Initial steps of myogenesis in somites are independent of influence from axial structures

Transcription

1 Development 120, (1994) Printed in Great Britain The Company of Biologists Limited Initial steps of myogenesis in somites are independent of influence from axial structures Eva Bober 1, *, Beate Brand-Saberi 2, Cecilia Ebensperger 2, Jörg Wilting 2, Rudi Balling 3, Bruce M. Paterson 4, Hans-Henning Arnold 1 and Bodo Christ 2 1 Department of Cell and Molecular Biology, Technical University of Braunschweig, Spielmannstrasse 7, Braunschweig, Germany 2 Institute of Anatomy, University of Freiburg, PO. Box 111, Freiburg i. Brsg., Germany 3 Max-Planck-Institute of Immunobiology, Department of Developmental Biology, Stübeweg 51, Freiburg i. Brsg., Germany 4 Laboratory of Biochemistry, National Cancer Institute NIH, Bethesda, MD, 20892, USA *Author for correspondence SUMMARY Formation of paraxial muscles in vertebrate embryos depends upon interactions between early somites and the neural tube and notochord. Removal of both axial structures results in a complete loss of epaxial myotomal muscle, whereas hypaxial and limb muscles develop normally. We report that chicken embryos, after surgical removal of the neural tube at the level of the unsegmented paraxial mesoderm, start to develop myotomal cells that express transcripts for the muscle-specific regulators MyoD and myogenin. These cells also make desmin, indicating that the initial steps of axial skeletal muscle formation can occur in the absence of the neural tube. However, a few days following the extirpation, the expression of MyoD and myogenin transcripts gradually disappears, and becomes almost undetectable after 4 days. From these observations we conclude that the neural tube is not required for the generation of the skeletal muscle cell lineage, but may support the survival or maitenance of further differentiation of the myotomal cell compartment. Notochord transplanted medially or laterally to the unsegmented paraxial mesoderm leads to a ventralization of axial structures but does not entirely prevent the early appearance of myoblasts expressing MyoD transcripts. However, the additional notochord inhibits subsequent development and maturation of myotomes. Taken together, our data suggest that neural tube promotes, and notochord inhibits, the process of myogenesis in axial muscles at a developmental step following the initial expression of myogenic bhlh regulators. Key words: somites, myogenic bhlh factor genes, neural tube, notochord, myotome, muscle differentiation INTRODUCTION Somites, which give rise to the axial skeleton, skeletal muscle, and the dermis of the back, develop from the segmental plate mesoderm in a metameric fashion along the longitudinal axis of the vertebrate embryo (for reviews see, Rong et al., 1990; Christ and Wilting, 1992). It has been reported that removal of the neural tube in early chick embryos results in defects of somite differentiation leading to complete loss of epaxial muscles (intrinsic deep back muscles), whereas hypaxial musculature, such as muscles of the ventrolateral body wall and limb muscles develop normally (Rong et al., 1992; Christ et al., 1992). It has also been demonstrated that differentiation of myotomal muscle depends on early influences from neural tube and/or notochord (Rong et al., 1992). These influences appear to be required only for a certain time period during the formation of somites, as removal of the neural tube at later stages of somite development has no inhibitory effect on subsequent muscle differentiation (Rong et al., 1992). These and other observations suggest that somites need signals from the axial structures in order to differentiate (Watterson et al., 1954; Strudel, 1955; Christ et al., 1972; Jacob et al., 1974; Christ et al., 1992; Rong et al., 1992). In these studies, the influence of the neural tube on somite differentiation has been assessed primarily by morphological criteria and the appearance of molecular markers for late stages in the differentiation process, e.g. the formation of myotomes and vertebrae and the immunohistochemical staining of somites with muscle-specific antibodies. The earliest muscle-specific markers known to be expressed in the somites are members of the myogenic regulatory gene family encoding basic helix-loop-helix (bhlh) transcription factors (Sassoon et al., 1989; Ott et al., 1991; Bober et al., 1991; Pownall and Emerson, 1992). Their spatiotemporal expression pattern during embryonic development is consistent with their pivotal role in myoblast determination and/or differentiation (for reviews see, Buckingham, 1992; Arnold and Braun, 1993; Emerson, 1993). In particular, the myf-5

