Notch signalling acts in postmitotic avian myogenic cells to control MyoD activation

Transcription

1 Development 128, (2001) Printed in Great Britain The Company of Biologists Limited 2001 DEV Notch signalling acts in postmitotic avian myogenic cells to control MyoD activation Estelle Hirsinger 1, Pascale Malapert 1, Julien Dubrulle 1, Marie-Claire Delfini 2, Delphine Duprez 2, Domingos Henrique 3, David Ish-Horowicz 4 and Olivier Pourquié 1, * 1 Laboratoire de Génétique et de Physiologie du Développement (LGPD), Developmental Biology Institute of Marseille (IBDM), CNRS-INSERM-Université de Méditerranée-AP de Marseille, Campus de Luminy Case 907, Marseille Cedex 09, France 2 Institut d Embryologie Cellulaire et Moléculaire du CNRS et du Collège de France, 49 bis avenue de la Belle Gabrielle, Nogent sur Marne, France 3 Instituto de Histologia e Embriologia, Faculdade de Medicina, Universidade de Lisboa, Av. Prof. Egas Moniz, 1699 Lisboa Codex, Portugal 4 Imperial Cancer Research Fund, PO Box 123, 44 Lincoln s Inn Fields, London WC2A 3PX, UK *Author for correspondence ( pourquie@ibdm.univ-mrs.fr) Accepted 23 October; published on WWW 27 November 2000 SUMMARY During Drosophila myogenesis, Notch signalling acts at multiple steps of the muscle differentiation process. In vertebrates, Notch activation has been shown to block MyoD activation and muscle differentiation in vitro, suggesting that this pathway may act to maintain the cells in an undifferentiated proliferative state. In this paper, we address the role of Notch signalling in vivo during chick myogenesis. We first demonstrate that the Notch1 receptor is expressed in postmitotic cells of the myotome and that the Notch ligands Delta1 and Serrate2 are detected in subsets of differentiating myogenic cells and are thus in position to signal to Notch1 during myogenic differentiation. We also reinvestigate the expression of MyoD and Myf5 during avian myogenesis, and observe that Myf5 is expressed earlier than MyoD, consistent with previous results in the mouse. We then show that forced expression of the Notch ligand, Delta1, during early myogenesis, using a retroviral system, has no effect on the expression of the early myogenic markers Pax3 and Myf5, but causes strong down-regulation of MyoD in infected somites. Although Delta1 overexpression results in the complete lack of differentiated muscles, detailed examination of the infected embryos shows that initial formation of a myotome is not prevented, indicating that exit from the cell cycle has not been blocked. These results suggest that Notch signalling acts in postmitotic myogenic cells to control a critical step of muscle differentiation. Key words: Notch, Delta, MyoD, Myf5, Myogenesis, Somite, Muscle, Chick embryo INTRODUCTION The Notch signalling pathway is involved in a variety of developmental processes. In Drosophila, Notch is a large transmembrane receptor molecule that recognises two sets of transmembrane ligands called Delta and Serrate. Upon recognition of the ligand, the Notch receptor undergoes a set of proteolytic cleavages that result in the liberation of its cytoplasmic domain that translocates to the nucleus and associates with the transcription factor RBP-Jk/Su(H). This complex then activates the expression of downstream targets such as the transcriptional repressors of the Enhancer of split complex that ultimately control differentiation choices in the receiving cell (Artavanis-Tsakonas et al., 1995). This molecular pathway has been largely conserved through evolution and several Notch receptors, ligands and target genes have been identified in vertebrates. In the fly, Notch signalling is implicated in the control of cell fate decisions during myogenesis and neurogenesis (reviewed in Baylies et al., 1998; Jan and Jan, 1993). In both cases, Notch signalling inhibits differentiation by activation of transcriptional repressors that antagonise differentiationpromoting genes. The Notch pathway acts through a process referred to as lateral inhibition that results in the segregation of individual precursors from a cluster of equivalent cells. In these precursors, Notch activation is prevented, allowing them to embark on the neuronal or myogenic pathway. Notch signalling via the lateral inhibition process has been shown to be conserved in the vertebrate nervous system where it is used to select neuronal precursors from the pool of dividing ventricular cells (reviewed in Lewis, 1996). During vertebrate neurogenesis, Notch activation appears to control the progression of neurogenesis, during which nascent neurones signal to neighbouring progenitor cells to prevent their exit from the cell cycle and thus their differentiation. The role of Notch signalling during vertebrate myogenesis is less well understood. Experiments in C2 cultured myoblast cells show that Notch activation can inhibit expression of

2 108 E. Hirsinger and others myogenic factors (MyoD, myogenin and Mrf4) and of molecular markers of terminal myogenic differentiation (troponint) as well as the morphological appearance of fullydifferentiated multinucleated myotubes (Jarriault et al., 1998; Kato et al., 1997; Kopan et al., 1994; Kuroda et al., 1999; Lindsell et al., 1995; Luo et al., 1997; Nofziger et al., 1999; Shawber et al., 1996). The molecular details of this blockade have been worked out and are likely to involve the activity of the negative regulator Hes1 (Jarriault et al., 1998; Kuroda et al., 1999). The blockade also involves interference with activators of myogenesis such as the Mef2 family members since it has been shown that constitutively-activated Notch blocks the cooperative activation of myogenesis by Mef2c and the myogenic bhlh factors (Wilson-Rawls et al., 1999). However, in mouse mutants, genes of the Notch pathway such as Notch1 or Delta1, segmentation of the muscle derivatives is affected but no clear-cut effect on skeletal muscle differentiation has been reported (Conlon et al., 1995; de la Pompa et al., 1997; Hrabe de Angelis et al., 1997). These results could be due to functional redundancy between Notch receptors and ligands since mouse Notch1, Notch2, Delta1 and Delta3 are all expressed in the paraxial mesoderm tissue that contains the muscle progenitors (reviewed in Pourquié, 1999). In support of this argument, a unique RBP-Jk gene has been isolated so far and the knock out of this gene leads to the blockade of myogenin activation (Oka et al., 1995). This result is not expected if one assumes that Notch signalling represses expression of myogenic factors (as in Drosophila), but it could be explained if RBP-Jk acts as a repressor (Morel and Schweisguth, 2000). Nevertheless, activating or blocking Notch