RBM24 is a direct target of MyoD and is required for myogenesis during Xenopus early development

MECHANISMS OF DEVELOPMENT

1 2 7 ( 2 0 1 0 ) 2 8 1 –2 9 1

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/modo

The RNA-binding protein Seb4/RBM24 is a direct target of MyoD and is required for myogenesis during Xenopus early development Hong-Yan Li

a,b

, Audrey Bourdelas a, Cle´mence Carron a, De-Li Shi

a,*

a

Groupe de Biologie Expe´rimentale, Laboratoire de Biologie du De´veloppement, CNRS UMR 7622, Universite´ Pierre et Marie Curie, 9 quai Saint-Bernard, 75005 Paris, France b Key Laboratory of Marine Genetics and Breeding, Ocean University of China, Ministry of Education, Qingdao 266003, China

A R T I C L E I N F O

A B S T R A C T

Article history:

RNA-binding proteins play an important role to post-transcriptionally regulate gene

Received 2 October 2009

expression. During early development they exhibit temporally and spatially regulated

Received in revised form

expression pattern. The expression of Xenopus laevis Seb4 gene, also known as RBM24 in

17 March 2010

other vertebrates, is restricted to the lateral and ventral mesoderm during gastrulation

Accepted 19 March 2010

and then localized to the somitic mesoderm, in a similar pattern as XMyoD gene. Using a

Available online 23 March 2010

hormone-inducible form of MyoD to identify potential direct MyoD target genes, we find that Seb4 expression is directly regulated by MyoD at the gastrula stage. We further show

Keywords: RNA-binding protein Seb4

that a 0.65 kb X. tropicalis RBM24 regulatory region contains multiple E boxes (CANNTG), which are potential binding sites for MyoD and other bHLH proteins. By injecting a RBM24 reporter construct into the animal pole of X. laevis embryos, we find that this reporter gene is indeed specifically activated by MyoD and repressed by a dominant negative

RBM24

MyoD mutant. Knockdown of Seb4 produces similar effects as those obtained by the dom-

E box Myogenesis MyoD

inant negative MyoD mutant, indicating that it is required for the expression of myogenic genes and myogenesis in the embryo. In cultured ectodermal explants, although overexpression of Seb4 has no obvious effect on myogenesis, knockdown of Seb4 inhibits the

Xenopus

expression of myogenic genes and myogenesis induced by MyoD. These results reveal that Seb4 is a target of MyoD during myogenesis and is required for myogenic gene expression.  2010 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

Skeletal muscles of the vertebrate body and limbs are derived from progenitor cells present in the somites, which are formed at regular intervals in an anterior to posterior sequence from the posterior unsegmented presomitic mesoderm (reviewed in McGrew and Pourquie´, 1998). The specification of skeletal muscle fate involves temporally and spatially regulated expression and function of several bHLH myogenic regulatory genes including Myf5, MyoD, Myogenin

and Mrf4 (Buckingham, 2006). In Xenopus laevis embryo, XMyf5 and XMyoD are earlier markers of muscle development and are expressed in the precursors of muscle mesoderm, they act together to regulate the earliest events of myogenesis (Hopwood et al., 1989, 1991, 1992; Frank and Harland, 1991). When expressed in a wide variety of cell types, different myogenic factors can convert these cells to skeletal muscle, albeit with varying efficiency (Weintraub et al., 1989). These suggest that myogenic factors are differentially involved in the specification and differentiation of skeletal muscle. Although the

* Corresponding author. Tel.: +33 1 44272772; fax: +33 1 44273451. E-mail address: [email protected] (D.-L. Shi). 0925-4773/$ - see front matter  2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mod.2010.03.002

282

MECHANISMS OF DEVELOPMENT 127 (2010)

expression pattern and general functions of each myogenic factor are clearly distinct and their activation leads to the expression of a wide range of genes with a wide range of biological function, such as transcription factors, muscle structural genes and cell cycle regulators (reviewed in Berkes and Tapscott, 2005), the specific target genes of each myogenic factor are not known (Chanoine et al., 2004). In vitro, the basic regions of MyoD act as sequence-specific DNA-binding domains that recognize a binding site with a simple core consensus sequence of CANNTG, termed E box, which is present in the promoters and enhancers of muscle-specific genes (Blackwell and Weintraub, 1990). RNA-binding proteins play an important role in gene regulation in different species (Burd and Dreyfuss, 1994; Perrone-Bizzozero and Bolognani, 2002). One of the largest families of RNA-binding proteins is characterized by the presence of RNA recognition motifs. The RNA-binding specificity and biological activity significantly differ among different members of this family, which may be regulated through a cooperation of different functional modules within each protein (Lunde et al., 2007). These proteins are in general all involved in RNA metabolism, however, the tissue expression pattern of different RNA-binding proteins varies considerably, implying that they may be involved in distinct developmental process. Both in X. laevis and mouse embryos, distinct RNA-binding proteins were shown to be specifically expressed in heart, somitic mesoderm and nervous system, and play an important role in the specification of these tissues (Fetka et al., 2000; Gerber et al., 2002; Boy et al., 2004; Mientjes et al., 2004; Huot et al., 2005; Spagnoli and Brivanlou, 2006). This suggests that they are implicated in early development in a tissue-specific manner. Here we report the regulation and function of Seb4 during early myogenesis. Seb4 protein is also known as RBM24 in other vertebrates. Except the characteristic RNA recognition motif, it differs from other RNA-binding proteins, in particular the closely related RBM38, by the presence of two alanine-rich regions at the carboxyl-terminal half. X. laevis Seb4 exhibits strong overlapping expression pattern with XMyoD at various stages of development, however, its regulation and activity during myogenesis is not clear. We show that Seb4 may be a direct target gene of MyoD protein. Its expression is induced by MyoD and is repressed by a dominant negative MyoD mutant. In addition, myogenic factors specifically activate a RBM24 promoter reporter construct. Although forced expression of Seb4 alone is not sufficient to activate myogenic program, it is required for the expression of a subset of mesodermal genes and myogenesis. Furthermore, we find that Seb4 acts downstream of MyoD in myogenesis. These results suggest that Seb4 is likely implicated in maintaining myogenic gene expression at different steps of myogenesis.

