Identification of a critical control element directing expression of the muscle-specific transcription factor MRF4 in the mouse embryo

Skeletal muscle development in the vertebrate embryo critically depends on the myogenic regulatory factors (MRFs) including MRF4 and Myf5. Both genes exhibit distinct expression patterns during mouse embryogenesis, although they are genetically closely linked with multiple regulatory elements dispersed throughout the common gene locus. MRF4 has a biphasic expression profile, first in somites and later in foetal skeletal muscles. Here, we demonstrate by transgenic analysis that elements within a 7.5-kb promoter fragment of the MRF4 gene are sufficient to drive the embryonic wave of expression very similar to the endogenous gene in somites of mouse embryos. In contrast, a 3-kb fragment of the proximal promoter fails to support expression in the myotome, suggesting that essential cis-acting elements are located between -7.5 and -3 kb upstream of MRF4. Further analysis of this sequence delimits an essential region between -6.6 and -5.6 kb that together with the 3-kb promoter fragment directs transgene expression in the epaxial myotome of all somites during the appropriate developmental period. These data provide evidence that the partly overlapping expression patterns of Mrf4 and Myf5 in somites are controlled by distinct regulatory elements. We also show that 11.4 kb sequence upstream of MRF4, including the promoter and the somitic control region identified in this study, is not sufficient to elicit target specificity towards the strong Myf5 (-58/-48 kb) enhancer, suggesting that additional yet unidentified elements are necessary to convey promoter selectivity and protect the MRF4 gene from this enhancer.     

Myf5 and MyoD activation define independent myogenic compartments during embryonic development

Gene targeting has indicated that Myf5 and MyoD are required for myogenic determination because skeletal myoblasts and myofibers are missing in mouse embryos lacking both Myf5 and MyoD. To investigate the fate of Myf5:MyoD-deficient myogenic precursor cells during embryogenesis, we examined the sites of epaxial, hypaxial, and cephalic myogenesis at different developmental stages. In newborn mice, excessive amounts of adipose tissue were found in the place of muscles whose progenitor cells have undergone long-range migrations as mesenchymal cells. Analysis of the expression pattern of Myogenin-lacZ transgene and muscle proteins revealed that myogenic precursor cells were not able to acquire a myogenic fate in the trunk (myotome) nor at sites of MyoD induction in the limb buds. Importantly, the Myf5-dependent precursors, as defined by Myf5(nlacZ)-expression, deficient for both Myf5 and MyoD, were observed early in development to assume nonmuscle fates (e.g., cartilage) and, later in development, to extensively proliferate without cell death. Their fate appeared to significantly differ from the fate of MyoD-dependent precursors, as defined by 258/-2.5lacZ-expression (-20 kb enhancer of MyoD), of which a significant proportion failed to proliferate and underwent apoptosis. Taken together, these data strongly suggest that Myf5 and MyoD regulatory elements respond differentially in different compartments.     

Vascular smooth muscle cells spontaneously adopt a skeletal muscle phenotype: a unique Myf5(-)/MyoD(+) myogenic program.

Smooth and skeletal muscle tissues are composed of distinct cell types that express related but distinct isoforms of the structural genes used for contraction. These two muscle cell types are also believed to have distinct embryological origins. Nevertheless, the phenomenon of a phenotypic switch from smooth to skeletal muscle has been demonstrated in several in vivo studies. This switch has been minimally analyzed at the cellular level, and the mechanism driving it is unknown. We used immunofluorescence and RT-PCR to demonstrate the expression of the skeletal muscle-specific regulatory genes MyoD and myogenin, and of several skeletal muscle-specific structural genes in cultures of the established rat smooth muscle cell lines PAC1, A10, and A7r5. The skeletal muscle regulatory gene Myf5 was not detected in these three cell lines. We further isolated clonal sublines from PAC1 cultures that homogeneously express smooth muscle characteristics at low density and undergo a coordinated increase in skeletal muscle-specific gene expression at high density. In some of these PAC1 sublines, this process culminates in the high-frequency formation of myotubes. As in the PAC1 parental line, Myf5 was not expressed in the PAC1 sublines. We show that the PAC1 sublines that undergo a more robust transition into the skeletal muscle phenotype also express significantly higher levels of the insulin-like growth factor (IGF1 and IGF2) genes and of FGF receptor 4 (FGFR4) gene. Our results suggest that MyoD expression in itself is not a sufficient condition to promote a coordinated program of skeletal myogenesis in the smooth muscle cells. Insulin administered at a high concentration to PAC1 cell populations with a poor capacity to undergo skeletal muscle differentiation enhances the number of cells displaying the skeletal muscle differentiated phenotype. The findings raise the possibility that the IGF signaling system is involved in the phenotypic switch from smooth to skeletal muscle. The gene expression program described here can now be used to investigate the mechanisms that may underlie the propensity of certain smooth muscle cells to adopt a skeletal muscle identity.(J Histochem Cytochem 48:1173-1193, 2000)     

Transdifferentiation of esophageal smooth to skeletal muscle is myogenic bHLH factor-dependent

Previously, coexpression of smooth and skeletal differentiation markers, but not myogenic regulatory factors (MRFs), was observed from E16.5 mouse fetuses in a small percentage of diaphragm level esophageal muscle cells, suggesting that MRFs are not involved in the process of initiation of developmentally programmed transdifferentiation in the esophagus. To investigate smooth-to-skeletal esophageal muscle transition, we analyzed Myf5nlacZ knock-in mice, MyoD-lacZ and myogenin-lacZ transgenic embryos with a panel of the antibodies reactive with myogenic regulatory factors (MRFs) and smooth and skeletal muscle markers. We observed that lacZ-expressing myogenic precursors were not detected in the esophagus before E15.5, arguing against the hypothesis that muscle precursor cells populate the esophagus at an earlier stage of development. Rather, the expression of the MRFs initiated in smooth muscle cells in the upper esophagus of E15.5 mouse embryos and was immediately followed by the expression of skeletal muscle markers. Moreover, transdifferentiation was markedly delayed or absent only in the absence of Myf5, suggesting that appropriate initiation and progression of smooth-to-skeletal muscle transdifferentiation is Myf5-dependent. Accordingly, the esophagus of Myf5(-/-):MyoD(-/-)embryos completely failed to undergo skeletal myogenesis and consisted entirely of smooth muscle. Lastly, extensive proliferation of muscularis precursor cells, without programmed cell death, occurred concomitantly with esophageal smooth-to-skeletal muscle transdifferentiation. Taken together, these results indicate that transdifferentiation is the fate of all smooth muscle cells in the upper esophagus and is normally initiated by Myf5.