Myotubes were isolated from enzymically disaggregated embryonic muscles and examined with light microscopy. Primary myotubes were seen as classic myotubes with chains of central nuclei within a tube of myofilaments, whereas secondary myotubes had a smaller diameter and more widely spaced nuclei. Primary myotubes could also be distinguished from secondary myotubes by their specific reaction with two monoclonal antibodies (MAbs) against adult slow myosin heavy chain (MHC). Myonuclei were birthdated with [3H]thymidine autoradiography or with 2-bromo-5 ′-deoxyuridine (BrdU) detected with a commercial monoclonal antibody. After a single pulse of label during the 1 –2 day period when primary myotubes were forming, some primary myotubes had many myonuclei labelled, usually in adjacent groups, while in others no nuclei were labelled. If a pulse of label was administered after this time labelled myonuclei appeared in most secondary myotubes, while primary myotubes received few new nuclei. Labelled and unlabelled myonuclei were not grouped in the secondary myotubes, but were randomly interspersed. We conclude that primary myotubes form by a nearly synchronous fusion of myoblasts with similar birthdates. In contrast, secondary myotubes form in a progressive fashion, myoblasts with asynchronous birthdates fusing laterally with secondary myotubes at random positions along their length. These later-differentiating myoblasts do not fuse with primary myotubes, despite being closely apposed to their surface. Furthermore, they do not generally fuse with each other, as secondary myotube formation is initiated only in the region of the primary myotube endplate.

Skeletal muscle fibres constitute a spectrum of cell types sharing a common theme of a sarcomeric arrangement of sliding filaments and actin – myosin interactions. The varied proportions of muscle fibre types within different muscles allow the matching of individual physiological demands for endurance, strength and speed of contraction. Within particular muscles, different fibre types have an ordered spatial distribution with an adaptive relation to factors such as the frequency of use of particular motor units or uneven restrictions in intramuscular blood flow when the muscle contracts. These patterns arise early in muscle development (McLennan, 1983; Butler etal. 1982) so that they reflect developmental patterning, rather than adaptation to use. Muscle fibres form by fusion of mononucleate myoblasts into multinucleate myotubes (Schwann, 1847), and the emergence of different muscle fibre types may depend on their origins from different populations of myoblasts (reviewed by Stockdale and Miller, 1987). There are two major generations of myotubes: primary myotubes form first, and are followed after a delay by secondary myotubes, which will form the majority of muscle fibres in the adult tissue (Couteaux, 1937, 1941; Kelly and Zacks, 1969; Harris, 1981a; Ross et al. 1987a). These two generations have characteristic differences in their sequence of expression of MHC isoforms, as discussed in the preceding paper (Harris et al. 1989).

Specificity of fusion within individual classes of myoblast in a mixed population has been shown in vitro, and there is indirect evidence to suggest that it may occur in vivo (reviewed by Stockdale and Miller, 1987). We describe here experiments designed to test whether distinct populations of myoblasts can be detected during normal muscle development in vivo. Our results indicate that the nerve-dependent population of ‘late’ myoblasts that generates secondary myotubes do not fuse with primary myotubes, at least in the early stages of the foetal period of muscle development. Foetal rat and mouse muscles were disaggregated into their component myotubes which could then be identified as primary or secondary myotubes by criteria of morphology, time of appearance, and MHC isoform composition. Nuclear birth dating was used to determine the times of origin of individual nuclei within a myotube to see if we could distinguish different populations of myoblasts contributing to the cell lineages of primary and secondary myotubes.

If the hypothesis that the two classes of myotube have different cellular origins is correct, one might predict that myoblasts present during the time of formation of secondary myotubes would specifically fuse with secondary myotubes alone. If, on the other hand, myoblasts fused randomly with other cells, some definite predictions on their distribution can be made. We have already excluded the possibility of random homophilic fusion between secondary myoblasts, as new secondary myotubes form only in association with primary myotube endplates, and not randomly through the muscle despite uniform distribution of myoblasts throughout the tissue and equal access of myoblasts to both primary and secondary myotubes (Duxson et al. 1989). In the early stages of secondary myotube formation, during E16 –17 in rat diaphragm muscle (Harris, 1981a), myoblasts would have a much greater probability of encountering primary rather than secondary myotubes, and few labelled nuclei should be found in secondary myotubes. Only by E18, when the number of secondary myotubes exceeds primary myotubes by more than 2:1 and many secondary myotubes extend the full length of the muscle, would random fusion be expected to give rise to similar numbers of labelled nuclei entering both secondary and primary myotubes.

Duxson et al. (1989) showed also, in confirmation of earlier work (Kelly and Zacks, 1969, Ontell and Dunn, 1978), that secondary myotubes grow longitudinally along the surface of a single primary myotube, thus distinguishing their pattern of growth from that of primary myotubes which reach from tendon to tendon from an early stage. We now ask whether this pattern of growth is due to myoblasts fusing only with the growing ends of the secondary myotube, or whether myotubes are receptive to fusion along their whole length.

The recent commercial production of monoclonal antibodies to 2-bromo-5 ′-deoxyuridine (BrdU) makes possible rapid birthdating of cell nuclei using immunological rather than autoradiographic marking of labelled nuclei (Gratzner, 1982). BrdU is incorporated into DNA in place of thymidine, and the labelled DNA can be replicated so that the label is diluted over successive cell generations. Our interpretation of the nuclear birthdating experiments is based on the theory (Bischoff and Holtzer, 1969) that myoblasts become fusion-competent immediately after a mitosis, so that after incorporating label a myoblast will divide and then one or both daughter cells may promptly fuse, or may reenter the cell cycle; heavily labelled cells, which remained dormant in the developing muscle without either dividing or fusing, would be defined as satellite cells, not myoblasts.

Incorporation of BrdU into chick myoblasts in tissue culture early in S-phase has been shown to reduce their capacity for fusion (Lough and Bischoff, 1976), so that they must undergo one further mitotic division before they become fusion-competent, indicating that care must be taken in interpreting results obtained with this label. Accordingly, all experimental results were verified with parallel applications of [3H]thymidine followed by autoradiography.

Experiments were done on both rats and mice. Rats (white Wistar) were mated by coupling overnight in wire-bottomed cages, and pregnancies dated by the presence of a copulation plug at 9:00h the following morning (E0). Mice (strain 129/J) were coupled between 12:00 and 16:00h, and pregnancies confirmed by the presence of a copulation plug at 16:00h (E0). Thus, when comparing the timing of events between mice and rats, mouse embryos at a given ‘embryonic day’ were about 12 h younger than rats.