2 3074 E. Bober and others gene in mouse and its functional homologues qmf-1 in quail and CMDI in chicken are possible candidates for myogenic determinators, as they are the first myogenic regulatory genes to be expressed in the dorsomedial quadrant of the epithelial, still undifferentiated somites (Ott et al., 1991; Pownall and Emerson, 1992; Bober et al., unpublished observations). This is also the region where the first myotomal cells arise (Kaehn et al., 1988). Based on time and position of its expression, it is resonable to hypothesize that the activation of the myf-5 gene, and by analogy, qmf-1 and CMDI, may depend on influences from the dorsal part of the neural tube including the neural crest. In order to examine the influences of the axial structures on muscle formation in vivo we interfered with this interaction by microsurgical manipulations and studied the expression of marker genes for early somite differentiation (CMDI, myogenin, Pax-1). The operations included extirpation of the neural tube, one half of the neural tube, and implantation of an extra notochord. Our results show that removal of the axial structures does not prevent the appearance of myotomal cells that express transcripts for the muscle-specific regulators CMDI and myogenin. However, this expression disappears a few days after the operation. Therefore, we conclude that the neural tube is not important for the initial steps of myogenesis, but supports the survival and subsequent development of axial skeletal muscle. We also present evidence that notochord transplanted adjacent to the unsegmented paraxial mesoderm does not prevent the first activation of the myogenic regulatory factor CMDI in myotomal cells, although it later exerts a ventralizing effect and inhibits further development of axial muscles. MATERIAL AND METHODS Microsurgical procedures The eggs of chick (Gallus gallus) and quail (Coturnix coturnix japonica) embryos were incubated at 37.8 C and 80% humidity for 2 days up to HH stages (16-22 somites; Hamburger and Hamilton, 1951). The neural tube, one half of the neural tube, or neural tube and notochord were excised at the level of the unsegmented paraxial mesoderm as described previously (Fig. 1; Christ et al., 1992). In a control experiment, the chicken neural tube was exchanged with a neural tube from the same region and developmental stage of a quail embryo. Operated embryos were allowed to develop in ovo for 1-4 days following the extirpation. Notochords were prepared from quail embryos (stages HH) and grafted into chick host embryos of similar developmental stages (Brand-Saberi et al., 1993). The operated embryos were re-incubated for 1 or 2 days. Histology and immunohistochemistry Embryos were fixed in Serra s solution for hours. After dehydration, the embryos were embedded in paraffin for serial sections. Immunohistochemistry was performed using the indirect immunoperoxidase staining method with anti-desmin antibody (Laboserv Diagnostica) diluted 1:300, anti HNK-1 antibody (Becton- Dickinson) diluted 1:10, MF20 antibody which stains skeletal myosin (Bader et al., 1982; gift from Dr J. Cossu) diluted 1:100, antiquail antibody (a gift from Dr. B. Carlson) and the anti-chondroitin- 6-sulfate-proteoglycan antibody (ICN, Meckenheim) diluted 1:70. The latter was applied to cryosections of embryos embedded in Tissue Tec (Leica). In situ hybridization In situ hybridizations were performed on 7 µm thick paraffin sections as described previously (Sassoon et al., 1988; Bober et al., 1991; Brand-Saberi et al., 1993). The hybridization probes were prepared as follows. A 900 bp PstI-EcoRI restriction fragment from the 3 end of CMDI cdna (Lin et al., 1989) and a 660 bp EcoRI-partial PstI fragment of the chicken myogenin cdna (Fujisawa-Sehara et al., 1990) were subcloned into the pks + plasmid vector (Stratagene). To generate hybridization probes the CMDI and myogenin constructs were linearized with PstI and XbaI, respectively, and crna was synthesized with T3 RNA polymerase and 35 S-UTP as described previously (Ott et al., 1991). The in situ hybridization using the mouse Pax-1 probe was performed as described by Deutsch et al. (1988). The HincII-SacI paired box fragment, which was cloned into the SmaI site of the pspt18 plasmid, was linearized with HindIII and transcribed with T7 RNA polymerase to generate a 377 nucleotide antisense riboprobe (Koseki et al., 1993). RESULTS Muscle-specific bhlh regulators are expressed in somitic cells in the absence of the neural tube The expression pattern of the myogenic regulatory genes CMDI and myogenin, the earliest known markers in prospective myotomal cells in avian