2.

Results

2.1.

Direct regulation of Seb4 by MyoD during gastrulation

At the initial step of myogenesis different genes are expressed in the presumptive somitic mesoderm in a very similar or identical pattern as myogenic genes such as Myf5 and MyoD. Although it has been established that a genetic interaction

2 8 1 –2 9 1

between Myf5 and MyoD occurs at this stage, whether and how they interact with other genes remains largely unclear. In particular, the specific target genes for each myogenic factor are not known. To identify a potential interaction between myogenic factors and other genes expressed in the presomitic mesoderm during X. laevis early development, we first overexpressed XMyoD in ectodermal cells by injection of the synthetic mRNA (500 pg) into the animal pole region at two-cell stage. Ectodermal explants were dissected and cultured to early gastrula stage. RT-PCR was used to analyze the expression level of different genes that exhibit overlapping expression pattern with XMyoD at different stages of myogenesis. Among a variety of genes analyzed, the expression level of Seb4, encoding a RNA-binding protein (Fetka et al., 2000), was strongly upregulated (not shown). The expression pattern of Seb4 strongly overlaps that of XMyoD at different stages (Fig. 1A and B; Fetka et al., 2000), raising the possibility that they may participate in the same myogenic process. This in vitro observation was further confirmed in the embryo by in situ hybridization. Wild-type XMyoD mRNA (500 pg) or a dominant negative mouse MyoD mutant (MyoD-ENR) mRNA (500 pg) was coinjected with Lac Z mRNA (1 ng) at four-cell stage into one of the two dorsal blastomeres in the equatorial region. The result clearly showed that overexpression of XMyoD strongly induced Seb4 expression in a cell autonomous manner (Fig. 1C). Conversely, inhibition of MyoD function by the dominant negative mutant strongly blocked Seb4 expression at the site of injection (Fig. 1D). These results thus indicate that Seb4 expression is regulated by MyoD. To assay whether Seb4 may be directly regulated by MyoD, we used a hormone-inducible form of MyoD in which mouse MyoD was fused with the ligand-binding domain of the human glucocorticoid receptor (Wittenberger et al., 1999). The activity of this fusion protein (MyoD-GR) can be induced at appropriate stages by incubation of explants or whole embryos in dexamethasone (DEX). We first examined whether Seb4 expression could be induced by MyoD-GR in the embryo. Lac Z mRNA was injected as a cell lineage tracer alone or coinjected with MyoD-GR mRNA (200 pg) into one of the two dorsal blastomeres in the equatorial region at four-cell stage. When injected embryos reached stage 10.5 (early gastrula) and 11.5 (mid-gastrula), they were treated with 10 lM DEX for 1.5 h. In situ hybridization analysis showed that Seb4 expression was strongly induced in embryos injected with MyoD-GR mRNA and treated with DEX (Fig. 1E–H). We next confirmed this result by an in vitro analysis in ectodermal explants. Synthetic mRNA corresponding to MyoD-GR was injected at twocell stage in the animal pole region and ectodermal explants were dissected at stage 10.5. They were then treated with DEX for 1.5 h in the absence or presence of 10 lg/ml of the protein synthesis inhibitor, cycloheximide (CHX). As in the experiment using whole embryos, the short-time induction by DEX and the presence of CHX should allow only the expression of MyoD direct target gene. Indeed, RT-PCR analysis indicated that Seb4 expression was strongly induced by MyoD-GR after 1.5 h of induction in DEX, either in the absence or presence of CHX (Fig. 1I), while only background expression level of other mesodermal markers like Xbra and Xwnt8 was detected in these explants. We further confirmed that CHX treatment is actually inhibiting protein synthesis. Synthetic

MECHANISMS OF DEVELOPMENT 127 (2010)

2 8 1 –2 9 1

283

b Fig. 1 – (A–H) MyoD regulates the expression of Seb4 in whole embryos at different stages analyzed by in situ hybridization. (A and B) Expression of Seb4 (A) and XMyoD (B) in the lateral mesoderm of a stage 11 gastrula. (C) Overexpression of XMyoD induces ectopic Seb4 expression on the injected side of an early gastrula. (D) A dominant negative MyoD mutant, MyoD-ENR, blocks Seb4 expression. (E) Injection of Lac Z mRNA and treatment with DEX do not induce the expression of Seb4 in the early gastrula. (F) An early gastrula previously injected on one side with MyoD-GR and Lac Z mRNAs and treated with DEX for 1.5 h shows ectopic expression of Seb4. (G) Injection of Lac Z mRNA and treatment with DEX for 1.5 h do not induce ectopic Seb4 expression at stage 12. (H) Injection of MyoD-GR and Lac Z mRNAs and treatment with DEX for 1.5 h strongly induce ectopic Seb4 expression at stage 12. (I) Induction of Seb4 expression in ectodermal explants by MyoD-GR in the absence of protein synthesis. Both uninjected and injected explants were treated with DEX or CHX, or both, for 1.5 h. The expression of Seb4, Xbra and Xwnt8 was analyzed by RT-PCR. Seb4 expression is induced in explants injected with MyoD-GR mRNA and treated with DEX in the presence of CHX, while the expression of Xbra and Xwnt8 is not induced. ODC (ornithine decarboxylase) was used as a loading control. (J) Effect of CHX on the translation of injected Seb4MT mRNA.

4 h with CHX. Western blot analysis indicated that the presence of CHX strongly inhibited the translation of Seb4MT protein (Fig. 1J), suggesting that CHX treatment is actually inhibiting protein synthesis. These observations, together with the overlapping expression pattern between Seb4 and XMyoD at different stages, strongly argue for a direct regulation of Seb4 by the myogenic factor MyoD.