Frozen sections

Diaphragm muscles were dissected from pithed embryos and pinned out on Sylgard in a Petri dish. They were fixed in 4 % paraformaldehyde in 0.1 M-phosphate buffer, pH7.4, for 30min and then washed in phosphate-buffered saline (PBS). Just before freezing they were rinsed for 2 min in distilled water, and then placed horizontally on the cryotome chuck on a flat base cut from frozen Tissue-Tek, so that longitudinal sections could be made (Harris, 1981b). Sections were collected on subbed slides, and processed to reveal mitoses or BrdU labelling.

Preparation of muscle disaggregates

Disaggregates were prepared from a variety of muscles including diaphragm, stemo-mastoid, intercostals, soleus and extensor digitorum longus, from both rats and mice. The majority of experiments were done on diaphragm muscle. Embryos were chilled on ice for anaesthesia and their muscles dissected and pinned, well stretched, to incubate in azide-Ringers (140mM-NaCl, 5mM-KCl, 2mM-CaC12, 4mM-Hepes, 0.05% sodium azide) for 30min at room temperature. The muscles were fixed in 30 mM-dimethylsuberimidate dihydrochloride (DMS) (Sigma) in 100mM-NaCl, 50mM-Tris, pH 7.4, (TBS) at 37°C for 30 min. Next they were incubated in collagenase-neutral protease (Bischoff, 1986) in 140 mM-NaCl, 5 mM-KCl, 2 mM-CaCl2 at 37 °C for 1 –2 h, at which time myotubes could be gently separated from their tendons and any unwanted tissue fragments removed. The remaining tissue was incubated for 30 –60 min more, until gentle pipetting with a fire-polished Pasteur pipette sufficed to disaggregate the cells. In early experiments, myotubes were separated from mononucleate cells by 1G sedimentation (Bischoff, 1986), but it was later found more useful to retain both mononucleate and multinucleate cells. The disaggregation procedure released a variable number of free nuclei which caused some contamination by adhering to the surfaces of the disaggregated cells. The muscle disaggregates were postfixed in 4% paraformaldehyde in 100 mM-phosphate buffer, pH 7.4; centrifuged, and resuspended in distilled water. They were then spread on subbed slides and air dried at 4 °C.

Immunohistochemistry

After being dried onto slides, disaggregated muscle cells were prepared for immunohistochemistry by a wash in 0.1M-glycine in 0.1 M-NaCl, 20mM-phosphate buffer, pH7.4 (PBS), rinsed in PBS, and given a mild digestion in protease K (Merck) 2 μgml-1 in 0.1M-Tris buffer, 5mM-EDTA, pH7.4, for 5 min at room temperature, followed by 5 min washes in 20mM-Tris, 0.1 M-NaCl (TBS) plus 4mM-CaCl2 to stop action of the protease, and in PBS.

All primary antibodies were aliquotted, quickfrozen in liquid nitrogen, and stored at – 80° C. After thawing for use, hybridoma supernatants were diluted with an equal quantity of 120mM-phosphate buffer, pH7.4, and further diluted, as required, in PBS containing 0.2% Triton X-100 and 0.5% BSA (ID). All the results presented in this paper are with the monoclonal antibodies NOQ7.5.4D or NOQ7.1.1A, two clones raised against adult human muscle slow myosin heavy chain (MHC) (Draeger etal. 1987) which bind a MHC isoform present in all newly formed rat primary myotubes (Narasuwa et al. 1987; Harris et al. 1989). Second antibodies included biotin anti-mouse (DAKO or Amersham) followed by streptavidin – HRP or streptavidin – biotin – HRP complex (Amersham); or FITC or RHO anti-mouse (DAKO). HRP activity was revealed using diaminobenzidine as chromagen.

Location of mitotic figures

Pregnant rats were anaesthetized with ether and the uterus exposed by laparotomy. Taxol (gift from Dr J. Douros) was kept frozen at – 80°C as a 2.5 DIM stock solution in DMSO, and diluted 1:1000 in sterile saline just before use. Hoechst 33258 dye was stored at 4°C in the dark as a 1:1000 solution in distilled water, and further diluted 1:500 in PBS just before use. Individual embryos were injected intraperitoneally (i.p.) with 4 μl of taxol to arrest mitoses, and examined 6h later. Frozen sections were incubated in rhodamine-labelled alphabungarotoxin (RHO-α-BTX, gift from Dr R. Bloch) to locate endplates, and incubated for 4min in Hoechst 33258 to reveal chromosomes. Sections were examined with a 40 · oil immersion objective in a Zeiss fluorescence microscope.

Birthdating with 5-bromo-2 ′-deoxyuridine

The drug (purchased from Sigma) was stored desiccated at – 20°C and just before use was freshly dissolved in sterile saline. The standard procedure was i.p. injection of 5 mg into a pregnant rat (12 –25 mg kg-1) or 1 mg into a pregnant mouse (20 –40 mg kg-1). In experiments comparing [3H]thymidine autoradiography with BrdU immunohistochemistry, foetuses in one uterine horn were individually injected with 50 μg of BrdU and those in the other horn with 5 μCi of [3H]thymi-dine, as described below.

Slides of BrdU-treated myotube preparations were immersed in 1 M-HCI at 60°C for 8 min, or 2M-HC1 at 37°C for 30 min. Time and temperature were carefully controlled, as some nonspecific nuclear labelling occurred in tissues processed in 1 M-HCI at higher temperatures or for longer times. The slides were then rinsed in 0.1 M-phosphate buffer followed by two changes of PBS. They were then incubated overnight with anti-BrdU (purchased from Becton-Dickinson), 1:50 dilution in ID, washed in two changes of PBS, and the anti-BrdU revealed with a fluorescent second antibody (FITC or RHO-antimouse, DAKO, diluted 1:50 in ID). Our routine preparation for birth dating analysis was to distinguish myotube types with anti-slow MHC Ab marked with the HRP sandwich technique, and then to reveal nuclei containing BrdU with a fluorescent second antibody.

[3H]fihymidine birthdating, and autoradiography

[6-3H]thymidine (Amersham, 5mCimmole-1, 1mCiml-1) was stored at 4°C as a sterile solution in distilled water. Dated pregnant rats were anaesthetized with ether and the uterus exposed with a laparotomy. Individual foetuses were visualized through the uterine wall and injected intraperitoneally with 5 μCi of the sterile solution.

Disaggregated myotube preparations containing [3H]thy-midine-labelled nuclei were incubated with anti-slow MHC antibody which was then marked with HRP by the biotinstreptavidin sandwich technique, and the HRP revealed using diaminobenzidine as a chromogen. Slides were rinsed in PBS and then brominated in freshly prepared 10mM-H2SO4, 20 mM-KBr, 10 mM-H2O2 for 20 min to prevent chemography. They were coated with Kodak NTB2 emulsion, exposed for 2 –3 weeks and developed for 6 min in half-strength D19 developer. The slides were then stained for 20 min with Harris’ haematoxylin to mark nuclei, and permanently mounted.

