Dach2-Hdac9 Signaling Regulates Reinnervation of Muscle Endplates Peter C

STEM CELLS AND REGENERATION RESEARCH ARTICLE Dach2-Hdac9 signaling regulates reinnervation of muscle endplates Peter C. D. Macpherson, Pershang Farshi and Daniel Goldman*

ABSTRACT highly expressed in innervated muscle and suppressed following Muscle denervation resulting from injury, disease or aging results in muscle denervation (Méjat et al., 2005; Tang and Goldman, 2006; impaired motor function. Restoring neuromuscular communication Tang et al., 2009). They mediate activity-dependent suppression of requires axonal regrowth and endplate reinnervation. Muscle activity Myog, a muscle-specific transcription factor that is induced in inhibits the reinnervation of denervated muscle. The mechanism by denervated muscle (Buonanno et al., 1992) and an activator of genes which muscle activity regulates muscle reinnervation is poorly encoding endplate-associated proteins, like nicotinic acetylcholine understood. Dach2 and Hdac9 are activity-regulated transcriptional receptors (nAChRs) and Musk (Méjat et al., 2005; Tang and co-repressors that are highly expressed in innervated muscle and Goldman, 2006; Tang et al., 2006). Thus, Dach2 and Hdac9 are suppressed following muscle denervation. Dach2 and Hdac9 control candidates for mediating the inhibitory effects of muscle activity on the expression of endplate-associated genes such as those encoding muscle reinnervation. nicotinic acetylcholine receptors (nAChRs). Here we tested the idea In order for muscle activity to regulate muscle reinnervation it is that Dach2 and Hdac9 mediate the effects of muscle activity on expected to affect the expression of secreted factors that impact muscle reinnervation. Dach2 and Hdac9 were found to act in a motor axons and their ability to reform neuromuscular connections. collaborative fashion to inhibit reinnervation of denervated mouse Previous studies have suggested that secreted factors like Igf, Bdnf, skeletal muscle and appear to act, at least in part, by inhibiting Gdnf, Nt-3 and Nt-4 (also known as Ntf3 and Ntf5, respectively) are denervation-dependent induction of Myog and Gdf5 gene released from muscle and might be involved in motor neuron expression. Although Dach2 and Hdac9 inhibit Myog and Gdf5 survival, axonal sprouting and maturation (Smith et al., 1985; mRNA expression, Myog does not regulate Gdf5 transcription. Thus, Rassendren et al., 1992; Caroni, 1993; Koliatsos et al., 1993; Myog and Gdf5 appear to stimulate muscle reinnervation through Funakoshi et al., 1995; Wang et al., 1995; Gould et al., 2008); parallel pathways. These studies suggest that manipulating the however, none have been shown to regulate the reinnervation of Dach2-Hdac9 signaling system, and Gdf5 in particular, might be a endplates. good approach for enhancing motor function in instances where Here we report that Dach2 and Hdac9 collaborate to inhibit neuromuscular communication has been disrupted. muscle reinnervation of pre-existing endplates. We show that Myog is regulated in a Dach2/Hdac9-dependent fashion and that loss of KEY WORDS: Muscle, Nerve, Regeneration, Denervation, Myog expression results in delayed muscle reinnervation. Neuromuscular junction, Myogenin, Dach2, Hdac9, Gdf5, Importantly, we found that Dach2-Hdac9 signaling also inhibits Brachypodia, Motor nerve Gdf5 expression and that Gdf5 is a muscle tissue-derived factor that appears to act in a retrograde fashion to stimulate endplate INTRODUCTION reinnervation. Communication between nerve and muscle takes place at the neuromuscular junction (NMJ) where presynaptic motor axons RESULTS release neurotransmitter that is bound by receptors concentrated in Dach2 and Hdac9 inhibit reinnervation of denervated muscle the muscle’s endplate region. This communication is often endplates disrupted in people suffering from neuromuscular diseases and To investigate if Dach2 and Hdac9 controlled muscle reinnervation, age-related sarcopenia, and results in reduced motor function. we first used a nerve transfer model where the distal end of the tibial Therefore, devising strategies for restoring neuromuscular nerve innervating the soleus muscle is severed and transplanted to communication might help people with these conditions. Indeed, the soleus’s periphery, and then allowed to re-grow over the muscle miR206 was found to promote regeneration of neuromuscular surface where it can form ectopic endplates and also reconnect synapses and this appeared to slow disease progression and increase with old endplates (Payne and Brushart, 1997). These experiments survival in a mouse model of amyotrophic lateral sclerosis (ALS) were performed on wild-type (Wt), Dach2−/−, Hdac9−/− and (Williams et al., 2009). Dach2−/−/Hdac9−/− mice. Unlike the predominantly slow fiber Although mechanisms controlling muscle reinnervation are composition of rat soleus muscle, the mouse soleus muscle is mixed poorly understood, it is clear that muscle activity plays an with a slightly higher percentage of fast fibers over slow fibers inhibitory role (Jansen et al., 1973; Cangiano et al., 1980; (Wigston and English, 1992). Upon dissection of soleus muscles at Hennig, 1987). Dach2 and Hdac9 are transcriptional co-repressors 6 weeks post nerve transfer, Dach2−/−/Hdac9−/− muscles appeared to display less relative atrophy than those from Wt, Dach2−/− and −/− Molecular and Behavioral Neuroscience Institute, Department of Biological Hdac9 animals (Fig. 1A, Fig. S1A). Quantification of muscle Chemistry, University of Michigan, Ann Arbor, MI 48109, USA. mass confirmed this observation (Fig. 1B). By contrast, when −/− *Author for correspondence ([email protected]) soleus muscles were not allowed to reinnervate in Wt and Dach2 / Hdac9−/− mice, a similar amount of denervation-dependent atrophy

Received 23 April 2015; Accepted 1 October 2015 was observed (Fig. 1C). This suggested that the reduced atrophy DEVELOPMENT

