Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
1443 Development 120, 1443-1452 (1994) Printed in Great Britain © The Company of Biologists Limited 1994 Myogenic specification in somites: induction by axial structures Nicholas Buffinger and Frank E. Stockdale* Stanford University, School of Medicine, Room M211, Stanford CA 94305-5306, USA *Corresponding author SUMMARY Specification of the myogenic phenotype in somites was examined in the early chick embryo using organotypic explant cultures stained with monoclonal antibodies to myosin heavy chain. It was found that myogenic specification (formation of muscle fibers in explants of somites or segmental plates cultured alone) does not occur until Hamburger and Hamilton stage 11 (12-14 somites). At this stage, only the somites in the rostral half of the embryo are myogenically specified. By Hamburger and Hamilton stage 12 (15-17 somites), the three most caudal somites were not specified to be myogenic while most or all of the more rostral somites are specified to myogenesis. Somites from older embryos (stage 13-15, 18-26 somites) showed the same pattern of myogenic specification — all but the three most caudal somites were specified. We investigated the effects of the axial structures, the notochord and neural tube, on myogenic specification. Both the notochord and neural tube were able to induce myogenesis in unspecified somites. In contrast, the neural tube, but not the notochord, was able to induce myogenesis in explants of segmental plate, a structure which is not myogenic when cultured alone. When explants of specified somites were stained with antibodies to slow or fast MyHC, it was found that myofiber diversity (fast and fast slow fibers) was established very early in development (as early as Hamburger and Hamilton stage 11). We also found fiber diversity in explants of unspecified somites (the three most caudal somites from stage 11 to 15) when they were recombined with notochord or neural tube. We conclude that myogenic specification in the embryo results in diverse fiber types and is an inductive process which is mediated by factors produced by the neural tube and notochord. INTRODUCTION (Teillet and Le Douarin, 1983; Rong et al., 1992), though the subsequent development of the musculature of the appendages seems unaffected. Interestingly, the notochord appears able to substitute for the neural tube in that it will support maturation of somites (Rong et al., 1992), but the implantation of ectopic notochord is reported to inhibit myogenesis in somites (Pourquie et al., 1993). Using an in vitro model, Kenny-Mobbs and Thorogood (1987) showed that the neural tube/notochord complex can induce muscle fiber formation in explants of brachial somites. Understanding the molecular events leading to myogenic differentiation has been advanced in recent years by the discovery of the MyoD family of myogenic regulatory factors (MRFs) (Davis et al., 1987; Wright et al., 1989; Edmondson and Olson, 1989; Braun et al., 1989; Rhodes and Konieczny, 1989). Any of the members of this family (MyoD, myogenin, myf5 and MRF4) can convert many non-myogenic cells to a myogenic phenotype in vitro. Studies of early embryos show that MRFs are expressed in somites long before there are overt signs of myogenic differentiation (Sassoon et al., 1989; Charles de la Brousse and Emerson, 1990; Piette et al., 1992). However, the exact role these genes play in myogenesis in vivo is unclear. Gene knock out experiments in transgenic mice have yielded mixed results. Mice lacking the MyoD gene have a normal phenotype, and the major defect in myf5 knock out Inductive interactions play key roles throughout early development of vertebrate embryos. Indeed, the formation of one of the three germ layers in amphibia, the mesoderm, is the result of an inductive interaction between the animal and vegetal hemispheres of the embryo (Nieuwkoop, 1969). Later in development, inductive events are required for the formation of many of the tissues and organs of the embryo. The formation of the heart, optic lens, kidney and vertebral cartilage are all dependent on inductive events (Bacon, 1945; Spemann, 1901; Grobstein, 1955; Holtzer and Detwiler, 1953). The axial structures of the embryo, the neural tube and notochord, are important elements in several of these inductive interactions. The neural tube is involved in the induction of kidney and vertebral cartilage while the notochord also plays a role in vertebral cartilage induction, as well as in specification of cell fate in the neural tube (Yamada et al., 1993). The axial structures have also been implicated in the induction of somitic myogenesis. The somites, which lie directly adjacent to the neural tube and notochord, are the source of all of the skeletal muscle precursor cells of the trunk and limbs of the developing vertebrate embryo (Chevallier et al., 1977). Extirpation of the neural tube and notochord results in the absence of myotomal structures in the adjacent somites Key words: mesoderm, neural tube, notochord, muscle, somite, induction, myogenesis 1444 N. Buffinger and F. E. Stockdale mice is confined to abnormalities of the skeleton (Rudnicki et al., 1992; Braun et al., 1992), although at least one of the two genes is required, as mice lacking both genes have a complete absence of muscle formation (Rudnicki et al., 1993). The myogenin gene also appears to be essential for normal myogenesis, as two groups have shown that transgenic mice lacking the myogenin gene form muscle fibers, but exhibit gross defects in muscle formation and die as embryos (Hasty et al., 1993; Nabeshima et al., 1993). In contrast, it has been reported that there is a population of myoblasts in the somites of early mouse which, when cultured, differentiate even though they do not express MyoD or myogenin proteins (Cusella-De Angelis et al., 1992). To better define the earliest events of myogenesis, we have used an in vitro organ culture system to examine when somitic cells are first specified to myogenesis and which tissues are involved in inducing the myogenic phenotype in somitic mesoderm. We found that myogenic cells first appear in explants formed from somites from the 13-somite embryo (Hamburger and Hamilton stage 11), and