Criteria for scoring labelled nuclei

FITC-or RHO-antiBrdU-labelled nuclei were distinguished under the microscope by the subjective criterion of being obviously fluorescent. The technique used for labelling is very sensitive, and the results were comparable to those using [3H]thymidine labelling (Table 1). The total number of nuclei in a myotube segment was counted using phase-contrast illumination, and then fluorescent nuclei counted under epifluorescence illumination. Because of bleaching it was usual to examine only one or two segments in a field of view. Photographs were made for illustration only, and not used for quantitative analysis.

Table 1.

Percentage of nuclei in primary and secondary myotubes labelled with birthdating markers[3H]thymidine or BrdU

Percentage of nuclei in primary and secondary myotubes labelled with birthdating markers[3H]thymidine or BrdU
Percentage of nuclei in primary and secondary myotubes labelled with birthdating markers[3H]thymidine or BrdU

With [3H]thymidine autoradiography, most labelled nuclei were heavily labelled under the conditions of labelling and exposure used; too heavily, for example, to be useful for analysis of timing of the cell cycle. The threshold criterion for labelling was 5 or more grains per nucleus; the average background density was always equivalent to much less than 1 grain per nucleus.

Fluorescence microscopy

Slides were examined using Leitz Labolux, Zeiss Universal or VANOX epifluorescence microscopes with 200W mercury illuminators, or a Zeiss Standard epifluorescence microscope with a 75 W xenon illuminator. Unlabelled nuclei were identified with phase contrast. Colour photography was with Ektachrome (ASA 200) or Fuji (ASA 1600) colour negative film; most records were made on Ilford HP5 black and white film.

Location of mitotic figures in muscle

Individual foetal rats were injected in utero with the antimitotic drug taxol, and frozen sections incubated in the DNA label Hoechst 33258. The incidence of mitotic figures was plotted along the length of the muscles with respect to the location of the endplates, which were located with RHO-α-BTX. Diaphragm muscles from E18 and E20 foetuses were examined: in both cases, the distribution was uniform, with no detectable excess in the endplate region of the muscle. We could not distinguish between mitoses in myoblasts and mitoses in fibroblasts. In preliminary experiments, we also looked at BrdU-labelled nuclei in longitudinal frozen sections of foetal diaphragm muscles. These typically contained large numbers of labelled nuclei and it was impossible to tell whether they were in mononucleate or multinucleate cells. We found no significant inhomogeneity in their longitudinal distribution along the muscle sections.

Preparation of isolated myotubes

We used the technique of Bischoff (1986) to prepare isolated myotubes. Cells isolated from disaggregated living embryonic muscles generally were supercontracted with their nuclei compacted together. Next, we tried a ‘natural’ form of fixation, inducing rigor by room temperature incubation in Mg2+-free saline containing 0.02% sodium azide before exposing tissues to the disaggregating enzymes. This worked quite well with muscles from older foetuses, but new-formed myotubes still supercontracted. We finally used the hetero-bifunctional reagent dimethylsuberimidate dihydrochloride (DMS) (Expert-Bezanfon et al. 1977), which provided fixation adequate to prevent contraction, while leaving the tissue susceptible to the enzymes used for disaggregation of the cells. This reagent has the further advantage of minimally masking or denaturating antigenic determinants.

Bischoffs technique, applied to adult muscle, is intended to separate muscle fibres with satellite cells still attached to their surface beneath the basal lamina. Foetal muscles contain patchy and diffuse basal lamina (Duxson et al. 1989), and we hoped to obtain myotubes free of mononucleate cells. Most results presented here come from myotubes with central nuclei, so there is little risk of confusion with nuclei in tightly attached mononucleate cells.

Timing of BrdU injections

Muscle disaggregates prepared 3 h and 5.5 h after BrdU injection contained large numbers of heavily labelled nuclei, but none of these were in multinucleate cells. Seven hours after injection, a few labelled myonuclei were found (mouse injected at E14; 70% of myotubes had no labelled nuclei; 23 % had a single labelled nucleus. Labelled nuclei in myotubes were more faintly labelled than most mononucleate cell nuclei). In muscles disaggregated 14 h after injection, there was substantial labelling of nuclei in multinucleate cells. Most experiments allowed 24 h between injection and examination of muscles. Control experiments showed no labelling when antibodies were applied to tissues that had not been exposed to BrdU, and no labelling when the second antibody was applied alone, without previous exposure to anti-BrdU.

Ordered appearance of muscle fibre types: fibre size

Myotube preparations from rat diaphragms disaggregated on E15 or E16, or mouse diaphragms disaggregated on E14 or E15, contained a uniform population of large diameter (15 – 20 μm) myotubes (Fig. 1). One day later, on E17 (rat) or E16 (mouse), the populations became bimodal with the appearance of a new, smaller (5 – 10 μm) diameter class of myotube (Figs 2, 3). The small diameter myotubes initially had a beaded appearance with rounded nuclei separated by narrow threads of cytoplasm, fitting Couteaux’ (1941) description of young secondary myotubes as ‘like a pearl necklace’ (Figs 2B, 3). With time, e.g. by E18 in rat diaphragm or stemomastoid muscles, older secondary myotubes had become elongated and uniform in diameter with central nuclei, but still remained smaller in diameter than the primary myotubes.

Fig. 1.

Primary myotubes isolated from embryonic muscles. (A) Fibres from an E16 rat diaphragm muscle stained with antislow MHC. All fibres in the disaggregate were stained, but mononucleate cells were not. Fluorescence micrographs, rhodamine-labelled second antibody. (B) Primary myotube isolated from an E16 rat diaphragm following injection of BrdU on E15 (–25 h). All nuclei in this segment are marked by anti-BrdU (FITC-labelled second antibody). (C) Primary myotube isolated from an E15 mouse diaphragm, following injection of [3H]thymidine on E14 (–24 h). Autoradiograph viewed with Nomarski optics. Labelled nuclei are marked by arrows. Calibration bars, 20 μm.

Fig. 1.

Primary myotubes isolated from embryonic muscles. (A) Fibres from an E16 rat diaphragm muscle stained with antislow MHC. All fibres in the disaggregate were stained, but mononucleate cells were not. Fluorescence micrographs, rhodamine-labelled second antibody. (B) Primary myotube isolated from an E16 rat diaphragm following injection of BrdU on E15 (–25 h). All nuclei in this segment are marked by anti-BrdU (FITC-labelled second antibody). (C) Primary myotube isolated from an E15 mouse diaphragm, following injection of [3H]thymidine on E14 (–24 h). Autoradiograph viewed with Nomarski optics. Labelled nuclei are marked by arrows. Calibration bars, 20 μm.