embryos (Pownall and Emerson, 1992; Bober et al., unpublished data) has been investigated in somites developing in the absence of neural influence. To this end, we microsurgically removed the neural tube at the level of the segmental plate mesoderm in 2-day chicken embryos (HH stage 12-14, somites). The operation procedure is depicted in Fig. 1 and has been described previously (Christ et al., 1992). Following the removal of neural tube, embryos were incubated for 1-4 additional days and CMDI and cmgn expression were determined by in situ hybridization on serial sections using the respective cdna probes. Fig. 1 shows the results of two individual experiments, which are representative of 15 operated embryos examined 1 day after the excision of the neural tube. Signs of cell death around the region of the removed neural tube were sometimes observed (Fig. 1B). However, in all embryos, expression of CMDI and myogenin was detected in somites that were characteristically fused owing to the lack of the neural tube, as described previously (Christ, 1970). Frequently, a layer of myogenic cells reminiscent of newly forming myotomes was seen underneath the dermatome (Figs 1E,F, 2A). In addition, CMDI-positive cells were also found in the loose mesenchyme which appeared to correspond to the sclerotome (Fig. 1F). However, in most cases we found early sclerotomal cells, which expressed transcripts for Pax-1 (Koseki et al., 1993), underneath the myotomal cells in a region that did not extensively overlap with CMDI expression (Fig. 1H,I). This suggests that sclerotome, although sometimes slightly dislocated, was forming normally in a spatial distribution distinct from myotome. However, the appearance of both compartments, myotome and sclerotome, was frequently abnormal in operated embryos with less confined myotomal boarders as compared to controls. Regions with hybridization signals for myogenin transcripts generally coincided with those for CMDI (Fig. 1). At later

3 Influence of axial structures on myogenesis 3075 Fig. 1. Somite development in a chicken embryo 1 day after extirpation of the neural tube. The operation procedure is presented in A. All serial sections were performed on embryos 1 day after the operation. The neural tube was removed from a 19-somite stage embryo (HH 13) in a region extending from the youngest somites more caudally. Normal myotomes are shown on transverse sections of a control embryo with phase contrast optics (C) and by in situ hybridization with thecmdi probe (D). Unpaired somites, fused at the midline in operated embryos are shown by phase contrast microscopy (E), H/E staining (B), and in situ hybridization with the CMDI probe (F). Myotomal-like cells (m) express CMDI (small arrows). Note also CMDI-positive cells further ventrally (large arrow). Note multiple pycnotic cells around the operation field (B). (G,H,I) Transverse sections of an independently operated embryo in phase contrast (G), in situ hybridization with the myogenin probe (H) and the Pax-1 probe (I). Magnifications: 235 (B-F), and 125 (G-I). nt, neural tube; no, notochord; d, dermatome; m, myotome.

4 3076 E. Bober and others Fig. 2. In situ hybridizations with CMDI and myogenin probes and anti-desmin staining of chicken embryos 2 and 3 days after removal of the neural tube. Myotomal cells expressing CMDI (A; magnification 125 ) and desmin-positive, spindle-shaped myoblasts (B; magnification 235 ) are shown in an embryo incubated for 2 days after the operation. A myoblast in atypical perpendicular orientation to the longitudinal axis is marked by the arrow. In situ hybridizations on embryo 3 days after operation with the myogenin probe (C) and the CMDI probe (D). Transverse sections were taken from the wing level. Note high concentrations of transcripts in lateral muscles compared to axial muscles. Magnification: (C,D) 55. stages we also observed single desmin-expressing cells in a region corresponding to the myotomal cell layer (Fig. 2B). The desmin-positive cells in operated embryos appeared scattered in an irregular pattern in contrast to the longitudinally arranged myoblasts in control myotomes (Fig. 2B). Similar results were obtained when both, neural tube and notochord were removed together (data not shown). To exclude the possible destructive influence of the operation procedure on myotome development, we performed an interspecific control grafting experiment (Fig. 3). The host chick neural tube was exchanged for a donor quail neural tube. Fig. 3A depicts the quail neural tube and neural crest-derived structures stained with a quail-specific antibody. Significantly, muscle development proceeds normally following this operation