signalling in the frog also disorganises the segmental arrangement of skeletal muscles without affecting their differentiation (Jen et al., 1997). Thus, there is a marked discrepancy between the effects of gain or loss of vertebrate Notch function during myogenesis in vivo, and the in vitro analysis of its role in myogenic differentiation of C2 cells. Evidence in favour of a conserved role of Notch signalling between fly and vertebrate myogenesis therefore remains weak. In vertebrates, all skeletal muscles of the body derive from somites, more precisely from their dorsal epithelial part, the dermomyotome (reviewed in Hirsinger et al., 2000). As somitic maturation proceeds, muscle progenitors located in the dermomyotome translocate into the underlying myotome. In the trunk, the myotome ultimately gives rise to the axial and body wall musculature. During this process, muscle progenitors are thought to downregulate the Pax3 transcription factor and upregulate the myogenic regulatory factors MyoD and Myf5 as well as to exit the cell cycle (Williams and Ordahl, 1994). Expression of MyoD and Myf5 seems to indicate the definitive moment when the cells become determined towards the muscle differentiation pathway. These two genes are thought to control commitment to the myogenic lineage, since forcing their expression in various cell types is sufficient to induce muscle differentiation (Weintraub et al., 1991). In addition, loss of Myf5 or of Myf5 and MyoD function in mouse cells that are normally fated to give rise to muscles leads those cells to adopt other fates (Kablar et al., 1999; Tajbakhsh et al., 1996). The onset of expression of Myf5 and MyoD has been reported to differ between chick and mouse. In the mouse, Myf5 is expressed early in the presomitic mesoderm and epithelial somite whereas MyoD expression is activated two days later (reviewed in Buckingham, 1992). In the chick, MyoD has originally been reported to be activated first, at the level of the newly formed somite, and Myf5 soon thereafter (Borycki et al., 1997; Pownall and Emerson, 1992). More recently, however, Hacker and Guthrie described activation of Myf5 expression earlier in development than MyoD in chick, and they reported that expression of these genes was not dependent on the maturity of somites but on the age of the embryo (Hacker and Guthrie, 1998). The relative spatiotemporal expression patterns of those genes is therefore a controversial issue. Although expression of the genes of the Notch signalling pathway has been extensively described in many developmental systems (Morrison et al., 1999; Vargesson et al., 1998), the expression patterns during vertebrate myogenesis have not been analysed in great detail. In this paper, we show that the chick Notch1 receptor and two of its ligands, chick Delta1 and chick Serrate2, are strongly expressed during chick myogenesis. To address the role of Notch signalling in myogenesis, we overexpressed the Delta1 ligand during early chick myotome development. Our results indicate that, as in vertebrate and fly neurogenesis as well as in fly myogenesis, widespread activation of Notch during early myogenesis blocks cellular differentiation. But, in contrast to the vertebrate nervous system (Dorsky et al., 1997; Henrique et al., 1997), widespread activation of Notch during early myogenesis does not block the exit of precursor cells from the cell cycle. Rather, Notch appears to act in post-mitotic Myf5-expressing myogenic cells, preventing them from activating the MyoD gene and thereby blocking myogenesis. MATERIALS AND METHODS Overexpression of the wild-type form of Delta1 in the fertilized paraxial mesoderm Fertile White Leghorn eggs (Ferme Avicole HAAS, Kaltenhouse, France) were incubated at 37 C until the embryos reached stage HH 11 (Hamburger and Hamilton, 1992). Surgical experiments were performed in ovo on 823 embryos. The retroviral constructs and the generation of retrovirally infected O-Line cells are described in Henrique et al. (Henrique et al., 1997). When O-Line cells were 100% infected, the cultures were transferred to a bacterial Petri dish for 24 hours to aggregate the cells. These aggregates were then grafted unilaterally in the presomitic mesoderm at the level of Hensen s node in stage 11 chick embryos, the non-operated side of the embryo therefore providing an internal control. 48 hours later, at stage 20-21, the zone of retroviral infection was located at the inter-limb level and encompassed the length of two to three somites. Control embryos grafted with RCAS-Alkaline Phosphatase OLine aggregates developed normally (data not shown). Antibodies and probes The Notch1, Delta1, Serrate1 and Serrate2 probes are described in Myat et al. (Myat et al., 1996). Chick Myf5 probe is described in Saitoh et al. (Saitoh et al., 1993). Probes for the avian MyoD and Pax3 genes are described in Pourquié et al. (Pourquié et al., 1996). The retroviral infection was followed with the anti-p27 Gag antibody (Life Sciences). The Delta1 overexpression was monitored by whole-mount in situ hybridisation with the Delta1 probe. Terminally differentiated muscle cells were detected with the Mf20 antibody directed against sarcomeric myosin heavy chain (MHC). This antibody developed by

3 Notch signalling in postmitotic myogenic cells 109 Fig. 1. Notch pathway members are expressed during somitic myogenesis. (A,F,I) Whole-mount chick embryos hybridised with Notch1 (stage 19, A), Delta1 (stage 18, F) and Serrate2 (stage 21, I) probes. (B) Transverse section of a stage 19 embryo hybridised with Notch1 showing strong expression in the ventral dorsal lip and the nascent myotome. (C) Transverse section through a stage 21 embryo hybridised with the Notch1 probe, and stained with anti-brdu (in red). Notch1-expressing cells of the myotome are essentially postmitotic. (D,E). Transverse sections of a stage 25 embryo hybridised with the Notch1 probe (D) and immunostained with the Mf20 antibody and counterstained with Hoechst to indicate the nuclei (E). Notch1 is expressed in immature myotomal cells located between the dorsal lip and more differentiated myotubes expressing Mf20. (G) Frontal section showing Delta1 expression in a stage 13 embryo. Delta1 is essentially restricted in the caudal lip of the dermomyotome but is also detected in scattered cells in the dermomyotome (arrowhead). (H) Transverse section through a stage 18 embryo hybridised with Delta1 probe. Expression is seen in cells of the ventral part of the dorsomedial lip (arrowhead) and in the nascent myotome (arrowhead). (J) Transverse section through a stage 19 embryo showing Serrate2 expression in the mature myotome fibres. Note that Serrate2 is absent from