4038 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

Fig. 1. Recovery of soleus muscle mass and function reflects − − − − muscle reinnervation in Dach2 / /Hdac9 / mice. (A) Representative − − − − images of dissected soleus muscles from Wt, Dach2 / , Hdac9 / and − − − − Dach2 / /Hdac9 / mice at 6 weeks post muscle denervation and nerve transplant. Scale bar: 2 mm. (B) Quantification of reinnervated soleus muscle mass normalized to the contralateral innervated control. Error bars are s.d.; n=3 for 14 and 25 days and n=5 for 42 days post nerve transplant. *P<0.05; **P<0.01 relative to Wt. (C) Quantification of denervated soleus muscle mass normalized to the contralateral innervated control. Error bars are s.e.m.; n=3. noted in Dach2−/−/Hdac9−/− muscles after nerve transfer might have resulted from an enhanced rate of muscle reinnervation. Encouraged by the above results, we quantified the number of reinnervated endplates in soleus muscles of Wt, Dach2−/−, Hdac9−/− and Dach2−/−/Hdac9−/− mice after muscle denervation with subsequent nerve transfer. For this analysis βIII tubulin Fig. 2. Dach2 and Hdac9 inhibit reinnervation of endplates in denervated immunofluorescence was used to visualize regenerating nerves soleus muscle following nerve transfer. (A) βIII tubulin staining of and presynaptic terminals, and fluorescent α-bungarotoxin (α-Btx) motor nerve terminals overlaps precisely with αBtx stained endplates. was used to visualize muscle endplates (Fig. 2A). Depending on the (B) Representative images and (C) quantification of reinnervated endplates at extent of βIII tubulin and α-Btx overlap, endplates were scored as 6 weeks post nerve transfer. βIII tubulin+ regenerating tibial nerve branches are + completely reinnervated (100% overlap), partially reinnervated green and αBtx endplates are red. Arrows point to fully innervated endplates. Arrowheads point to denervated endplates. Scale bar: 50 µm. Error bars are (∼10-80% overlap) or denervated (no overlap). Within the first −/− −/− −/− −/− −/− −/− s.e.m.; n=4 for Wt and Dach2 ; n=3 for Hdac9 and Dach2 /Hdac9 . couple of days of muscle denervation, Wt and Dach2 /Hdac9 *P<0.05; **P<0.01 relative to Wt. (D) Soleus force measurements following mice exhibited a similar rate of motor nerve degeneration with indirect stimulation through the transplanted tibial nerve. Error bars are s.e.m.; − − − − ∼90% of the muscle endplates exhibiting no association with βIII n=6 for Wt and n=5 for Dach2 / /Hdac9 / .*P<0.05; **P<0.01 relative to Wt. tubulin immunofluorescence and a complete loss of SV2 Inn, innervated; Den, denervated. presynaptic terminal staining (data not shown). When denervated soleus muscles were evaluated for endplate reinnervation at 6 weeks significant endplate fragmentation was noted. However, consistent post nerve transfer, we found significant differences at the original with previous studies suggesting that Dach2 and Hdac9 mediate pre-existing endplates, which were more completely innervated activity-dependent suppression of Myog and genes encoding in Dach2−/−/Hdac9−/− mice compared with Wt, Dach2−/−,or nAChR proteins (Méjat et al., 2005; Tang and Goldman, 2006), Hdac9−/− mice (Fig. 2B,C). The enhanced reinnervation of we found their combined deletion in innervated and denervated Dach2−/−/Hdac9−/− muscle was reflected by a stronger muscle muscle generally increased the expression of these genes more so twitch and tetanic force when stimulating the transplanted than their individual deletion (Fig. S1C,D). Taken together, these nerve (Fig. 2D). Consistent with these observations, we also data suggested that Dach2 and Hdac9 collaborate to inhibit muscle found a tendency for an increase in the total number of ectopic reinnervation. αBtx+ plaques formed in Dach2−/−/Hdac9−/− muscles when they In the above model of muscle reinnervation, the interpretation of were evaluated 14 days post nerve transfer (Wt=433±132 versus the results could have been influenced by the precise location of the Dach2−/−/Hdac9−/−=580±161). Over time it became increasingly nerve transplant as well as the degree of muscle damage induced difficult to visualize ectopic endplates as the transplanted nerve during transplantation. Therefore, we next tested if Dach2 and preferentially migrated toward the original endplates. Hdac9 also inhibited endplate reinnervation using a nerve crush When our analysis of presynaptic nerve terminal overlap with model in which the tibial nerve branch to the soleus muscle is endplates was applied to innervated muscle, we found no significant specifically directed to original endplates by regenerating through differences between Wt and Dach2−/−/Hdac9−/− mice (Fig. 3B). its perineural sheath. As in the nerve transplant experiments, we Similarly, endplate surface areas and muscle fiber widths were found a significant enhancement of endplate reinnervation in −/− −/− −/− unaffected by Dach2 and Hdac9 deletion (Fig. S1B), and no Dach2 /Hdac9 mice compared with Wt, Dach2 ,or DEVELOPMENT

4039 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

results in expression that is restricted to electroporated muscle fibers (Kong et al., 2004; Sadasivam et al., 2005). Consistent with these studies, we find GFP confined to muscle and no indication of electroporation of Schwann cells or nerve (Fig. S1G). We also note that in contrast to denervated muscle, there are no macrophages in innervated muscle so they too cannot be electroporated (Fig. S1H). We generally observe ∼30% of the soleus fibers are electroporated at the site of DNA injection (Fig. 4B). At 12 days post nerve crush, the number of fully innervated and denervated endplates on GFP+ electroporated fibers was quantified. Consistent with the idea that Dach2 and Hdac9 act on muscle gene expression programs to regulate endplate reinnervation, forced expression of Dach2 and/or Hdac9 in denervated Dach2−/−/Hdac9−/− soleus muscle suppressed reinnervation (Fig. 4C,D). The incomplete inhibition of endplate reinnervation by Dach2 and Hdac9 overexpression in muscle might indicate insufficient expression levels, the contribution of other cell types, or the involvement of additional inhibitory pathways. Nonetheless, these data suggest Dach2 and Hdac9 act in a redundant fashion to suppress muscle reinnervation.