that the neural tube or notochord are able to induce myogenesis in somites. We have also found that the segmental plates and the three most newly formed somites, which do not autonomously form muscle fibers in vitro (stages 10-15), make an excellent test system for further characterization of myogenic induction by the neural tube or notochord. MATERIALS AND METHODS Nomenclature Somites are numbered utilizing a system that refers to the age of a somite rather than the position of a somite. Somites are numbered, using roman numerals, from the caudal end of the embryo to the rostral end, with the most newly formed somite designated as somite I, regardless of the stage of the embryo. Somites are designated as two types, depending on whether muscle fibers form within them when explanted in vitro. Somites are designated as specified somites if when incubated alone as explants, they form muscle fibers. When somites form muscle fibers only when recombined with other tissues in vitro, such as the notochord or neural tube, they are designated as unspecified somites. Explant culture Fertile chicken eggs (white Leghorn) were obtained from a local supplier (Western Scientific, Sacramento, CA) and incubated at 38°C for the time appropriate for the desired stage. Embryos were staged according to Hamburger and Hamilton (1951). Eggs were cracked into a dish containing Pannett and Compton saline (Pannett and Compton, 1924) and embryos were freed from the yolk, transferred to a dish containing Tyrode’s solution and staged. Embryos of the appropriate stages were transferred to a second dish and held in place with the aid of a glass cloning ring. Several drops of a 2.5% solution of trypsin (Difco, 1:250) in saline G (Gibco) were added to the surface of the embryo. After a brief period of digestion, the trypsin was diluted by the addition of a large volume of Pannett and Compton saline. Segmental plates, somites and portions of the neural tube and notochord were dissected using finely sharpened stainless steel insect pins. The dissected structures were transferred to dishes containing medium (10% horse serum (Gibco), 5% chick embryo extract, with penicillin and streptomycin in DMEM) until the explant cultures could be formed. Explant cultures were made in 12well tissue culture plates (Falcon or Costar) which had been lined with a 1:1 mixture of medium and 0.5% agarose. Several types of explant cultures were performed: individual somites were recombined in clusters of 10 somites (Fig. 1A); 4 somites were recombined with a small segment of either notochord (Fig. 1B) or neural tube (Fig. 1C) from the same staged embryo; 4 individual segmental plates were cultured alone (Fig. 1D); or 2 segmental plates were recombined with notochord (Fig. 1E) or neural tube (Fig. 1F) from the same stage embryo. Unless otherwise stated, the notochord and neural tube used in the explants was from the same stage embryo as the somites. The segments of neural tube and notochord used in the explants were from portions adjacent to the somites used in the explants. The individual tissue elements in an explant fused together into a single structure within 2 hours of being placed in culture. At this time sufficient medium to cover the explant was added. Cultures were incubated at 37°C for 4 days, with feeding every day. Immunohistochemistry At the end of the culture period, the explants were transferred to small glass vials for whole-mount immunohistochemical staining. The explants were washed once with Pannett and Compton saline and fixed for 30 minutes at room temperature in 70% ethanol. The explants were washed twice with TBS, then incubated in a blocking solution of 20% horse serum in TBS for 2 to 3 hours. Hybridoma supernatants containing monoclonal antibodies (mAb) for fast (mAb F59) and/or slow (mAb S58) myosin heavy chain (Crow and Stockdale, 1986) were added directly to the blocking solution to a final concentration of 10% each. Explants were incubated overnight at 4°C in primary antibodies, then washed 5 to 6 times with TBS. Explants were incubated with secondary antibodies (fluorescein-conjugated anti-mouse IgA and Texas red-conjugated anti-mouse IgG, Vector Labs and Zymed), diluted 1:100 or 1:50 in 20% horse serum, overnight at 4°C. Explants were washed 5 to 6 times with TBS, and mounted under a coverslip with 2.5% DABCO in PBS/glycerol 1:9. Explants were surveyed under epifluorescent optics, and images were collected with a laser scanning confocal microscope. RESULTS Myogenic specification in paraxial mesoderm Evidence of specification of mesodermal cells to a myogenic fate was sought in explant cultures of early stage somites or segmental plates. Explants of paraxial mesoderm tissues (somites or segmental plate) were isolated from the early chick embryo and cultured in isolation in an organ culture system that preserved tissue integrity. Explants of somites (the source of skeletal muscle precursor cells of the limb and trunk in the embryo) and segmental plates (the source of somites) were assayed (Fig. 1). In each experiment, somites or segmental plates were dissected from 5-10 embryos of a specific stage and were held as separate pools of segmental plates or numbered somites in separate dishes containing culture medium until use. Explants of 4 segmental plates or 10 somites of the same numbered group were assembled in random orientation. It should be noted that individual somites or segmental plates were dissected free from other somites before recombination into explants. The explants were incubated on agarose pads that did not permit outgrowth and were assayed for myogenesis at day 4 by staining as whole mounts for myosin heavy chains using monoclonal antibodies directed against fast (mAb F59) and slow (mAb S58) isoforms of myosin heavy chain (MyHC). Myogenesis was first detected in explanted somites from stage 11 embryos, whereas myogenesis did not occur in explants of segmental plate from any aged embryo (Fig. 2). No Myogenic induction in somites 1445 A B C D E F Fig. 1. Somite nomenclature and types of explant cultures. Somites are numbered from the caudal to the rostral end of the embryo. Only the 12 most caudal somites are used except for stage 10 which has only 9-11 somites. (A) Somites