AVP Induces Myogenesis through the Transcriptional Activation of the Myocyte Enhancer Factor 2

Abstract

The neurohypophyseal nonapeptide Arg⁸ vasopressin (AVP) promotes differentiation of cultured L6 and L5 myogenic cell lines and mouse primary satellite cells. Here, we investigated the molecular mechanism involved in the induction of the myogenic program by AVP. In L6 cells, AVP treatment rapidly induces Myf-5, myogenin, and myocyte enhancer factor 2 (MEF2) mRNAs, without affecting the expression of known myogenic growth factors such as IGF-I, IGF-II, or their receptors. In the presence of cycloheximide, AVP up-regulates the expression of MEF2, but not of myogenin, indicating that the synthesis of a protein intermediate is not necessary for MEF2 induction. Notably, AVP treatment activates a calcium/calmodulin kinase signaling pathway that induces cytosolic compartmentalization of the histone deacetylase 4, a mechanism related to the transcriptional activation of MEF2. The activity of chloramphenicol acetyltransferase reporter constructs carrying the Myo184 and Myo84 fragments of the myogenin promoter is also induced by AVP. Mutation of the MEF2 site completely abolishes the response to AVP, whereas deletion of the E1 site present in pMyo84 does not impair this response. Together, these results show that AVP induces myogenic differentiation through the transcriptional activation of MEF2, a mechanism that is critical for myogenesis.

MYOGENESIS IS CHARACTERIZED by the transcriptional activation of muscle-specific genes encoding for contractile proteins, metabolic enzymes, ion channels, and neurotransmitter receptors (1–3). Transcription of muscle-specific genes is regulated by a set of basic helix-loop-helix muscle regulatory factors (bHLH-MRFs) that include Myf-5, MyoD, myogenin, and MRF4 (4). Other more general factors such as the bHLH proteins E12 and E47 and the myocyte enhancer factor 2 (MEF2) are essential for muscle differentiation (5–7). During skeletal muscle development, Myf-5 and MyoD are responsible for the specification and maintenance of myoblast identity, whereas myogenin regulates the transcriptional activation of skeletal muscle structural genes (8).

The myogenic bHLH-MRFs are activated during myogenic differentiation and are sensitive to growth factors and oncogenic signals (9). Fibroblast growth factor (FGF) and TGFβ inhibit differentiation (10, 11), whereas IGFs are potent inducers of myogenic proliferation, differentiation, and hypertrophy (12–18).

We have previously showed that the neurohypophyseal nonapeptide AVP and related peptides constitute a novel family of positive regulators of terminal differentiation of myogenic cell lines (L6 and L5) and primary satellite cells (19–21). By interacting with V1 type receptor, AVP induces activation of phospholipases C and D, regulates cAMP levels, increases cytosolic Ca²⁺ concentrations and up-regulates Myf-5 and myogenin expression, both at the mRNA and at the protein level (22–24). The up-regulation of MRF genes is followed by stimulation of myotube formation with concomitant accumulation of myosin, increase of creatine-kinase activity, and number of acetylcholine receptor sites (20). Notably, in a chemically defined medium, which eliminates the interference of serum components, AVP induces myogenin expression and myogenic differentiation to a higher extent than that achieved with insulin or IGFs (21). However, the molecular mechanisms that mediate the effect of IGFs and of AVP are not fully understood.

Recent findings suggest that MyoD and MEF2 are present at specific myogenic gene promoters sites in protein:protein complexes that recruit the histone acetylases p300/CBP and P/CAF to activate gene expression (25–28). These nuclear histone acetyltransferases (HAT) promote myogenesis by acting as cotranscriptional activating factors that induce chromatin remodeling thus increasing the accessibility of specific DNA-binding factors and basal transcriptional machinery. A link between acetylation and transcriptional activity, and between hypoacetylation and transcriptional repression, has been strengthened by the observation that transcriptional coactivators possess HAT activity, whereas transcriptional corepressors are part of multisubunit complexes with histone deacetylase (HDAC) activity. The enzymatic activities of HATs and HDACs have been shown to play an important role in transcriptional activation and repression of eukaryotic genes, respectively, thereby establishing a molecular paradigm of transcriptional regulation based on modulation of chromatin structure by reversible acetylation of histone tails (29, 30). Recent work on HDACs indicated that some members of the class II HDACs (HDACs 4-5-7) are expressed in skeletal muscle cells, interact with MEF2, and specifically inhibit skeletal myoblast differentiation by repressing MEF2 activity on muscle-specific genes (31–36). These studies also show that HDAC 4-5-7 proteins shuttles from the nucleus to the cytoplasm when myoblasts differentiate into postmitotic myotubes (31–36). Thus, the nucleo-cytoplasmic trafficking of class II HDACs provides a novel mechanism for the transcriptional activation of MEF2 within the nucleus and induction of myogenesis.

