Tissue development requires the controlled regulation of cell-differentiation programs. In muscle, the Mef2 transcription factor binds to and activates the expression of many genes and has a major positive role in the orchestration of differentiation. However, little is known about how Mef2 activity is regulated in vivo during development. This study characterized a gene, Holes in muscle (Him), which is part of this control in Drosophila. Him expression rapidly declines as embryonic muscle differentiates, and consistent with this, Him overexpression inhibits muscle differentiation. This inhibitory effect is suppressed by mef2, implicating Him in the mef2 pathway. Him downregulates the transcriptional activity of Mef2 in both cell culture and in vivo. Furthermore, Him protein binds Groucho, a conserved, transcriptional corepressor, through a WRPW motif and requires this motif and groucho function to inhibit both muscle differentiation and Mef2 activity during development. Together, these results identify a mechanism that can inhibit muscle differentiation in vivo. It is concluded that a balance of positive and negative inputs, including Mef2, Him, and Groucho, controls muscle differentiation during Drosophila development and suggest that one outcome is to hold developing muscle cells in a state with differentiation genes poised to be expressed (Liotta, 2007).

Analysis of mef2 function during Drosophila muscle development has shown that a major aspect of its role is in the differentiation pathway downstream of the genes that specify muscle. However, Mef2 protein expression precedes muscle differentiation. It is first expressed in the mesoderm at gastrulation, approximately 3 hr after egg laying (AEL). This is approximately 7 hr before the activation at stage 13 (10 hr AEL) of the expression of many known Mef2 target genes, e.g., Mhc, Mlc1, and wupA. This delay implies that the activity of Mef2 is restrained and that other regulatory proteins operate in the control of muscle differentiation during this period. However, little is known about these other proteins nor about how the gene expression at stage 13 is coordinated. This study addresses these unanswered questions through an analysis of the Him gene in muscle differentiation. Him was described in a computational screen (Rebeiz, 2000), and it was isolated separately in an expression screen (Taylor, 2000), but its function has not been analyzed. This study shows that Him is an inhibitor of Mef2 activity and muscle differentiation, and on the basis of this phenotype, it was called Holes in muscle (Him) (Liotta, 2007).

Him has a striking, transient pattern of expression during Drosophila embryogenesis. It is first expressed broadly in the mesoderm during stage 9. This expression then refines, and at stage 12 it is specifically expressed in the precursors of the somatic musculature and of the heart. Him expression then rapidly declines in the somatic mesoderm, such that in 90 min it has disappeared from the differentiating somatic muscle (stage 13). However, it persists in the adult muscle precursors (AMPs), which are set aside in the somatic mesoderm and which remain undifferentiated at this stage, and also in the developing heart. Him protein expression closely resembles that of Him RNA. The disappearance of Him coincides with the expression of Myosin, a classic marker of muscle differentiation. Double labeling with a Him-GFP fusion gene demonstrates that Myosin is expressed only after Him disappears from the developing muscle. The expression of Him in the progenitors of the somatic muscle and its disappearance from differentiating muscle are consistent with a role for Him as an inhibitor of muscle differentiation (Liotta, 2007).

To test whether Him is an inhibitor of muscle differentiation, it was overexpressed in the developing mesoderm by using the Gal4/UAS system. This induced a dramatic reduction in the number of Myosin-expressing cells and thereby produced large gaps or holes in the musculature. It was then asked when in muscle development Him has this effect. Up to stage 13 (10 hr AEL) muscle development proceeds similarly to that of the wild-type. At this stage, developing muscles are seen in the wild-type as small syncytia, which express founder cell markers, e.g., Kruppel, and Mef2, surrounded by Mef2-expressing myoblasts. When Him is overexpressed, the expression of these markers is similar. Subsequently, immunostaining for Mef2 reveals disrupted differentiation at stage 15, and there is increased cell death at stage 16. Together, these findings demonstrate that Him inhibits the differentiation phase, of muscle development, that occurs from stage 13 onward and that produces the morphologically distinct muscles of the functional musculature by the end of embryogenesis. To explore this function further, Him was knocked down by using RNAi from a splice-activated UAS hairpin vector. Although the musculature develops similarly to that of the wild-type, in the knockdown there is impaired muscle differentiation as revealed by disrupted muscle morphology (Liotta, 2007).

Overexpression of Him during muscle development phenocopies the mef2¹¹³ hypomorphic allele. Development of the musculature is inhibited similarly, and many of the residual muscles have a similar, abnormal morphology. This suggests that the two genes function in a common pathway. Consistent with this, Him and Mef2 are coexpressed in somatic muscle progenitors at stage 12, prior to the activation of muscle-differentiation markers such as Myosin. To test whether Him and mef2 genetically interact, both genes were overexpressed together. Strikingly, the inhibition of muscle differentiation caused by Him is rescued toward the wild-type by mef2. Furthermore, overexpression of Him alone induces lethality, and under the conditions of this experiment only 18% survive. This lethality is suppressed by mef2, and there are more than twice as many survivors. Together, the phenotypic analysis and genetic interaction findings indicate that Him functions in the Mef2 pathway that controls muscle differentiation (Liotta, 2007).