alone; (D) segmental plate alone; (B) somites plus notochord; (E) segmental plate plus notochord; (C) somites plus neural tube; (B) somites plus notochord. explant of somites from the stage 10 embryo, nor explants of segmental plates from stage 10-15 embryos formed muscle fibers when cultured alone. Only one segmental plate explant of the 50 studied was positive for myosin heavy chain (MyHC) at the end of the culture period; this one was most likely due to the inadvertent inclusion of a small amount of the neural tube with segmental plate during dissection. Because somites form in a rostral-caudal fashion, there exists a gradient of ages in the somites of an embryo. We divided somites into four groups of three somites each and tested each group of three somites from stage 10 (9-11 somite stage) through stage 15 (24-26 somite stage). The groups were chosen with regard to the time elapsed since the somites had formed from the segmental plate. We have used a numbering system for the somites which reflects this, in which the most newly formed somite at any given stage is designated somite I, the second youngest somite is somite II, and so on, counting from the caudal to the rostral end of the series of somites. The groups were: somites I-III, somites IV-VI, somites VII-IX, and somites X-XII (for stage 10 embryos, the latter group consisted of somites X-XI only). Ten somites from a particular group were recombined in explant culture. A total of 26 explants were assayed from stage 10 embryos, none of which stained for MyHC. Stage 11 (12-14 somite stage) was the first stage at which somite cells specified to a myogenic fate were detected in explant culture. Only explants formed from stage 11 somites VII-IX and X-XII were positive for MyHC after the 4 day culture period. By stage 12, there was an expansion in the groups of somites which formed myogenic cells when placed Cumulative percentage of explants that express MyHC 0 100 Somites X-XII Somites VII-IX Somites IV-VI Somites I-III Segmental Plate Stage 10 36 11 12 13 14 15 200 300 400 500 51 36 41 24 29 10 11 12 13 14 15 Alone 29 35 45 28 38 32 With notochord 17 22 21 11 5 5 10 11 12 13 14 15 0 100 200 300 400 500 With neural tube Fig. 2. Myogenesis in explants of paraxial mesoderm from stages 1015. Myogenesis was assayed by whole-mount staining with mAb F59, which stains fast MyHC(s). All four groups of somites and the segmental plate were assayed at each stage for each type of explant. Note that addition of notochord to the explants induces myogenesis in somites I-III from stages 11-15, and in the three rostral groups of somites from stage 10. Neural tube induces myogenesis in all unspecified somites as well as the segmental plate. The number at the top of each column indicates the total number of explants assayed. 1446 N. Buffinger and F. E. Stockdale in explant culture. At this stage, explants of somites IV-VI, as bined with segments of the notochord. Also, explants of stage well as VII-IX and X-XII, were now positive for MyHC when 10 somites I-III recombined with notochord did not produce cultured alone (Fig. 3). The results of explant cultures for muscle fibers. Nevertheless, notochord induced myogenesis in stages 13, 14, and 15 were similar to those of stage 12 (Fig. 2, all other groups of somites that did not exhibit myogenesis Alone). when cultured alone. These included somite groups VI-VI, We refer to groups of somites as either ‘specified’ or VII-IX, and X-XI from stage 10 and all unspecified somites ‘unspecified’. Those groups of somites that did not form from stages 11-15. myogenic cells when cultured alone (all somites from stage 10, Somites formed in isolation from the neural tube can be somites I-VI from stage 11, and somites I-III from stages 12induced to be myogenic by the notochord. To determine if mat15) are unspecified, and groups of somites that did form uration of the segmental plate must occur in the presence of myogenic cells (somites VII-XII from stage 11 and somites IVthe neural tube for the resulting somites to be responsive to a XII from stage 12-15) are specified. It should be noted that myogenic signal from the notochord, the segmental plate was somites I-III at stages 10-15 did not autonomously form allowed to form somites in isolation from the neural tube and myogenic cells when recombined in explant culture. Only one of 43 explants of somites I-III from stage 1015 embryos was found to be positive for MyHC at the end of the culture period. Thus only after somites segmented from the segmental plate and matured beyond Somites the third somite position did specification to myogenesis X-XII occur. Myogenic induction by axial structures To determine the role of tissue-tissue interactions in myogenic specification, we made recombinant explant cultures, recombining isolated segmental plates or somites with isolated segments of neural tube or notochord from the same stage embryo (Fig. 1). Cells in explants containing neural tube in combination with segmental plate or with any group of somites from all stages tested were always myogenic. Somites from any of the groups of unspecified somites at stage 11-15 and including all somites at stage 10, formed muscle fibers when recombined with neural tube. Segmental plate from stage 10-15 also formed myogenic cells when recombined in explant culture with neural tube (Fig. 2, with Neural Tube; Fig. 4 C,F). The segments of neural tube used for the explants were isolated from levels adjacent to the somites. These observations suggest that the neural tube induces myogenic specification in segmental plate and unspecified somites. To determine if only paraxial mesoderm can be specified to a myogenic fate by interacting with the neural tube, we cultured recombined isolated lateral plate mesoderm (including both somatopleure and splanchnopleure) with segments of the neural tube. Only three of 20 such explants were positive for MyHC (data not shown). The staining for MyHC in these three positive explants was confined to single small patches of cells, which we believe are the result of contamination of the lateral plate explants with small numbers of somitic cells. This observation suggests that only the paraxial mesoderm, rather than all mesoderm, is responsive to the neural