The expression of Myf5 in the developing mouse embryo is controlled by discrete and dispersed enhancers specific for particular populations of skeletal muscle precursors.

The development of skeletal muscle in vertebrate embryos is controlled by a transcriptional cascade that includes the four myogenic regulatory factors Myf5, Myogenin, MRF4 and MyoD. In the mouse embryo, Myf5 is the first of these factors to be expressed and mutational analyses suggest that this protein acts early in the process of commitment to the skeletal muscle fate. We have therefore analysed the regulation of Myf5 gene expression using transgenic technology and find that its control is markedly different from that of the other two myogenic regulatory factor genes previously analysed, Myogenin and MyoD. We show that Myf5 is regulated through a number of distinct and discrete enhancers, dispersed throughout 14 kb spanning the MRF4/Myf5 locus, each of which drives reporter gene expression in a particular subset of skeletal muscle precursors. This region includes four separate enhancers controlling expression in the epaxial muscle precursors of the body, some hypaxial precursors of the body, some facial muscles and the central nervous system. These elements separately or together are unable to drive expression in the cells that migrate to the limb buds and in some other muscle subsets and to correctly maintain expression at late times. We suggest that this complex mechanism of control has evolved because different inductive signals operate in each population of muscle precursors and thus distinct enhancers, and cognate transcription factors, are required to interpret them.

Formation of primary and secondary myotubes in aneural muscles in the mouse mutant peroneal muscular atrophy.

The role of motor innervation in supporting and regulating muscle development was studied using aneural muscles in the hindlimb of the mouse mutant peroneal muscular atrophy (pma). This is a single-locus autosomal mutation where homozygous animals lack the common peroneal nerve, so that muscles in the anterolateral compartment of the lower leg develop entirely without innervation. In adults, these muscles are extremely atrophied, and the mice display a clubfoot deformity. The mutant animals provide a preparation in which aspects of muscle formation can be studied in muscles that have never been exposed to direct contact with somatic motor or sensory axons, without pharmacological or surgical intervention. Using quantitative electron microscopy, we found that normal numbers of primary myotubes formed in aneural pma EDL muscles, but a greater than normal proportion degenerated during the first 2 days after their formation. Secondary myotubes appeared at their normal time and position within the muscle, initially in normal numbers, so that the ratio of secondary to primary myotubes initially was greater in pma than in CF1 control strain mice. No abnormalities in ultrastructure were seen until the time of birth, when retardation in development was obvious, together with invading macrophages and degenerating myofibres. The results show that secondary myotube formation in the mouse, as in the chick (B. J. Fredette and L. T. Landmesser, Dev. Biol. 143, 19-35, 1991) is not directly dependent on innervation. In control muscles, secondary myotubes first form in the vicinity of endplates on primary myotubes. No aggregations of ACh receptors or acetylcholinesterase were present in the aneural muscles, showing that these are neurally induced in the mouse, but secondary myotubes formed in their normal position indicating that positional information related to endplate formation is present in aneural muscles.

Regulation of myogenesis in paralyzed muscles in the mouse mutants peroneal muscular atrophy and muscular dysgenesis.

The roles of innervation, muscle electrical activity, and muscle contraction in regulating the formation and survival of primary and secondary myotubes during embryonic and fetal development of skeletal muscle were studied using the mouse mutants peroneal muscular atrophy (pma) and muscular dysgenesis (mdg). The pma phenotype includes the absence of the peroneal division of the sciatic nerve, so muscles in the anterior compartment of the lower hindlimb are aneural throughout development. Muscles in mdg mice are paralyzed due to the absence of excitation-contraction coupling and hyperinnervated due to suppression of motoneuron death in consequence of their paralysis, but otherwise are electrically excitable and receive synaptic transmission. In a quantitative comparison between control and mutant extensor digitorum longus (EDL) muscles at E15, primary myotube numbers were depressed by 20-30% in both mutants and in paralyzed or denervated muscles from control strain animals. The number of secondary myotubes, however, was normal in pma mutants and two and a half times greater than normal in the hyperinnervated mdg EDL muscles, so that the ratio of secondary to primary myotubes was increased by 300% in the mutant with respect to heterozygous or -/- littermates. Chronic paralysis with tetrodotoxin (TTX) caused no further depression of primary myotube numbers in aneural pma muscles, but secondary myotube numbers were reduced by 40%, reducing the ratio of secondary to primary myotubes by 35%. We conclude that during normal development the generation of secondary myotubes depends on neurally evoked electrical activity in primary myotubes, which stimulates mitosis of secondary myoblasts. The effect of TTX shows that aneural pma primary myotubes discharge spontaneous myogenic action potentials, while mdg muscles may receive greater than normal electrical activation due to their hyperinnervation, explaining the presence and numbers of secondary myotubes in the mutant mouse muscles.