Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nature. Author manuscript; available in PMC 2014 May 13.
Published in final edited form as:
PMCID: PMC4019331

Death brings new life to muscle


Fusion of muscle cells called myoblasts underlies the generation and maintenance of skeletal muscle throughout an animal’s life. Emerging data indicate that cell death acts as a signal to enhance these processes in mammals.

The multinucleated myofibres that comprise skeletal muscle are generated from the iterative fusion of mononucleated myoblasts. During muscle development, myoblast fusion is mediated by an evolutionarily conserved signalling pathway that alters the actin cytoskeleton1,2 — a network of filaments that controls cellular movement and organization. A second, but related, pathway that regulates the actin cytoskeleton drives phagocytosis, the process by which phagocytic cells engulf dead cells3. On page 263 of this issue, Hochreiter-Hufford et al.4 show that the phagocytic pathway also contributes to myoblast fusion. Specifically, they find that the presence of cells dying by the process of apoptosis, and a receptor that recognizes these dying cells, potentiate myoblast fusion during muscle development, regeneration and repair*.

The phospholipid phosphatidylserine is normally present in the inner leaflet of the cell membrane, but serves as an ‘eat me’ signal when exposed on the outer leaflet in apoptotic cells (Fig. 1). On phagocytic cells, BAI1, the seven-pass transmembrane receptor for phosphatidylserine, signals through the conserved ELMO/Dock180/Rac1 signalling pathway to regulate the downstream actin cytoskeleton network and promote engulfment of dying cells5.

Figure 1
Helping hand from a dying cell

In the fruitfly Drosophila, four transmembrane receptors mediate myoblast adhesion, which subsequently activates the ELMO/Myoblast city/Rac signalling module. (Myoblast city is the Drosophila version of Dock180.) At the site of myoblast–myoblast contact, this module remodels the actin cytoskeleton to drive fusion. The ELMO/Dock180/Rac1 module is conserved in zebrafish and mammals. In mammals, however, numerous receptors have been found that mediate myoblast–myoblast contact and adhesion between myoblasts and the developing myofibres called myotubes, with related and novel proteins continuing to be identified1,2. It remains unclear how the different receptors contribute to and coordinate the fusion process.

Previous studies68 have linked exposure of phosphatidylserine on myoblasts to the enhancement of fusion, yet the receptor mediating this signal has not been identified. Hochreiter-Hufford et al. asked whether BAI1 might transduce the phosphatidylserine signal during mammalian myoblast fusion. They find that the levels of this receptor increase during fusion in vitro, and that it is expressed in actively fusing myotubes in vivo. Moreover, increasing the expression of BAI1 enhanced fusion in an ELMO/Dock180/Rac1-dependent manner in vitro, indicating that the signalling pathway downstream of this receptor is intact (Fig. 1).

The authors also report that BAI1 seems to be required for muscle development and repair in vivo. The muscles of mice lacking the Bai1 gene were fully formed but were smaller than those of control animals. And after injury, the regenerating muscles of the mutant mice remained smaller and contained fewer centrally localized nuclei; centrally positioned nuclei are a hallmark of muscle repair9. That muscles can still undergo some fusion in the absence of BAI1 suggests that the BAI1-related proteins BAI2 and/or BAI3, which are also expressed in muscle, may compensate for its loss.

Contact between apoptotic cells and myotubes has been documented4,7 during normal muscle development in vivo. Hochreiter-Hufford and co-workers show that phosphatidylserine externalization, together with the activity of the caspase enzymes that mediate apoptosis, are necessary to enhance fusion. In agreement with previous work10, they find that inhibiting caspase activity using a pan-caspase inhibitor blocks fusion in vitro. Strikingly, the researchers could overcome this inhibition by adding apoptotic myoblasts, or even apoptotic thymocytes — unrelated cells belonging to the blood lineage. Thus, various types of apoptotic cell can provide a cue to healthy muscle cells to promote fusion.