In the present study, we investigated the molecular mechanism involved in the induction of the myogenic program by AVP. Here, we report that neither the secretion of IGFs nor the expression of IGF receptors (IGFRs) appears to be modulated by AVP. Rather, AVP induces myogenesis through a mechanism involving the transcriptional activation of MEF2 at the MEF2 site of the myogenin promoter. In addition, we show that AVP through a calcium-calmodulin kinase (CaMK) signaling pathway stimulates MEF2 activity by dissociating HDAC4 from the DNA binding domain, a potential mechanism that might be critical for the early event of myogenesis.

RESULTS

AVP Treatment Induces Myf-5, Myogenin, and MEF2 Expressions in L6 Cells

We have previously shown that AVP induced myogenesis in L6 cells by up-regulating the expression of myogenic bHLH-MRF Myf-5 and myogenin. MyoD and MRF4 are not expressed or induced by AVP in these cells (20, 21). In L6 cells cultured in serum-free medium, 0.1 μm AVP induced a transient increase of Myf-5 mRNA expression within 2 h (Fig. 1A). Myf-5 expression declined and reached basal levels after 6 h of AVP treatment concomitant to a progressive increase of myogenin mRNA expression (Fig. 1A and data not shown). The increase of myogenin was also detectable by Western blot analysis starting at 6 h of AVP treatment (Fig. 1B).

Figure 1.

AVP Treatment Induces Myf-5, Myogenin, and MEF2 Expression in L6 Cells L6 cells were plated in DMEM + 10% FCS, shifted in serum-free medium after 24 h and treated with AVP (0.1 μm). At the indicated times expression of Myf-5, myogenin and MEF2 in untreated (C) or in AVP-treated cells were detected by: A, Northern blots analysis of total RNA samples (30 μg) using the ³²P-labeled cDNA probes described in the method section; and B, Western blots performed using 30 μg proteins separated by SDS-PAGE and probed with anti-MEF2 and anti-myogenin F5D antibodies. C, L6 cells were plated in DMEM + 10% FCS, shifted (−serum) or not (+serum) in serum-free medium (DMEM + 1% BSA) after 24 h and treated for additional 24 h with AVP (0.1 μm) in the presence or in the absence of d(CH₂)₅Tyr(Me)AVP, [1-(β-mercapto-β,β-cyclopentamethylenepropionic acid),2-O-methyltyrosine]AVP, an antagonist of the V1 receptor.

Basic HLH-MRFs regulate the myogenic program by acting synergistically with other more general factors such as the myocyte enhancer factors MEF2, which in turn regulate each other’s expression (37–41). Interestingly, we found that treatment with 0.1 μm AVP induced the expression of MEF2 mRNA and protein in a time-dependent manner and with similar kinetics to the induction of myogenin expression, as shown by Northern blot and immunoblot analyses (Fig. 1, A and B). In addition, we found that in L6 cells cultured in serum-free media the induction of both MEF2 and myogenin by AVP could be inhibited by the addition of a potent and specific V1 antagonist [1-(β-mercapto-β,β-cyclopentamethylenepropionic acid),2-O-methyltyrosine]AVP (Fig. 1C). As shown in Fig. 1C, MEF2 and myogenin expressions resulted barely detectable after 24 h exposure to AVP of proliferating L6 cells cultured in serum-containing medium (DMEM + 10% FBS). These results are consistent with our previous findings indicating that a V1 receptor mediates the AVP-dependent induction of MEF2 and that the induction of myogenin by AVP occurs at earlier time in L6 cells cultured in serum-free medium than in cells cultured in the presence of serum (20, 21). Thus, it appears that AVP represents a pure differentiation factor that is active at the early stages of the myogenic program also in the absence of proliferative signals.

To test this possibility, we investigated whether the effect of AVP on myogenesis could be mediated by the autocrine expression of other known myogenic differentiation factors such as IGFs or IGFRs (42). Indeed, a reciprocal relationship between Myf-5, myogenin, and MEF2 mRNA expression levels was previously reported in L6 cells and human primary myoblasts after treatment with insulin or IGFs (14, 15, 18). Total L6 cellular RNA was collected during a 0.1 μm AVP 7-d time course. Northern blot analysis showed that AVP treatment did not affect the expression of IGF-I, IGF-II, IGF-IR, or IGF-IIR in L6 cells grown in serum-free or in complete medium (data not shown). These results indicate that AVP-dependent L6 cell differentiation is not mediated by the induction of the expression of these myogenic differentiation agonists or their receptors.