the dorsal lip. ant, anterior; dl, dorsal lip; dm, dermomyotome; m, myotome; nt, neural tube; s, somite; scl, sclerotome. D. A. Fishman was obtained from the Developmental Studies Hybridoma Bank maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA Proliferating cells having incorporated BrdU (Sigma or Amersham) were revealed with the monoclonal anti-brdu antibody (Sigma or Amersham). In situ hybridisation procedures and histology Whole-mount E2 to E5 embryos were processed through in situ hybridisations with the Notch1, Delta1, Serrate1, Serrate2, Myf5, MyoD and Pax3 probes according to Henrique et al. (1995). Some were then sectioned using either a Leica VT1000E vibratome after gelatin/albumin embedding (50 µm section thickness) or using a Leica CM3000 cryostat after gelatin/sucrose inclusion (25 µm section thickness). Immunolabeling of cryosections with antibodies directed against BrdU, p27 Gag or MHC was then conducted. E10 embryos were sectioned using a Leica RM2035 microtome after paraffin embedding. 10 µm sections were immunolabeled with the Mf20 antibody and counterstained with Alcian Blue (15 mg/100 ml in distilled water, 5% acetic acid) which labels cartilage. Whole-mount stained embryos were photographed with a Leica Wild M10 stereomicroscope. The sections were photographed using a Leica DM microscope with Nomarski optics. BrdU experiments For E2 embryos, a 10 mm solution of BrdU (Amersham) was applied 15 minutes before fixation. The embryos were then sectioned on a cryostat as described above. The 20 µm sections were processed by in situ hybridisation with the Myf5 or MyoD probes and by immunolabelling for BrdU according to the manufacturer instructions (Amersham). For E4 operated embryos, 50 µl of a 10 mm solution of BrdU solution (Sigma) was applied 1 hour before fixation. The embryos were processed through whole-mount in situ hybridisation, cryosectioned and then immunolabeled for BrdU incorporation, according to Sechrist and Marcelle (Sechrist and Marcelle, 1996). RESULTS Expression of Notch1, Delta1, Serrate1 and Serrate2 during early stages of myogenesis in the chick embryo We have compared the expression of Notch1, Delta1, Serrate1 and Serrate2 during somite differentiation in the chick embryo. As previously reported, both Notch1 and Delta1 are strongly expressed in the presomitic mesoderm (PSM) and the caudal compartment of the newly formed somites throughout somitogenesis (Palmeirim et al., 1998). When the sclerotome forms, expression of both genes was retained only in the dermomyotome, which also expresses Pax3 and contains the muscle progenitors. While Notch1 was strongly expressed in

4 110 E. Hirsinger and others Fig. 2. Myf5 is expressed earlier and in more immature cells than MyoD. (A-C) Whole-mount stage 12 embryo (A) hybridised with the Myf5 probe and the corresponding transverse vibratome sections (B,C). The levels of the sections are indicated by the black broken arrows in A. Myf5 is expressed at a low level in the presomitic mesoderm (A), in the dorsal part of the epithelial somite (C) and then in the dermomyotome (B). It is expressed at a higher level in the developing myotome (B). (D) Transverse sections of E2 embryos hybridised with both the Myf5 (in blue) and the MyoD (in orange) probes. Only Myf5 is expressed in the epithelial dorsal lip. (E,F) Sections of stage 13 embryos hybridised with the Myf5 (E; transverse section, in blue) or MyoD (F; frontal section, in blue) probes and then immunostained with the anti-brdu antibody (in brown). Myf5 expression is activated in the proliferative dermomyotome and persists in the postmitotic myotomal cells (E), overlapping in the latter with MyoD expression (D,F). ant, anterior; dl, dorsal lip; dm, dermomyotome; m, myotome; nt, neural tube; s, somite; scl, sclerotome. the nascent myotome (Fig. 1A-E), Delta1 expression was limited to the caudal aspect of the dermomyotome (Fig. 1F-H) and Serrate2 to the differentiating myotome (Fig. 1I,J). Between stage HH 14 and 22, Notch1 expression was observed weakly throughout the dermomyotome and more strongly in cells of the myotome (Fig. 1A,B). BrdU labelling experiments demonstrated that cells strongly expressing Notch1 in the myotome were postmitotic (Fig. 1C). Notch1 is very strongly expressed in the medial most part of the myotome, in the region underlying the dorsal lip (Fig. 1A,B). Notch1 expression became progressively weaker in fibres located more laterally, which according to Denetclaw and Ordahl (Denetclaw and Ordahl, 1997) are older than those located medially at that stage. During these stages, most Delta1 expression was seen in the caudal dermomyotome lip (Fig. 1F,G). Delta1 expression was also detected in the ventral part of the caudal dorsal lip (Fig. 1H, arrowhead). In addition, a few Delta1-expressing cells, albeit at a lower level, were detected throughout the dermomyotome at this stage (Fig. 1G, arrowhead). Serrate2 was first detected in the forming myotome of stage HH embryos in cells that do not incorporate BrdU (data not shown). As somites mature, Serrate2 was observed in a stripe of cells located in the middle of the myotome (Fig. 1I,J). Based on their location and their morphology, these cells correspond to differentiated myocytes. From stage HH onwards, the epithelial dermomyotome is no longer present and Notch1 expression was observed in cells located in the ventral part of the dorsal lip (Fig. 1D). These Notch1-positive cells did not express the sarcomeric myosin heavy chain (MHC), which marks terminal muscle differentiation (Fig. 1E). Notch1 was also detected in cells forming the inner layer of the myotome, which according to Kahane et al. (Kahane et al., 1998) and to Denetclaw et al. (Denetclaw and Ordahl, 2000), correspond to the most immature postmitotic cells of this structure at that stage (data not shown). From stage 23 to 26, Delta1 was still weakly detected in the remnant of the caudal dermomyotome lip whereas Serrate2 expression remained in a stripe of differentiated myotubes spanning the whole myotome (data not shown). At stage HH 26, scattered cells expressing Notch1 and Serrate2 lay throughout the forming muscles whereas Delta1 expression was absent from the myotome (data not shown). Expression of Serrate1 mrna was only detected in sclerotomal derivatives (data not shown). Therefore, Notch1 is transiently expressed weakly in proliferative dermomyotomal cells and then more strongly in postmitotic myotomal cells where it is downregulated before they undergo terminal differentiation. Delta1 is detected in cells of the dorsal and caudal lips of the dermomyotome and Serrate2 is expressed in differentiated myotomal cells. Thus both Serrate2- and