as shown by myotomes stained with the sarcomeric myosin monoclonal antibody MF20 (Fig. 3B). Our results provide compelling evidence that myotomal muscle cells can differentiate in somites that have not been in contact with the neural tube during segmentation. Therefore, the initial expression of the myogenic regulatory factors and desmin are not dependent on neural structures during the time of somitogenesis. Continous development of myotomal cells depends on axial structures Previous studies have shown that epaxial muscle in the chick embryo fails to develop when neural tube and notochord were removed (Rong et al., 1992; Christ et al., 1992). Yet, myotomal cells appear in early somites shortly after segmentation in the absence of the neural tube. The fate of these myotomal cells was studied 3 and 4 days after the operation and sections were analyzed by in situ hybridization with probes for the myogenic regulators. While 3 days after the removal of neural tube CMDI and myogenin transcripts were still readily detectable in axially located muscle structures (compare Figs 2C,D and 4), little or no signals remained after 4 days (Fig. 4, small arrows). Already 3 days after the operation there was a relative decrease of CMDI and myogenin transcripts in myotomal regions compared to premuscle masses in the limbs and ventrolateral body wall (Fig. 2C,D). These data are consistent with earlier findings showing that hypaxial muscle differentiation is not affected by the absence of the neural tube. The gradual disappearance of transcription factor expression in the axially located myotomes was found in somites that had fused due to

5 Influence of axial structures on myogenesis 3077 the extirpation of the neural tube, while all paired somites located rostrally or caudally to the operation area developed normal myotomes (Fig. 1C and data not shown). After the extirpation of one half of the neural tube and a reincubation period of 3 days, the epaxial part of the myotome failed to develop on the operated side, whereas the hypaxial part of this myotome and the myotome at the contralateral side remained unchanged (Fig. 5). As already demonstrated previously (Christ et al., 1992) and shown in Fig. 5C, no neural tube- or neural crest-derived structures were detectable on the operated site, whereas HNK-1-positive cells Fig. 3. Interspecific grafting of the neural tube does not affect muscle development. The neural tube were normally present on the from the region of unsegmented paraxial mesoderm was exised from a stage 14 chicken embryo and contralateral site. This confirms replaced by a corresponding part of a stage-matched quail embryo. Staining with an anti-quail antibody that the extirpation of the one shows that neural tube- and neural crest-derived structures are of quail origin (A). The myotome half of the neural tube has been development proceeds normally in the operated embryo as documented by early skeletal myosin performed prior to migration of staining with MF20 antibody (B). Magnification 60. the neural crest cells. Taken together, these results demonstrate that either the maintenance or the developmental progression of epaxial muscle cells, but not their initial appearance depends on the neural tube. Ectopically transplanted notochord does not entirely prevent activation of the CMDI gene in early somites It has been demonstrated previously that transplantation of notochord medially or laterally to the unsegmented paraxial mesoderm prevents myotome formation (Pourquie et al., 1993; Brand-Saberi et al., 1993). This observation is consistent with the concept that ventralization of somitic derivatives depends on signals emanating from the notochord. Here, we examined the effect of an extra notochord on the early accumulation of CMDI transcripts. Significant expression of CMDI was observed in the most dorsal part of the paraxial mesoderm 1 day after notochord transplantation (Fig. 6C-H). We also noticed that CMDIpositive myoblasts arise even very close to the transplanted notochord (see Fig. 6C-F). It should be mentioned, however, that the distance between the newly introduced notochord and the presumptive myotome seemed to affect myotomal development. When the notochord was placed further ventrally to the somites, the morphological appearance of the myotome was less affected (Fig. 6G,H). Fig. 4. CMDI expression on embryo sections 4 days after the operation. Transverse sections at the wing (A) and pelvic (B) level hybridized with the CMDI probe. Small arrows indicate residual signals around the vertebral column. Large arrows show relatively high expression in the shoulder (A) and abdominal muscles (B). Magnification 25.

