Skeletal muscle differentiation relies on signals that induce the transcriptional activity of MyoD, a basic helix-loop-helix transcription factor that acts as a key regulator of the muscle determination program. Muscle differentiation also depends on chromatin-modifying factors that acetylate and remodel histones to unravel chromatin and allow further access of transcriptional machinery to promoters1. The differentiation-activated p38α/β signaling pathway also stimulates the transcriptional activity of MyoD2, but it was not clear whether this occurred directly, through MyoD phosphorylation, or indirectly, perhaps through chromatin modifiers. On page 738, Simone and colleagues3 show that p38α/β signals directly to the SWI-SNF chromatin remodeling complex, thereby recruiting SWI-SNF to specific target promoters, where it cooperates with MyoD to activate transcription of muscle-specific genes.

Modifying chromatin

Previous work showed that inhibiting the p38α/β pathway with the drug SB203580 prevents myoblast differentiation due to the inability of MyoD to activate transcription2,4,5. To further clarify the mechanism and test the hypothesis that p38α/β indirectly regulates MyoD activity, Simone et al. examined the effects of SB203580 on chromatin modifications at the promoters of the differentiation genes myogenin (Myog) and muscle creatine kinase (Ckm). They found that the acetylation status of histones H3 and H4 was unaltered by inhibition of p38α/β signaling. Consistent with this, the interaction between MyoD and the histone acetyltransferases p300 and PCAF6 still occurred at these promoters in the absence of p38α/β signaling. These results suggest that, even when functional MyoD-acetyltransferase complexes are present on these promoters, activation of gene expression requires a further signal from p38α/β.

MyoD contains two domains that mediate the remodeling of chromatin7, and SWI-SNF chromatin remodeling activity is also necessary for MyoD-mediated differentiation8. Simone et al. therefore tested whether p38α/β signaling affects chromatin remodeling at the Myog and Ckm promoters. Indeed, they found that chromatin was not remodeled in the absence of p38α/β signaling, suggesting several possibilities: (i) p38α/β signals directly to MyoD to activate its intrinsic chromatin remodeling activity; (ii) p38α/β signaling recruits the SWI-SNF complex to promoters; (iii) p38α/β activates the SWI-SNF complex already present on promoters; or (iv) some combination of the above.

Targeting SWI-SNF

To narrow down the possibilities, Simone et al. examined the association of BRG1 and BRM, the ATPase subunits of the SWI-SNF complex, with the Myog and Ckm promoters. They found that inhibition of p38α/β prevented the association of BRG1 and BRM with MyoD, implying that p38α/β signaling is involved in recruiting the SWI-SNF complex to specific myogenic promoters. They also showed that p38α/β phosphorylates the BAF60 subunit of the SWI-SNF complex in vitro, suggesting a direct link between p38α/β and SWI-SNF activity.

These studies show that blocking p38α/β activity results in a failure to recruit the SWI-SNF complex to myogenic promoters. What happens when myoblasts are forced to differentiate by the overexpression of the upstream p38α/β activator MKK6? Simone et al. found that forced activation of the p38α/β pathway in myoblasts resulted in premature association of MyoD with BRG1, p300 and PCAF on the Myog promoter, with only the MyoD-BRG1 association dependent on p38α/β signaling.

To determine whether p38α/β signals directly to MyoD in addition to the SWI-SNF complex, Simone et al. transfected SW13 cells deficient in BRG1 and BRM with constructs expressing Gal4 fused to full-length MyoD or to a truncated form of MyoD lacking the chromatin-remodeling domains. They found that neither Gal4 fusion protein could activate a Gal4-dependent reporter further in response to MKK6-induced p38α/β signaling. In fact, further activation of the reporter was not observed until either BRG1 or BRM was added back to these cells. These results provide compelling evidence that p38α/β does not signal directly to MyoD or affect its intrinsic remodeling activity.

Mef2 proteins cooperate with MyoD to induce gene expression9 and are also targets of p38α/β signaling2,5,10. Could Mef2 proteins be mediating some of the observed effects? Notably, the truncated version of MyoD, which lacks the ability to associate with Mef2 (ref. 11), retains the ability to activate the reporter in the presence of BRG1 or BRM and p38α/β signaling. These findings argue against an essential role for Mef2 under the assay conditions used, but further studies are required to examine this question in detail.

Pathways to activation

This study by Simone et al. provides evidence that the p38α/β pathway promotes muscle differentiation by indirectly signaling to MyoD through the SWI-SNF chromatin-remodeling complex, resulting in expression of Myog and Ckm (Fig. 1). Increasing evidence over the past few years suggests that, in addition to targeting transcription factors, kinases can alter gene expression by targeting chromatin-modification factors. Simone et al. have demonstrated this for the first time with the p38α/β pathway in muscle.

Figure 1: Regulation of MyoD-dependent gene expression by p38α/β.
figure 1

(a) The MyoD-p300-PCAF complex is bound to DNA but unable to activate transcription in the absence of further signals. (b) p38α/β targets SWI-SNF to the MyoD-p300-PCAF complex, resulting in remodeling of chromatin at the Myog and Ckm promoters and activation of gene expression.

Full size image

As p38α/β regulates only a small proportion of MyoD target genes12, it will be interesting to determine if other promyogenic signaling molecules are responsible for recruiting chromatin modifiers to different loci. It will also be important to determine exactly how BAF60 phosphorylation influences the selective recruitment of SWI-SNF to specific target loci. Are all p38α/β regulated targets in muscle activated by this mechanism, or do Mef2 proteins mediate some of the observed effects of p38α/β signaling on muscle-specific transcription? Finally, it will be important to determine how this signaling mechanism influences muscle formation in vivo. Mutating the phosphorylated residue(s) of BAF60 or the residue(s) important for the interaction between BRG1 and MyoD, and targeting these mutants to endogenous loci in mice, could be used to determine the importance of this signaling mechanism in regulating the muscle differentiation program.