2.2.

mRNA (500 pg) encoding myc-tagged Seb4 protein (Seb4MT) was injected in four-cell stage embryos. Injected embryos were allowed to develop for 1 h, they were then treated for

MyoD activates RBM24 promoter

To directly examine the regulation of Seb4 gene by MyoD, we analyzed the X. tropicalis RBM24 regulatory sequence. At the protein level, X. laevis Seb4 and X. tropicalis RBM24 exhibit 92% overall identity, and the coding nucleotide sequences show 93% identity (not shown). We found that a 0.65 kb fragment containing the putative regulatory region of RBM24 contained multiple E boxes, which are potential binding sites for MyoD and other bHLH proteins. Six core consensus sequences of CANNTG, named E1–E6, can be identified in the sequence upstream of the translation start site (Fig. 2A). To see whether MyoD indeed activates the RBM24 promoter, we fused the 0.65 kb promoter sequence to a luciferase reporter gene (RBM24-luc). This reporter DNA was then either injected alone or coinjected with synthetic mRNA corresponding to XMyoD (500 pg), mouse MyoD (500 pg), XMyf5 (500 pg), mouse Myf5 (500 pg), XMyogenin (500 pg) and VegT (100 pg) in the animal pole region at two-cell stage. Luciferase assay was performed using ectodermal explants dissected at early gastrula stage. The result showed clearly that XMyoD, and also XMyogenin, significantly increased the luciferase activity,

284

MECHANISMS OF DEVELOPMENT 127 (2010)

2 8 1 –2 9 1

Fig. 2 – (A) Nucleotide sequence of the 0.65 kb fragment of the Xenopus tropicalis RBM24 regulatory region. The first 24 nucleotides of the RBM24 coding sequence with its deduced amino acids and 633 nucleotides of 5 0 RBM24 upstream sequence are shown. The core consensus E boxes (CANNTG) are over-lined and identified above the sequence, as E1, E2, E3, E4, E5 and E6. The arrow indicates a possible transcription start site. (B) Activation of the RBM24 promoter by different bHLH myogenic factors and repression by MyoD-ENR. VegT has no effect on the reporter activity. The luciferase reporter assay was performed in triplicates. Error bars indicate standard deviation.

while X. laevis and mouse Myf5 moderately increased the luciferase activity. Injection of VegT RNA had no effect, indi-

cating that the activation of RBM24 promoter by myogenic factors should be specific and direct. Furthermore, injection

MECHANISMS OF DEVELOPMENT 127 (2010)

2 8 1 –2 9 1

285

b Fig. 3 – Knockdown of Seb4 inhibits myogenic gene expression at the gastrula stage. Embryos at four-cell stage were injected in the dorso-lateral region with CoMO, Seb4MO or MyoD-ENR mRNA. In situ hybridization was performed at the early gastrula stage using probes indicated on the right side. Injection of CoMO (A) does not affect Xbra expression, while injection of Seb4MO (B) or MyoD-ENR mRNA (C) similarly inhibits, but does not completely block, Xbra expression. (D) Injection of CoMO has no effect on XMyoD expression, while both Seb4MO (E) and MyoD-ENR (F) inhibit XMyoD expression. (G) XMyf5 expression is not affected by CoMO, but Seb4MO (H) and MyoD-ENR (I) inhibit its expression. Xwnt8 expression is not affected by CoMO (J), Seb4MO (K) and MyoD-ENR (L). Similar chordin expression pattern in embryos injected with CoMO (M), Seb4MO (N) and MyoD-ENR mRNA (O). (P) Inhibition of the Myf5 promoter reporter (XTM1-luc) activity by Seb4MO and MyoD-ENR in stage 10.5 early gastrula. The luciferase reporter assay was performed in triplicates. Error bars indicate standard deviation.

gene is directly regulated by, and may represent a potential target gene of MyoD at the initial stage of myogenesis.

2.3. Seb4 is required for myogenic gene expression and myogenesis

of MyoD-ENR mRNA (500 pg) strongly blocked the basal activity of luciferase (Fig. 2B). This is consistent with a previous study showing that MyoD and Myogenin may have a common set of promoters (Cao et al., 2006). However, the activation of RBM24 reporter by XMyoD was relatively low. It was postulated that XMyoD is under negative control in X. laevis embryo, while mouse MyoD is not regulated by this negative signal, due to a slight sequence difference (Rupp et al., 1994). We thus tested the activity of mouse MyoD to activate the RBM24 promoter. Consistent with previous study, mouse MyoD indeed more strongly activated the RBM24 promoter than XMyoD (Fig. 2B). This result indicated that Seb4/RBM24

The function of Seb4 during early development is presently not clear despite its restricted expression in the somitic mesoderm. Its direct regulation by MyoD led us to analyze how it is involved in myogenic gene expression and myogenesis using morpholino knockdown approach. We also compared the effects of Seb4 knockdown with those produced by overexpression of MyoD-ENR. Seb4 morpholino oligonucleotide (Seb4MO and MO2) or MyoD-ENR mRNA was injected into the two dorsal blastomeres at four-cell stage, in situ hybridization using a panel of mesoderm markers was performed at early gastrula stage. Since Seb4MO and MO2 were similarly effective (see Fig. 5B), we used Seb4MO in most of our experiments. The result showed that injection of 10 ng control morpholino (CoMO) had no effect on the expression of Xbra (n = 37), XMyoD (n = 35), XMyf5 (n = 35), Xwnt8 (n = 38) and chordin (n = 31) in all injected embryos (Fig. 3A, D, G, J and M), while 78% of embryos (n = 34) injected with 10 ng Seb4MO and 85% of embryos (n = 32) injected with 500 pg MyoD-ENR mRNA had Xbra expression partially inhibited at the sites of injection (Fig. 3B and C). The expression of XMyoD was inhibited in 87% of embryos (n = 63) injected with Seb4MO and in 92% of embryos (n = 69) injected with MyoD-ENR mRNA (Fig. 3E and F). A similar inhibition of XMyf5 expression was observed in 92% of embryos (n = 72) injected with Seb4MO and in 97% of embryos (n = 81) injected with MyoD-ENR mRNA (Fig. 3H and I). The expression pattern of Xwnt8 (Fig. 3K and L) and chordin (Fig. 3N and O) was not significantly affected by Seb4MO or MyoD-ENR mRNA in all injected embryos. A BLAST search of the NCBI nucleotide

286

MECHANISMS OF DEVELOPMENT 127 (2010)

