parlakian, A., C. Charvet, B. Escoubet, M. Mericskay, J. D. Molkentin, G. Gary-Bobo, L. J. De Windt, M. A. Ludosky, D. Paulin, D. Daegelen, D. Tuil, and Z. Li. 2005. Temporally controlled onset of dilated cardiomyopathy through disruption of the SRF gene in adult heart. Circulation 112:2930-2939.
26. Parlakian, A., D. Tuil, G. Hamard, G. Tavernier, D. Hentzen, J. P. Concordet, D. Paulin, Z. Li, and D. Daegelen. 2004. Targeted inactivation of serum response factor in the developing heart results in myocardial defects and embryonic lethality. Mol. Cell. Biol. 24:5281-5289.
27. Rabinovsky, E. D., E. Gelir, S. Gelir, H. Lui, M. Kattash, F. J. DeMayo, S. M. Shenaq, and R. J. Schwartz. 2003. Targeted expression of IGF-1 transgene to skeletal muscle accelerates muscle and motor neuron regeneration. FASEB J. 17:53-55.
28. Rosenblatt, J. D., A. I. Lunt, D. J. Parry, and T. A. Partridge. 1995. Culturing satellite cells from living single muscle fiber explants. In Vitro Cell Dev. Biol. Anim. 31:773-779.[Medline]
29. Schiaffino, S., and C. Reggiani. 1996. Molecular diversity of myofibrillar proteins: gene regulation and functional significance. Physiol. Rev. 76:371-423.
30. Schmitt, A., J. Guichard, J. M. Masse, N. Debili, and E. M. Cramer. 2001. Of mice and men: comparison of the ultrastructure of megakaryocytes and platelets. Exp. Hematol. 29:1295-1302.[CrossRef][Medline]
31. Schroder, J. M., H. Durling, and N. Laing. 2004. Actin myopathy with nemaline bodies, intranuclear rods, and a heterozygous mutation in ACTA1 (Asp154Asn). Acta Neuropathol. 108:250-256.[Medline]
32. Schwander, M., M. Leu, M. Stumm, O. M. Dorchies, U. T. Ruegg, J. Schittny, and U. Muller. 2003. Beta1 integrins regulate myoblast fusion and sarcomere assembly. Dev. Cell 4:673-685.[CrossRef][Medline]
33. Selvaraj, A., and R. Prywes. 2003. Megakaryoblastic leukemia-1/2, a transcriptional co-activator of serum response factor, is required for skeletal myogenic differentiation. J. Biol. Chem. 278:41977-41987.
34. Shang, Y., X. Hu, J. DiRenzo, M. A. Lazar, and M. Brown. 2000. Cofactor dynamics and sufficiency in estrogen receptor-regulated transcription. Cell 103:843-852.[CrossRef][Medline]
35. Shavlakadze, T., N. Winn, N. Rosenthal, and M. D. Grounds. 2005. Reconciling data from transgenic mice that overexpress IGF-I specifically in skeletal muscle. Growth Horm. IGF Res. 15:4-18.[CrossRef][Medline]
36. Soulez, M., C. G. Rouviere, P. Chafey, D. Hentzen, M. Vandromme, N. Lautredou, N. Lamb, A. Kahn, and D. Tuil. 1996. Growth and differentiation of C2 myogenic cells are dependent on serum response factor. Mol. Cell. Biol. 16:6065-6074.[Abstract]
37. Sparrow, J. C., K. J. Nowak, H. J. Durling, A. H. Beggs, C. Wallgren-Pettersson, N. Romero, I. Nonaka, and N. G. Laing. 2003. Muscle disease caused by mutations in the skeletal muscle alpha-actin gene (ACTA1). Neuromuscul. Disord. 13:519-531.[CrossRef][Medline]
38. Vincent, C. K., A. Gualberto, C. V. Patel, and K. Walsh. 1993. Different regulatory sequences control creatine kinase-M gene expression in directly injected skeletal and cardiac muscle. Mol. Cell. Biol. 13:1264-1272.
39. Wang, G., H. I. Yeh, and J. J. Lin. 1994. Characterization of cis-regulating elements and trans-activating factors of the rat cardiac troponin T gene. J. Biol. Chem. 269:30595-30603.
40. Wynes, M. W., and D. W. Riches. 2003. Induction of macrophage insulin-like growth factor-I expression by the Th2 cytokines IL-4 and IL-13. J. Immunol. 171:3550-3559.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||