Delta1-expressing cells are located in close contact with the Notch1-expressing cells and are in position to signal to early differentiating myocytes. Early expression of Myf5 and MyoD resembles that observed in the mouse Reported expression of Myf5 and MyoD differs between mouse and chick or quail (reviewed in Borycki et al., 1997; Buckingham, 1992; Denetclaw and Ordahl, 2000; Ott et al., 1991; Pownall and Emerson, 1992). In the mouse, Myf5 appears well before MyoD in myogenic precursors whereas the opposite pattern has been described in the chick. We have reinvestigated the expression profiles of Myf5 and MyoD in the developing chick somite. As previously described, we observed strong Myf5 expression in myogenic cells of the myotome from E2 to E5 (data not shown). However, we also found earlier expression at lower levels and in novel sites. Like Hacker and Guthrie (Hacker and Guthrie, 1998), we observed an early onset of Myf5 expression in six-somite embryos at the level of the rostral somites (data not shown). However, in contrast to their observations, we observed a differential expression of Myf5 and MyoD that was dependent on the maturation stage of the somites. Myf5 expression is found in a stripe of cells in the PSM at the level of somite I (Fig. 2A). From stage 10 onward, Myf5 is detected in the dorsal epithelial somite and then in the

5 Notch signalling in postmitotic myogenic cells 111 Fig. 3. Overexpression of the Delta1 ligand affects expression of the Notch pathway members. (A) Schematic representation of the experimental procedure. RCAS-alkaline phosphatase or RCAS-Delta1-transfected O-Line cells are aggregated by plating them on a bacterial Petri dish for 24 hours. The retrovirus-infected cellular aggregates are then unilaterally grafted in the caudal-most region of the presomitic mesoderm of stage 11 embryos. The phenotype is analysed 2 or 8 days after the graft. (B) Whole-mount stage 20 embryo infected with the RCAS-Delta1 virus and hybridised with the Delta1 probe. The infected area is delimited by the bracket. After 2 days, the infected region encompassing two to three somites is heavily infected and expresses high levels of the Delta1 mrna. (C-F) Transverse sections of stage 20 RCAS-Delta1-infected embryos hybridised with the Notch1 (C) and the Serrate2 (E) probes, and then immunostained with an antibody against the retroviral protein p27 Gag (D,F). In (C,E), the operated, infected, side is on the right, the non-operated, control, side is on the left. (D,F) show the infected side only. The p27 Gag staining identifies the infected region and thereby correlates the eventual phenotype with the presence of the retrovirally expressed Delta1. On the infected side (D,F), Notch1 expression is upregulated in the dermomyotome and the myotome (C, arrowheads) while Serrate2 expression is downregulated in the myotome (E, arrowheads). ant, anterior; dm, dermomyotome; m, myotome; nt, neural tube; scl, sclerotome. dermomyotome (Fig. 2A-C) where it is co-expressed with Pax3. Myf5 remains expressed in the dorsal lip of the dermomyotome (Fig. 2D), which remains epithelial and retains Pax3 expression, after desepithelialisation of the dermomyotome. BrdU analysis shows that these Myf5-positive cells of the dorsal lip are proliferative (Fig. 2E). These patterns of expression thus resemble the sequence previously described for Myf5 in developing mouse embryos (reviewed by Buckingham, 1992). As previously described, MyoD expression is first detected in stage 10 chick embryos, in the dorsal lip of the dermomyotome and in the nascent myotome, where it partly overlaps with the domain of Myf5 expression (Fig. 2D). BrdU analysis of MyoD-expressing cells in the somite indicate that they are mostly postmitotic (Fig. 2F). Thus, contrary to previous claims, Myf5 expression in developing chick muscle precursor cells, occurs prior to that of MyoD, as has been reported in the mouse. Delta1 overexpression in the developing somite affects expression of Notch1, Serrate1 and Serrate 2 To examine the role of Notch signalling during myogenesis in vivo, we used an RCAS retroviral vector to overexpress Delta1 in the developing somites. Previous studies during chick neuroretinal differentiation have shown that activation of Notch signalling using this retroviral construct leads to inhibition of neural differentiation (Henrique et al., 1997). Grafts were performed using embryos ranging from 10 to 15 somites.

6 112 E. Hirsinger and others Fig. 4. Activation of Notch signalling blocks expression of MyoD, but not that of Pax3 and Myf5, without preventing the myoblasts from exiting the cell cycle. (A,E,F) Stage 20 RCAS-Delta1-infected embryos hybridised in whole mount with the Pax3 (A), Myf5 (E) and MyoD (I) probes. The brackets delineate the infected region. (B-D,F-H,J-L) Corresponding transverse sections of those embryos immunostained with the anti-p27 Gag antibody (D,H,L) and the anti-brdu antibody (F,G,J,K, in brown). The levels of the sections are indicated by the black broken arrows in A,E,F. (B,F,J) The non-operated, control, side of the embryo. (C,D,G,H,J-L). The operated, infected, side of the embryo. In the three cases (D,H,L), the dermomyotomes and myotomes are heavily infected. On the infected side, Pax3 expression is slightly upregulated (A,C) in somites that did not undergo the epithelio-mesenchymal transition of the dermomyotome observed on the control side (C, arrowheads). Myotomal expression of Myf5 is not perturbed (E,G, arrowheads) while that of MyoD is abolished (I,K, arrowheads). The BrdU staining of the infected side (G,K) is not different from that on the control side (F,J), showing that overexpression of the Delta1 ligand does not prevent the cells from exiting the cell cycle. The infected myotome is therefore composed of Myf5-positive postmitotic cells (G). ant, anterior; dm, dermomyotome; m, myotome; scl, sclerotome. Aggregates of O-line fibroblasts transfected with the RCASalkaline phosphatase (AP) or the RCAS-Delta1 construct were grafted in the caudal-most part of the embryo, within the paraxial mesoderm (Fig. 3A). These implanted pellets act as a local viral source that, by 48 hours, has infected wide territories spanning several somites in the embryo, thereby resulting in widespread ectopic expression of Delta1 mrna in the interlimb region (Fig. 3B). The extent of the retroviral infection can also be followed by detection of the p27 Gag retroviral protein (Figs 3D,F and 4D,H,L). Strong expression of the Delta1 mrna in the newly formed somites of this region is already detected 24 hours after the graft (data not shown), indicating that these somites develop in an environment in which all cells are exposed to excess ligand. As the Notch1 receptor is expressed in the myotome throughout its development, the ectopic Delta1 expression is likely to result in constitutive activation of Notch signalling in these cells. In many cellular contexts, binding of Delta to its receptor Notch triggers a feedback loop that sensitises receiving cells and desensitises signalling cells to Notch signalling, resulting in the modification of the expression levels of Notch pathway members (reviewed in Lewis, 1996). We examined whether, in our experiments, Delta1 overexpression affects expression of Notch1 (n=4), Serrate1 (n=3) and Serrate2 (n=4). 