Fig. 2.

Secondary myotubes isolated from maturing rat and mouse muscles. (A) Phase-contrast photomicrograph of fibres isolated from an E18 mouse diaphragm illustrating a primary (larger diameter) myotube and a secondary (smaller diameter) myotube. (B) Secondary myotube from an E17 rat diaphragm: BrdU was injected at 2h intervals from – 24 h to – 12 h previously, in order to cumulatively label all nuclei undergoing mitosis in that time interval, in contrast to the pulse labelling used in the experiments whose results are summarized in Table 1. Although all secondary myotube nuclei were labelled by this procedure, few labelled nuclei appeared in primary myotubes in this disaggregate. Fluorescence micrograph. (C) Primary and secondary myotubes from an E18 rat diaphragm, labelled with a single injection of [3H]thymidine on E17. A single labelled nucleus is present in the primary myotube, while labelled nuclei appear frequently in the secondary myotube. (D) Primary and secondary mytotubes from an E15.5 mouse, labelled with [3H]thymidine on E14.5. Labelled nuclei are seen only in secondary myotubes (arrows). (C and D): Nomarski optics. Calibration bars, 20 μm.

Fig. 2.

Secondary myotubes isolated from maturing rat and mouse muscles. (A) Phase-contrast photomicrograph of fibres isolated from an E18 mouse diaphragm illustrating a primary (larger diameter) myotube and a secondary (smaller diameter) myotube. (B) Secondary myotube from an E17 rat diaphragm: BrdU was injected at 2h intervals from – 24 h to – 12 h previously, in order to cumulatively label all nuclei undergoing mitosis in that time interval, in contrast to the pulse labelling used in the experiments whose results are summarized in Table 1. Although all secondary myotube nuclei were labelled by this procedure, few labelled nuclei appeared in primary myotubes in this disaggregate. Fluorescence micrograph. (C) Primary and secondary myotubes from an E18 rat diaphragm, labelled with a single injection of [3H]thymidine on E17. A single labelled nucleus is present in the primary myotube, while labelled nuclei appear frequently in the secondary myotube. (D) Primary and secondary mytotubes from an E15.5 mouse, labelled with [3H]thymidine on E14.5. Labelled nuclei are seen only in secondary myotubes (arrows). (C and D): Nomarski optics. Calibration bars, 20 μm.

Fig. 3.

The earliest secondary myotubes in a disaggregate of an embryonic rat diaphragm muscle. A primary and a new-formed secondary myotube from an E16 rat embryo injected with BrdU on E15 (– 21 h). A bright-field view of cells stained with anti-slow MHC, revealed with an HRP-linked biotin-streptavidin system. Every primary myotube reacted with the antibody, but few primary myotube nuclei were labelled by anti-BrdU. The very small number of new-formed secondary myotubes present were not stained by anti-slow MHC and the majority of their nuclei were marked by anti-BrdU (inset, fluorescence micrograph showing FITC-linked second antibody), showing they had formed from myoblasts which had divided and fused since E15. Calibration bar, 20 μm.

Fig. 3.

The earliest secondary myotubes in a disaggregate of an embryonic rat diaphragm muscle. A primary and a new-formed secondary myotube from an E16 rat embryo injected with BrdU on E15 (– 21 h). A bright-field view of cells stained with anti-slow MHC, revealed with an HRP-linked biotin-streptavidin system. Every primary myotube reacted with the antibody, but few primary myotube nuclei were labelled by anti-BrdU. The very small number of new-formed secondary myotubes present were not stained by anti-slow MHC and the majority of their nuclei were marked by anti-BrdU (inset, fluorescence micrograph showing FITC-linked second antibody), showing they had formed from myoblasts which had divided and fused since E15. Calibration bar, 20 μm.

Antibody staining

The early appearing larger-diameter myotubes were stained by the two anti-slow MHC MAbs NOQ7.5.4D or NOQ7.1.1A (Fig. 1). Mononucleate cells present at the same time did not stain with these MAbs, or with the muscle-specific antibody anti-desmin. In rat muscles, a small number of myotubes which did not stain with anti-slow MHC (but did stain with antidesmin) appeared on E16 (Fig. 3), and were more numerous on E17. These were smaller-diameter fibres, and were still consistently of the smaller-diameter class on E18 (Fig. 4). By E19, some approached the dimensions of primary myotubes. The timing of appearance of these two classes of myotube correlates well with structural descriptions of the time of origin of primary and secondary myotubes, respectively, and we will use this nomenclature from here on. We conclude that primary myotubes, but not secondary myotubes, reacted with the anti-slow MHC MAbs, confirming the conclusions reached in the accompanying paper (Harris et al. 1989) from experiments with frozen sections.

Fig. 4.

Young secondary myotubes are not stained by anti-slow MHC antibody. All illustrations are from disaggregates of E18 rat diaphragm muscles following injection of BrdU or pHjthymidine on E17. The anti-slow MHC MAb is revealed with a biotin-streptavidin –HRP sandwich technique, and anti-BrdU marked with a rhodamine-labelled second antibody. (A) Bright-field and fluorescence pair of views, showing a primary myotube marked by HRP (top) and secondary myotubes containing nuclei marked by anti-BrdU (bottom). (B) More extended view of a secondary myotube, showing scattered labelled and unlabelled nuclei. (C, D and E) HRP-staincd primary myotubes, and unstained secondary myotubes. (C) Autoradiograph viewed with Nomarski optics, (D) and (E) bright-field microscopy. Calibration bars, 20 μm.

Fig. 4.

Young secondary myotubes are not stained by anti-slow MHC antibody. All illustrations are from disaggregates of E18 rat diaphragm muscles following injection of BrdU or pHjthymidine on E17. The anti-slow MHC MAb is revealed with a biotin-streptavidin –HRP sandwich technique, and anti-BrdU marked with a rhodamine-labelled second antibody. (A) Bright-field and fluorescence pair of views, showing a primary myotube marked by HRP (top) and secondary myotubes containing nuclei marked by anti-BrdU (bottom). (B) More extended view of a secondary myotube, showing scattered labelled and unlabelled nuclei. (C, D and E) HRP-staincd primary myotubes, and unstained secondary myotubes. (C) Autoradiograph viewed with Nomarski optics, (D) and (E) bright-field microscopy. Calibration bars, 20 μm.

Nuclear birthdating

When BrdU or [3H]thymidine was injected into a pregnant mouse or rat, or directly into the peritoneal cavity of an individual foetus in utero, many nuclei in myotubes disaggregated from muscles of the embryos could be marked with anti-BrdU or autoradiography, respectively.