tube-derived myogenic signal. We found that the notochord was also able to initiate myogenesis when combined with unspecified somites (Fig. 2, with Notochord; Fig. 4B). However, the myogenic-inducing effect of the notochord differed from that of the neural tube, in that the notochord did not induce myogenesis in explants of segmental plates (Fig. 4E) or explants of one group of somites. Segmental plates from all stages did not form muscle cells when recom- A B C D E F G H I J Somites VII-IX Somites IV-VI Somites I-III Segmental Plate Fig. 3. Myogenic specification in explants of somites or segmental plates of stage 12 embryos. Differential interference contrast (A,C,E,G,I)) or confocal fluorescence (B,D,F,H,J)) microscopy of explants double stained with monoclonal antibodies to myosin heavy chain. (I,J) Segmental plate; (G,H) somites I-III; (E,F) somites IV-VI; (C,D) somites VII-IX; (A,B) somites X-XII. Explants of specified somites (B,D,F) all stained with an antibody to fast MyHC (mAb F59), as indicated by red fluorescence. Note that staining for slow MyHC (mAb S58, green fluorescence) is found principally in explants with a large number of fibers (B,D). Images were collected on a laser scanning confocal microscope with a 20× objective. Myogenic induction in somites 1447 With Notochord Alone A B With Neural Tube C A B C D D EE FF Somites I-III Segmental Plate Fig. 4. Induction of myogenesis in unspecified somites by notochord and neural tube. Somites I-III (A) and the segmental plate (D) are unspecified, as shown by the absence of staining for MyHC. The notochord induces myogenesis in somites I-III (B), but not the segmental plate (E). The neural tube induces myogenesis in somites I-III (C) and the segmental plate (F). Explants were formed with the neural tube or notochord flanked by 4 somites (2 on each side) and incubated for 4 days. The explants were whole mounted double stained with antibodies to fast and slow MyHC (mAb F59 and mAb S58), and detected with Texas Red- and FITC-labeled secondary antibodies, respectively. Images were collected on a laser scanning confocal microscope with a 20× objective. notochord. In these experiments, portions of embryos containing the segmental plate and lateral plate mesoderm, and the over- and underlying ectoderm and endoderm, but lacking the neural tube and notochord, were isolated and permitted to segment in vitro (Packard and Jacobson, 1976). The somites formed in these cultures were then dissected free from the ectoderm, endoderm and lateral plate mesenchyme and were recombined with segments of notochord. These somites became myogenic when they were recombined with the notochord in explant culture (7 of 9 explants). The remaining unsegmented segmental plates were also recombined with the notochord and none of these explants formed muscle fibers (n=5) (data not shown). These observations suggest that association of the Fig. 5. Muscle fiber diversity in an explant of unspecified somites recombined with neural tube. Somites of group I-III from a stage 12 embryo were recombined with neural tube in organ culture for 4 days, then whole-mount stained with antibodies to fast and slow MyHC (mAb F59 and mAb S58). Fluorescein indicates slow MyHC and Texas Red indicates fast MyHC(s). Note the presence of a fast fiber (carets), a fast/slow fiber (arrows) and a slow fiber (arrowheads). This image was collected on a laser scanning confocal microscope with a 60× objective. 1448 N. Buffinger and F. E. Stockdale segmental plate with neural tube during segmentation of somites is not required for somites to acquire responsiveness to myogenic specification by notochord, although we could not determine if it is segmentation per se, or a maturation event, that permits segmental plates that undergo segmentation in vitro to respond to the notochord. Muscle fiber diversity in somite and segmental plate explants It is proposed that diversity among muscle fibers is an early event in myogenesis and this early diversity has its origins in the diversity of myoblasts (Stockdale, 1992). It has been shown in both birds and mammals that myoblasts within the early embryo have different fates because one can isolate clonal populations of myoblasts that form colonies of muscle fibers in cell culture that express fast or fast and slow isoforms of MyHC (Miller and Stockdale, 1986a,b; Stockdale and Miller, 1987; Vivarelli et al., 1988). To determine whether the muscle fibers formed within explants (autonomously or in response to neural tube or notochord) were of diverse types, specified somites cultured alone, and unspecified somites cultured with neural tube or notochord, were stained with antibodies to fast and slow isoforms of MyHC. The muscle fibers that formed autonomously in specified somites in explant culture always contained fast isoforms of MyHC, and most, but not all explants also contained muscle fibers that expressed slow MyHC isoforms (Fig. 6, Alone). Explants of specified somites from stage 12-15 embryos expressed fast isoforms of MyHC in 100% of the explants and slow MyHC in about 88% of the explants (42 of 48 expressed slow MyHCs). However, explants of specified somites X-XII from the earlier stage 11 embryo, and explants of specified somites that formed very small, rather than large numbers of fibers, generally did not express slow MyHC (Fig. 3F). It appears that explants which contained many fibers were more likely to contain fibers that express slow MyHC (Fig. 3B,D). Thus when specified somites undergo myogenesis in organ culture, they autonomously form fibers of more than one type. When unspecified somites or segmental plates were recombined with segments of neural tube, they always expressed slow and fast MyHCs (80 of 80 explants) (Fig. 5, 6). However, when explants of unspecified somites were recombined with notochord (somites I-III from stage 11-15), fewer muscle fibers formed than with neural tube, and the percentage of explants expressing slow MyHC was lower (24 of 36 explants, 66%). Thus, in cultures of somites that have been specified in vivo, or segmental plate or unspecified somites that become specified in vitro in response to neural tube or notochord, both fast and fast/slow fibers appear. The frequency of differing types of fibers