The Induction of Myogenin by AVP, But Not MEF2, Requires de Novo Protein Synthesis

We then investigated whether the increase of myogenin and MEF2 mRNA expression levels required protein synthesis. This could indicate whether AVP acts on expression of these genes by transcriptional or posttranscriptional mechanisms. Proliferating L6 cells were treated with cycloheximide (15 μg/ml) 30 min before the addition of 0.1 μm AVP and throughout the 24 h of AVP treatment. As illustrated in Fig. 2A, the induction of myogenin transcripts by AVP was blocked by cycloheximide, whereas the MEF2 transcript was unaffected. These results provide evidence that in L6 cells up-regulation of MEF2 mRNA and protein by AVP does not require de novo protein synthesis, whereas newly synthesized transcription factor/s are necessary for the induction of myogenin expression.

Figure 2.

Role of MEF2 in the Up-Regulation of the Myogenin Promoter Activity Induced by AVP A, L6 cells were treated with cycloheximide (15 μg/ml) 30 min before the addition of 0.1 μm AVP and throughout the AVP treatment. After 24 h, total RNA was collected and analyzed by Northern blot analysis using ³²P-labeled cDNA probes for myogenin and MEF2. Equal loading of samples was verified by hybridization of the blots with a ribosomal 18S RNA. B, Trans-activation of the pMyo184CAT reporter construct transfected into L6 cells treated or not with the indicated concentrations of AVP for 48 h. CAT activity was assayed and normalized to β-gal activity as described in the method section. C, Effect of 48-h AVP treatment on the transactivation activity of wild-type pMyo84-CAT, mMEF2-CAT (which contains a mutation in the MEF2 site of pMyo84-CAT), -E1-CAT (in which the E1-box has been deleted) and mMEF2-E1-CAT (containing a mutation of the MEF2 site combined with the E1-box deletion) transiently transfected into L6 cells.

AVP Up-Regulates the Myogenin Promoter Activity in L6 Cells

The transcriptional activation of MEF2 gene and the presence of a MEF2 binding site in the myogenin promoter (43) raised the possibility of a direct role for MEF2 in the activation of myogenin gene by AVP in L6 cells. We then tested whether AVP treatment could induce the activity of the myogenin promoter using a thymidine kinase-chloramphenicol acetyltransferase (CAT) reporter gene driven by the 184-bp sequence preceding the transcription initiation site of the human myogenin promoter (pMyo184CAT), transiently transfected into L6 cells. The pMyo184CAT reporter plasmid retains the two E-boxes (E1 and E2), the MEF2 site, and the TATA-box that confer the muscle specificity, growth factor responsiveness, and cross-regulation by myogenic bHLH-MRF to the myogenin gene (40). An induction of the pMyo184CAT activity above the basal transcription of the thymidine kinase promoter was detectable at a concentration of 1 nm AVP and increased in a dose-dependent fashion (Fig. 2B). Notably, the active concentrations were in the range of those necessary for the induction of myoblast fusion by AVP (20, 21).

Up-Regulation of the Myogenin Promoter by AVP Occurs through the MEF2 Site

To establish the regulatory element that would mediate transactivation of myogenin promoter by AVP, we transiently transfected into L6 cells a CAT reporter construct carrying the 84-bp sequence containing the E1-box, the MEF2 site, and the TATA-box of the myogenin promoter (43). As for the pMyo184CAT, 0.1 μm AVP increased the wild-type pMyo84CAT activity about 4-fold (Fig. 2C), suggesting the presence of an AVP-responsive element within this myogenin promoter fragment. We next tested the ability of AVP to induce the CAT activity of pMyo84CAT deletion mutants reporter plasmids transiently transfected into L6 cells. The pMyo84(mMEF2) carries a mutation in the MEF2 site which prevents MEF2 binding (43). In the pMyo84(-E1) the E1-box site was deleted (43). In pMyo84(mMEF2-E1), the MEF2 site is mutated and the E1-box is deleted (43). As shown in Fig. 2C, the capacity of AVP to transactivate the 84-bp fragment of the myogenin promoter was abolished by mutation of the MEF2 site [pMyo84(mMEF2)CAT] but not by deletion of the E1-box [pMyo84(-E1)CAT]. Mutation of MEF2 site combined with E1-box deletion [pMyo84(mMEF2-E1)CAT] completely blocked the ability of the promoter to be induced by AVP. Thus, within the context of the 84-bp promoter, an intact MEF2 site is required for the induction of the myogenin promoter activity by AVP in L6 cells.