6 3078 E. Bober and others Fig. 5. Myotome development after removing one lateral half of the neural tube. The operation procedure is shown in A. Desmin staining illustrates the absence of most of the myotome 2 days after the operation (B). The ventral bud region giving rise to body wall muscles is unaffected. The HNK-1 antibody staining demonstrates the lack of neural tissue on the operated site (C). Magnification 100. In embryos analysed 2 days after the operation, myotomes were strongly reduced in size or totally absent on the operated side (Fig. 7). At this time Pax-1-positive cells, indicative for early sclerotome formation, were observed in a wider spatial distribution (Fig. 7B; Brand-Saberi et al., 1993). This ventralizing effect was accompanied by the expression of chondroitin- 6-sulfate-proteoglycan, a component of the extracellular matrix, around the transplanted notochord (Fig. 6B) and a down-regulation of chondroitin-6-sulfate-proteoglycan in the surrounding tissue, similar to the perinotochordal area in normal development. These observations indicate that the notochord does not entirely prevent early myogenesis at the level of CMDI activation, although it leads to ventralization of the somites. This process results eventually in the loss of myoblast differentiation and enhanced development of sclerotomal derivatives. DISCUSSION Microsurgical removal of the axial organs, the neural tube or the neural tube and the notochord, was performed at the level of the unsegmented paraxial mesoderm, caudal to the youngest newly formed somites. The structures were excised over a distance of several presumptive somites. This technique ensures that somites arising during the postoperation period are certainly deprived of any influence by the neural tube or neural crest cells. However, unsegmented presomitic paraxial mesoderm may have received earlier signals from axial structures (see below). The absence of neural tissue was confirmed by the fact that in the operated area no cells positive for the HNK-1 antibody were present. The segmentation process of the paraxial mesoderm was found not to be affected by the operation, which is in agreement with previous observations (Christ et al., 1972; Primmet et al., 1989). The significant result of our experiments is that cells expressing CMDI and myogenin do arise in somites that have not been in contact with the neural tube, suggesting that the activation of these genes, and presumably the initiation of myogenesis, do not depend on any kind of neural induction during or after somite formation. This, however, does not exclude determination events prior to the segmentation of the paraxial mesoderm. Indeed, it has been shown that gastrulation is an important step during specification of the myogenic cell lineage (von Kirschhofer et al., 1994). During further development, the cells with myogenic potential are exclusively layed down in the paraxial but not the lateral plate mesoderm (Krenn et al., 1988; Wachtler et al., 1982). However, determi-

7 Influence of axial structures on myogenesis 3079 Fig. 6. CMDI expression in somites after transplantation of an additional notochord. The operation procedure is shown in A. The arrow in B marks chondroitin-6-sulfateproteoglycan deposition around the newly inserted notochord. In situ hybridizations with CMDI probe (D,F,H) and phase contrast micrographs (C,E,G) of embryos operated at HH stage 14 (C-F) or HH stage 13 (G,H). Embryos were incubated for one additional day following the transplantation. The arrows illustrate the position of newly inserted notochords. Magnification 190 (B); 125 (C-H).

8 3080 E. Bober and others Fig. 7. Effect of a transplanted notochord on myotome and sclerotome development. The transverse section shown in A was stained with antidesmin antibody. The additional notochord is marked by an asterisk. No myotome is present at the side of operation. (B) An adjacent section hybridized with Pax-1 probe. The sclerotome appears extended dorsally at the place of notochord insertion (arrow). This embryo was incubated for 2 days after transplantation. Magnification 95. nation processes occurring during the early developmental stages, up to the formation of the paraxial mesoderm, have to be considered as permissive regarding somite specification, since somitic cells remain plastic for a certain time period after segmentation and differentiate according to their acquired position, not to their original fate (Christ et al., 1992; Aoyama, 1993). These results suggest that there is local signalling that leads to the formation of myotome-derived muscle. Here, we show the capacity of cells to enter the myogenic pathway in the absence of local cues from the neural tissue. Nevertheless, the maintenance and/or further development of myogenic cells clearly depends on interactions with surrounding tissues. The critical role of the axial organs on differentiation of muscle cells in somites has been established previously by several investigators (Christ et al., 1992, Teillet and Le Douarin, 1983; Rong et al., 1992). The main conclusions from these