Mouse E13/14 or rat E14/15

Label injected into mice on E13 or E14, or into rats on E14 or E15, was incorporated into nuclei in the large diameter myotubes stained by anti-slow MHC (Table 1). Following a single injection of label at these times disaggregated diaphragm muscle myotubes examined 24 –48 h later included some myotube segments in which all the nuclei were labelled (Fig. 1B), and others in which no nuclei were labelled. In partially labelled myotube segments, the labelled nuclei often were present as a coherent group, rather than being mixed with nonlabelled nuclei. The timing of the injection was critical: for example, injecting a pregnant rat late on E15 and examining the myotubes 24 h later might result in few labelled nuclei in primary myotubes, while most mononucleated cells were labelled, and a very small number of secondary myotubes was present, all with a large proportion of nuclei labelled (Fig. 3).

Mouse E15 or E16, rat El6 or El 7

Injection of either birthdating reagent at these later times labelled a much smaller proportion of nuclei in primary myotubes, but labelled nuclei were present in the new-formed small-diameter slow MHC-negative myotubes (secondary myotubes) (Fig. 4, Table 1). Nearly all secondary myotube segments in diaphragm or stemomastoid muscle disaggregates prepared 1 –3 days after injecting BrdU at these times contained labelled nuclei. Timing of the injection was not critical: for example, in two experiments, diaphragm myotube preparations were made from rat foetuses injected with BrdU on E16 and examined 24 h later on E17. 75 % and 90%, respectively, of primary myotube segments (defined by reaction with anti-slow MHC) had no labelled nuclei and the others typically had no more than one, whereas 85 % and 88 %, respectively, of secondary myotube segments (slow MHC –ve) had labelled nuclei and typically these were in the majority. Similarly, in 2 examples of rat diaphragm myotubes labelled on E17 and examined 24 h later on E18, 82% and 93%, respectively, of primary myotube segments had no labelled nuclei, and 75% and 100%, respectively, of the secondary myotube segments were labelled. The labelled nuclei in secondary myotubes were randomly scattered amongst unlabelled nuclei (Fig. 5).

Fig. 5.

Random positions of insertion of new nuclei into secondary myotubes. (A) E16 (evening) mouse diaphragm secondary myotubes following injection of BrdU 13 h earlier. Top: a single labelled nucleus (arrow) amongst unlabelled nuclei. Bottom: scattered brightly and faintly labelled nuclei in a segment of a secondary myotube. (B) Secondary myotubes from an E18 mouse diaphragm, BrdU injected on E16 (–50h). There is a random scattering of brightly labelled, faintly labelled and unlabelled nuclei. Anti-BrdU revealed with FITC-labelled second antibody. (C and D) Autoradiographs of secondary myotubes from E18 rat diaphragm muscles, [3H]thymidine injected on E17 (– 24h). (C) Bright-field; (D) Nomarski optics. Calibration bars, 20 μm.

Fig. 5.

Random positions of insertion of new nuclei into secondary myotubes. (A) E16 (evening) mouse diaphragm secondary myotubes following injection of BrdU 13 h earlier. Top: a single labelled nucleus (arrow) amongst unlabelled nuclei. Bottom: scattered brightly and faintly labelled nuclei in a segment of a secondary myotube. (B) Secondary myotubes from an E18 mouse diaphragm, BrdU injected on E16 (–50h). There is a random scattering of brightly labelled, faintly labelled and unlabelled nuclei. Anti-BrdU revealed with FITC-labelled second antibody. (C and D) Autoradiographs of secondary myotubes from E18 rat diaphragm muscles, [3H]thymidine injected on E17 (– 24h). (C) Bright-field; (D) Nomarski optics. Calibration bars, 20 μm.

Only about 5 % of the nuclei in primary myotubes were labelled on E16, E17, or E18 (Figs 2C, 6) compared to 30 –50 % if label was injected on E14 or E15 (Table 1). In one experiment, [3H]thymidine was injected into rat embryos on E16 and the myotubes examined on E19. 10% of the nuclei in primary myotubes were labelled, as compared to 37% of the nuclei in secondary myotubes. This time allowed for considerable dilution of label (i.e. by 2 or more mitoses occurring after labelling and before cell fusion). However, nuclei from myoblasts that had fused following their first mitosis after labelling could be identified by their high density of autoradiographic grains. Heavily labelled nuclei were identified under dark-field microscopy, and then their location in a primary or a secondary myotube determined with bright-field microscopy. In this case, 14% of the heavily labelled nuclei were in primary myotubes. If it can be assumed that the disaggregated fibre preparation preserves primary and secondary myotubes in the same proportions as in the muscle, then this experiment indicates that on E16, 86% of the differentiated myoblasts fused into secondary myotubes, and not with primary myotubes. As secondary myotubes are just beginning to form at this time, no model of random fusion with any available myotube could account for this observation.

Fig. 6.

Labelled nuclei in an autoradiograph of a primary myotube (stained with anti-slow MHC) from an E18 rat diaphragm, labelled with [3H]thymidine on E17. This myotube segment contained 52 nuclei, of which 2 were labelled. Note also the free nuclei released from cells damaged during the disaggregation procedure, adhering to the side of the myotube. Calibration bar, 20 μm.

Fig. 6.

Labelled nuclei in an autoradiograph of a primary myotube (stained with anti-slow MHC) from an E18 rat diaphragm, labelled with [3H]thymidine on E17. This myotube segment contained 52 nuclei, of which 2 were labelled. Note also the free nuclei released from cells damaged during the disaggregation procedure, adhering to the side of the myotube. Calibration bar, 20 μm.

Patterns of fusion of myoblasts with primary and secondary myotubes

It has long been accepted that the ‘primary generation’ of myotubes (Tello, 1917) forms during a brief interval of time (Kelly and Zacks, 1969, Ontell and Kozecka, 1984; Ross et al. 1987a), while the appearance of the secondary generation is protracted (Couteaux, 1941; Betz et al. 1980; Harris, 1981a). The classical description of primary myotube formation (reviewed by Swatland, 1984) is that myoblasts assemble in rows, and then fuse more or less simultaneously to form the myotubes. Secondary myotubes, on the other hand, grow longitudinally along the surface of their parent primary myotube (Duxson et al. 1989). Because myoblasts are uniformly distributed throughout the muscle (Duxson et al. 1989), this pattern of growth could be explained by the extending tip of the myotube being particularly receptive to receiving a fusing myoblast. This hypothesis predicts that birthdating label should mark nuclei near the ends of secondary myotubes, and a pulse of label should mark a group of adjacent nuclei.