Myog stimulates muscle reinnervation Our data suggested that Dach2 and Hdac9 might act by repressing the expression of genes contributing to enhanced reinnervation of denervated muscle. Previous studies have shown Dach2 and Hdac9 repress Myog expression and that Myog regulated the expression of genes that are associated with endplate formation, like those encoding nAChRs and Musk (Méjat et al., 2005; Macpherson et al., 2006; Tang and Goldman, 2006; Tang et al., 2006, 2009). Interestingly, miR206, a gene previously shown to participate in endplate reinnervation (Williams et al., 2009) is also regulated in a Dach2/Hdac9-dependent manner (Fig. S1E). Because Myog is a transcription factor that has the potential to Fig. 3. Dach2 and Hdac9 inhibit reinnervation of endplates in denervated regulate gene expression programs impacting synapse regeneration, soleus muscle following nerve crush. (A) Representative images and (B,C) we investigated whether Myog deletion in adult muscle repressed quantification of innervated and denervated NMJs at various times post nerve + + endplate reinnervation. In innervated muscle, expression of nAChR- crush. βIII tubulin regenerating tibial nerve branches are green and αBtx endplates are red. Arrows point to fully innervated endplates. Arrowheads point encoding genes is repressed (Goldman et al., 1988; Goldman and to denervated endplates. Scale bar: 50 µm. Error bars are s.e.m.; n=3. Staple, 1989; Buonanno et al., 1992) and Myog deletion modestly *P<0.05; **P<0.01; ***P<0.001 relative to Wt. reduced this basal expression (Fig. S2A). Importantly, endplate surface areas and muscle fiber widths were unaffected by Myog Hdac9−/− mice (Fig. 3A-C). This enhancement was not restricted to deletion (Fig. S2B) and no significant endplate fragmentation was the soleus muscle, as denervated Dach2−/−/Hdac9−/− EDL muscle noted. By contrast, Myog deletion prior to nerve crush resulted in also exhibited a greater degree of reinnervation compared with Wt significant delays in the rate of endplate reinnervation (Fig. 5) and muscle at 14 days post peroneal nerve crush (Fig. S1F). The this was reversed by forced expression of Myog (Fig. S2C). Thus, quantitative differences noted in muscle reinnervation when nerves Myog is under control of the Dach2-Hdac9 signaling system and its were allowed to grow through their perineural sheath versus across induction following soleus muscle denervation enhances endplate the muscle surface might reflect differences in these two reinnervation. environments that are further amplified in mutant muscle. Gdf5 is induced in denervated muscle in a Dach2/Hdac9- Dach2 and Hdac9 act on muscle to inhibit endplate dependent manner reinnervation Our data are consistent with the idea that Dach2 and Hdac9 suppress In addition to their expression in muscle, we found that Dach2 was the expression of genes that stimulate muscle reinnervation in adult expressed in motor neurons and cells associated with the motor muscle. Interestingly, even the reduced levels of Dach2 and Hdac9 nerves, whereas Hdac9 was induced in cells associated with injured found in denervated muscle are sufficient to inhibit muscle motor nerves (Fig. 4A). This raised the possibility that Dach2 and reinnervation (Figs 2, 3). Based on these observations we Hdac9 might act on both muscle and nerve to regulate hypothesized that genes mediating the effects of Dach2 and muscle reinnervation. To test if Dach2 and Hdac9 acted on Hdac9 would be induced following muscle denervation and muscle to regulate endplate reinnervation, we electroporated further increased in denervated muscle of Dach2−/−/Hdac9−/− Dach2−/−/Hdac9−/− soleus muscle with sCMV expression mice. Furthermore, because the products of these genes stimulate vectors, allowing forced expression of Dach2 and Hdac9 muscle reinnervation, we suspected they would encode secreted individually or in combination (Fig. 4C,D). Electroporations also proteins that could act on motor neurons. Previously identified included sCMV:GFP so electroporated fibers could be identified. candidate genes included Gdnf, Bdnf, Cntf, Nt-3 and Fgfbp1

Previous studies have shown that soleus muscle electroporation (Caroni, 1993; Koliatsos et al., 1993; Funakoshi et al., 1995; DEVELOPMENT

4040 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

Fig. 4. Dach2/Hdac9 inhibits muscle reinnervation. (A) RT-PCR analysis of Dach2, Hdac9 and γ-Actin mRNA in innervated (Inn) and denervated (Den) muscle; intact or axotomized (Axot) motor neurons and intact or crushed motor nerves and associated cells (Crush). (B) Representative image of soleus muscle cross section electroporated with CMV:GFP expression vector and stained for GFP immunofluorescence. Dashed line marks the muscle borders. − − − − (C) Representative images and (D) quantification of reinnervated endplates at 12 days post nerve crush in Wt and Dach2 / /Hdac9 / soleus muscles that were electroporated with GFP or Dach2, Hdac9 and GFP expression vectors. GFP marks electroporated muscle fibers green, αBtx marks endplates red and βIII tubulin marks regenerating axons cyan. Arrows point to fully and partially innervated endplates, whereas arrowheads point to denervated endplates. Error bars are s.d.; − − − − − − − − − − − − n=3 for Wt+GFP; n=4 for Dach2 / /Hdac9 / +GFP, n=3 for Dach2 / /Hdac9 / + GFP + Hdac9 and Dach2 / /Hdac9 / + GFP + Dach2 + Hdac9. **P<0.01; − − − − ***P<0.001 relative to Dach2 / /Hdac9 / + GFP. Scale bar: 50 µm.