studies were that epaxial skeletal musculature requires the influence of neural tube and/or notochord in order to develop normally, whereas muscles in limbs and the body wall musculature form independently of the neural tube/notochord complex. Although all of these muscle tissues are derived from somites, their progenitors obviously differ in their response to signals emanating from the axial organs. Incidentally, progenitor cells migrating from the lateral dermomyotome to give rise to the limb muscles, fail to express myogenic regulatory factors of the MyoD family, in contrast to precursor cells for trunk muscles (Buckingham, 1992). It has been shown recently that the Pax-3 gene may serve as a marker for cells with myogenic potential. This gene is first expressed throughout the whole paraxial mesoderm and then becomes restricted to the dermomyotome and cells migrating into the limb buds (Goulding et al., 1993; Bober et al., 1994; Williams and Ordahl, 1994). It is noteworthy that myf-5 in mouse and qmf-1/cmdi in avian embryos begin to be expressed very early during somitogenesis in the dorsomedial region of epithelial somites in close proximity to the neural tube. This is about the region of the somites that gives rise to the myotome (Ordahl and Le Douarin, 1992). However, it is important to note that myotome formation is a continuous process, starting from the craniomedial edge of the somite and proceeding laterally (Kaehn et al., 1988). Thus, activation of the mouse myf-5 or chicken CMDI gene may be related to the determination process of myotomal progenitor cells. Interestingly, Rong et al. (1992) have shown recently that the effect of the neural tube/notochord complex on myotome differentiation is restricted to a distinct time window of approximately 10 hours following the appearance of somites, a period closely related to activation of the first myogenic regulatory genes (Ott et al., 1991; Pownall and Emerson, 1992). This temporal expression pattern of myf- 5/CMDI and its local coincidence with myogenic regions that develop under the influence of the neural tube prompted us to examine CMDI expression in chicken embryos in the absence of neural tube or in the absence of both, notochord and neural tube. Despite the fact that myogenic regulatory genes are activated in somites that have not been in contact with the neural tube, further expression of the bhlh factors and development of paraxial muscles does not proceed normally. Thus, additional interactions have to be considered to be important for stable establishment and/or survival of the epaxial muscle compartment. Based on the presented experiments we are unable to discriminate between cessation of the myogenic factor expression and subsequent failure of myotomal cells to differentiate and myotomal cell death. Although we have observed pycnotic cells around the operation field, we are not certain whether potential cell death afflicts the population of myogenic precursors. There are several possibilities as to how the neural tube may exert its supporting effect. The maintenance of the myotome may depend on trophic factors originating from the neural tube or the neural crest cells, since immigration of these cells into somites and development of the myotome are temporarily and spatially correlated (Erickson and Perris, 1993). One possible role for such factors could be the regulation of the prolifera-

9 Influence of axial structures on myogenesis 3081 tive capacity of the myogenic cell population. The proper size of this compartment may be important for its further development and differentiation. The importance of cell number and cell-cell interactions for muscle cell development, termed the community effect, has been shown before (Gurdon et al., 1993). It has also been demonstrated that the proliferative capacity of early somites depends on environmental cues, yet, the source and nature of these putative signals remains unknown (Tam and Tan, 1992). Alternatively, as already mentioned, down-regulation of CMDI expression may cause a block of myotomal differentiation. Another possibly indirect role of the neural tube could be its influence on the composition and assembly of components of the extracellular matrix, which is important for the appropriate organization of cells within each compartment. In our experiments we continuously observed a grossly disturbed organization of the myotomal compartment, which may have prevented further maturation of committed cells. In this context, it is interesting that a regulatory role for the extracellular matrix components and/or cell adhesion molecules in somitogenesis has been documented recently (Griffith and Sanders, 1991; Mills et al., 1990). It furthermore seems possible that the neural tube provides a differentiation promoting factor which functions in addition to the myogenic bhlh regulators, controlling later steps of epaxial myogenesis. Using the expression of sarcomeric myosin heavy chain as muscle-specific marker, Vivarelli and Cossu (1986) demonstrated that differentiation of embryonic day 10 mouse somites in vitro can be promoted