48 hours after grafting, expression of Notch1 was strongly upregulated in the infected dermomyotome when compared with the control side (Fig. 3C). In similar experiments, Serrate2 was downregulated in the myotome (Fig. 3E) whereas Serrate1 was upregulated in the sclerotome (data not shown). These results suggest that overexpressing Delta1 in the developing somite results in wide ectopic activation of Notch signalling and they are consistent with a regulatory loop linking the different components of the pathway in the developing somite. Constitutive Notch activation during myogenesis prevents MyoD expression in Myf5 postmitotic myotomal cells We then assessed the effect of Notch activation on the expression of the myogenic factors, MyoD and Myf5, 48 hours after infection. In infected somites, MyoD expression was strongly down-regulated in the myotome (Fig. 4I; compare Fig. 4J with 4K, n=39), whereas Myf5 was still normally expressed (Fig. 4E; compare Fig. 4F with 4G, n=22). Pax3 was also unaffected in most embryos (n=13) and even slightly upregulated in three cases (Fig. 4A; compare Fig. 4B with 4C). The infected dermomyotome maintained its epithelial structure

7 Notch signalling in postmitotic myogenic cells 113 after it should have undergone an epithelio-mesenchymal transition, allowing the release of dermal precursors (Fig. 4C, arrowheads). In contrast, the infected myotome appeared morphologically similar to its uninfected counterpart on the contralateral side (Figs 3C,E, 4C,G,K). Myf5 was expressed in proliferative cells of the dermomyotome and the dorsal lip in addition to the myotome, whereas MyoD was essentially found in the postmitotic cells of myotome (Fig. 2B,D). The absence of MyoD in the infected embryos could be due to an accumulation of proliferative Myf5-expressing cells that are unable to proceed further in their differentiation. This situation would be reminiscent of that in the nervous system where widespread Delta1 overexpression blocks exit of neural progenitor cells from the cell cycle (Henrique et al., 1997). To examine whether myogenic progenitors are also prevented from exiting the cell cycle, we monitored BrdU incorporation together with the expression of MyoD (n=1) and Myf5 (n=3) in infected embryos. Postmitotic myogenic cells were found in both infected and uninfected myotomes (Fig. 4; compare Fig. 4F,J with 4G,K), indicating that ectopic Notch signalling does not block exit from the cell cycle in this context. This is consistent with the retention of normal, and not dramatically widespread, Pax3 and Myf5 expression in the dermomyotome and myotome. The loss of MyoD expression but maintenance of Myf5 expression in postmitotic cells in the myotomal layer implies that constitutive Notch activation does not affect the production of postmitotic Myf5-expressing cells, but specifically blocks subsequent MyoD expression by these cells. Constitutive activation of the Notch pathway during somite differentiation prevents subsequent differentiation of somite muscle derivatives The results described above indicate that Notch signalling does not interfere with production of postmitotic muscle precursor cells. To assess the capacity of these precursor cells to differentiate terminally despite the absence of MyoD, infected embryos were examined at E10 using the Mf20 antibody to monitor the effect of Delta1 overexpression on MHC expression and muscle formation. In the infected embryos, the epaxial and hypaxial muscles were almost entirely absent and replaced by a loose mesenchyme (Fig. 5A,B; n=6). This indicates that widespread Notch activation in the somite results in a complete block of myogenic differentiation. DISCUSSION Conservation of the temporal expression profile of Myf5 and MyoD between birds and mammals Our findings demonstrate that in contrast to previously published observations (Borycki et al., 1997; Pownall and Emerson, 1992), Myf5 is expressed earlier than MyoD during the myogenic differentiation sequence in birds. As reported in mouse (Ott et al., 1991), we find that Myf5 is initially detected as a stripe in the rostral PSM and then in the dermomyotome where it is co-expressed with Pax3. Its expression is then strongly upregulated in cells of the newly formed myotome. We confirm the previously described temporal profile of MyoD expression in which the gene is first detected from the level of somite I in the dorso-medial quadrant and then in the forming Fig. 5. Activation of Notch signalling results in the blockade of skeletal muscle formation. (A,B) Transverse sections of a stage 36 RCAS-Delta1-infected embryo, immunostained with the Mf20 antibody and counterstained in Alcian Blue. The muscle masses appear in brown and the skeletal elements in blue. The operated side is on the right. The asterisks indicate the position of muscles that are entirely or partially missing. Overexpression of the Delta1 ligand massively blocks muscle formation on the operated side. myotome. MyoD expression was never observed in the PSM or in the dermomyotome. Myf5 is activated early in Pax3-positive cells of the proliferating compartment of the dermomyotome, whereas MyoD expression is predominantly restricted to the postmitotic cells of the underlying myotome. Since apparently all dermomyotome cells express Myf5, all muscle progenitors must activate Myf5 early in their differentiation. This expression is maintained and upregulated when the cells enter the postmitotic myotomal compartment, at which time they downregulate Pax3 (Williams and Ordahl, 1994). Subsequently, myotomal cells switch on MyoD. A similar differentiation sequence has recently been reported in the mouse embryo (Venters et al., 1999). Thus, whereas previous reports have implied that the temporal hierarchy linking Myf5 and MyoD varies between species, our findings argue in favour of an evolutionary conservation of this molecular network. Widespread Notch activation blocks myogenesis Widespread activation of Notch during early myogenesis results in a downregulation of MyoD expression in the postmitotic myotomal fibres ultimately leading to a blockade of myogenesis. Similar to what has been described during larval or adult myogenesis in Drosophila (Anant et al., 1998; Brennan et al., 1999), these observations suggest that Notch activation must be relieved for the cells to activate MyoD and to proceed with differentiation. Genetic analysis in the mouse has led to the proposal that myogenesis is controlled by two parallel molecular pathways, one relying on Myf5 and the other on Pax3 and MyoD (Kablar