appears to correlate, in general, with the number of fibers that form in the explants. DISCUSSION These studies show that specification to myogenesis of unspecified somites from the early chick embryo can be induced by the neural tube or the notochord. Axial mesenchyme need not be segmented into somites for it to become specified to a myogenic fate by interaction with the neural tube, although Cumulative percentage of Explants that express slow MyHC 0 100 200 300 400 500 Somites X-XII Somites VII-IX Somites IV-VI Somites I-III Segmental Plate Stage 10 11 12 13 14 15 Alone 10 11 12 13 14 15 With notochord 10 11 12 13 14 15 With neural tube 0 100 200 300 400 500 Fig. 6. Muscle fiber diversity in explants of paraxial mesoderm. Muscle fiber diversity is indicated by expression of slow MyHC, which is detected by staining with mAb S58. Note that in explants of somites alone, muscle fiber diversity is found principally in explants of the more rostral somites — explants which produce a large number of fibers. Explants with notochord have a higher incidence of muscle fiber diversity. Explants with the neural tube showed muscle fiber diversity and larger numbers of fibers in almost all cases. Myogenic induction in somites 1449 only after segmentation of the segmental plate does the notochord induce myogenesis. The muscle fibers that form in segmental plate or somite explants in response to the neural tube and notochord express fast or fast and slow isoforms of myosin heavy chain. None of the somites of the stage 10 embryo will form muscle fibers when placed in explant culture. Somites that form muscle fibers autonomously in explant culture are first found in the rostral somites at stage 11 of development. We designate these somites as specified somites, and those that do not form muscle fibers when placed in explant culture as unspecified somites. None of the somites of the stage 10 embryo are specified when cultured alone, but the rostral somites (somite group VII-IX and X-XII) of stage 11 embryos are. At later stages (stages 12-15), the pattern of specification expands, and somite groups IV-VI become specified. Because the somites are pooled into groups, we cannot categorize a specific somite within a somite group which lies at the unspecified/specified somite boundary (the boundary between somites I-III and somites IV, V or VI at stages 12-15). Although as a group somites IV-VI are specified, it could be that the only somite VI or somites V and VI are specified, as might be inferred by the small number of fibers observed in explants of this group. The three most newly formed somites are not specified at any stage tested (stage 10 through 15). We also found that the segmental plate is not specified at any of the stages tested. Thus from stage 11 through 15 only the segmental plate and the three youngest somites are unspecified — all other somites autonomously form muscle fibers in explant culture. Thus beyond stage 11, there do not appear to be stage-specific events that lead to specification of a somite, but only events associated with the ‘age’ of a somite. On the other hand, before stage 11 all somites tested were unspecified. Therefore, there are two phases in myogenic specification of somites, that in which the embryo matures past stage 10, and that in which the somite matures past somite III. That we find unspecified somites at any stage indicates that myogenesis is not a ‘default’ fate for somitic cells as has been recently suggested (Pourquie et al., 1993). The results reported here are in good agreement with the work of Kenny-Mobbs and Thorogood (1987) and Rong and co-workers (1992). Kenny-Mobbs and Thorogood used several explant culture systems to investigate the myogenic specification of the six brachial somites (somites number 15-21) in stages 12, 15, and 18 embryos. They found that the brachial somites at stage 12 (which would be somite group I-III) were not myogenic when cultured alone, and that the brachial somites from stage 15 (which would be included in somite groups IV-VI, VII-IX, and X-XII) were myogenic. Rong and co-workers (1992) used a monolayer tissue culture system to investigate myogenic specification of somites. Their results for somites from stage 10-13 embryos are virtually identical to ours. However, they report slightly different findings on somites from older embryos. While we do not see myogenesis in somites I-III from stage 14 and 15 embryos, Rong et al. (1992) find small numbers of cells which stain with a monoclonal antibody to an unidentified protein expressed in smooth as well as striated muscle. The basis for this difference is unclear, but may be due to the markers of myogenesis used in the different experiments. Explants of segmental plate and the first three somites of early embryos thus become an excellent test system to determine the importance of tissue interactions in specification of somites to myogenesis. The neural tube and the notochord can induce myogenesis in unspecified somites of the early embryo. However, there are differences in the inductive effects of the neural tube and notochord. The neural tube can induce myogenesis in all unspecified somites and segmental plates tested from stage 10-15 embryos. The notochord can induce myogenesis in all somites tested from stage 11-15 embryos, but only in the rostral somites (somite groups IV-VI, VII-IX and X-XI) of stage 10 embryos and not in segmental plate of any staged embryo. It is not clear if this indicates a quantitative (where the neural tube produces more of the myogenic factor(s) than the notochord) or a qualitative (where the neural tube and notochord produce different myogenic signals) difference in the myogenic signals from the neural tube and notochord. The differences between the effects of the neural tube and notochord highlight another distinction between the results reported here and those reported by Rong and coworkers (1992). While Rong et al. (1992) find that the notochord and neural tube are equivalent in initiating myogenesis, we find that the myogenic effects of the neural tube have a broader potency compared to the effects of the notochord. However, both the notochord and neural tube appear to have a positive effect on somitic myogenesis. Jessell’s