Fig. 4 – Knockdown of Seb4 inhibits myogenesis. In situ hybridization using indicated probes was performed on embryos injected with CoMO or Seb4MO and cultured to stage 30. (A) Myosin light chain (MLC) expression in the somitic mesoderm of a CoMO-injected embryo. (B) A Seb4MO-injected embryo with reduced and disorganized somitic mesoderm. (C) XMyf5 expression in a CoMO-injected embryo. (D) Injection of Seb4MO inhibits XMyf5 expression in the anterior and trunk region. (E) XMyoD expression in a CoMO-injected embryo. (F) Injection of Seb4MO inhibits XMyoD expression. (G) Tbx6 expression in the tail-bud of a CoMO-injected embryo. (H) Injection of Seb4MO reduces the expression level of Tbx6 in the tail-bud. (I) HoxB9 expression in the spinal cord of a CoMO-injected embryo. (J) Injection of Seb4MO does not significantly affect HoxB9 expression.

data base using X. laevis Seb4 coding region as a query sequence indicated that there are at least two copies of Seb4 gene, with one nucleotide difference in the Seb4MO targeted sequence (not shown). Although this difference should not significantly reduce the efficiency of Seb4MO, the maternal expression of Seb4 (Fetka et al., 2000) could account for, at least to some extent, the incomplete blockade of XMyoD and XMyf5 expression in Seb4 morphants. In order to quantitatively analyze the inhibitory effect of Seb4MO and MyoD-ENR on myogenic gene expression, we injected the X. tropicalis Myf5 promoter reporter construct (XTM1-luc) alone or coinjected with Seb4MO or MyoD-ENR mRNA in the equatorial region at four-cell stage. The XTM1 promoter sequence was shown to contain the Myf5 regula-

2 8 1 –2 9 1

tory region necessary for the correct expression of Myf5 (Polli and Amaya, 2002). When luciferase assay was performed at the early gastrula stage, we found that the activation of XTM1-luc was strongly inhibited (Fig. 3P). This result suggests that inhibition of Seb4 and MyoD function similarly affects the expression of myogenic genes, and that Seb4 is required for the expression of myogenic genes during gastrulation. When injected embryos were allowed to develop until tailbud stage, they exhibited short and bent axis, likely caused by gastrulation defects following the inhibition of Xbra expression, which also plays a role in cell movements during gastrulation (Tada and Smith, 2000). Analysis of different tissue-specific markers indicated that the expression of XMyf5, XMyoD and myosin light chain (MLC) was down-regulated (Fig. 4A–F), as well as the presomitic marker Tbx6 (Fig. 4G and H), while the expression of the spinal cord marker HoxB9 did not seem to be affected (Fig. 4I and J). This shows that Seb4 function is required for myogenesis. The effectiveness and specificity of Seb4MO was assayed by Western blotting and by rescue experiment. First, synthetic mRNA corresponding to Seb4MT or Seb4MT with 4 nucleotide substitutions in the morpholino targeting sequence (mutSeb4MT), without change of the amino acid sequence, was injected alone or coinjected with CoMO (10 ng) or Seb4MO (10 ng) at two-cell stage. Western blot performed at early gastrula stage using whole embryo extract clearly showed that Seb4MO specifically blocked the translation of Seb4MT (Fig. 5A, lane 2), while it had no effect on the translation of mutSeb4MT (Fig. 5A, lane 4). In addition, Seb4MO did not affect the translation of an unrelated protein Xdsh-myc (Sokol, 1996), which was coinjected and used as a loading control (Fig. 5A, lanes 1–4). This indicates that the targeting effect of Seb4MO should be specific. Thus, we used mutSeb4MT to rescue the effect of Seb4MO on Xbra and XMyf5 expression. When Seb4MO (10 ng) was coinjected with mutSeb4MT mRNA (100 pg) and in situ hybridization was performed at the early gastrula stage, the result showed that it indeed significantly rescued the normal expression pattern of Xbra in 72% of embryos (n = 41) and XMyf5 in 67% of embryos (n = 47) at the early gastrula stage (Fig. 5B). Furthermore, when MO2 (10 ng) was coinjected with Lac Z mRNA, a similar inhibition of Xbra and XMyf5 expression was observed (Fig. 5B). This suggests that the effects produced by Seb4MO results from a specific inhibition of Seb4 activity.

2.4. Seb4 is required for the function and expression of MyoD during myogenesis The in vivo analyses suggest that Seb4 is required for myogenesis. We further wanted to know if it is able to induce myogenic gene expression and myogenesis in vitro. Different amounts of Seb4 mRNA (0.2–2 ng) were injected into animal pole region at two-cell stage and ectodermal explants dissected at blastula stage were cultured to early gastrula and late neurula stage. RT-PCR analysis showed that overexpression of Seb4 did not induce the expression of any myogenic genes at different stages (not shown), indicating that it is not sufficient for myogenic gene expression. We thus assayed if it is required for myogenesis downstream of MyoD. In

MECHANISMS OF DEVELOPMENT 127 (2010)

2 8 1 –2 9 1

287

Fig. 5 – Effectiveness and specificity of Seb4MO. (A) Western blot analysis. Seb4MT (lanes 1 and 2) or mutSeb4MT (lanes 3 and 4) mRNA was coinjected with CoMO (lanes 1 and 3) or Seb4MO (lanes 2 and 4). Xdsh-myc mRNA was coinjected as an unrelated mRNA in each condition and used as a loading control. CoMO has no effect on Seb4MT translation (lane 1), while Seb4MO completely blocks Seb4MT translation (lane 2), but not the coinjected Xdsh-myc mRNA. Neither CoMO (lane 3) nor Seb4MO (lane 4) has any effect on mutSeb4MT translation. (B) Coinjection of mutSeb4MT mRNA rescues Seb4MO-inhibited Xbra and XMyf5 expression in whole embryos at the early gastrula stage, and MO2 similarly inhibits Xbra and XMyf5 expression as Seb4MO.