At all ages studied, the opposite was true; labelled nuclei appeared in secondary myotubes at random points (Figs 5,7). This was regardless of whether the secondary myotubes were new-formed and small, or whether they were nearing maturity. Fusion patterns were analyzed (Fig. 8) by measuring the probability of a labelled secondary myotube nucleus being next to 1, 2 or 3 other labelled nuclei. The probabilities were those expected by random chance. In new-formed primary myotubes (labelled on E14), by contrast, a labelled nucleus was more likely to be next to 2 or 3 other labelled nuclei than can be accounted for by chance, indicating that groups of myoblasts with similar birthdates had fused together. Labelling on E15 produced a more nearly random pattern, indicating that primary myotubes, once formed, grew by addition of further myoblasts at random points.

Fig. 7.

Schematic illustration of the pattern of labelling of secondary myotube nuclei. Nuclei labelled with [3H]thymidine are represented by filled circles. (A) Examples of E17 secondary myotube segments, labelled on E16. (B) Examples of E18 secondary myotube segments, labelled on E17.

Fig. 7.

Schematic illustration of the pattern of labelling of secondary myotube nuclei. Nuclei labelled with [3H]thymidine are represented by filled circles. (A) Examples of E17 secondary myotube segments, labelled on E16. (B) Examples of E18 secondary myotube segments, labelled on E17.

Fig. 8.

Grouping of labelled nuclei in primary and secondary myotubes from rat embryos and foetuses. Each histogram represents one experiment, and is labelled with the day of injection of label and the day of examination. The numbers of labelled nuclei with 0, 1, 2 or 3 labelled nuclei adjacent to them were counted (dark bars), and the distributions compared with those expected by chance (slashed bars). Nuclei in primary myotubes labelled on E14 were more likely to be in groups than expected from random chance. Nuclei in primary myotubes labelled on E15, and in secondary myotubes labelled on E16 or E17 were distributed as expected by random chance, indicating that myoblasts fused randomly at any point along the surface of the growing myotube.

Fig. 8.

Grouping of labelled nuclei in primary and secondary myotubes from rat embryos and foetuses. Each histogram represents one experiment, and is labelled with the day of injection of label and the day of examination. The numbers of labelled nuclei with 0, 1, 2 or 3 labelled nuclei adjacent to them were counted (dark bars), and the distributions compared with those expected by chance (slashed bars). Nuclei in primary myotubes labelled on E14 were more likely to be in groups than expected from random chance. Nuclei in primary myotubes labelled on E15, and in secondary myotubes labelled on E16 or E17 were distributed as expected by random chance, indicating that myoblasts fused randomly at any point along the surface of the growing myotube.

Labelling secondary myotube precursors

Secondary myotubes also were labelled in muscles disaggregated from E16 (mice) or E17 (rats) onwards, following injection of BrdU at times that labelled primary myotubes. The labelling was less intense than in the primary myotubes, and than in many of the mononucleate cells (Fig. 9). These results indicate that the precursors of myoblasts destined to fuse into secondary myotubes also were mitotically active at the time that myoblasts destined to fuse into primary myotubes were undergoing their last S-phase, with the difference that secondary myoblast precursors reentered the cell cycle and their label was diluted. The brightly labelled mononucleate cells might include differentiating fibroblasts, and satellite cell precursors that had withdrawn from the cell cycle.

Fig. 9.

Secondary myotube nuclei faintly stained following injection of BrdU to mark primary myotube nuclei. Primary and secondary myotubes from an E17 rat embryo diaphragm muscle following injection of BrdU on E14. (A) Primary myotube with brightly marked nuclei. (B) Secondary myotube with faintly marked nuclei together with brightly stained nuclei in many mononucleate cells. (C) Primary (marked by sets of double asterisks) and secondary myotubes in the same field of view to illustrate contrast in nuclear marking. The secondary myotube nuclei, faintly labelled and unlabelled, are marked by arrows. Calibration bar, 20 μm.

Fig. 9.

Secondary myotube nuclei faintly stained following injection of BrdU to mark primary myotube nuclei. Primary and secondary myotubes from an E17 rat embryo diaphragm muscle following injection of BrdU on E14. (A) Primary myotube with brightly marked nuclei. (B) Secondary myotube with faintly marked nuclei together with brightly stained nuclei in many mononucleate cells. (C) Primary (marked by sets of double asterisks) and secondary myotubes in the same field of view to illustrate contrast in nuclear marking. The secondary myotube nuclei, faintly labelled and unlabelled, are marked by arrows. Calibration bar, 20 μm.

Recent tissue culture studies of the origins of different muscle fibre types indicate that these result from homophilic fusion of distinct groups of myoblasts within a heterogeneous population of myogenic stem cells (reviewed by Stockdale and Miller, 1987). The birth dating experiments in embryonic and foetal mice and rats, which we present here, confirm that at least two myogenic cell populations can be distinguished in vivo. These include myoblasts that fuse with each other to form primary myotubes and myoblasts that fuse with secondary but not primary myotubes during the prolonged progressive phase of secondary myotube growth.

Does BrdU inhibit differentiation?

Incorporation of BrdU into chick myoblasts in tissue culture early in S-phase has been shown to reduce their capacity for fusion (Lough and Bischoff, 1976), so that they must undergo one further mitotic division before they become fusion-competent, indicating that care must be taken in interpreting results obtained with this label. We made a direct comparison between foetuses injected with [3H]thymidine and BrdU by injecting those in one uterine horn with one reagent, and those in the other horn with the other. Groups of foetuses injected on E16 and examined on E17, that is during the peak period of production of secondary myotubes in diaphragm muscle (Harris, 1981a), were identical in mean body weight (P>0.5). Preparations of disaggregated myotubes from each group of animals showed no qualitative differences. Both comprised a mixture of primary and secondary myotubes, and the incidence of labelled nuclei in secondary myotubes was similar (Table 1). Our results are drawn from nuclei formed in myoblasts which had differentiated and expressed their capacity to fuse. Thus, even if the BrdU had slightly reduced the numbers of fusion-competent myoblasts, this should not affect our conclusions on the question of specificity of fusion between myoblasts and primary or secondary myotubes.

Each technique had its own advantages: speed of analysis in the case of BrdU, and suitability for quantitative analysis in the case of [3H]thymidine autoradiography where both labelled and unlabelled nuclei could be distinguished, and the intensity of labelling measured by grain counting.