Hochreiter-Hufford et al. did not observe fusion between apoptotic myoblasts and myotubes, whereas earlier work7,8 indicated that even non-apoptotic myoblasts show a feature of apoptosis, transiently externalizing phosphatidylserine before fusion. It could be that different cell populations contribute to fusion: fusion-capable myoblasts transiently expose phosphatidylserine without undergoing apoptosis; and dying myoblasts (and other cell types) enhance the fusion of healthy myoblasts without participating in fusion themselves. Indeed, as noted earlier, apoptotic mouse thymocytes, a cell type that does not fuse, could restore fusion in caspase-inhibited myoblasts. It is not clear how phosphatidylserine is presented by apoptotic cells and non-apoptotic cells so as to allow fusion of the latter but not of the former. Additional phosphatidylserine-binding proteins are required for myoblast fusion in vitro11, suggesting that phosphatidylserine may affect parallel, yet different, pathways during myoblast fusion.

If the apoptotic cells are not fusing with myotubes, how is the phosphatidylserine–BAI1 signal integrated with the other elements of the fusion machinery? It is likely that, in this case, BAI1 is not participating in the regulation of actin necessary for cell fusion, despite the fact that its influence on fusion depends on ELMO/Dock180/Rac1 signalling within the myotube. Because signalling by BAI1 can restore fusion capacity in caspase-inhibited myoblasts, which have decreased expression of several regulators of muscle-gene transcription10, this receptor may target genes essential for myoblast differentiation.

Transient externalization of phosphatidylserine has been observed in other cell-fusing systems — for instance, in trophoblast cells during placental development12. So the exposure of this phospholipid may be part of a conserved fusion mechanism. But is phosphatidylserine–BAI1 signalling common to all fusing systems? Moreover, apoptotic epithelial cells have previously been linked to compensatory proliferation and regeneration in a variety of organisms — from Hydra and planarians to Drosophila and mice13. It would be interesting to know whether phosphatidylserine– BAI1 signalling also induces proliferation and differentiation of satellite cells, the resident stem cells of skeletal muscle.

The present work provides a new dimension to the fields of myoblast fusion, muscle differentiation and muscle regeneration. A better understanding of the link between apoptosis, BAI1 signalling and fusion could lead to interventions that aid muscle recovery, be it after intense physical training or under conditions of disease-associated muscle wasting, prolonged bed rest or in the elderly.


*This article and the paper under discussion were published online on 24 April 2013.


1. Rochlin K, Yu S, Roy S, Baylies MK. Dev. Biol. 2010;341:66–83. [PMC free article] [PubMed]
2. Abmayr SM, Pavlath GK. Development. 2012;139:641–656. [PubMed]
3. Gumienny TL, et al. Cell. 2001;107:27–41. [PubMed]
4. Hochreiter-Hufford AE, et al. Nature. 2013;497:263–267. [PMC free article] [PubMed]
5. Park D, et al. Nature. 2007;450:430–434. [PubMed]
6. Sessions A, Horwitz AF. FEBS Lett. 1981;134:75–78. [PubMed]
7. van den Eijnde SM, et al. J. Cell Sci. 2001;114:3631–3642. [PubMed]
8. Jeong J, Conboy IM. Biochem. Biophys. Res. Commun. 2011;414:9–13. [PMC free article] [PubMed]
9. Chargé SBP, Rudnicki MA. Physiol. Rev. 2004;84:209–238. [PubMed]
10. Fernando P, Kelly JF, Balazsi K, Slack RS, Megeney LA. Proc. Natl Acad. Sci. USA. 2002;99:11025–11030. [PubMed]
11. Leikina E, et al. J. Cell Biol. 2013;200:109–123. [PMC free article] [PubMed]
12. Gauster M, Moser G, Orendi M, Huppertz B. Placenta. 2009;30:S49–S54. [PubMed]
13. Bergmann A, Steller H. Sci. Signal. 2010;3:re8. [PMC free article] [PubMed]