Griesbeck et al., 1995; Wang et al., 1995; Gould et al., 2008; For this analysis, we compared the transcriptomes from Wt Williams et al., 2009); however, none of these met our innervated muscle, Wt denervated muscle and Dach2−/−/Hdac9−/− criterion of being super-induced in 3-day denervated muscle from denervated muscle. Candidate genes were identified by focusing on Dach2−/−/Hdac9−/− mice (Fig. S3A). Therefore, we developed an differentially expressed genes with an abundance greater than five RNAseq-based screen to identify genes that fit the above criterion. fragments per kilobase of exon per million fragments mapped

− − Fig. 5. Myog stimulates endplate reinnervation. (A) Representative images and (B,C) quantification of innervated and denervated endplates in Wt and Myog / soleus muscle at different times post nerve crush. βIII tubulin+ regenerating motor nerve branches are green and αBtx+ endplates are red. Arrows point to partially or fully innervated endplates and arrowheads point to denervated endplates. Scale bar: 50 µm. Error bars are s.d.; n=3 for 7 days and n=4 for 14, 21 and 28 days post nerve crush. ***P<0.001 relative to Wt. DEVELOPMENT

4041 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

(FPKM), induced at least two-fold after Wt muscle denervation, Hdac9 were deleted (Fig. 6A,B). Electroporation of denervated super-induced in denervated muscle of Dach2−/−/Hdac9−/− mice muscle with either a combination of sCMV:GFP, sCMV:Dach2 and and encoded a secreted protein (http://spd.cbi.pku.edu.cn/). This sCMV:Hdac9 expression vectors or an sCMV:GFP vector alone, analysis identified Serpina3m, Serpina3n and Gdf5 (Fig. S3B). Of also suggested Dach2/Hdac9 inhibit denervation-dependent Gdf5 these, the most highly induced gene was Gdf5, a member of the bone mRNA expression (Fig. S3C). Gdf5 mRNA levels peaked within morphogenetic protein (Bmp) family (Nishitoh et al., 1996). 3 days post denervation of soleus muscle and then decreased over Although Bmps have not been previously associated with the next 10 days (Fig. 6C,D). This timeframe corresponds to when mammalian NMJs, they have been demonstrated to serve as most endplates begin to reinnervate, as only ∼30% of the endplates retrograde signals that mediate activity-dependent remodeling of are fully denervated at 7 days post nerve crush and this is reduced to the Drosophila NMJ (Berke et al., 2013; Piccioli and Littleton, 2014). less than 10% at 14 days post nerve crush (Fig. 3C). Because of its high induction in Dach2−/−/Hdac9−/− denervated Using nAChRε (also known as Chrne) mRNA as a positive control muscle and this evolutionary connection, we decided to investigate if for an RNA localized to NMJs (Goldman and Staple, 1989), we Gdf5 was a muscle tissue-derived factor that could regulate muscle found Gdf5 mRNA is ∼3-fold more abundant in junctional than reinnervation in an activity-dependent manner. extrajunctional regions of denervated muscle (Fig. 6E,F), which We first confirmed the RNAseq data using RT-PCR and qPCR. might suggest a gradient of Gdf5 expression emanating from This analysis revealed ∼80-fold induction of Gdf5 mRNA at 3 days endplates. Surprisingly, denervation-dependent Gdf5 induction was post denervation which was further enhanced when Dach2 and independent of Myog expression (Fig. 6G). When comparing nerve

Fig. 6. Gdf5 gene expression is regulated by muscle activity in a Dach2/Hdac9-dependent manner. (A) RT-PCR and (B) qPCR show Gdf5 mRNA − − − − levels in innervated (Inn) and denervated (Den) soleus muscle from Wt and Dach2 / /Hdac9 / mice. Error bars are s.d.; n=3. **P<0.01. (C) RT-PCR and (D) qPCR shows changes in Gdf5 mRNA at various times post muscle denervation. Error bars are s.d.; n=3. **P<0.01; ***P<0.001. (E) RT-PCR and (F) qPCR show Gdf5 and nAChRε mRNA expression in junctional and extrajunctional regions of innervated and denervated soleus muscle. mRNA values are relative to innervated nAChRε mRNA (Y-axis on the left) and to innervated Gdf5 mRNA (Y-axis on the right). Error bars are s.d.; n=4. *P<0.05; **P<0.01; ***P<0.001. − − (G) qPCR show Gdf5 mRNA levels in innervated (Inn) and denervated (Den) soleus muscle from Wt and Myog / mice. Error bars are s.d.; n=3 for Wt and n=2 for − − Myog / . (H) RT-PCR and (I) qPCR shows expression levels of the indicated mRNAs in innervated and denervated soleus muscle, uninjured and axotomized (Ax) motor neurons and cells associated with uninjured or crushed (Cr) sciatic nerve. For qPCR values were normalized to γ-Actin for muscle and Gapdh for motor neuron and sciatic nerve. Error bars are s.d.; n=3. *P<0.05; **P<0.01 relative to uninjured for each tissue. DEVELOPMENT