We then examined the ability of nuclear extracts prepared from L6 cells treated or not with 0.1 μm AVP for 16 h to bind oligonucleotides representing the MEF2 and/or the E1 box site of the myogenin promoter. An EMSA showed that nuclear extracts from AVP treated L6 cells produced MEF2 shifted complexes that were efficiently inhibited by an excess of unlabeled MEF2 oligonucleotides and could be supershifted by an anti-MEF2 antibody (Ab) (Fig. 3A). These AVP-induced protein:DNA complexes appeared specific because they were not present in untreated L6 cells and not formed in the presence of oligonucleotides representing a mutated MEF2 or the E1-box site of the myogenin promoter (Fig. 3A). Interestingly, the addition of oligonucleotides representing the E1 box of the myogenin promoter induced the appearance of specific protein:DNA complexes that were not modified by AVP treatment. Thus, AVP treatment of L6 cells induced the formation of specific nuclear complexes at the MEF2 site of myogenin promoter.

Figure 3.

Specific Nuclear Complexes Are Induced by AVP at the MEF2 Site of Myogenin Promoter A, EMSA analysis of nuclear proteins from L6 cells treated or not with 0.1 μm AVP for 24 h. ³²P-Labeled oligonucleotides representing the MEF2 site, the MEF2 mutated site of the myogenin promoter, the E1-box site, or the E1-box site of the myogenin promoter were used as probes. Hundred-fold molar excess competing oligonucleotides (competitor) or the anti-MEF2 Ab were added to the extracts as indicated. Arrows indicate shifted complexes. B, Expression of different MEF2 family members in L6 cells. RNA was extracted from L6 cells treated or not with 0.1 μm AVP for 24 h and subjected to RT-PCR analysis using MEF2A, MEF2B, MEF2C, and MEF2D specific primers as described in the method section. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression was used as an internal control for the RT-PCR analysis.

MEF2A, C, and D Transcripts, But Not MEF2B, Are Expressed in L6 Cells

It has been shown that MEF2 proteins are not sufficient to induce myogenesis but cooperatively increase the activity of myogenic bHLH transcription factors. The MEF2 family member MEF2C appears specifically involved with myogenin in the regulatory loop resulting in myogenic differentiation of P19 cells, whereas the association between MEF2A and HDAC4 within the nucleus results in the transcriptional repression of MEF2A (31, 38). The full-length MEF2C cDNA or the Ab directed against MEF2 could not distinguish between different MEF2s due to the high homology in their NH2-terminal MADS box domain and in the adjacent motif known as the MEF2 domain (6). Thus, to identify the MEF2 family member involved in the AVP induced myogenesis of L6 cells, we performed a semiquantitative RT-PCR analysis using specific primers for the MEF2A, B, C and D. By RT-PCR, we showed that MEF2A, C, and D transcripts, but not MEF2B transcripts, are expressed in L6 cells (Fig. 3B and data not shown). In addition, we found that MEF2A and MEF2C expression are up-regulated by 24-h AVP treatment, suggesting their potential role in AVP-induced myogenesis of L6 cells.

AVP Treatment Induces Nuclear Export of HDAC4

Recent work on HDAC4 indicated its role in myogenesis linked to MEF2 in that a MEF2 activated reporter resulted repressed by this enzyme (31, 34). We then investigated whether MEF2 could be associated with HDAC4 and whether AVP treatment could affect this association. We first performed immunofluorescence analysis using the anti-MEF2 Ab (which recognizes MEF2A and, to a lesser extent, MEF2C and MEF2D), the anti-HDAC4 Ab, and the antiacetylated histone H3 Ab (to evaluate the status of acetylation of histone H3 in L6 cells). In untreated L6 cells, MEF2 was detectable at low level in the nucleus, whereas HDAC4 showed a diffuse cellular pattern (Fig. 4A). Treatment of L6 cells with AVP increased MEF2 expression at 8 and 24 h treatment (Fig. 4A, and data not shown). At these time points, HDAC4 became easily detectable as a discrete speckled perinuclear pattern in the cytoplasm of L6 cells, although a residual faint nuclear staining was still present (Fig. 4A). Notably, treatment with AVP strongly induced the appearance of acetylated forms of histone H3 in the nucleus of L6 cells that were undetectable in untreated cells (Fig. 4A). The AVP-induced accumulation of MEF2 in the nucleus and the concomitant relocalization of the MEF2 repressor HDAC4 from the nucleus to the cytosol was also detectable by immunoblot analysis of cytosolic and nuclear extracts prepared from L6 cells (Fig. 4B). To demonstrate an in vivo association between MEF2 and HDAC4, coimmunoprecipitation experiments were performed using whole cell lysates prepared from L6 cells treated or not with AVP for 24 h (input, Fig. 4C). Extracts were either immunoprecipitated with an anti-MEF2 Ab and immunoblotted with an anti-HDAC4 Ab, or immunoprecipitated with an anti-HDAC4 Ab and immunoblotted with an anti-MEF2 Ab (Fig. 4C). Interestingly, the addition of oligonucleotides representing the MEF2 site of myogenin promoter to L6 cell lysates resulted critical for the isolation of the endogenous MEF2-HDAC4 complex by immunoprecipitation with either anti-HDAC4 or anti-MEF2 antibodies. In fact, in the absence of oligonucleotides the MEF2-HDAC4 complex was hardly detectable (data not shown), indicating that structural modifications of MEF2 and/or HDAC4 might occur at the MEF2 site of myogenin promoter. Figure 4C shows that MEF2 and HDAC4 coimmunoprecipitated in extracts prepared from control L6 myoblasts. However, MEF2-HDAC4 interaction was strongly decreased in extracts from AVP-treated L6 cells.