by explants of the neural tube, whereas somites isolated from older embryos developed muscle cells without the addition of the neural tube. Rong et al. (1992) also provided evidence that myogenesis in culture may be controlled by the neural tube and the notochord. In the light of the in vivo experiments presented here, we suggest that the results obtained in these tissue culture experiments probably reflect a survival effect of the neural tube on myotomal cells, rather than an inductive phenomenon. Conflicting results have been obtained as to the influence of the notochord on early myotome formation. According to Rong et al. (1992) the notochord exerts a positive influence on myogenic cell development. In agreement with previous observations (Pourquie et al., 1993; Brand-Saberi et al., 1993) we found that no mature myotome develops in the vicinity of ectopically grafted notochord. In contrast, an enhancing effect on sclerotome formation, as demonstrated by the enlarged Pax- 1 expression domain, can be seen in these experiments. The grafted notochord does not inhibit muscle development at the level of the activation of the myogenic regulatory genes and, therefore, is likely to prevent more downstream events, presumably associated with proliferation or survival of the early myoblasts. In no case, however, did we observe a rescuing effect on myotomes by the notochord. Rong et al. (1992) discussed the possibility that the induction and/or survival of epaxial muscles in the absence of the neural tube could be achieved by the notochord alone. In our experiments we have not been able to show such a function for the notochord. Development of the myotomes was always equally disturbed when the neural tube alone or neural tube and notochord were removed. It has recently been shown that the notochord induces the expression of the Pax-1 gene in the ventral parts of somites (Koseki et al., 1993; Brand-Saberi et al., 1993). After induction of the floor plate by the notochord, the former has the same inductive capacity (Koseki et al., 1993). This capacity clearly interferes with myotomal cell development and the formation of epaxial muscles (Brand-Saberi et al., 1993). It therefore seems likely that only the most dorsal parts of the neural tube and the neural crest cells or the dorsal ectoderm (Kenny-Mobbs and Thorogood, 1987) promote myogenic differentiation. Frequently, we observed myoblasts only in the most dorsal edge of the somites after grafting an additional notochord. In summary, our studies demonstrate that the first steps of myotomal differentiation are independent of local cues from the neural tube. The notochord inhibits myogenesis, the initial steps of myogenic differentiation however, may take place even in the vicinity of the notochord. We would like to thank Sonja Hoffmann, Lidia Koschny and Maria Schüttoff for excellent technical assistance. The work presented here was supported by grants Ch 44/12-1 to B. C. and Ar 115/8-2 to H. H. A. from the Deutsche Forschungsgemeinschaft. REFERENCES Aoyama, H. (1993). Developmental plasticity of the prospective dermatome and the prospective sclerotome region of an avian somite. Dev. Growth Differ. 35, Arnold, H. H. and Braun, T. (1993). The developmental control genes in myogenesis of vertebrates. In Advances in Developmental Biochemistry (ed. P. M. Wassarman), vol. 2, pp London: JAI Press Inc. Bader, D., Masaki, T. and Fischman, D. A. (1982). Immunochemical analysis of myosin heavy chain during avian myogenesis in vivo and in vitro. J. Cell. Biol. 95, Bober, E., Lyons, G., Braun, T., Cossu, G., Buckingham, M. and Arnold, H. H. (1991). The muscle regulatory gene, Myf-6, has a biphasic pattern of expression during early mouse development. J. Cell Biol. 113, Bober, E., Franz, T., Arnold, H. H., Gruss, P. and Tremblay, P. (1994). Pax-3 is required for the development of limb muscles: a possible role for the migration of dermomyotomal muscle progenitor cells. Development 120, Brand-Saberi, B., Ebensperger, C., Wilting, J., Balling, R. and Christ, B. (1993). The ventralizing effect of the notochord on somite differentiation in chick embryos. Anat. Embryol. 188, Buckingham, M. (1992). Making muscle in mammals. Trends Genet. 8, Christ, B. (1970). Experimente zur Lageentwicklung der Somiten. Verh. Anat. Ges. 64, Christ, B., Jacob, H. J. and Jacob, M. (1972). Experimentelle Untersuchungen zur Somitenentstehung beim Hühnerembryo. Z. Anat. Entwicklungsgesch. 138, Christ, B., Brand-Saberi, B., Grim, M. and Wilting, J. (1992). Local signalling in dermomyotomal cell type specification. Anat. Embryol. 186, Christ, B. and Wilting, J. (1992). From somites to vertebral column. Ann. Anat. 174, Deutsch, U., Dressler, G. R. and Gruss, P. (1988). Pax-1, a member of a paired box homologous murine gene family, is expressed in segmented structures during development. Cell 53, Emerson, C. P. (1993). Skeletal myogenesis: genetics and embryology to the fore. Curr. Opin. Genet. Dev. 3, Erickson, C. A. and Perris, R. (1993). The role of cell-cell and cell-matrix interactions in the morphogenesis of the neural crest. Dev. Biol. 159, Fujisawa-Sehara, A., Nabeshima, Y., Hosoda, Y., Obinata, T. and Nabeshima, Y. (1990). Myogenin contains two domains conserved among myogenic factors. J. Biol. Chem. 265, Goulding, M. D., Lumsden, A. and Gruss, P. (1993). Signals from the notochord and floor plate regulate the region-specific expression of two Pax genes in the developing spinal cord. Development 117,