laboratory (Basler et al., 1993; Yamada et al., 1993) has described signals involved in specification of cell fate in the developing neural tube that occur in the same developmental time frame as those described here. These signals, some diffusible and others cell-cell contact-mediated, are produced by both the notochord and neural tube. Specification of the dorsal-ventral pattern within the neural tube may be analogous, if not related to, the system of signals involved in the specification of myogenesis in somites. It has been shown in the quail embryo that somites II and III express mRNA for qmf1 (the quail homologue of MyoD) as early as stage 12 (Pownall and Emerson, 1992). However, none of the explants formed from somites I-III from stage 12 embryos, or any staged embryo tested, formed muscle fibers when cultured alone. This suggests that expression of MyoD, at least at the RNA level, is not sufficient for myogenic specification in vitro. It may be that the signals from the neural tube and notochord sustain the continuous expression of MyoD in developing somites, suggesting that the function of the notochord and neural tube is to maintain myogenic regulatory gene expression. If this is the case, then after their removal from the embryo, somites I-III would stop expressing MyoD resulting in their subsequent failure to undergo myogenesis. This implies that either the MyoD mRNA is not translated in the early somites or that the cross- and auto-activation by MyoD shown to operate in tissue culture model systems (Thayer et al., 1989) may not be operative in the animal, or that certain thresholds of myogenic regulatory factor protein are required to initiate cross and auto-activation (reviewed in Weintraub, 1993). Another possibility is that continued induction by the neural tube and notochord is required to either turn off an inhibitory factor, such as helix-loop-helix protein inhibitor Id (Benezra et al., 1990) or to turn on a second required myogenic factor, for example myogenin (Hollenberg et al., 1993). While the results described here can neither support nor refute any of these possibilities, results from John 1450 N. Buffinger and F. E. Stockdale Gurdon’s laboratory (Hopwood et al., 1989, 1991; Hopwood and Gurdon, 1990) suggest that this phenomenon is not restricted to avian embryos. In early Xenopus embryos, both MyoD and myf5 are expressed even before somites form, and forced expression of either or both of these genes in mesoderm is not sufficient to induce myogenesis. Our data suggest that myogenic specification in paraxial mesoderm by neural tube does not require segmentation. Cells from segmental plates undergo myogenesis when recombined with the neural tube even though they have not segmented into somites. It seems unlikely that segmentation is occurring in the paraxial mesoderm of these explants, as segmental plates isolated from ectoderm and endoderm do not segment under our culture conditions (Buffinger, unpublished data). However, it appears that either a maturational event or segmentation of the segmental plate is required for specification of myogenesis by the notochord. The notochord does not induce myogenesis when recombined with segmental plates, but will induce myogenesis in most somites from stage 10 and all somites from stage 11-15. In addition, somites formed when segmental plates are incubated in isolation from the neural tube and notochord are induced to form muscle by the notochord in explant culture. These data also demonstrate a difference between the sensitivity of segmental plate and somites to the myogenic factor(s) coming from the neural tube and notochord. It is currently unclear whether these differences between somite and segmental plate response are due to differences in the myogenic factors produced by notochord and neural tube, or due to differences in sensitivity of somites and segmental plate to the same factor. We found muscle fiber type diversity in explants of specified somites as well as in somites combined with neural tube or notochord. Muscle fibers that formed in explants were almost always of different types — those that expressed only fast and those that expressed both fast and slow MyHC isoforms. Using explants of specified somites cultured alone, slow MyHC (as indicated by staining by mAb S58) was detected as early as stage 11, the stage at which myogenic specification is first observed. We found that, when large number of fibers were formed in explants of specified somites cultured alone, there usually were muscle fibers that contained slow MyHCs. When unspecified somites from stage 10 embryos were cultured with notochord or neural tube, muscle fibers appeared that stained with mAb S58, indicating that induction produces fibers of diverse types. Even in explants of segmental plate from stages as early as stage 10 recombined with neural tube, muscle fibers were produced that expressed slow MyHC. Models of myogenic induction One model for the induction of myotomal muscle implicates neural crest cells in myogenic induction (Christ et al., 1992). These authors found that ablation of the neural tube (and thus the neural crest) at early stages led to the absence of a myotome in adjacent somites. They also noted that myotome formation is preceded by invasion of neural crest cells (Bronner-Fraser, 1986) and that neural crest cells migrate through the cranial half of the somite (Rickmann et al., 1985), which coincides with the portion of the somite in which the myotome begins to form (Christ et al., 1992). More recently, Pourquie and colleagues (1993) have implicated the notochord in dorsal/ventral patterning of the somite. Using segments of notochord grafted ectopically into early chick embryos, they found that myogenesis in the myotome was blocked, and that the entire somite became chondrogenic. They found the same effects when ectopic floor plate was grafted to the lateral or dorsal aspect of the neural tube. Combining these observations with those of van Straaten and Hekking (1991), which showed that ablation of the notochord had no effect on the differentiation of the myotome, Pourquie et al. (1993) concluded that ‘dorsal’ structures of the somite (the myotome and dermatome) form by default, that is that the notochord, by