Figure 4.

Subcellular Distribution and Dissociation of HDAC4 from MEF2 by AVP Treatment in L6 Cells A, Immunofluorescence analysis of L6 cells treated with 0.1 μm AVP for 24 h, fixed in methanol/acetone, and stained with anti-MEF2, anti-HDAC4, and anti-Ac-H3 antibodies. B and C, Immunoblot analysis of cytoplasmic and nuclear extracts (B) and whole extracts (C, input) from L6 cells treated or not with 0.1 μm AVP for 24 h. Whole cell extracts were coimmunoprecipitated with an anti-MEF2 Ab or an anti-HDAC4 Ab in the presence of oligonucleotides representing the MEF2 site of myogenin promoter, and immunoblotted with anti-HDAC4 and anti-MEF2 antibodies, respectively, as described in the Materials and Methods section.

Inhibition of the Calcium/CaMK Signaling Blocks the AVP Effect on Myogenesis

We next asked whether the activation of the CaMK pathway was involved in regulating the effect of AVP on myogenesis. Indeed, AVP-dependent increase of cytosolic calcium levels (19) might possibly activate CaMK signaling; this in turn could mediate the nuclear export of HDAC4–5 and stimulate myogenesis by relieving their repression on MEF2 transcriptional activity (34, 36, 44). To test this possibility, L6 cells cultured under serum-free conditions were treated with two CaMK inhibitors KN62 and KN93 (45) as sole agents or in combination with AVP. Treatment with KN62 and KN93 as sole agents did not affect the expression of MEF2 and myogenin or L6 cells differentiation as shown by immunoblot, immunofluorescence, and morphological analyses (Fig. 5, A–C, and data not shown). No fusion was detectable in control culture, whereas AVP stimulated myoblast fusion as shown by the appearance of large myotubes that were not present in cells treated with 5 μm KN62 or KN93 (Fig. 5C). The inactive analog KN92 at concentrations of 5 μm did not inhibit the AVP effect on myogenesis. Both KN62 and KN93 were found able to decrease, in a dose-dependent manner the AVP induced up-regulation of MEF2. Remarkably, the AVP-induced nuclear export of HDAC4 (data not shown), myogenin expression, and L6 cells differentiation were significantly inhibited by 5 μm CaMK inhibitors KN62 or KN93, but not by the inactive analog KN92 (Fig. 5, A–C). Together, these results suggest that the activation of a CaMK signaling pathway by AVP is involved in the transcriptional activation of MEF2 and in the myogenic action of AVP.

Figure 5.

Effect of CaMK Inhibitors on AVP-Induced MEF2 and Myogenin Expression and L6 Cell Differentiation L6 cells were plated in DMEM + 10% FCS. After 24 h, cells were shifted in serum-free medium and treated with AVP (0.1 μm) in the presence or in the absence of CaMK inhibitors (KN62 and KN93), or the inactive analog KN92 (Calbiochem, La Jolla, CA) at the indicated concentrations. Untreated cells are shown as controls (C). A, Immunoblot analysis of MEF2 and myogenin expression in whole cell extracts prepared from L6 cells treated as indicated for 24 h. Note that both KN62 and KN93, but not the inactive analog KN92, at the concentration of 5 μm, strongly inhibited AVP induction of MEF2 and myogenin expression. Expression of α-tubulin was used to verify the equal loading of samples. B, A KN62 dose-dependent assay indicates a good correlation between the ability of this CaMK inhibitor to block AVP-induced myogenin expression and L6 cells differentiation showed in panel C. C, After 6 d treatment, L6 cells morphological differentiation was evaluated by Wright-Giemsa staining (magnification, ×10).