4042 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674 and muscle components, we found Gdf5 predominantly in Although Gdf5bp-4j mice exhibited a trend towards decreased denervated muscle, whereas Gdf5 receptors (Bmpr1a, 1b and 2) nAChRα (also known as Chrna1) and increased nAChRε gene are in muscle, motor nerve and sciatic nerve-associated cells expression these were not statistically significant changes (Fig. S4C). (Fig. 6H,I). However, motor neurons appeared to preferentially At 7 days post tibial nerve crush, the regenerating Gdf5bp-4j tibial express high levels of the preferred Gdf5 receptor complex Bmpr1b/ nerve appeared to exhibit excessive branching, which was quantified Bmpr2 (Fig. 6H) (Nishitoh et al., 1996). These data are consistent by counting the number of βIII tubulin+ nerve branches that with the idea that Gdf5 is a factor produced by muscle tissue that can contacted or crossed over an individual endplate (Fig. 8A,B). act on both muscle and nerve. Because of the excessive branching noted in the regenerating tibial nerve of Gdf5bp-4j mice, we were concerned that βIII tubulin staining Gdf5 stimulates muscle reinnervation might have difficulty in distinguishing nerves growing over We next investigated if Gdf5 knockdown in denervated muscle endplates from those forming NMJs. Therefore, we used SV2 affected muscle reinnervation. For this analysis we first immunofluorescence to quantify synapse regeneration at 7 days post electroporated muscle with a CMV:GFP expression vector and nerve crush and found more fully denervated endplates in Gdf5bp-4j one of two different Gdf5-targeting siRNAs. We then crushed the mice (Fig. 8C,D; Fig. S4D-F). Although many endplates had nerve tibial nerve branch leading to the soleus muscle. Dissection of GFP+ branches growing over them in the Gdf5bp-4j mice at 7 days post fibers and qPCR for Gdf5 mRNA indicated robust Gdf5 knockdown nerve crush (Fig. 8A,B), most (∼80%) were scored as denervated as (Fig. 7A,B). Although Gdf5 knockdown had no effect on NMJs in they lacked differentiated SV2-positive nerve terminals (Fig. 8C,D). innervated muscle (Fig. 7D), nerve crush experiments showed that By contrast, most nerve branches contacting endplates in Wt mice at muscle fibers with Gdf5 knockdown (GFP+ fibers) did not 7 days post nerve crush exhibited some SV2 staining (Fig. 8C,D). It reinnervate endplates as well as Wt muscle (Fig. 7C,D). A control is noteworthy that even at 7 days post nerve crush in Wt muscle, βIII siRNA targeting E. coli lacZ mRNA had no effect on muscle tubulin and SV2 staining gave comparable results (∼35% fully reinnervation (Fig. 9A). denervated endplates; Fig. 3C, Fig. 8D). At later time points βIII To further investigate a role for Gdf5 in muscle reinnervation and tubulin staining proved reliable in labeling nerve terminals in Gdf5bp- confirm the above results, we obtained Gdf5bp-4j mice from the 4j mice and demonstrated that the lag in muscle reinnnervation in Jackson Laboratory (stock 014183) that harbor a Gdf5 mutation that Gdf5bp-4j mice persisted out to 21 days post nerve crush (Fig. 8D). causes brachypodism (shortened digits and limbs). Inspection of the Importantly, Gdf5bp-4j motor nerve terminals degenerated at a similar soleus muscle in these animals revealed thinner muscle fibers and rate as that observed in Wt and Dach2−/−/Hdac9−/− mice. We smaller endplates that despite their diminished size, maintained their electroporated Gdf5bp-4j soleus muscle with either CMV:GFP alone appropriate proportionality and were fully innervated (Fig. S4A,B). or a combination of CMV:GFP and CMV:Gdf5 expression vectors

Fig. 7. Gdf5 knockdown inhibits muscle reinnervation. (A) RT-PCR and (B) qPCR shows siRNAs 24 and 36 suppress Gdf5 mRNA levels. Error bars are s.d.; n=3 for control, n=1 for siRNA 24 and n=2 for siRNA 36. (C) Representative images of reinnervated endplates at 21 days post nerve crush in Wt soleus muscle electroporated with sCMV:GFP or a combination of sCMV:GFP and Gdf5-targeting siRNA24. αBtx marks endplates red; βIII tubulin marks regenerating axons cyan; and GFP identifies electroporated fibers. Arrows point to innervated endplates; arrowheads point to denervated endplates. Scale bar: 50 µm. (D) Quantification of regenerated neuromuscular synapses with and without Gdf5 knockdown. Error bars are s.d.; n=4 for GFP-treated 14 days post nerve crush and n=3 for 21 days post nerve crush. *P<0.05; **P<0.01 relative to GFP control. DEVELOPMENT

4043 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

Fig. 8. Gdf5 stimulates muscle reinnervation. (A) Representative images showing increased axonal branching over the soleus muscle in Gdf5bp-4j mice at 7 days post nerve crush. βIII tubulin+ regenerating motor nerve branches are green and αBtx+ endplates are red. Arrowheads point to a single axonal branch innervating endplates, whereas arrows point to multiple axonal branches coursing over the endplate. Scale bar: 100 µm. (B) Quantification of the number of regenerating axonal branches/endplate in Wt and Gdf5bp-4j mice. Error bars are s.d.; n=3. *P<0.05; **P<0.01; ***P<0.001 relative to Wt. (C) Representative images showing reduced differentiation of axon terminals at endplates in Gdf5bp-4j mice at 7 days post nerve crush compared with Wt mice. SV2+ differentiated axon terminals are cyan, βIII tubulin+ regenerating motor nerves are green and αBtx+ endplates are red. Arrowheads point to SV2+ differentiated axon terminals that are also βIII tubulin+ and innervating αBtx+ endplates. (D) Quantification of denervated endplates in Wt and Gdf5bp-4j mice. Error bars are s.d.; n=3 for 7 and 14 days post nerve crush; n=4 for 21 days post nerve crush. **P<0.01; ***P<0.001 relative to Wt.

and found that forced expression of Gdf5 in muscle rescued the poor experimental siRNA targeting mouse Gdf5 mRNA. Fourteen days reinnervation phenotype of Gdf5bp-4j mice (Fig. S4G). This analysis post muscle denervation, muscles were sectioned and fully is consistent with the idea that denervation-dependent Gdf5 innervated endplates were quantified. This analysis showed that expression in muscle enhances muscle reinnervation and synapse knockdown of Gdf5 reduced the number of fully innervated formation. endplates by ∼40% (Fig. 9A,B). Thus, Gdf5 contributes to both the Our analysis of muscle reinnervation in Dach2−/−/Hdac9−/− normal reinnervation of endplates in Wt mice and the enhanced mice revealed enhanced rates of muscle reinnervation (Figs 2, 3) reinnervation of endplates noted in Dach2−/−/Hdac9−/− mice. that was correlated with Gdf5 mRNA super-induction (Fig. 6A,B). To directly test if Gdf5 mRNA induction contributed to the Identification of the Gdfbp-4j mutation causing brachypodism − − − − enhanced muscle reinnervation noted in Dach2 / /Hdac9 / mice, and reduced endplate reinnervation we electroporated denervated Dach2−/−/Hdac9−/− muscle with The Gdf5bp-4j mouse harbors a spontaneous recessive mutation that either a control siRNA targeting E. coli lacZ mRNA or an maps to the brachypodism locus on chromosome 2 and

