Developmental biology

One source for muscle

Producing muscle as an embryo, and making or repairing it as an adult, could be considered to be quite different processes. But it seems that cells that share a common origin carry out both of these tasks.

The generation and repair of skeletal muscle is a highly ordered, multi-step process that requires many progenitor cells and continues throughout embryonic, fetal and postnatal life. In this issue, Gros et al. (page 954)1 and Relaix et al. (page 948)2 show that a common progenitor cell not only maintains muscle growth during late embryonic development, but also seems to be the origin of the satellite cells responsible for postnatal muscle growth and repair.

The cellular origin of skeletal muscle is the somites — segmented blocks of tissue that form on either side of the developing spinal cord in the embryo (Fig. 1a). But the exact mechanism by which somitic cells generate the total skeletal-muscle mass is poorly understood. During early embryonic development the somites, resembling a tightly clustered ball of epithelial cells, undergo a programme of maturation and specialization (‘differentiation’) to produce the sclerotome, which forms the skeleton, and the dermomyotome, which will form both the skin (dermis) and skeletal muscle (Fig. 1b).

Figure 1: The making of muscle.

a, Somites form on either side of the developing spinal cord (notocord) of the embryo. One side is shown in cross-section. b, The dermomyotome, which will give rise to muscle and skin, develops from the somite. c, Proliferating muscle precursor cells ingress from the borders of the dermomyotome to form a sheet of non-proliferating, differentiating myoblasts — the primary myotome. d, Pax3- and Pax7-expressing cells originate from the central domain of the dermomyotome1,2. They migrate into the primary myotome, proliferate and do not express markers of myogenic differentiation.

This initial generation of muscle relies on proliferating muscle precursor cells from the borders of the dermomyotome, which ingress beneath the dermomyotome to form what we should now refer to as the primary myotome, a sheet of non-proliferating muscle-precursor cells called myoblasts (Fig. 1c)3. Notably, the dermomyotome is a transient structure that can produce only a limited number of myoblasts. It cannot support the progressive and extensive growth of muscle throughout embryogenesis, and so how the total muscle mass is generated remains a puzzle.

During later stages of embryogenesis, cells termed satellite cells are observed. These cells form a self-replenishing pool of muscle-specific stem cells responsible for postnatal muscle growth and repair. Their embryonic origin is unclear, as traditional embryological studies seem to indicate a somitic origin4, although more recent evidence implies they might be derived from the embryonic vasculature5,6.

Gros et al.1 and Relaix et al.2 have used different cell-labelling strategies to track the movement and persistence of a population of previously undefined skeletal-muscle progenitor cells that originate from the central domain of the dermomyotome. Following cell division, one daughter cell migrates into the primary myotome and continues to proliferate (Fig. 1d), without expressing proteins that signal differentiation into specialized muscle cells. This is contrary to previous theories that the myotome cells become specialized and do not divide — the newly defined cells proliferate in the myotome to produce a source of skeletal muscle.

The authors found that, unlike the cells that initially form the myotome, the skeletal-muscle progenitor cells do not ingress from the borders of the dermomyotome. Instead, they move directly from the central dermomyotome region as it undergoes a transition from epithelial to mesenchymal cells, a process that signals the onset of dermis formation7 and the disintegration of the dermomyotome. Interestingly, around 90% of the skeletal-muscle progenitor cells express both the gene for Pax7, a distinct molecular marker of satellite cells, and that for Pax3, which is implicated in determining the fate of muscle cells. A smaller proportion (less than 10%) express either one gene or the other. The Pax3- and Pax7-expressing cells persist throughout the later stages of development, and go on to form skeletal muscle. In embryos in which both Pax3 and Pax7 are mutated, the cells that would normally coexpress Pax3 and Pax7 adopt non-muscular fates, such as bone or cartilage2.

The satellite cells that emerge during late embryonic development account for the vast majority of muscle progenitor cells. Significantly, the authors of both papers observed that the skeletal-muscle progenitor cells from the central dermomyotome move to where satellite cells reside. Given that these cells accounted for more than 90% of cells in this location, this strongly suggests that most, if not all, satellite cells are derived from the central dermomyotome of the somite. The demonstration that muscle progenitor cells can contribute to the satellite-cell population need not exclude a contribution from other sources; however, it could be argued that any alternative source will be less significant.

Further characterization of these novel skeletal-muscle progenitor cells should be informative. In particular, it would be useful to know when they coexpress Pax3 and Pax7 as opposed to expressing either gene alone, and the effect this has on their fate — do they form skeletal muscle immediately or become satellite cells? Given that only a few cells that are mutated in both Pax3 and Pax7 adopt non-muscle fates, these muscle-progenitor cells may well represent a mixed population of cells with differing developmental potential. That said, these studies1,2 are highly significant because they describe both the second stage of myotome formation (vital for the generation of muscle precursors required for continued muscle growth) and a developmental route for the origin of satellite cells. Remarkably, both sets of progenitors share the same cellular origin — the hitherto unappreciated central portion of the dermomyotome.


  1. 1

    Gros, J., Manceau, M., Thomé, V. & Marcelle, C. Nature 435, 954–958 (2005).

  2. 2

    Relaix, F., Rocancourt, D., Mansouri, A. & Buckingham, M. Nature 435, 948–953 (2005).

  3. 3

    Christ, B. & Ordahl, C. P. Anat. Embryol. (Berl.) 191, 381–396 (1995).

  4. 4

    Armand, O. et al. Arch. Anat. Microsc. Morphol. Exp. 72, 163–181 (1983).

  5. 5

    De Angelis, L. et al. J. Cell Biol. 147, 869–878 (1999).

  6. 6

    Minasi, M. G. et al. Development 129, 2773–2783 (2002).

  7. 7

    Olivera-Martinez, I., Coltey, M., Dhouailly, D. & Pourquié, O. Development 127, 4611–4617 (2000).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

McKinnell, I., Rudnicki, M. One source for muscle. Nature 435, 898–899 (2005).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.