Nuclear labelling

The major results in this study concern labelled myotube nuclei, observed 24h after pulse labelling. After finishing DNA replication, a labelled myoblast must have undergone mitosis, and one or both daughter cells fused with a myotube. We observed that a minimum period of 7h was required before any labelled nuclei were seen in myotubes; these early fusing cells presumably were labelled late in S-phase. Within 24 h of labelling, few myoblasts would have had time to reenter the mitotic cycle and replicate a second time before fusion. In the case of myotubes left 48 h before analysis, it was possible for nuclear label to have been diluted by 2 or possibly 3 divisions subsequent to the initial labelling. This gives rise to the possibility of overestimating the number of myotube nuclei whose parents were in S-phase at the time of injecting the label, by counting nuclei from myoblast daughter cells that reentered the mitotic cycle before finally fusing. As we are concerned with the relative numbers of labelled nuclei entering primary and secondary myotubes, respectively, this possibility does not disturb the interpretation of our results, i.e. that separate populations of myoblasts contributed nuclei to primary and secondary myotubes. A further check, in myotubes examined 3 days after labelling, was to locate dense grain clusters under dark-field illumination (to identify heavily labelled nuclei) and then to identify the type of myotube with bright-field illumination; these nuclei must have entered the myotubes within 24 h of labelling.

Formation of primary myotubes

There was a 2 day time window (E13 –14, mice; E14 –15, rats) during which injection of a pulse of birthdating label marked a substantial proportion of nuclei in some primary myotubes (Fig. 1, Table 1). At the same time, other primary myotubes might have no labelled nuclei, or might contain a discrete group of neighbouring nuclei all of which were labelled (Fig. 8). For several days after this time, injection of BrdU or [3H]thymidine resulted in few labelled nuclei appearing in primary myotubes. As primary myotubes increased in both length and diameter during this time their growth must primarily have been due to synthesis rather than accretion of new cytoplasm and nuclei by cell fusion.

In disaggregates of embryonic muscles, primary myotubes appeared as classic myotubes (Schwann, 1847; Couteaux, 1941) with nuclei lying within a tube of contractile filaments (Figs 1, 2, 4, 6). Very shortly after their formation primary myotubes reacted strongly with the anti-slow MHC MAbs NOQ7.5.4D and NOQ7.1.1A. A quantitative study of the formation of the rat IVth lumbrical muscle has previously shown that primary myo tubes all appeared within 48 h, and could possibly have been generated within a 24 h interval (Ross et al. 1987a).

Initiation of secondary myotube formation

Our previous work has shown that formation of secondary myotubes is initiated only in the vicinity of an endplate, although there is no absolute requirement for a nerve terminal to be physically present at the precise moment when a secondary myotube forms (see Discussion in Duxson et al. 1989).

The irreversible reduction in secondary myotube number that follows maternal undemutrition (Wilson et al. 1988) suggests there is a critical period in development, preceding the actual appearance of new secondary myotubes, when their potential maximum number is determined. We cannot say whether new myotubes are initiated from a special cell type present in developmentally controlled numbers, or whether an inductive signal from nerve or extracellular matrix is required to trigger the rare myoblast –myoblast fusion event that marks formation of a new secondary myotube.

Secondary myotube growth differed from that of primaries in relying on the progressive addition of new cytoplasm and nuclei as myoblasts fused laterally at random positions along the length of the myotube. The secondary myotubes also maintained a characteristically smaller diameter until near the time of birth, even after they had elongated sufficiently to connect with the muscle tendons and split laterally from their primary.

Mechanisms of primary and secondary myotube formation

Our results indicate that primary and secondary myotubes differ not only in their times of development, their structural relations with one another, the presence or absence of plasticity in cell numbers, and the types of muscle fibre into which they will differentiate in the adult (Rubinstein and Kelly, 1981; Harris et al. 1989) but also in the manner in which their precursor myoblasts fuse together. Primary myotubes form in consequence of a well-synchronised mitosis and then fusion of their myoblasts. We have not studied this process structurally, but there are many reports in the light microscope literature (reviewed by Swatland, 1984) of alignment of myoblasts in rows followed by their fusing together.

Secondary myotubes form initially as binucleate cells, and then progressively elongate and increase their number of nuclei, over several days. Our birthdating experiments show that some nuclei in a secondary myotube are labelled regardless of when the label is injected, at least up until the time of birth in the rat. These nuclei are inserted at random points along the length of the myotube, so that extension of the extremities of secondary myotubes is due to growth rather than to an accretion of new myoblasts at their ends. Again this contrasts with primary myotubes, which display a period of several days after their formation when they receive few new nuclei.

Specificity in myogenic cell fusion

There already exists much evidence for specificity in cell fusion within developing muscles. Primary myotube numbers are tightly regulated (Harris, 1981a; Ross et al. 1987b; Wilson et al. 1988), showing that fusion between myoblasts is directed rather than random. The accompanying ultrastructural study (Duxson et al. 1989) shows that secondary myoblasts do not fuse with each other (except in a directed fashion in proximity to primary myo tube endplates), but only with existing secondary myotubes. Furthermore, despite their intimate cell –cell contacts, primary and secondary myotubes do not fuse with each other. The current study adds to this the observation that there is a period in muscle development when myoblasts selectively fuse with secondary, and not primary myotubes.

The abrupt change in nuclear labelling patterns in foetal rat muscles between E15 and E16 can only be explained by specificity in fusion. Myoblasts within muscle bundles frequently contact primary and secondary myotubes simultaneously (Duxson et al. 1989) and, early in the course of secondary myotube formation, myoblasts away from the endplate region could only have contacted the primary myotube. If fusion was not directed, but a random process with committed myoblasts fusing with any adjacent myotube, then the onset of secondary myotube formation would be gradual, rather than the observed abrupt phenomenon. Also, the proportion of nuclei labelled in primary myotubes should decline slowly with time, rather than the observed abrupt decline.

In a freeze-fracture study of myoblast fusion in tissue culture, Kalderon and Gilula (1979) showed that myoblasts initially were metabolically coupled via gap junctions but the incidence of coupling was reduced to <50% at the time fusion began. Fusion was initiated by the appearance of regions of membrane without intra-membranous particles, and no gap junctions were seen in the fusion regions. Their results suggest that there is an intercellular recognition event which precedes fusion, and that recognition and fusion may rely on separate mechanisms.

Specificity in fusion of myogenic cells may most simply be explained by assuming the existence on myoblasts of fusion proteins analogous to those described for some forms of virus –cell fusion. A viral envelope protein binds to specific receptors on its target cell (Dalgleish et al. 1984) with subsequent exposure of a hydrophobic fusion protein on the virus which contacts the host cell membrane and initiates virus –host fusion. This model would require different types of myotube to express different fusion receptors, specific for each myoblast binding protein type; the hydrophobic fusion proteins themselves need not possess specificity, but would be exposed only after the specific interaction between myoblast agonist and myotube fusion receptor.

This work was supported by the New Zealand Medical Research Council, the Muscular Dystrophy Association of America (grants to A.J.H. and F.R.), the Vernon Willey Foundation, INSERM and CNRS. A.J.H. thanks Professor D. Paulin and the Université Paris 7 for a visiting professorship and M. Fardeau for hospitality in his laboratory. Support for the preparation of monoclonal antibodies NOQ7.5.4D and NOQ7.7.1A by Dr Fitzsimons came from the Postgraduate Committee in Medicine of the University of Sydney.