DISCUSSION

Although it is accepted that several agonists can activate myogenesis through the induction of muscle-specific gene expression, the underlying mechanism is still unclear. We have previously shown that AVP treatment induces myogenesis in cell lines and in primary cultures of mouse satellite cells (20, 21). In the present study, we investigated the molecular mechanism involved in the myogenic action of AVP. Our results demonstrate that myogenic bHLH-MRF proteins in the nucleus can be the direct targets for intracellular signal transduction pathways activated by AVP through V1 receptors at the myoblast cell surface. We show that in L6 cells cultured in serum-free medium, AVP treatment caused a sequential induction of Myf-5 and myogenin mRNA expression that are not mediated by an autocrine induction of other known myogenic differentiation factors such as IGFs or IGFRs (42, 46). These results are consistent with previous reports indicating that during skeletal muscle development Myf-5: 1) is the first of the myogenic bHLH-MRF to be expressed and rapidly extinguished; 2) is responsible for the specification and maintenance of myoblast identity; and 3) acts upstream of myogenin, which has a crucial role in regulating the terminal differentiation of myoblasts (47, 48). Thus, AVP seems to act as a pure differentiation factor and its action on L6 cells myogenesis offers a significant in vitro model to study the mechanism of MRF activation.

Several lines of evidence are consistent with the notion that bHLH-MRF mediate muscle gene transcription and transactivation when bound to the consensus sequences CANNTG (E-box), present in the control regions of skeletal muscle genes (1, 49, 50). However, other transcriptional activators are critical for myogenesis. Among these are the four members of the MEF2 family of MADS (MCM1 agamous deficiens, and serum response factors)-box transcription factors (in vertebrates MEF2A-D) (2, 3, 7). These factors recognize specific DNA elements (MEF2 sites) present in the promoter and enhancer regions of numerous muscle-specific genes (2, 7). MEF2s can also interact with myogenic and neurogenic bHLH factors and modulate their transcriptional activities (2, 3, 7).

In this study, we show that MEF2A and C expression are induced by AVP during the myoblast to myotube transition in L6 cells cultured in the absence of serum. In addition, MEF2 induction by AVP appears concomitant to myogenin expression. However, treatment with cycloheximide blocks the AVP induction of myogenin mRNA but not of MEF2, suggesting that the induction of MEF2 by AVP is independent on protein synthesis and that MEF2 protein might control myogenin expression. Notably, AVP activates a MEF2 binding site that regulates the transcriptional activity of the myogenin promoter (43). Together, these findings outline a possible mechanism for AVP-induced myoblast differentiation, whereby myogenic bHLH-MRF and MEF2 can amplify and maintain one another’s expression in an autoregulatory loop that positively influences myogenesis. Interestingly, by EMSAs we also found that specific nuclear complexes can be induced by AVP at the MEF2 site of myogenin promoter.

Recent results obtained in transfected HeLa cells, skeletal myoblasts, and in cardiomyocytes indicate that MEF2 interact directly with HDAC 4-5-7 (31–36). HDAC4, HDAC5, and HDAC7 are referred as class-II HDACs. Class II HDACs, differently than the class I HDACs (HDAC 1-2-3), shuttle between the nucleus and the cytosol. The association of MEF2A and MEF2C products with HATs and HDACs activities in skeletal and cardiac muscle highlights the important role of these enzymatic activities in myogenesis. Strikingly, we found that in L6 cells AVP treatment induces MEF2A and MEF2C expression and cytosolic compartmentalization of HDAC4, events that have been found associated with the transcriptional activation of MEF2 within the nucleus (31).

It has been shown that the activation of CaMK signaling is sufficient to mediate the cytosolic compartmentalization of class II HDACs that relieves the HDAC4-mediated transcriptional repression of MEF2 and stimulate myogenesis (34, 36, 44). Notably, we found that KN62 and KN93, two inhibitors of CaMK signaling (45), inhibited the ability of AVP to induce the nuclear export of HDAC4 (data not shown), myogenin expressions and L6 cells myogenic differentiation. These evidences reinforce the view that AVP stimulates myogenesis by enhancing the transcriptional activity of MEF2 at critical DNA binding sites present on target genes such as myogenin.