8 114 E. Hirsinger and others et al., 1997; Maroto et al., 1997; Tajbakhsh et al., 1997). It has been hypothesised that in the absence of Myf5, muscle cells will differentiate via the Pax3/MyoD pathway whereas, in the absence of Pax3 or MyoD, cells will differentiate through the Myf5 pathway. This model explains the lack of a muscular phenotype in MyoD and Myf5 single mutant embryos, and the complete absence of muscles seen in MyoD/Myf5 and Pax3/Myf5 double mutants. Our results do not support the existence in chick of these two independent pathways since in the infected embryos, muscle differentiation is completely inhibited although Myf5, and Pax3 expression is not affected in the myotomal fibres. Either Notch activation blocks a factor required for Myf5 activity or, more likely, Myf5 alone is insufficient to drive muscle differentiation. Our results are in line with a recent study analysing triple mutants of MyoD/myogenin/Mrf4 (Valdez et al., 2000). These animals express only Myf5 and no muscle differentiation occurs. Although we have no data about the expression of Mrf4 in embryos infected with Delta1, we observed a downregulation of myogenin expression (data not shown; n=7), suggesting that those embryos lack MyoD and myogenin expression, similarly to the mouse triple knockout embryos. Notch acts in postmitotic myotomal cells during myogenesis Together with the published in vitro evidence, our overexpression studies suggest that a normal role of Notch activation in myogenesis is the control of MyoD activation. This raises the question of when and where does this occur in myogenic differentiation. Conflicting models for myotome formation exist. In one, myogenic cells are produced by the dorsal lip of the dermomyotome and then translocate into the myotome (Denetclaw and Ordahl, 2000). In an alternative proposal, a first lineage of postmitotic cells called pioneer cells is set aside early in development and provides a scaffold in which cells produced by the rostral and caudal lip of the dermomyotome intercalate (Kahane et al., 1998). In this model, myogenic cells are also produced by the dorsal lip but are required to translocate to the rostral and caudal lips before elongating in the myotome. In this controversial context, since localisation of muscle precursors and their migration is still unclear, it is therefore difficult to definitively correlate gene expression patterns with precise morphogenetic and differentiation events. According to the Denetclaw and Ordahl model, myotomal cells are sequentially produced by the dorsal lip and thus older fibres are located more laterally in the myotome. Therefore, the mediolateral arrangement of cells in the myotome reflects their differentiation status. Our observations can be interpreted more easily in the context of this model. First, we observe that Myf5 is expressed earlier than MyoD in myogenic differentiation. Accordingly, Myf5 expression in seen in proliferating cells of the dorsal lip region whereas MyoD is only detected more laterally in post-mitotic cells of the myotome (Fig. 2D). Second, Notch1-expressing cells are located in a domain including the ventral part of the dorsal lip and extending in the medial-most myotome (Fig. 1B-D). These Notch1-positive cells of the myotome are postmitotic (Fig. 1C), but do not yet express myosin heavy chain, suggesting that they are still immature (Fig. 1D-E). Our observations suggest the following differentiation sequence in the early myotome: proliferating myotomal precursor cells expressing Pax3 and Myf5 in the dorsal lip upregulate Notch1 and downregulate Pax3 upon entering the ventral part of the lip domain. Slightly more laterally in the myotome they start to express MyoD while maintaining Notch1 and Myf5 expression. More laterally still, the cells downregulate Notch1 and start expressing MHC indicating that they undergo terminal differentiation. According to this differentiation sequence, Notch1 could play a role in regulating MyoD activation in vivo. Which ligand is responsible for activating Notch in differentiating myotome cells? Although in the mouse, Delta1 is expressed in the cells of the myotome (Bettenhausen et al., 1995), no such expression was observed in the chick. During myogenesis, Delta1, which is initially expressed in the caudal somite, becomes progressively restricted to the caudal lip and to the caudal dorsal lip of the dermomyotome. Since the Notch1-positive myotomal fibres elongate to span the whole extent of the myotome they will rapidly contact the caudal lip of the dermomyotome from where they could receive the Delta1 signal. Alternatively, a Delta1 signal could be provided by the cells of the caudal dorsal lip or by the scattered cells observed in the dermomyotome. The effects of ectopic Delta1 could be mimicking endogenous signalling by Serrate2. Serrate2 is also expressed in a position to activate Notch1 and so could also be involved in signalling. However, it is expressed too late to regulate MyoD activation and so may play a later role. Finally, we cannot rule out the possibility that Notch1 interacts with an as yet unidentified ligand in the developing myotome. This differential expression of Notch ligands suggests that Notch, expressed initially in the dermomyotome and then in the myotome, may act at multiple times and sites to control several differentiation steps during muscle development, as it has been shown during Drosophila myogenesis (reviewed by Baylies et al., 1998). Thus, Notch1 expression is detected in immature postmitotic fibres of the myotome in an expression domain overlapping with that of MyoD. Our data also indicate that myogenic cells first express Myf5 when proliferative and then MyoD when post-mitotic and that Notch activation can block the latter step in myotomal cells (Fig. 6). Therefore, the in vivo role of Notch in myogenesis is likely to take place during the earliest stages of myotome formation rather than at the time of myoblast exit from the cell cycle. Notch activation does not block exit of myotomal cells from the cell cycle Neurogenesis and myogenesis share a number of common strategies in Drosophila and vertebrates (reviewed by Baylies et al., 1998; Jan and Jan, 1993). Both differentiation processes require the specification of a pool of progenitor cells which exit the cell cycle to ultimately yield highly specialised postmitotic cells. Both processes use related molecular determinants such as the bhlh MRFs in myogenesis and the proneural genes in neurogenesis (reviewed by Jan and Jan, 1993). In the fly, both differentiation processes rely on lateral inhibition to select progenitors by regulating proneural bhlh. In vertebrate neurogenesis, lateral inhibition contributes to controlling the cell cycle exit of neuroblast progenitors (reviewed in Lewis, 1998). For example, in the developing chick retina, overexpression of Delta1, using the same