Betz
,
W. J.
,
Caldwell
,
J. H.
and
Ribchester
,
R. R.
(
1980
).
The effects of partial denervation at birth on the development of muscle fibres and motor units in rat lumbrical muscle
.
J.Physiol., Lond
.
303
,
265
279
.
Bischoff
,
R.
(
1986
).
Proliferation of muscle satellite cells on intact myofibers in culture
.
Devi Biol
.
115
,
129
139
.
Bischoff
,
R.
and
Holtzer
,
H.
(
1969
).
Mitosis and the processes of differentiation of myogenic cells in vitro
.
J. Cell Biol
.
41
,
188
200
.
Butler
,
J.
,
Cosmos
,
E.
and
Brierley
,
J.
(
1982
).
Differentiation of muscle fiber types in aneurogenic brachial muscles of the chick embryo
.
J. exp. Zool
.
224
,
65
80
.
Couteaux
,
R.
(
1937
).
L’origine des “myotubes secondaires” chez les embryons de mammifdres
.
C. r. hebd. Soc. Biol
.
128
,
990
992
.
Couteaux
,
R.
(
1941
).
Rech érches sur I’histog énese du muscle stri é des mammiferes et la formation des plaques motrices
.
Bull. Biol. Franco-Belg
.
75
,
101
239
.
Dalgleish
,
A. G.
,
Beverley
,
P. C. L.
,
Clapham
,
P. R.
,
Crawford
,
D. H.
,
Greaves
,
M. F.
and
Weiss
,
R. A.
(
1984
).
Nature
312
,
763
767
.
Draeger
,
A.
,
Weeds
,
A. G.
and
Fitzsimons
,
R. B.
(
1987
).
Primary, secondary and tertiary myotubes in developing skeletal muscle: a new approach to the analysis of human myogenesis
.
J. Neurol. Sci
.
81
,
19
43
.
Duxson
,
M. J.
,
Usson
,
Y.
and
Harris
,
A. J.
(
1989
).
The origin of secondary myotubes in mammalian skeletal muscles: ultrastructural studies
.
Development submitted
.
Expert-Bezanon
,
A.
,
Barritault
,
D.
,
Mjlet
,
M.
,
Guerin
,
M.-F.
and
Hayes
,
D. H.
(
1977
).
Identification of neighbouring proteins in Escherichia coli 30 S ribosome subunits
.
J. molec. Biol
.
122
,
603
629
.
Gratzner
,
H. G.
(
1982
).
Monoclonal antibody to 5-bromo and 5-iododeoxyuridine: a new reagent for detection of DNA replication
.
Science
218
,
474
475
.
Harris
,
A. J.
(
1981a
).
Embryonic growth and innervation of rat skeletal muscles. I. Neural regulation of muscle fibre numbers
.
Phil. Trans. Roy. Soc., Lond. B
.
293
,
257
277
.
Harris
,
A. J.
(
1981b
).
Embryonic growth and innervation of rat skeletal muscles. II. Neural regulation of junctional and extra-junctional acetylcholine receptor clusters
.
Phil. Trans. Roy. Soc., Lond. B
.
293
,
287
314
.
Harris
,
A. J.
,
Fitzsimons
,
R. B.
and
McEwan
,
J.
(
1989
).
Neural control of the sequence of expression of myosin heavy chain isoforms in foetal mammalian muscles
.
Development submitted
.
Kalderon
,
N.
and
Gilula
,
N. B.
(
1979
).
Membrane events involved in myoblast fusion
.
J. Cell Biol
.
81
,
411
425
.
Kelly
,
A. M.
and
Zacks
,
S. I.
(
1969
).
The histogenesis of rat intercostal muscle
.
J. Cell Biol
.
42
,
135
153
.
Lough
,
J.
and
Bischoff
,
R.
(
1976
).
Differential sensitivity to 5-bromodeoxyuridine during the S phase of synchronized myogenic cells
.
Devi Biol
.
50
,
457
475
.
McLennan
,
I. S.
(
1983
).
Neural dependence and independence of myotube production in chicken hindlimb muscles
.
Devi Biol
.
98
,
287
294
.
Narusawa
,
M.
,
Fitzsimons
,
R. B.
,
Izumo
,
S.
,
Nadal-Ginard
,
B.
,
Rubinstein
,
N. A.
and
Kelly
,
A. M.
(
1987
).
Slow myosin in developing rat skeletal muscle
.
J. Cell Biol
.
104
,
447
459
.
Ontell
,
M.
and
Dunn
,
R. F.
(
1978
).
Neonatal muscle growth: a quantitative study
.
Am. J. Anat
.
152
,
539
556
.
Ontell
,
M.
and
Kozeka
,
K.
(
1984
).
The organogenesis of murine striated muscle: a cytoarchitectural study
.
Am. J. Anat
.
171
,
133
148
.
Ross
,
J. J.
,
Duxson
,
M. J.
and
Harris
,
A. J.
(
1987a
).
Formation of primary and secondary myotubes in rat lumbrical muscles
.
Development
100
,
383
394
.
Ross
,
J. J.
,
Duxson
,
M. J.
and
Harris
,
A. J.
(
1987b
).
Neural determination of muscle fibre numbers in embryonic rat lumbrical muscles
.
Development
100
,
395
409
.
Rubinstein
,
N. A.
and
Kelly
,
A. M.
(
1981
).
Development of muscle fiber specialization in the rat hind limb
.
J. Cell Biol
.
90
,
128
144
.
Schwann
,
T.
(
1847
).
Microscopical Researches into the Accordance in the Structure and Growth of Animals and Plants. (Translated by H. Smith)
, pp.
129
140
.
The Sydenham Society
,
London
.
Stockdale
,
F. E.
and
Miller
,
J. B.
(
1987
).
The cellular basis of myosin heavy chain isoform expression during development of avian skeletal muscles
.
Devi Biol
.
123
,
1
19
.
Swatland
,
H. J.
(
1984
).
Structure and Development of Meat Animals
.
Prentice-Hall
,
Englewood Cliffs, New Jersey
.
Tello
,
J. F.
(
1917
).
Genesis de las terminaciones nerviosas, motrices y sensitivas. I. En el sistema locomotor de los vertebrados superiores
.
Trab. del Lab. de Invest, biol
.
15
,
101
199
.
Wilson
,
S. J.
,
Ross
,
J. J.
and
Harris
,
A. J.
(
1987
).
A critical period for formation of secondary myotubes defined by prenatal undernourishment in rats
.
Development
102
,
815
821
.