X. laevis, two XMyoD genes are expressed during early development (Rupp and Weintraub, 1991). We first injected XMyoDb mRNA (500 pg) alone or coinjected XMyoDb mRNA with Seb4MO (10 ng), injected ectodermal explants were cultured to stage 10.5 and 20, respectively. XMyoDb was able to induce the expression of XMyoDa and muscle actin (Fig. 6), however, it did not induce the expression of the pan-mesoderm gene Xbra, indicating that the induction of XMyoDa expression by XMyoDb was independent of a general mesoderm formation. Coinjection of XMyoDb mRNA (500 pg) with Seb4MO led to a significant decrease of the expression level of XMyoDa and XMyf5 at the early gastrula stage. It also inhibited, but not completely blocked, the expression of XMyoDa and muscle actin gene at late stage (Fig. 6), indicating that Seb4 acts downstream of MyoD, and is also required for the function and expression of myogenic genes. A similar result was obtained using mouse MyoD (not shown). Together these results sug-

gest that Seb4 is induced by, and cooperates with MyoD during myogenesis.

3.

Discussion

3.1. Seb4/RBM24 is a direct target gene of MyoD during myogenesis In this study we have first used a hormone-inducible form of MyoD (Wittenberger et al., 1999) to identify direct MyoD target genes at the initial stage of myogenesis. Both in vivo and in vitro analyses indicate that the expression of X. laevis Seb4 gene may be directly induced following incubation of whole embryos or ectodermal explants expressing MyoD-GR in DEX. This suggests that Seb4 is a potential target gene of MyoD. Furthermore, analysis of the proximal regulatory region of X. tropicalis RBM24, which is an ortholog gene of X. lae-

288

MECHANISMS OF DEVELOPMENT 127 (2010)

2 8 1 –2 9 1

cifically induced the expression of the d-subunit of the acetylcholine receptor gene, but not the c-subunit gene. Conversely, XMyoD specifically induced the expression of the c-subunit gene, but not the d-subunit gene (Charbonnier et al., 2003). Unlike XMyf5, XMyogenin activates the RBM24 promoter similarly as XMyoD, this is consistent with the observation that MyoD and Myogenin target a common set of promoters (Cao et al., 2006). Thus, our result raises the possibility that Seb4/RBM24 gene represents a specific target of MyoD during myogenesis.

3.2. Seb4 is required for the function and expression of MyoD during myogenesis

Fig. 6 – Seb4 function is required for myogenic gene expression and myogenesis induced by XMyoDb in animal cap explants. Embryos at two-cell stage were injected with XMyoDb mRNA alone or coinjected with Seb4MO, ectodermal explants were dissected and cultured to different stages. RTPCR analysis of myogenic gene expression was performed at stage 10.5 and 20. Seb4MO inhibits the expression of XMyf5 and XMyoDa at both stages. It also inhibits the expression level of muscle actin (Ms-Act) at stage 20. Note that overexpression of XMyoDb does not induce the expression of the pan-mesoderm gene Xbra. ODC was used as a loading control.

vis Seb4, reveals the presence of multiple E boxes, which are potential binding sites of MyoD and other bHLH proteins (Lassar et al., 1989; Blackwell and Weintraub, 1990). In addition, the presence of multiple E boxes is consistent with the observation that MyoD forms a relatively stable complex with DNA if at least two E boxes are present (Weintraub et al., 1990). Indeed, XMyoD is able to activate the RBM24 promoter, conversely, a dominant negative MyoD mutant strongly blocks the reporter activity, suggesting that MyoD may directly regulate Seb4/RBM24 expression during different stages of myogenesis. Consistent with previous observation showing that XMyoD may be under negative regulation in X. laevis early embryo (Rupp et al., 1994), we find that mouse MyoD more potently activated the RBM24 promoter. Together with the strong overlapping expression pattern between XMyoD and Seb4 genes at different stages of X. laevis embryos, our results are consistent with a direct regulation of Seb4 by MyoD. We also find that VegT, which shows significant overlapping expression pattern with Seb4 at the early gastrula stage, does not activate the RBM24 luciferase reporter gene. However, both X. laevis and mouse Myf5, a related bHLH protein to MyoD, activate the RBM24 reporter driven by the RBM24 E boxes to a lesser extent than MyoD. One possibility is that MyoD and Myf5 exhibit distinct specificity for a subset of target genes. In this respect, it has been shown that XMyf5 spe-

Overexpression of Seb4 in X. laevis ectodermal explants is not able to induce the expression of any mesodermal markers, however, Seb4 knockdown inhibits the expression of myogenic gene expression and myogenesis both in vivo and in vitro. This effect is closely similar to that produced by inhibition of MyoD activity using a dominant negative MyoD mutant. This suggests that the function of Seb4, although not sufficient, is necessary for myogenic gene expression and myogenesis. In addition, the similarity of effects produced by inhibition of Seb4 and MyoD activity indicates that they may indeed participate in the same myogenic process. Furthermore, we find that knockdown of Seb4 inhibits, but not completely blocks, the activity of MyoD in myogenesis, this is consistent with the observation that Seb4 is a target of MyoD and indicates that it is required for the function of MyoD in myogenesis. The effect of Seb4 knockdown on myogenic gene expression may be both dependent and independent of its effect on mesoderm induction. Since both Xbra and myogenic gene expression is inhibited in Seb4 morphants, it is unclear whether Seb4 knockdown directly affects myogenic gene expression in whole embryo. However, in cultured ectodermal explants expressing MyoD, no significant Xbra expression is detected, whereas myogenic gene expression and myogenesis are inhibited by Seb4 knockdown. This suggests that Seb4 may exert, at least to some extent, a direct regulation on myogenic gene expression. Together with the observation that overexpression of Seb4 is not sufficient to induce myogenic gene expression, our result suggests that Seb4 should have a fine-tuning function during myogenesis.

3.3.