Fig. 9. Gdf5 knockdown inhibits endplate − − − − reinnervation in Dach2 / /Hdac9 / mice. (A) Quantification of the percentage of fully innervated endplates at 14 days post nerve − − − − crush (dpc) in Dach2 / /Hdac9 / mice whose denervated soleus muscles were electroporated with a CMV:GFP expression vector and a siRNA targeting either E. coli lacZ mRNA or mouse Gdf5 mRNA. Error bars are s.d.; n=3. ***P<0.001. (B) Representative images of CMV:GFP and siRNA electroporated − − − − Dach2 / /Hdac9 / muscle fibers. GFP (green) identifies electroporated fibers, αBtx (red) identifies muscle endplates and βIII tubulin (cyan) identifies motor nerves. Arrowheads point to partially innervated and denervated endplates, whereas arrows point to fully innervated endplates. Scale bar: 50 µm. DEVELOPMENT

4044 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674 complementation tests with other brachypodism alleles proved this The reinnervation of denervated muscle is a multistep process mutation is an allele of Gdf5 (Jackson Laboratory, stock 014183). where motor axons must regrow over the muscle surface, recognize However, the nature of this mutation remained unknown. RT-PCR and approach denervated endplates, and then grow over this confirmed normal denervation-dependent Gdf5 mRNA expression in structure so axon terminals are in perfect register with the Gdf5bp-4j mouse muscle (Fig. 10A). To identify the Gdf5bp-4j underlying endplate. Our data suggests that the Dach2-Hdac9- mutation, we sequenced Gdf5 cDNA prepared from denervated Wt Gdf5 signaling system might participate in many of these events. and Gdf5bp-4j muscle. This analysis identified a 4 bp deletion in the The observation that Gdf5 mRNA is expressed in a gradient Gdf5 mRNA from Gdf5bp-4j mice that causes a reading frame shift in emanating from denervated endplates might indicate that the the carboxy-terminal region (Fig. 10B,C). Amino acid changes in this regenerating motor axons use this gradient to identify old region have previously been implicated in causing brachypodism in endplates. Furthermore, the reduced differentiation of axon the Gdf5bp-5j mouse (T484N) (Jackson Laboratory, stock 021333) terminals and increased axonal branching in Gdf5bp-4j mice and in humans (C498S) (Everman et al., 2002). These data suggest suggests that the Gdf5 signaling gradient might contribute to the possibility that human mutations causing brachypodism might endplate recognition by the axon, axonal growth arrest and axon also inhibit muscle reinnervation. terminal differentiation. Dach2 and Hdac9 are transcriptional corepressors that are highly DISCUSSION expressed in innervated muscle and decrease following muscle The work reported here identifies a Dach2-Hdac9 signaling system denervation (Méjat et al., 2005; Tang and Goldman, 2006; Tang that mediates the inhibitory effects of muscle activity on muscle et al., 2009). Using two models of nerve injury we show that Dach2 reinnervation and appears to do so, at least in part, by inhibiting the and Hdac9 reduce the rate of muscle reinnervation. These factors expression of RNAs encoding Myog, miR206 and Gdf5. We found appear to act in a collaborative fashion as their individual deletion that Gdf5 is specifically expressed in muscle tissue and appears to had only small effects. Remarkably, even in denervated muscle, act in a retrograde fashion to stimulate endplate reinnervation. Our residual Dach2 and Hdac9 are sufficient to inhibit muscle analysis of Gdf5 mRNA expression suggests that Gdf5 is enriched reinnervation. Our data suggest that Dach2 and Hdac9 mediate in endplate regions of the denervated muscle fiber. Finally, we their effects, at least in part, by controlling Myog and Gdf5 identified a 4-nucleotide deletion in the Gdf5 gene that is expression. Both Myog and Gdf5 mRNAs are induced in Wt responsible for both brachypodia and reduced muscle denervated muscle and super-induced in denervated muscle from reinnervation in Gdf5bp-5j mice. Dach2/Hdac9 null mice. Importantly, Myog deleted mice and

bp-4j Fig. 10. Identification of Gdf5 gene mutation in Gdf5 mice. (A) RT-PCR shows normal denervation-dependent induction of Gdf5 mRNA in the Gdf5bp-4j mouse soleus muscle. (B) Diagram of Gdf5 protein sequence and DNA sequence of the Gdf5 carboxy-terminal region cDNA prepared from denervated Wt and Gdf5bp-4j muscle. A four nucleotide deletion is identified that alters the Gdf5 reading frame (C) in Gdf5bp-4j mice (amino acids in red). Mutations known to cause brachypodism in this region of the Gdf5 protein are indicated with arrows above the protein sequence. Cysteines that contribute to intrachain disulfide bonds are indicated with asterisks above the protein sequence. DEVELOPMENT