10 3082 E. Bober and others Gurdon, J. B., Lemaire, P. and Kato, K. (1993). Community effects and related phenomena in development. Cell 75, Griffith, C. M. and Sanders, E. J. (1991). Effects of extracellular matrix components on the differentiation of chick embryo tail bud mesenchyme in culture. Differentiation 47, Hamburger, V. and Hamilton, H. L. (1951). A series of normal stages in the development of the chick embryo. J. Morphol. 88, Jacob, H. J., Christ, B. and Jacob, M. (1974). Die Somitogenese beim Hühnerembryo. Experimente zur Lageentwicklung des Myotom. Verh. Anat. Ges. 68, Kaehn, K., Jacob, H.J., Christ, B., Hinrichsen, K. and Poelman, R.E. (1988). The onset of myotome formation in the chick. Anat. Embryol. 177, Kenny-Mobbs, T. and Thorogood, P. (1987). Autonomy of a differentiation in avian brachial somites and the influence of adjacent tissues. Development 100, Koseki, H., Wallin, J., Wilting, J., Mizutani, Y., Kispert, A., Ebensperger, C., Herrmann, B. G., Christ, B. and Balling, R. (1993). A role for Pax-1 as a mediator of notochord signals during the dorsoventral specification of vertebrae. Development 119, Krenn, V., Gorka, P., Wachtler, F., Christ, B. and Jacob, H. J. (1988). On the origin of cells determined to form skeletal muscle in avian embryos. Anat. Embryol. 179, Lin, Z.-Y., Dechesne, C. A., Eldridge, J. and Paterson, B. M. (1989). An avian muscle factor related to MyoD1 activates muscle-specific promoters in nonmuscle cells of different germ-layer origin and in BrdU-treated myoblasts. Genes Dev. 3, Mills, C. L., Ariyo, O., Yamada, K. M., Lash, J. W. and Bellairs, R. (1990). Evidence for the involvement of receptors for fibronectin in the promotion of chick segmentation. Anat. Embryol. 180, Ordahl, C. P. and Le Douarin, N. M. (1992). Two myogenic lineages within the developing somite. Development 114, 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, Pourquie, O., Coltey, M., Teillet, M.-A., Ordahl, C. and Le Douarin, N. M. (1993). Control of dorsoventral patterning of somitic derivatives by notochord and floor plate. Proc. Natl. Acad. Sci. USA 90, Pownall, M. E. and Emerson, C. P. (1992). Sequential activation of three myogenic regulatory genes during somite morphogenesis in quail embryos. Dev. Biol. 151, Primmett, D. R. N., Norris, W. E., Carlson, G. J., Keynes, R. J. and Stern, C. D. (1989). Periodic segmental anomalies induced by heat shock in the chick embryo are associated with the cell cycle. Development 105, Rong, P. M., Teillet, M.-A. and Ziller, C. (1990). Differential influence of the neural primordium on muscle development in myotome and limb. In The Avian Model in Developmental Biology: from Organism to Genes. (eds. N. M. Le Douarin, F. Dieterlen-Lievre and J. Smith), pp , Paris: Editions du CNRS. Rong, P. M., Teillet, M.-A., Ziller, C. and Le Douarin, N. M. (1992). The neural tube/notochord complex is necessary for vertebral but not limb and body wall striated muscle differentiation. Development 115, Sassoon, D., Garner, I. and Buckingham, M. E. (1988). Transcripts for α- cardiac and α-skeletal actins are early markers for myogenesis in the mouse embryo. Development 104, Sassoon, D., Lyons, G., Wright, W. E., Lin, V., Lassar, A., Weintraub, H. and Buckingham, M. (1989). Expression of two myogenic regulatory factors myogenin and MyoDI during mouse embryogenesis. Nature 341, Strudel, G. (1955). L`action morphogene du tube nerveux et la chorde sur la differenciation des vertebres et das muscles vertebraux chez l`embryon de poulet. Arch. Anat. micr. Morph. exp. 44, Tam, P. P. L. and Tan, S.-S. (1992). The somitogenetic potential of cells in the primitive streak and the tail bud of the organogenesis-stage mouse embryo. Development 115, Teillet, M.-A. and Le Douarin, N. M. (1983). Consequences of neural tube and notochord excision on the development of the peripheral nervous system in the chick embryo. Dev. Biol. 98, Vivarelli, E. and Cossu, G. (1986). Neural control of early myogenic differentiation in culture of mouse somites. Dev. Biol. 117, von Kirschhofer, K., Grim, M., Christ, B. and Wachtler, F. (1994). Emergence of myogenic and endothelial cell lineages in avian embryos. Dev. Biol. 163, Wachtler, F., Christ, B. and Jacob, H. J. (1982). Grafting experiments on determination and migratory behavior of presomitic, somitic and somatopleural cells in avian embryos. Anat. Embryol. 164, Watterson, R. L., Fowler, I. and Fowler, B.J. (1954). The role of the neural tube and notochord in development of the axial skeleton in the chick. Am. J. Anat. 95, Williams, B. A. and Ordahl, C. P. (1994). Pax-3 expression in segmental mesoderm marks early stages in myogenic cell specification. Development 120, (Accepted 1 August 1994)