9 Notch signalling in postmitotic myogenic cells 115 described in other postmitotic cells such as differentiated neurones where it was demonstrated to play a role in neurite plasticity (Franklin et al., 1999; Sestan et al., 1999). Similarly, Notch could also play a late role in controlling important events of myogenic cell differentiation such as fibre extension, attachment or fusion. In conclusion, our results show in vivo that Notch signalling is involved in the control of chick somitic myogenesis. Notch activation prevents MyoD expression in postmitotic Myf5- positive cells, thereby blocking terminal muscle differentiation. We are grateful to Mike McGrew, Kim Dale, Miguel Maroto, Christophe Marcelle and Monte Westerfield for helpful comments on the manuscript. This work was supported by the CNRS, and grants from the AFM, the HFSPO, the FRM and the ARC. REFERENCES Fig. 6. Proposed model: Notch signalling acts upstream of MyoD in postmitotic cells to block muscle formation. Notch signalling seems to have an important function at a differentiation step located between the Myf5 and MyoD-linked maturation steps. It does not prevent the myoblasts from exiting the cell cycle and, considering the expression pattern data, is likely to act in the postmitotic myoblasts. In infected embryos, muscle formation is prevented, whereas Pax3 and Myf5 expression is not perturbed. These data, along with expression data, suggest the existence in chick of a single myogenic cascade in which Myf5 lies upstream of MyoD, in contrast to what has been described in mice. retrovirus as in this study, prevents neuronal precursors from embarking upon the neuronal differentiation pathway (Henrique et al., 1997) such that cells exposed to ectopic Delta1 remain in the proliferative ventricular epithelium. Our findings show that the response to Notch signalling in developing muscles contrasts strikingly with that observed in the nervous system, since strongly overexpressing Delta1 somites give rise to a normal postmitotic myotomal layer. Therefore, widespread activation of the Notch pathway in the somite does not prevent the myogenic precursors in the dermomyotome from exiting the cell cycle. Our results are consistent with those of in vitro experiments carried out in C2 cells that indicate that Notch activation results in a failure of MyoD activation and of later differentiation events (Jarriault et al., 1998; Kato et al., 1997; Kopan et al., 1994; Kuroda et al., 1999; Lindsell et al., 1995; Luo et al., 1997; Nofziger et al., 1999; Shawber et al., 1996). Nevertheless, we draw different conclusions from these in vitro experiments. Based on the absence of MyoD expression and of molecular and morphological markers of terminal differentiation, it was inferred that Notch activation in C2 cells blocks differentiation and causes cells to remain in an undifferentiated proliferative state, as observed in the nervous system. However, it is possible that in these experiments, Notch activation blocks the C2 cells in a Myf5-positive postmitotic state, similar to what we observe in vivo. In many systems, Notch activation is thought to maintain cells in an immature proliferative state (reviewed in Artavanis- Tsakonas et al., 1995). Our observations indicate a later role for Notch signalling during myogenesis, in the postmitotic cells of the myotome. Recently, Notch expression has been Anant, S., Roy, S. and VijayRaghavan, K. (1998). Twist and Notch negatively regulate adult muscle differentiation in Drosophila. Development. 125, Artavanis-Tsakonas, S., Matsuno, K. and Fortini, M. E. (1995). Notch signaling. Science 268, Baylies, M. K., Bate, M. and Ruiz Gomez, M. (1998). Myogenesis: a view from Drosophila. Cell 93, Bettenhausen, B., Hrabe de Angelis, M., Simon, D., Guenet, J.L. and Gossler, A. (1995). Transient and restricted expression during mouse embryogenesis of Dll1, a murine gene closely related to Drosophila Delta. Development 121, Borycki, A. G., Strunk, K. E., Savary, R. and Emerson, C. P. J. (1997). Distinct signal/response mechanisms regulate pax1 and QmyoD activation in sclerotomal and myotomal lineages of quail somites. Dev. Biol. 185, Brennan, K., Baylies, M. and Arias, A. M. (1999). Repression by Notch is required before Wingless signalling during muscle progenitor cell development in Drosophila. Curr. Biol. 9, Buckingham, M. (1992). Making muscle in mammals. Trends Genet. 8, Conlon, R. A., Reaume, A. G. and Rossant, J. (1995). Notch1 is required for the coordinate segmentation of somites. Development. 121, de la Pompa, J. L., Wakeham, A., Correia, K. M., Samper, E., Brown, S., Aguilera, R. J., Nakano, T., Honjo, T., Mak, T. W., Rossant, J. et al. (1997). Conservation of the Notch signalling pathway in mammalian neurogenesis. Development. 124, Denetclaw, W. F. and Ordahl, C. P. (2000). The growth of the dermomyotome and formation of early myotome lineages in thoracolumbar somites of chicken embryos. Development 127, Dorsky, R. I., Chang, W. S., Rapaport, D. H. and Harris, W. A. (1997). Regulation of neuronal diversity in the Xenopus retina by Delta signalling. Nature 385, Franklin, J. L., Berechid, B. E., Cutting, F. B., Presente, A., Chambers, C. B., Foltz, D. R., Ferreira, A. and Nye, J. S. (1999). Autonomous and nonautonomous regulation of mammalian neurite development by Notch1 and Delta1. Curr. Biol. 9, Hacker, A. and Guthrie, S. (1998). A distinct developmental programme for the cranial paraxial mesoderm in the chick embryo. Development 125, Hamburger, V. and Hamilton, H. L. (1992). A series of normal stages in the development of the chick embryo, Dev. Dyn. 195, Henrique, D., Adam, J., Myat, A., Chitnis, A., Lewis, J. and Ish-Horowicz, D. (1995). Expression of a Delta homologue in prospective neurons in the chick. Nature 375, Henrique, D., Hirsinger, E., Adam, J., Le Roux, I., Pourquie, O., Ish- Horowicz, D. and Lewis, J. (1997). Maintenance of neuroepithelial progenitor cells by Delta-Notch signalling in the embryonic chick retina. Curr. Biol. 7, Hirsinger, E., Jouve, C., Dubrulle, J. and Pourquie, O. (2000). Somite formation and patterning. Int. Rev. Cytol. 198, Hrabe de Angelis, M., McIntyre, J. and Gossler, A. (1997). Maintenance of