4045 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

Gdf5bp-4j mutant mice had a slower rate of soleus muscle that loss of Gdf5 inhibits muscle reinnervation, we found that reinnervation than Wt mice, and the enhanced reinnervation of denervation-dependent induction of Gdf5 was independent of endplates noted in Dach2−/−/Hdac9−/− mice is inhibited by Gdf5 Myog. This suggests that the Dach2-Hdac9 signaling system knockdown. Although our observations on the effect of Myog and participates in the regulation of at least two signaling pathways Gdf5 on muscle reinnervation were limited to soleus muscle, these that converge on endplate reinnervation. proteins are induced after denervation of other skeletal muscles Although our data suggest Gdf5 regulates muscle reinnervation, a (Tang and Goldman, 2006; Tang et al., 2009; Sartori et al., 2013); number of additional considerations regarding Gdf5 remain to be however, the generality of their effect on muscle reinnervation resolved. First, it is not known if the Gdf5 mRNA gradient is remains to be tested. reflected by Gdf5 protein expression and whether a gradient of Gdf5 The Dach2-Hdac9-Myog signaling system might contribute to expression is important to endplate reinnervation. Second, although endplate reinnervation by regulating endplate-associated genes, like our data demonstrates that Gdf5 is highly enriched in denervated those encoding Musk and nAChRs that are crucial components of the muscle tissue relative to nerve-associated cells and motor neurons, endplate complex. In addition, miR206 is an important regulator of we cannot currently rule out the possibility that this enrichment muscle reinnervation (Williams et al., 2009) and our data indicates occurs in nerve terminal associated Schwann cells, or some other that miR206 is regulated by the Dach2-Hdac9 signaling system. muscle-associated cell type. However, its expression in Interestingly, the miR206 promoter harbors Myog consensus E-box extrajunctional regions of the muscle does suggest that at least binding sites that control its induction in denervated muscle (Rao part of the Gdf5 signal is derived from muscle. Third, we cannot et al., 2006; Williams et al., 2009). Based on these regulatory unequivocally state that Gdf5 functions in a retrograde manner to properties and the observations that loss of Myog or miR206 promote muscle reinnervation. Our data indicate that receptors for expression results in attenuated rates of reinnervation, we propose that Gdf5 are expressed in both muscle tissue and motor neurons. Given the Dach2-Hdac9 signaling system inhibits muscle reinnervation, in the observation that Gdf5 expression attenuates the magnitude of part by suppressing a Myog-miR206 signaling cascade. In addition, denervation-induced atrophy (Sartori et al., 2013), it is possible that we suspect other factors will also impinge on these signaling the delay in reinnervation observed in Gdf5bp-4j mice was secondary pathways and two such candidates are Msy3 (also known as Ybx3) to the loss of Gdf5 signaling within the muscle itself. Resolving this and MyoD (also known as Myod1). Msy3 is induced in innervated issue will require experiments involving tissue-specific knockout of muscle and, like Dach2 and/or Hdac9, suppresses Myog gene Gdf5 receptors in both muscle and motor nerves. expression (Berghella et al., 2008) and thereby, might inhibit muscle It is interesting that the Dach2-Hdac9 signaling system regulates a reinnervation. MyoD is an activity-regulated transcription factor that variety of activity-dependent processes like muscle reinnervation, is induced in denervated muscle (Buonanno et al., 1992) and often muscle metabolism, muscle fiber type determination and muscle collaborates with Myog to regulate gene expression (Cao et al., 2006). atrophy (Méjat et al., 2005; Tang and Goldman, 2006; Tang et al., Myog induction has previously been noted to regulate 2006, 2009; Moresi et al., 2010; Macpherson et al., 2011). The denervation-dependent muscle atrophy, at least in part by coordinate activation of these processes in denervated muscle might activating atrophy-related genes (Moresi et al., 2010; Macpherson help promote muscle survival and reinnervation after nerve damage. et al., 2011). In innervated muscle of Dach2−/−/Hdac9−/− mice, we Thus, this signaling system appears to play a key role in maintaining find increased Myog without increased muscle atrophy. This might motor function in adult animals and might represent a crucial target suggest that Myog is insufficient to drive muscle atrophy on its own for enhancing motor function when neuromuscular communication and collaborates with other factors induced in denervated muscle, or is compromised. that the levels of Myog induction noted in Dach2−/−/Hdac9−/− In conclusion, our data suggest that a Dach2-Hdac9 signaling innervated muscle are below the threshold for stimulating muscle system mediates the inhibitory effects of muscle activity on muscle atrophy. Indeed, although Myog RNA levels are elevated in reinnervation. Dach2 and Hdac9 are highly expressed in innervated Dach2−/−/Hdac9−/− innervated muscle, they are around one-tenth muscle and reduced by muscle denervation. This reduction that observed after muscle denervation. correlates with enhanced reinnervation of denervated muscle and An analysis of the denervated muscle transcriptome for RNAs we found that forced overexpression of Dach2 and Hdac9 inhibits encoding secreted factors that were induced in denervated muscle this reinnervation. Our studies suggest that the Dach2-Hdac9 and super-induced in denervated muscle from Dach2/Hdac9 null signaling system inhibits muscle reinnervation by acting on at least mice identified Gdf5, Serpina3m and Serpina3n. Serpina3m and two different pathways; one controlled by Myog and another acting Serpina3n are members of a large family of secreted serine protease via Gdf5. Our studies suggest Gdf5 might represent a secreted factor inhibitors (Forsyth et al., 2003) that appear to be expressed at the that can be used to enhance motor function in people suffering from NMJ and are regulated by muscle denervation (Akaaboune et al., neuromuscular diseases and age-related sarcopenia. 1993, 1994). Although their function remains untested, they could influence the availability of trophic factors and modify the MATERIALS AND METHODS extracellular environment of the muscle so it is more conducive to Mouse lines − − − − muscle reinnervation. Dach2 / and Hdac9 / mice were previously described (Zhang et al., 2002; − − − − We chose to focus on Gdf5, a member of the Bmp and Tgf-β Méjat et al., 2005; Davis et al., 2006). Dach2 / /Hdac9 / double knockout mice were generated by breeding Dach2−/− and Hdac9−/− mice with family of secreted proteins that has been shown to regulate flox/flox mammalian bone and/or cartilage formation and muscle atrophy each other. Myog conditional knockout mice (Myog ;CAGG-CreER) (Storm et al., 1994; Jin and Li, 2013; Sartori et al., 2013). Whereas were described previously (Knapp et al., 2006). Tamoxifen-mediated recombination was performed as previously described (Knapp et al., 2006; Bmp family members have been shown to regulate activity- Macpherson et al., 2011). Gdf5bp-4j mice were obtained from The Jackson dependent formation and plasticity of the Drosophila NMJ (Berke Laboratory (Bar Harbor, Maine). All animal care and experimental et al., 2013; Piccioli and Littleton, 2014), our study is the first to procedures were carried out in accordance with guidelines and approval of suggest a role for these proteins at the mammalian NMJ. Although the University Committee on Use and Care of Animals at the University of we show that Gdf5 is super-induced in Dach2/Hdac9 null mice and Michigan. See supplementary materials and methods for details. DEVELOPMENT