The target RNAs of Seb4

The target RNAs of Seb4 remain to be identified. Recently, a related protein, Xseb4R, has been shown to bind to the 3 0 untranslated region of VegT mRNA and to regulate its stability. Overexpression of Xseb4R also induces the expression of a panel of mesendodermal genes in ectoderm cells (Souopgui et al., 2008). Although the amino-terminal half containing the RNA recognition motif is remarkably conserved between Seb4 and Xseb4R, the carboxyl-terminal half of these two proteins shows little overall identity. Moreover, Seb4 and Xseb4R exhibit distinct expression pattern (Fetka et al., 2000; Boy et al., 2004) and Seb4 is not able to induce the expression of mesoderm genes in ectodermal explants. Indeed, Seb4 protein is only 69% identical with Xseb4R, which is more likely

MECHANISMS OF DEVELOPMENT 127 (2010)

a X. laevis ortholog of RBM38. Thus, the identification of Seb4 binding RNAs should allow a further understanding of the mechanism whereby Seb4 regulates myogenesis. In Caenorhabditis elegans, the Seb4-related protein SUP-12 was shown to regulate the muscle-specific splicing of unc-60 pre-mRNA, which gives rise to either unc-60A or unc-60B, encoding nonmuscle and muscle isoforms of actin depolymerizing factor, respectively (Anyanful et al., 2004). However, in vertebrates, these two proteins are encoded by separate genes and exhibit distinct tissue distribution. Therefore, the target RNAs of Seb4 homolog may differ between invertebrates and vertebrates. A more recent study has demonstrated that SUP-12 cooperates with Fox-1 to regulate tissue-specific splicing of the fibroblast growth factor receptor, egl15, which eventually determines the ligand specificity of the receptor in vivo (Kuroyanagi et al., 2007). In this regard, it has been shown that FGF signalling is involved in muscle differentiation (Marics et al., 2002) and in the regulation of the epithelio-mesenchymal transition of the central dermomyotome in amniotes (Delfini et al., 2009), thus, it is of interest to identify whether Seb4/RBM24 regulates tissue-specific splicing of different fibroblast growth factor receptors in vertebrates. Recently, the human RNAbinding protein RNPC1, also known as RBM38, was shown to be a target of p53 and bind to the 3 0 -untranslated region in p21 transcript to maintain its stability (Shu et al., 2008). Consistent with this observation and our results, it has been shown that RBM38, but not RBM24, binds to p21 transcript to regulate myogenic differentiation in C2C12 cells (Miyamoto et al., 2009). This raises the possibility that Seb4/RBM24 might be also involved in the regulation of cell cycle during myogenesis, but through a distinct mechanism than RBM38. Therefore, further study will be needed in order to identify the RNA targets of Seb4/RBM24.

4.

Experimental procedures

4.1.

Xenopus laevis embryos and tissue explants

Xenopus laevis eggs were obtained from females injected with 500 IU of human chorionic gonadotropin (Sigma) and artificially fertilized. Eggs were dejellied with 2% cysteine hydrochloride (pH 7.8) and kept in 0.1· modified Barth solution (MBS). Microinjections of embryos were done in 0.1· MBS containing 3% Ficoll-400. After injections, embryos were kept in this solution for 3 h and then cultured in 0.1· MBS until they reached appropriate stages (Nieuwkoop and Faber, 1967). Dissection and culture of embryonic explants were performed in 1· MBS. After injection of MyoD-GR mRNA (200 pg), whole embryos or animal cap explants were cultured to different stages and then incubated in 10 l of dexamethasone in the presence or absence of 10 lg/ml of the protein synthesis inhibitor, cycloheximide, as we have previous described (Shi et al., 2002).

4.2. otides

Expression constructs and morpholino oligonucle-

The full-length X. laevis Seb4, XMyoD, XMyf5, XMyogenin and mouse MyoD and Myf5 coding sequences were PCR-amplified

2 8 1 –2 9 1

289

and cloned in the pCS2 vector (Rupp and Weintraub, 1991). Myc-tagged Seb4 (Seb4MT) was obtained by cloning the coding sequence into the pCS2MT vector inframe with the six myc epitopes. A mutant form of Seb4 (mutSeb4) and Seb4MT (mutSeb4MT) was created by introducing four nucleotide substitutions, without change of amino acid residues, in the first 25 nucleotides targeted by the morpholino. All constructs were then sequenced before use. Other constructs including Xdsh-myc (Sokol, 1996), MyoD-ENR, MyoD-GR (Wittenberger et al., 1999), Xwnt8 (Shi et al., 2002) and VegT (Li et al., 2006) were reported previously. Capped synthetic mRNA was made by in vitro transcription as described (Djiane et al., 2000). Two partially overlapping morpholino oligonucleotides against Seb4 (Seb4MO, 5 0 -TAGTGTCCTTCTGGGTGGTGTGCAT-3 0 ; MO2, 5 0 -CTTCTGGGTGGTGTGCATCTTGCCC-3 0 ) and the standard control morpholino were from Gene Tools, and suspended in sterile water at a concentration of 5 mg/ml as small aliquots.

4.3.

In situ hybridization and cell lineage tracing

Whole-mount in situ hybridization was performed according to standard protocol (Harland, 1991). Probes including Seb4, Xbra, Xwnt8, chordin, XMyoD, XMyf5, myosin light chain, Tbx6 and HoxB9 were previously described (Li et al., 2006) and were labelled using digoxigenin-11-UTP and appropriate RNA polymerase (Roche). Lac Z mRNA was injected as a cell lineage tracer and ß-galactosidase staining was performed using red-gal as substrate (Research Organics, Cleveland).

4.4.

RT-PCR

Extraction of total RNA from whole embryos or animal cap explants was performed using guanidine isothiocyanate/phenol followed by LiCl precipitation to remove genomic DNA and polysaccharides. RNA samples were further treated with RNAse-free DNAse I (Roche) and were reverse-transcribed using 200 U M-MLV reverse transcriptase (Invitrogen). PCR primers for Seb4 are as follow: 5 0 -GATAGAGAGGGTGCCGG-3 0 and 5 0 -ATTCGCTCAAAAGTCTAAGC-3 0 . All the other primers including Xbra, Xwnt8, XMyf5, XMyoDa, muscle actin and ornithine decarboxylase (ODC) were as described previously (Li et al., 2006). One-twentieth of the reverse-transcribed cDNA was used for PCR amplification in a reaction mixture containing 1 lCi of [a-32P]dCTP (GE Healthcare Biosciences). PCR products were resolved on a 5% non-denaturing polyacrylamide gel and visualized by a Phospho-Imager (Bio-Rad).