10 116 E. Hirsinger and others somite borders in mice requires the Delta homologue DII1. Nature 386, Jan, Y. N. and Jan, L. Y. (1993). HLH proteins, fly neurogenesis, and vertebrate myogenesis. Cell 75, Jarriault, S., Le Bail, O., Hirsinger, E., Pourquié, O., Logeat, F., Strong, C. F., Brou, C., Seidah, N. G. and Israël, A. (1998). Delta-1 activation of notch-1 signaling results in HES-1 transactivation. Mol. Cell Biol. 18, Jen, W. C., Wettstein, D., Chitnis, D. and Kintner, C. (1997). The Notch ligand, X-Delta-2, mediates segmentation of the paraxial mesoderm in Xenopus embryos. Development 124, Suppl., Kablar, B., Krastel, K., Ying, C., Asakura, A., Tapscott, S. J. and Rudnicki, M. A. (1997). MyoD and Myf-5 differentially regulate the development of limb versus trunk skeletal muscle. Development. 124, Kablar, B., Krastel, K., Ying, C., Tapscott, S. J., Goldhamer, D. J. and Rudnicki, M. A. (1999). Myogenic determination occurs independently in somites and limb buds. Dev. Biol. 206, Kahane, N., Cinnamon, Y. and Kalcheim, C. (1998). The cellular mechanism by which the dermomyotome contributes to the second wave of myotome development. Development. 125, Kato, H., Taniguchi, Y., Kurooka, H., Minoguchi, S., Sakai, T., Nomura- Okazaki, S., Tamura, K. and Honjo, T. (1997). Involvement of RBP-J in biological functions of mouse Notch1 and its derivatives. Development. 124, Kopan, R., Nye, J. S. and Weintraub, H. (1994). The intracellular domain of mouse Notch: a constitutively activated repressor of myogenesis directed at the basic helix-loop-helix region of MyoD. Development 120, Kuroda, K., Tani, S., Tamura, K., Minoguchi, S., Kurooka, H. and Honjo, T. (1999). Delta-induced Notch signaling mediated by RBP-J inhibits MyoD expression and myogenesis. J. Biol. Chem. 274, Lewis, J. (1996). Neurogenic genes and vertebrate neurogenesis. Curr. Opin. Neurobiol. 6, Lewis, J. (1998). Notch signalling and the control of cell fate choices in vertebrates. Semin. Cell Dev. Biol. 9, Lindsell, C. E., Shawber, C. J., Boulter, J. and Weinmaster, G. (1995). Jagged: a mammalian ligand that activates Notch1. Cell 80, Luo, B., Aster, J. C., Hasserjian, R. P., Kuo, F. and Sklar, J. (1997). Isolation and functional analysis of a cdna for human Jagged2, a gene encoding a ligand for the Notch1 receptor. Mol. Cell. Biol. 17, Maroto, M., Reshef, R., Münsterberg, A. E., Koester, S., Goulding, M. and Lassar, A. B. (1997). Ectopic Pax-3 activates MyoD and Myf-5 expression in embryonic mesoderm and neural tissue. Cell 89, Morel, V. and Schweisguth, F. (2000). Repression by suppressor of hairless and activation by Notch are required to define a single row of single-minded expressing cells in the Drosophila embryo. Genes Dev. 14, Morrison, A., Hodgetts, C., Gossler, A., Hrabe de Angelis, M. and Lewis, J. (1999). Expression of Delta1 and Serrate1 (Jagged1) in the mouse inner ear. Mech. Dev. 84, Myat, A., Henrique, D., Ish-Horowicz, D. and Lewis, J. (1996). A chick homologue of Serrate and its relationship with Notch and Delta homologues during central neurogenesis. Dev. Biol. 174, Nofziger, D., Miyamoto, A., Lyons, K. M. and Weinmaster, G. (1999). Notch signaling imposes two distinct blocks in the differentiation of C2C12 myoblasts. Development 126, Oka, C., Nakano, T., Wakeham, A., de la Pompa, J. L., Mori, C., Sakai, T., Okazaki, S., Kawaichi, M., Shiota, K., Mak, T. W. et al. (1995). Disruption of the mouse RBP-J kappa gene results in early embryonic death. Development. 121, Ott, M. O., Bober, E., Lyons, G., Arnold, H. and Buckingham, M. (1991). Early expression of the myogenic regulatory gene, myf-5, in precursor cells of skeletal muscle in the mouse embryo. Development. 111, Palmeirim, I., Dubrulle, J., Henrique, D., Ish-Horowicz, D. and Pourquié, O. (1998). Uncoupling segmentation and somitogenesis in the chick presomitic mesoderm. Dev. Genet. 23, Pourquié, O. (1999). Notch around the clock. Curr. Opin. Genet. Dev. 9, Pourquié, O., Fan, C. M., Coltey, M., Hirsinger, E., Watanabe, Y., Breant, C., Francis-West, P., Brickell, P., Tessier-Lavigne, M. and Le Douarin, N. M. (1996). Lateral and axial signals involved in avian somite patterning: a role for BMP4. Cell 84, Pownall, M. E. and Emerson, C. P. J. (1992). Sequential activation of three myogenic regulatory genes during somite morphogenesis in quail embryos. Dev Biol. 151, Saitoh, O., Fujisawa-Sehara, A., Nabeshima, Y. and Periasamy, M. (1993). Expression of myogenic factors in denervated chicken breast muscle: isolation of the chicken Myf5 gene. Nucleic Acids Res. 21, Sechrist, J. and Marcelle, C. (1996). Cell division and differentiation in avian embryos: techniques for study of early neurogenesis and myogenesis. Methods Cell Biol. 51, Sestan, N., Artavanis-Tsakonas, S. and Rakic, P. (1999). Contact-dependent inhibition of cortical neurite growth mediated by notch signaling. Science 286, Shawber, C., Nofziger, D., Hsieh, J. J., Lindsell, C., Bogler, O., Hayward, D. and Weinmaster, G. (1996). Notch signaling inhibits muscle cell differentiation through a CBF1- independent pathway. Development 122, Tajbakhsh, S., Rocancourt, D. and Buckingham, M. (1996). Muscle progenitor cells failing to respond to positional cues adopt non- myogenic fates in myf-5 null mice. Nature 384, Tajbakhsh, S., Rocancourt, D., Cossu, G. and Buckingham, M. (1997). Redefining the genetic hierarchies controlling skeletal myogenesis: Pax- 3 and Myf-5 act upstream of MyoD. Cell 89, Valdez, M. R., Richardson, J. A., Klein, W. H. and Olson, E. N. (2000). Failure of Myf5 to support myogenic differentiation without myogenin, MyoD, and MRF4. Dev. Biol. 219, Vargesson, N., Patel, K., Lewis, J. and Tickle, C. (1998). Expression patterns of Notch1, Serrate1, Serrate2 and Delta1 in tissues of the developing chick limb. Mech. Dev. 77, Venters, S. J., Thorsteinsdottir, S. and Duxson, M. J. (1999). Early development of the myotome in the mouse. Dev. Dyn. 216, Weintraub, H., Davis, R., Tapscott, S., Thayer, M., Krause, M., Benezra, R., Blackwell, T. K., Turner, D., Rupp, R. and Hollenberg, S. (1991). The myod gene family: nodal point during specification of the muscle cell lineage. Science 251, Williams, B. A. and Ordahl, C. P. (1994). Pax-3 expression in segmental mesoderm marks early stages in myogenic cell specification. Development. 120, Wilson-Rawls, J., Molkentin, J. D., Black, B. L. and Olson, E. N. (1999). Activated notch inhibits myogenic activity of the MADS-Box transcription factor myocyte enhancer factor 2C. Mol. Cell Biol. 19,