As to the mechanism of AVP-dependent CaMK activation, it is worth noting that cytosolic calcium concentration rapidly increases in AVP-stimulated myogenic cells (19). Indeed, AVP acts through V1 receptors coupled to phospholipase C signaling (19, 22). More recently, it has also been shown that AVP treatment causes phospholipase D activation and a reduction of basal cAMP concentration by activation of a specific cAMP-phosphodiesterase that plays an important role in the effect of AVP upon L6 cell differentiation (23). The contribution of each (and possibly other) signaling pathway to the myogenic effect of AVP, as well as the molecular mechanisms involved in the transmission of the AVP signal between the second messenger level and the nuclear regulatory factor level is currently being investigated in our laboratory. Experiments are also needed to clarify whether AVP induces p300 activity and/or specific posttranslational mechanisms [e.g. calcium-dependent signals, and MAPK signaling, and their role in myogenesis (15, 17, 32, 36, 44, 46)]. In summary, the results of this study suggest that AVP represents a myogenic differentiation factor of significant importance because it acts by inducing and activating MEF2 transcription, which in turn regulates myogenin expression, without affecting the expression of myogenic-differentiation-inducers, such as IGFs. IGFs stimulate both proliferation and differentiation of muscle cells, whereas AVP does not affect myoblast proliferation (20, 21, 46) but promotes and enhances the myogenic differentiation program. Notably, high levels of immunoreactive AVP, which declined as gestational age increased, were measured in human embryonic skeletal muscle extracts, suggesting a physiological relevance of the effect of AVP upon myogenic cell differentiation and development (51). Thus, it is intriguing to speculate that during embryonic development AVP or an AVP-related peptide controls myogenesis by stimulating nuclear export of HDACs, resulting in the activation of MEF2-dependent genes. AVP might also represent an interesting candidate for novel therapeutical approaches to muscle diseases based on the phenotypic conversion of embryonic stem cells.

MATERIALS AND METHODS

Cell Culture

Subcloning and characterization of L6 rat myogenic cells has been previously reported (19). Cells of subclone C5 (L6-C5) were cultured in DMEM supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin (DMEM), and 10% heat-inactivated FBS. Twenty-four hours after plating, cultures were extensively washed with DMEM, shifted to serum-free medium consisting of DMEM supplemented with 1% fatty acid-free BSA (Roche Molecular Biochemicals, Mannheim, Germany) (21), and treated with synthetic AVP (Sigma, St. Louis, MO) at different time points, as detailed in the text.

RNA Precipitation and Northern Blot Analysis

Total RNA was isolated from the cells by the guanidinium isothiocyanate-CsCl procedure (52) or using the Tri Reagent extraction (Sigma). Equal amounts of total RNA (30 μg) were separated by electrophoresis in 0.66 m formaldehyde-1.2% agarose slab gel, transferred to Nytran membranes (Schleicher and Shuell, Keene, NH) by capillary blotting and cross-linked to the blots by UV irradiation (52). The cDNA probes were radiolabeled with [α-³²P]deoxy-CTP (3,000 Ci/mmol; NEN Life Science Products, Boston, MA) via random priming using the kit and protocols supplied by Life Technologies, Inc. (Gaithersburg, MD). The QuickHyb (Stratagene, La Jolla, CA) was used as prehybridization and hybridization buffer. The murine myogenin probe (1,500 bp), and the human Myf-5 probe (1,250 bp) were the EcoRI restriction fragments of the respective cDNA clones (53, 54). The MEF2C expression vector was created by insertion of a full-length MEF2C cDNA into HindIII-XbaI site of cytomegalovirus promoter-directed expression vector pcDNAI (Invitrogen, Carlsbad, CA) (55). Equal loading of samples was verified by hybridization of the blots with an 18S rRNA probe (American Type Culture Collection, Manassas, VA).

RT-PCR

Two micrograms of total RNA prepared as described above were employed in each RT-PCR assay. Experimental procedures were performed using the Titan one tube RT-PCR system (Roche Molecular Biochemicals). Specific primers and amplification conditions for MEF2A, MEF2B, MEF2C, and MEF2D were previously described (56, 57). The following oligonucleotides were used to detect glyceraldehyde 3-phosphate dehydrogenase transcripts (used as an internal control): sense 5′ CGGGA AGCTT GTGAT CAATGG 3′ and antisense 5′ GGCAG TGATG GCATG GACTG 3′.

Transient Transfection of L6 Cells and CAT Assay

L6 cells were plated in 6-well 35-mm culture dishes at a density of 2.5 × 10⁵ cells/well. Twenty-four hours later, cells were transfected with 1.5 μg of expression vectors driven by the myogenin promoter fragments: pMyo84(-E1)CAT, pMyo84CAT, pMyo84(mMEF2)CAT, pMyo84(mMEF2-E1)CAT and pMyo84(mMEF2-E1)CAT (43) using the Lipofectamine reagent (Life Technologies, Inc.). Plasmid encoding β-galactosidase (βgal) (Rous sarcoma virus-βgal) (0.3 μg) was cotransfected to monitor transfection efficiency and for normalization of reactions. Twenty-four hours after transfections, cells were transferred to a serum-free medium (DMEM + 1% BSA) and treated with AVP for 48 h. CAT activity was determined as described (52). Reporter gene expression was calculated in arbitrary units, relative to β-gal expression.