4.5. assay

RBM24 and Myf5 reporter constructs and luciferase

A 0.65 kb fragment of X. tropicalis RBM24 promoter region was PCR-amplified using genomic DNA according to the sequence in the Ensembl database (www.ensembl.org). To fuse the isolated RBM24 promoter sequence to a luciferase reporter gene, an oligonucleotide (5 0 -TTACCATGGCCTCACGAAACGCTCAA-3 0 ) was designed, which introduces a NcoI site (underlined) and eliminates the translation start site. PCR amplification was performed using this oligonucleotide and an upstream primer including an internal XhoI site

290

MECHANISMS OF DEVELOPMENT 127 (2010)

(5 0 -GCATGGCTCGAGTTCTAGCGCAG-3 0 ) (underlined). The PCR product was then cloned into the XhoI and NcoI sites of the modified pGL23-luc3 reporter plasmid (Kiyama et al., 1998). The X. tropicalis Myf5 luciferase reporter was obtained in a similar manner. The XTM1 Myf5 genomic sequence (Polli and Amaya, 2002) was amplified by PCR using the following primers: 5 0 -ACCCTCGAGAGATTCCAAACTTG-3 0 and 5 0 -TGT CCATGGAATACTGCTGCTTGTAAG-3 0 , including a XhoI and NcoI site, respectively (underlined). The reporter plasmid (200 pg) was either injected alone or coinjected with different synthetic mRNAs in the animal pole region or equatorial region at two- to four-cell stage. Cell lysates were prepared from 10 ectodermal explants or five whole embryos at early gastrula stage and the luciferase activity was measured using the luciferase assay system (Promega) and a Lumat LB 9507 luminometer (Berthold). The experiments were performed in triplicates.

Acknowledgements The authors thank Drs. R.A. Rupp, N. Pollet, E. Amaya, M. Taira, D. Duprez and S. Sokol for providing the reagents used in this study. This work was supported by grants from the Association Franc¸aise contre les Myopathies (AFM) and the Agence Nationale de la Recherche (ANR-09-BLAN-026203) to D.L. Shi, and the National Natural Science Foundation of China (No. 30800114) to H.Y. Li, who was also supported by a post-doctoral fellowship from CNRS.

R E F E R E N C E S

Anyanful, A., Ono, K., Johnsen, R.C., Ly, H., Jensen, V., Baillie, D.L., Ono, S., 2004. The RNA-binding protein SUP-12 controls muscle-specific splicing of the ADF/cofilin pre-mRNA in C. elegans. J. Cell Biol. 167, 639–647. Berkes, C.A., Tapscott, S.J., 2005. MyoD and the transcriptional control of myogenesis. Semin. Cell Dev. Biol. 16, 585–595. Blackwell, T.K., Weintraub, H., 1990. Differences and similarities in DNA-binding preferences of MyoD and E2A protein complexes revealed by binding site selection. Science 250, 1104–1110. Boy, S., Souopgui, J., Amato, M.A., Wegnez, M., Pieler, T., Perron, M., 2004. XSEB4R, a novel RNA-binding protein involved in retinal cell differentiation downstream of bHLH proneural genes. Development 131, 851–862. Buckingham, M., 2006. Myogenic progenitor cells and skeletal myogenesis in vertebrates. Curr. Opin. Genet. Dev. 16, 525–532. Burd, C.G., Dreyfuss, G., 1994. RNA binding specificity of hnRNP A1: significance of hnRNP A1 high-affinity binding sites in premRNA splicing. EMBO J. 13, 1197–1204. Cao, Y., Kumar, R.M., Penn, B.H., Berkes, C.A., Kooperberg, C., Boyer, L.A., Young, R.A., Tapscott, S.J., 2006. Global and genespecific analyses show distinct roles for Myod and Myog at a common set of promoters. EMBO J. 25, 502–511. Chanoine, C., Gaspera, B.D., Charbonnier, F., 2004. Myogenic regulatory factors: redundant or specific functions? Lessons from Xenopus. Dev. Dyn. 231, 662–670. Charbonnier, F., Della Gaspera, B., Armand, A.S., Lecolle, S., Launay, T., Gallien, C.L., Chanoine, C., 2003. Specific activation of the acetylcholine receptor subunit genes by MyoD family proteins. J. Biol. Chem. 278, 33169–33174.