The Him protein sequence includes a putative bipartite nuclear localization signal (NLS). Consistent with this, colocalization with the transcription factor Twist in the AMP nuclei shows that Him is predominantly nuclear. The Him protein also has a WRPW motif at its C terminus. This tetrapeptide in this position is found in the Hairy group of transcriptional repressors and mediates their interaction with the corepressor Groucho (Gro). A pulldown assay was used to show that Him can also bind Gro. Moreover, this interaction requires the WRPW motif because Him with the WRPW motif deleted (HimΔWRPW) cannot bind Gro. To investigate the importance of the WRPW for Him function, HimΔWRPW was overexpressed in embryos and it was found that there was no dramatic loss of muscles, in contrast to the effect of full-length Him. Together, these results show that Him can bind Gro through its WRPW tetrapeptide and that this motif is required to inhibit muscle differentiation (Liotta, 2007).

The significance of the Him/Gro interaction in vivo during embryonic muscle development was investigated by overexpressing Him in a gro mutant background. Strikingly, the loss of gro function suppresses the inhibitory effect of Him, showing that Him requires gro to inhibit muscle differentiation. This result, together with the finding that mef2 can suppress the inhibitory effect of Him, indicates that Drosophila muscle differentiation in vivo is controlled by a balance between the activities of Him and Gro on the one hand and Mef2 on the other. The effect of overexpression of Him can be balanced by a reduction in Gro or by an increase in Mef2 (Liotta, 2007).

To further investigate the mechanism of action of Him, it was asked whether Him could inhibit Mef2 activity in cell culture in a direct Mef2-dependent gene-expression assay. When mef2 was transfected into S2 cells, it stimulated the expression of a Mef2-responsive luciferase reporter, and this effect was inhibited by cotransfection with Him. Then whether Him could also inhibit Mef2 activity was tested in the context of muscle development. The effect of Him overexpression on the expression of Mef2 and of β3-tubulin, which is a direct Mef2 target gene in somatic muscle, was analyzed. β3-tubulin expression is strongly reduced in the somatic mesoderm, whereas Mef2 protein expression is similar to that of the wild-type. This indicates that Him can downregulate Mef2 activity in vivo during embryonic development. It was further shown that Him with the Gro-interacting WRPW motif deleted does not affect β3-tubulin expression, nor does full-length Him in a groucho mutant background (Liotta, 2007).

Taken together, this combination of in vitro and in vivo assays reveals key features of Him's mechanism of action. They demonstrate that Him is found in the nucleus and requires its Gro-binding WRPW motif and gro function to inhibit both Mef2 activity and muscle differentiation during development. The previously characterized Drosophila proteins that have a C-terminal Gro-interacting WRPW motif are the Hairy group of HLH domain DNA-binding transcriptional repressors. However, Him is novel and does not have an HLH domain, suggesting that it does not bind DNA directly. Its mechanism of action may have parallels with Ripply1, which functions in vertebrate somitogenesis (Kawamura, 2005). Ripply1 also appears not to be an HLH protein and yet contains a functional Gro-interacting WRPW motif, although in this case near the N-terminus of the protein. Like Ripply1, Him may be part of a transcriptional-repressor protein complex. The precise mechanism by which Him targets Mef2 awaits analysis of this putative complex and the protein partners within it (Liotta, 2007).

Despite considerable progress, much remains to be learned about the regulation of muscle differentiation during animal development. Although studies in cell culture indicate that this control might include negative mechanisms, little is known about the identity and mode of action of specific molecules that inhibit muscle differentiation in vivo during development. This study identified and analyzed the targeting of Mef2 by Him. The inhibitory action of Him, coupled to its transient expression in developing muscle cells, is an explanation for the observation that Mef2 is present significantly before overt differentiation. It also offers an explanation for how a burst of expression of many Mef2 target genes at a specific phase (stage 13) of the differentiation program is coordinated. It is suggested that the rapid decrease in the expression of Him will lead to a concomitant increase in the activity of Mef2 and the ability to activate a cohort of these genes. Further studies will determine whether this will link reports that the ability of Mef2 to bind DNA is temporally regulated (Liotta, 2007).

The results also indicate that the inhibition of Mef2 activity by endogenous levels of Him is incomplete prior to stage 13. Thus, in normal muscle development, the Mef2 target gene β3-tubulin is expressed at stage 12, even though it was found that overexpression of Him can downregulate its expression then. This implies that in the wild-type embryo, there is some Mef2 activity at stage 12, and such activity is sufficient for β3-tubulin expression. This is consistent with other work that indicates that Mef2 regulates some gene expression at this stage and earlier and suggests that Him can provide one level of control of Mef2 activity during the muscle-differentiation program. Taken together, these results move the molecular analysis of muscle differentiation on from a simple model in which the key events are expression of pivotal positive regulators, for example Mef2. Rather they indicate that muscle differentiation in vivo is controlled by a balance of positive and negative regulators, including Him, Gro, and Mef2, that governs whether muscle precursors differentiate. In this model, one can think of Him and Gro as part of a mechanism holding the cells in a committed, but undifferentiated, state in which a cohort of muscle-differentiation genes is poised to be expressed. This might be a widespread strategy for coordinated gene expression in cell-differentiation programs. For example, it can be compared with melanocyte stem cell differentiation, where cells are primed to rapidly express terminal differentiation markers once Pax3/Groucho-mediated repression is relieved (Liotta, 2007).