EMSAs

Nuclear extracts (10 μg protein) from L6 cells treated with 0.1 μm AVP for 24 h and nontreated controls were prepared as described (56) and incubated with 5 μg poly(deoxyinosinic: deoxycytidylic acid) (Amersham Pharmacia Biotech, Arlington Heights, IL) for 20 min at room temperature in 10 mm Tris-HCl (pH 7.5), 50 mm KCl, 5 mm MgCl₂, 1 mm dithiothreitol (DTT), 1 mm EDTA, and 15% glycerol. ³²P-End-labeled oligonucleotides (0.5–1 ng) were added in the presence or in the absence of 100-fold excess of cold competitor oligonucleotides for 30 min. Samples were loaded onto a 5% polyacrylamide gel in 0.5× TBE (45 mm Tris base, 45 mm boric acid, and 4 mm EDTA) prerun for 1 h at room temperature. Gels were then fixed, dried in vacuum desiccator and exposed for autoradiography using Kodak (Rochester, NY) X-Omat AR films. Oligonucleotides representing the MEF2 and E1-box sites of myogenin promoter were used as probes and competitors as previously described (43, 56). Oligonucleotides were purified by electrophoresis on polyacrylamide gels, followed by electroelution and annealing in equimolar amounts as previously described (56).

Immunofluorescence and Immunoblot Analysis

Cells were fixed and permeabilized in cold methanol/acetone and incubated with anti-MEF2, anti-HDAC4 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and antiacetylated histone H3 antibodies (Upstate Biotechnology, Inc., Charlottesville, VA) as described (58). Immunofluorescence was detected using a Carl Zeiss (Jena, Germany) Axioplan fluoromicroscope. Immunoblot analysis was performed on total cell homogenates or nuclear and cytosolic extracts prepared as described (20, 56) using the antimyogenin Ab F5D (kindly provided by Dr. W. E. Wright, University of Texas, Dallas, TX); anti-MEF2, anti-HDAC4 (Santa Cruz Biotechnology, Inc.). Immunoreactivity was determined using the electrochemiluminescence method (Amersham Pharmacia Biotech).

Coimmunoprecipitations

Cells were treated or not with AVP for 24 h, rinsed twice with PBS, and lysed in 10 mm HEPES/KOH (pH 7.9), 400 mm NaCl, 1.5 mm MgCl₂, 0.5 mm DTT, 0.1 mm EGTA, and 5% glycerol. Samples were dialyzed overnight in 20 mm HEPES/KOH (pH 7.9), 75 mm NaCl, 0.5 mm DTT, 0.1 mm EDTA, and 20% glycerol. One milligram of whole cell extracts was used for each immunoprecipitation. Extracts were incubated or not with annealed double-strand oligonucleotides representing the MEF2 site of myogenin on ice for 1 h as described (59). After incubation at 4 C for 1 h with the primary Ab (anti-HDAC4 or anti-MEF2), protein G-Agarose beads (Roche Molecular Biochemicals) were added, and the samples were rocked overnight at 4 C. After washing with ice-cold 10 mm Tris-HCl (pH 7.5), 50 mm KCl, 5 mm MgCl₂, 1 mm DTT, 1 mm EDTA, and 15% glycerol immunoprecipitants were resolved on SDS-polyacrylamide gels followed by immunoblot analysis performed as described above.

Acknowledgments

This work was partially funded by Ministero dell’ Università e della Ricerca Scientifica e Tecnologica (to M.M., S.A., and C.N.), Associazione Italiana per la Ricerca sul Cancro (to C.N.), European Economic Community (to C.N.), Telethon and Consiglio Nazionale delle Richerche (to A.M.). B.M.S. is supported by a research collaboration contract of the University of Rome “La Sapienza.”

We are grateful to L. Benedetti for excellent technical help, to E. Olson for reagents, and to S. Ferrari, G. Cossu, N. Rosenthal, and B.M. Zani for reagents and discussion. This work is dedicated to the memory of Franco Tató.

Abbreviations:

 
• Ab,

  Antibody;

 
• AVP,

  Arg⁸-vasopressin;

 
• bHLH-MRFs,

  basic helix-loop-helix muscle regulatory factors;

 
• CaMK,

  calcium-calmodulin kinase;

 
• CAT,

  chloramphenicol acetyltransferase;

 
• DTT,

  dithiothreitol;

 
• FGF,

  fibroblast growth factor;

 
• βgal,

  β-galactosidase;

 
• HAT,

  histone acetyltransferases;

 
• HDAC,

  histone deacetylase;

 
• IGFR,

  IGF receptor;

 
• MEF2,

  myocyte enhancer factor 2.