2 8 1 –2 9 1

Delfini, M.C., De La Celle, M., Gros, J., Serralbo, O., Marics, I., Seux, M., Scaal, M., Marcelle, C., 2009. The timing of emergence of muscle progenitors is controlled by an FGF/ERK/SNAIL1 pathway. Dev. Biol. 333, 229–237. Djiane, A., Riou, J.F., Umbhauer, M., Boucaut, J.C., Shi, D.L., 2000. Role of frizzled 7 in the regulation of convergent extension movements during gastrulation in Xenopus laevis. Development 127, 3091–3100. Fetka, I., Radeghieri, A., Bouwmeester, T., 2000. Expression of the RNA recognition motif-containing protein SEB-4 during Xenopus embryonic development. Mech. Dev. 94, 283–286. Frank, D., Harland, R.M., 1991. Transient expression of XMyoD in non-somitic mesoderm of Xenopus gastrulae. Development 113, 1387–1393. Gerber, W.V., Vokes, S.A., Zearfoss, N.R., Krieg, P.A., 2002. A role for the RNA-binding protein, hermes, in the regulation of heart development. Dev. Biol. 247, 116–126. Harland, R.M., 1991. In situ hybridization: an improved whole mount method for Xenopus embryos. In: Kay, B.K., Peng, H.B. (Eds.), Methods in Cell Biology, vol. 36. Academic Press, San Diego, pp. 685–695. Hopwood, N.D., Pluck, A., Gurdon, J.B., 1989. MyoD expression in the forming somites is an early response to mesoderm induction in Xenopus embryos. EMBO J. 8, 3409–3417. Hopwood, N.D., Pluck, A., Gurdon, J.B., 1991. Xenopus Myf-5 marks early muscle cells and can activate muscle genes ectopically in early embryos. Development 111, 551–560. Hopwood, N.D., Pluck, A., Gurdon, J.B., Dilworth, S.M., 1992. Expression of XMyoD protein in early Xenopus laevis embryos. Development 114, 31–38. Huot, M.E., Bisson, N., Davidovic, L., Mazroui, R., Labelle, Y., Moss, T., Khandjian, E.W., 2005. The RNA-binding protein fragile X-related 1 regulates somite formation in Xenopus laevis. Mol. Biol. Cell 16, 4350–4361. Kiyama, T., Akasaka, K., Takata, K., Mitsunaga-Nakatsubo, K., Sakamoto, N., Shimada, H., 1998. Structure and function of a sea urchin orthodenticle-related gene (HpOtx). Dev. Biol. 193, 139–145. Kuroyanagi, H., Ohno, G., Mitani, S., Hagiwara, M., 2007. The Fox-1 family and SUP-12 coordinately regulate tissue-specific alternative splicing in vivo. Mol. Cell. Biol. 27, 8612–8621. Lassar, A.B., Buskin, J.N., Lockshon, D., Davis, R.L., Apone, S., Hauschka, S.D., Weintraub, H., 1989. MyoD is a sequencespecific DNA-binding protein requiring a region of myc homology to bind to the muscle creatine kinase enhancer. Cell 58, 823–831. Li, H.Y., Bourdelas, A., Carron, C., Gomez, C., Boucaut, J.C., Shi, D.L., 2006. FGF8, Wnt8 and Myf5 are target genes of Tbx6 during anteroposterior specification in Xenopus embryo. Dev. Biol. 290, 470–481. Lunde, B.M., Moore, C., Varani, G., 2007. RNA-binding proteins: modular design for efficient function. Nat. Rev. Mol. Cell Biol. 8, 479–490. Marics, I., Padilla, F., Guillemot, J.F., Scaal, M., Marcelle, C., 2002. FGFR4 signaling is a necessary step in limb muscle differentiation. Development 129, 4559–4569. McGrew, M.J., Pourquie´, O., 1998. Somitogenesis: segmenting a vertebrate. Curr. Opin. Genet. Dev. 8, 487–493. Mientjes, E.J., Willemsen, R., Kirkpatrick, L.L., Nieuwenhuizen, I.M., Hoogeveen-Westerveld, M., Verweij, M., Reis, S., Bardoni, B., Hoogeveen, A.T., Oostra, B.A., Nelson, D.L., 2004. Fxr1 knockout mice show a striated muscle phenotype: implications for Fxr1p function in vivo. Hum. Mol. Genet. 13, 1291–1302. Miyamoto, S., Hidaka, K., Jin, D., Morisaki, T., 2009. RNA-binding proteins Rbm38 and Rbm24 regulate myogenic differentiation via p21-dependent and -independent regulatory pathways. Genes Cells 14, 1241–1252.

MECHANISMS OF DEVELOPMENT 127 (2010)

Nieuwkoop, P.D., Faber, J., 1967. Normal Table of Xenopus laevis (Daudin). North Holland, Amsterdam. Perrone-Bizzozero, N., Bolognani, F., 2002. Role of HuD and other RNA-binding proteins in neural development and plasticity. J. Neurosci. Res. 68, 121–126. Polli, M., Amaya, E., 2002. A study of mesoderm patterning through the analysis of the regulation of Xmyf-5 expression. Development 129, 2917–2927. Rupp, R.A., Weintraub, H., 1991. Ubiquitous MyoD transcription at the midblastula transition precedes induction-dependent MyoD expression in presumptive mesoderm of X. laevis. Cell 65, 927–937. Rupp, R.A., Snider, L., Weintraub, H., 1994. Xenopus embryos regulate the nuclear localization of XMyoD. Genes Dev. 8, 1311–1323. Shi, D.L., Bourdelas, A., Umbhauer, M., Boucaut, J.C., 2002. Zygotic Wnt/beta-catenin signaling preferentially regulates the expression of Myf5 gene in the mesoderm of Xenopus. Dev. Biol. 245, 124–135. Shu, L., Yan, W., Chen, X., 2008. RNPC1, an RNA-binding protein and a target of the p53 family, is required for maintaining the stability of the basal and tress-induced p21 transcript. Genes Dev. 20, 2961–2972. Sokol, S.Y., 1996. Analysis of Dishevelled signalling pathways during Xenopus development. Curr. Biol. 6, 1456–1467.

2 8 1 –2 9 1

291

Souopgui, J., Rust, B., Vanhomwegen, J., Heasman, J., Henningfeld, K.A., Bellfroid, E., Pieler, T., 2008. The RNA-binding protein Xseb4R: a positive regulator of VegT mRNA stability and translation that is required for germ layer formation in Xenopus. Genes Dev. 22, 2347–2352. Spagnoli, F.M., Brivanlou, A.H., 2006. The RNA-binding protein, Vg1RBP, is required for pancreatic fate specification. Dev. Biol. 292, 442–456. Tada, M., Smith, J.C., 2000. Xwnt-11, a target of Xenopus Brachyury, regulates gastrulation movements via Dishevelled, but not through the canonical Wnt pathway. Development 127, 2227– 2238. Weintraub, H., Tapscott, S.J., Davis, R.L., Thayer, M.J., Adam, M.A., Lassar, A.B., Miller, A.D., 1989. Activation of muscle-specific genes in pigment, nerve, fat, liver, and fibroblast cell lines by forced expression of MyoD. Proc. Natl. Acad. Sci. USA 86, 5434– 5438. Weintraub, H., Davis, R., Lockshon, D., Lassar, A., 1990. MyoD binds cooperatively to two sites in a target enhancer sequence: occupancy of two sites is required for activation. Proc. Natl. Acad. Sci. USA 87, 5623–5627. Wittenberger, T., Steinbach, O.C., Authaler, A., Kopan, R., Rupp, R.A., 1999. MyoD stimulates delta-1 transcription and triggers notch signaling in the Xenopus gastrula. EMBO J. 18, 1915–1922.