inducing the floor plate, controls normal morphogenesis in the somite by inducing structures that are ‘ventral’ in character (i.e. sclerotome). Neither model is sufficient to explain the specification of myogenesis in somites reported here. Our experiments show that myogenesis can occur in somites which do not contain neural crest cells. It has been demonstrated by several techniques that neural crest cells begin to emigrate from the neural tube into somites that are rostral to the most newly formed somites (i.e. they migrate in at somite IV) (Weston and Butler, 1966; Bronner-Fraser, 1986; Serbedzija et al., 1989). However, we have shown that notochord can induce myogenesis in explants of both somites I-III, as well as somites that were allowed to form from the segmental plate in isolation from the neural tube (and thus neural crest cells), demonstrating somitic myogenesis in the absence of neural crest cells. This, however, does not rule out a role for neural crest cells in the morphogenesis of the myotome. The observations reported here support a positive role for the notochord in initiating myogenesis in somites. Our findings that neither somites I-III nor the segmental plate form muscle fibers when cultured in isolation from the notochord, suggest that the myotomal muscle does not form as a default fate for cells of the somite. In addition, the notochord is able to induce muscle fiber formation in these unspecified somites. These findings indicate that the notochord is able to act on its own in somite cell specification, and is not restricted solely to the induction of somites cells to ‘ventral’ type fates. We think that a complex pattern of signals is produced by the neural tube and notochord which compete to specify cell fates in the somites. We have formulated a model, based on our findings as well as those of Pourquie et al. (1993) and van Straaten and Hekking (1991). The results of Pourquie and colleagues show that induction of an additional floor plate leads to the induction of a wholly cartilaginous somite, suggesting that the floor plate produces (or causes to be produced) a cartilage-inducing factor that can induce cartilage (sclerotome) at the expense of muscle (myotome). Extirpation of the notochord leads to abnormal somites in the area of the ablation. The somites that form in the absence of the notochord and floor plate, lack a sclerotome, but do form a myotome (van Straaten and Hekking, 1991), suggesting that, while induction of the sclerotome requires the presence of the notochord and/or floor plate, myogenic induction of the myotome does not. However, we have shown that both the notochord and neural tube are able to induce myogenesis in explants of somites that would not form muscle if cultured alone. These observations lead us to propose a model in which factors that induce cartilage and factors that induce myogenesis compete to specify the fates of cells in the somites. Cells that lie near the notochord and neural tube are induced to form cartilage by non-diffusible factors Myogenic induction in somites 1451 from the notochord and factors that require close proximity for transmission from the neural tube (Lash et al., 1957). Somitic cells more distant from the notochord and ventral neural tube are specified to myogenesis by diffusible factors produced by the neural tube and notochord. Our experimental results confirm and extend the previously reported work of Kenny-Mobbs and Thorogood (1987) and Rong et al. (1992). We have shown that either the neural tube or notochord can induce myogenesis in somites that would not otherwise undergo myogenesis when cultured alone. Both the neural tube and notochord are also able to induce diverse types of muscle fibers. Moreover, the neural tube is able to induce myogenesis in paraxial mesoderm in the absence of segmentation of the paraxial mesoderm. These observations indicate that myogenic-inducing signals are communicated from neural tube and notochord to somite cells. We thank Sandra Conlon for technical assistance, and Drs W. Nikovits and J. DiMario for helpful discussions. This investigation was supported by PHS training grant number 5T32CA09302 awarded by the National Cancer Institute, DHHS and a research grant from the DHHS. REFERENCES Bacon, R. L. (1945). Self-differentiation and induction in the heart of Amblystoma. J. Exp. Zool. 98, 87-125. Basler, K., Edlund, T., Jessel, T. M. and Yamada, T. (1993). Control of cell pattern in the neural tube: regulation of cell differentiation by dorsalin-1, a novel TGFb family fember. Cell 73, 687-702. Benezra, R., Davis, R. L., Lockshon, D., Turner, D. L. and Weintraub, H. (1990). The protein id: a negative regulator of helix-loop-helix DNA binding proteins. Cell 61, 49-59. Braun, T., Buschhausen-Denker, G., Bober, E., Tannich, E. and Arnold, H. H. (1989). A novel human muscle factor related to but distinct from MyoD1 induces myogenic conversion in 10T1/2 fibroblasts. EMBO J. 8, 701-709. Braun, T., Rudnicki, M. A., Arnold, H.-H. and Jaenisch, R. (1992). Targeted inactivation of the muscle regulatory gene Myf-5 results in abnormal rib development and perinatal death. Cell 71, 369-382. Bronner-Fraser, M. (1986). Analysis of neural crest cell lineage and migration. Dev. Biol. 115, 44-55. Charles de la Brousse, F. C. and Emerson, C. P., Jr. (1990). Localized expression of a myogenic regulatory gene, qmf1, in the somite dermatome of avian embryos. Genes Dev. 4, 567-581. Chevallier, A., M. Kieny and A. Mauger (1977). Limb-somite relationships: origin of the limb musculature. J. Embryol. Exp. Morph. 41, 245-258. Christ, B., Brand-Saberi, B., Grim, M. and Wilting, J. (1992). Local Signaling in dermomyotomal cell type specification. Anat. Embryol. 186, 505-510. Crow, M. T. and Stockdale, F. E. (1986). Myosin expression and specialization among the earliest muscle fibers of the developing avian limb. Dev. Biol. 113, 238-254. Cusella-De Angelis, M. G., Lyons, G., Sonnino, C., De Angelis, L., Vivarelli, E., Farmer, K., Wright, W. E., Molinaro, M., Bouche, M., M. Buckingham, M. and Cossu, G. (1992). MyoD, myogenin independent differentiation of primordial myoblasts in mouse somites. J. Cell. Biol. 116, 1243-1255. Davis, R. L., Weintraub, H. and Lassar, A. B. (1987). Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell 51, 987-1000. Edmondson, D. G. and Olson, E. N. (1989). A gene with homology to the myc similarity region of MyoD1 is expressed during myogenesis and is sufficient to activate the muscle differentiation program. Genes Dev. 3, 628-640. Grobstein, C. (1955). Inductive interaction in the development of the mouse metanephros. J. Exp. Zool. 130, 319-40. Hamburger, V. and Hamilton, H. (1951). A series of normal stages in the development of the chick embryo. J. Morphol. 88, 49-92. Hasty, P., Bradley, A., Edmondson, D., Venuti, J., Olson, E. and Klein, W. (1993). Muscle deficiency and neonatal death in mice with a targeted mutation in the myogenin gene. Nature 364, 501-506. Hollenberg, S. M., Cheng, P. F. and Weintraub, H. (1993). Use of a conditional MyoD transcription factor in studies of MyoD transactivation and muscle determination. Proc. Natl. Acad. Sci. USA 90, 80288032. Holtzer, H. and Detwiler, S. R. (1953). An experimental analysis of the development of the spinal column. III. Induction of skeletogenous cells. J. Exp. Zool. 123, 335-70. Hopwood, N. D., Pluck, A. and Gurdon, J. B. (1989). MyoD expression in the forming somites is an early response to mesoderm induction in Xenopus embryos. EMBO J. 8, 3409-17 Hopwood, N. D. and Gurdon, J. B. (1990). Activation of muscle genes without myogenesis by ectopic expression of MyoD in frog embryo cells. Nature 347, 197-200. Hopwood, N. D., Pluck, A. and Gurdon, J. D. (1991). Xenopus Myf-5 marks early muscle cells and can activate muscle genes ectopically in early embryos. Development 111, 551-560. Kenny-Mobbs, T. and Thorogood, P. (1987). Autonomy of differentiation in avian brachial somites and the influence of adjacent tissues. Development 100, 449-462. Lash, J., Holtzer, S. and Holtzer, H. (1957). An experimental analysis of the development of the spinal column. Exp. Cell Res. 13, 292-303. Miller, J. B. and Stockdale, F. E. (1986a). Developmental origins of skeletal muscle fibers: clonal analysis of myogenic cell lineages based on fast and slow myosin heavy chain expression. Proc. Natl. Acad. Sci. USA 83, 38603864. Miller, J. B. and Stockdale, F. E. (1986b). Developmental regulation of the multiple myogenic cell lineages of the avian embryo. J. Cell. Biol. 103, 21972208. Nabeshima, Y., Hanaoka, K., Hayasaka, M., Esumi, E., Li, S. and Nonaka, I. (1993). Myogenin gene disruption results in perinatal lethality because of severe muscle defect. Nature 364, 532-535. Nieuwkoop, P. D. (1969). The Formation of the mesoderm in urodelean amphibians. I. Induction by the endoderm. Wilhelm Roux Arch. EntwMech. Org. 162, 341-73. Packard, D. S. J. and Jacobson, A. G. (1976). The influence of axial structures on chick somite formation. Dev. Biol. 53, 36-48. Pannett, C. A. and Compton, A. (1924). The cultivation of tissues in saline embryonic juice. Lancet 205, 381-4. Piette, J., Huchet, M., Duclert, A., Fujisawa-Sehara, A. and Changeux, J.-P. (1992). Localization of mRNAs coding for CMD1, myogenin and the α-subunit of the acetylcholine receptor during skeletal muscle development in the chicken. Mech. Dev. 37, 95-106. 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 llate. Proc. Natl. Acad. Sci. USA 90, 5242-5246. Pownall, M. E. and Emerson, C. P., Jr. (1992). Sequential activation of three myogenic regulatory genes during somite morphogenesis in quail embryos. Dev. Biol. 151, 67-79. Rhodes, S. J. and Konieczny, S. F. (1989). Identification of MRF4: a new member of the muscle regulatory factor gene family. Genes Dev. 3, 20502061. Rickmann, M., Fawcett, J. W. and Keynes, R. J. (1985). The migration of neural crest cells and the growth of motor axons through the rostral half of the chick somite. J. Embryol. Exp. Morph. 90, 437-455. Rong, P., Teillet, M., Ziller, C. and Le Douarin, N. (1992). The neural tube/notochord complex is necessary for vertebral but not limb and body wall striated muscle differentiation. Development 115, 657-672. Rudnicki, M. A., Braun, T., Hinuma, S. and Jaenisch, R. (1992). Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development. Cell 71, 383-390. Rudnicki, M.A., Schnegelsberg, P. N. J., Stead, R. H., Braun, T., Arnold, H.-H. and Jaenisch, R. (1993). MyoD or Myf-5 is required for the formation of skeletal muscle. Cell 75, 1351-1359. 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 MyoD1 during mouse embryogenesis. Nature 341, 303-307. Serbedzija, G. N., Bronner-Fraser, M. and Fraser, S. E. (1989). A Vital dye analysis of the timing and pathways of avian trunk neural crest cell migration. Development 106, 809-816. Spemann, H. (1901). Uber korrelationen in der entwicklung des anges. Anat. Anz. 15, 61-79. Stockdale, F. E. and Miller, J. B. (1987). The cellular basis of myosin heavy 1452 N. Buffinger and F. E. Stockdale chain isoform expression during development of avian skeletal muscles. Dev. Biol. 123, 1-9. Stockdale, F. E. (1992). Myogenic cell lineages. Dev. Biol. 154, 284-298. Teillet, M. 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, 192-211. Thayer, M. J., Tapscott, S. J., Davis, R. L., Wright, W. E., Lassar, A. B. and Weintraub, H. (1989). Positive autoregulation of the myogenic determination gene MyoD1. Cell 58, 241-248. van Straaten, H. W. and Hekking, J. W. (1991). Development of floor plate, neurons and axonal outgrowth pattern in the early spinal cord of the notochord-deficient Chick Embryo. Anat. Embryol. 184, 55-63. Vivarelli, E., Brown, W. E., Whalen, R. G. and Cossu, G. (1988). The expression of slow myosin during mammalian somitogenesis and limb bud differentiation. J. Cell. Biol. 107, 2191-2197. Weintraub, H. (1993).The MyoD family and myogenesis: redundancy, networks, and thresholds. Cell 75, 1241-1244. Weston, J. A. and Butler, S. L. (1966). Temporal factors affecting localization of neural crest cells in the chicken embryo. Developmental Biology 14, 246266. Wright, W. E., Sassoon, D. A. and Lin, V. K. (1989). Myogenin, a factor regulating myogenesis, has a domain homologous to MyoD. Cell 56, 607617. Yamada, T., Pfaff, S. L., Edlund, T. and Jessell, T. M. (1993). Control of cell pattern in the neural tube: motor neuron induction by diffusible factors from notochord and floor plate. Cell 73, 673-686. (Accepted 16 March 1994)