4046 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

Muscle denervation tibial nerve to the soleus muscle was crushed as described above. See Muscle denervations followed previously published protocols (Goldman supplementary materials and methods for details. et al., 1988; Macpherson et al., 2011). To prevent muscle reinnervation after nerve sectioning, the proximal nerve stump was implanted into the hip RNAseq and DNA sequencing muscle, which allows for permanent denervation of the lower leg (Carlson DNA sequencing was performed by the University of Michigan DNA and Faulkner, 1988). See supplementary materials and methods for details. Sequencing Core. RNA was isolated from innervated and 3 day denervated Wt soleus muscle and denervated Dach2−/−/Hdac9−/− Laser capture microdissection, RNA isolation and real-time PCR soleus muscle using Direct-zol RNA MiniPrep Kit (Zymo Research) The Arcturus AutoPix system was used to isolate motor neurons from adult and was treated with DNase (Invitrogen). RNA was submitted to the spinal cord sections. Sections were stained with Cresyl Violet as previously University of Michigan DNA Sequencing Core for the generation of described (Veldman et al., 2007). RNA was isolated from microdissected RNAseq libraries according to the Illumina TruSeq protocol. Two samples using a PicoPure RNA Isolation Kit according to manufacturer’s independent libraries were prepared for each sample. Sequencing was directions. For all other tissues, total RNA was isolated using TRIzol performed on an Illumina HiSeq 2000 sequencer. The University of (Invitrogen). A Turbo DNase Kit (Ambion) was used to remove residual Michigan’s Bioinformatics Core analyzed DNA sequencing data. DNA and total RNAwas quantified spectrophotometrically. One microgram CUFFDIFF2 (http://cole-trapnell-lab.github.io/cufflinks/) was used to of total RNA was used to generate cDNA with random primers and determine differentially expressed genes, with a false discovery rate of Superscript II reverse transcriptase (Invitrogen). Real-time PCR was 5%, fold-changes of at least two and abundance greater than five FPKM. performed with SYBR Green mix (1:20,000 SYBR Green) on an iCycler The secreted protein database (SPD) (http://spd.cbi.pku.edu.cn/) was used iQ real-time detection system (BioRad) according to the manufacturer’s to annotate differentially expressed genes. RNAseq data was deposited in instructions. Each reaction was performed in duplicate or triplicate on at GEO with accession number GSE58669. See supplementary materials least three samples for each condition in a volume of 25 μl. Real-time PCR and methods for details. results for muscle were normalized to γ-Actin or desmin mRNA whereas those for spinal cord and sciatic nerve were normalized to gapdh mRNA. To Statistics assay miR206, poly(A) tails were first added to total RNA (Ambion) and All experiments were performed a minimum of three times. Statistical subsequent reverse transcription was performed with Superscript II and a analysis was performed using two-tailed Student’s t-test and, where poly(T) adapter [5′-GCGAGCACAGAATTAATACGACTCACTATAGG- appropriate, ANOVA with post-hoc Bonferroni t-tests. The level of (T)12VN-3′]. Following cDNA synthesis real-time qPCR was performed as significance was set a priori at P<0.05. described above. PCR primers are listed in Table S1. Acknowledgements − − Immunofluorescence analysis of NMJs and nerves We thank Eric Olson, Graeme Mardon, and William Klein for providing Hdac9 / , − − Standard immunofluorescence protocols were used. Briefly, soleus muscles Dach2 / and Myogflox/flox;CAGG-CreER mice, respectively; The University of were fixed for 10 min in 4% paraformaldehyde and cut to 40 to 80 μm-thick Michigan DNA Sequencing Core for RNAseq library preparation and DNA longitudinal sections. After blocking, endplates were visualized by staining sequencing; Ana Rodrigues Grant and the University of Michigan Bioinformatics with Alexa Fluor 594-conjugated α-bungarotoxin (α-Btx; 1:2000; Core for bioinformatics support; Carol Davis and Susan Brooks for help with nerve ThermoFisher, B-13423) and rabbit anti-βIII-tubulin antibody (1:1000; stimulation and muscle force measurements; Robert Thompson for help with laser capture microscopy; Dennis Claflin for assistance with instrumentation for muscle BioLegend, 802001) or mouse anti-SV2 antibody (1:50; Developmental force assays; Curtis Powell for the CMV:Gdf5 construct; and Eli Cornblath for help in Studies Hybridoma Bank, SV2). Washed muscles were incubated in quantifying regenerated neuromuscular junctions. secondary antibodies (see the supplementary materials and methods). Following secondary antibody incubation, slides were washed, coverslipped Competing interests and examined with either a Zeiss Axiophot fluorescence microscope, Zeiss The authors declare no competing or financial interests. Axiophot, Axio Observer Z.1 or an Olympus FV1000 laser scanning confocal microscope. Quantification of muscle reinnervation in Wt Author contributions and knockout mice was done blindly by two different individuals using D.G. conceived the study, designed experiments and wrote the manuscript. software associated with the Axio Observer Z.1 or Olympus FV1000 P.C.D.M. designed and conducted experiments and edited the paper. P.F. microscopes. conducted experiments. Nerve branching analysis was performed on soleus muscles after tibial nerve crush. For this analysis the number of nerve branches that either Funding contacted or crossed over an individual endplate was counted. This work was supported by NIH Grant [RO1 NS059848] from National Institute of Approximately 100 endplates from each muscle were analyzed and the Neurological Disorders and Stroke (D.G.). Deposited in PMC for release after 12 months. data represent the average number of nerve branches per endplate for three muscles. See supplementary materials and methods for details. Supplementary information Supplementary information available online at Muscle functional measurements http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.125674/-/DC1 Soleus muscles were surgically freed from surrounding muscle and connective tissue, the distal tendon was secured to the lever arm of a References Akaaboune, M., Ma, J., Festoff, B. W., Greenberg, B. D. and Hantaï, D. (1993). servomotor (Cambridge Instruments) with 5-0 suture and the hindlimb was The influence of denervation on beta-amyloid protein precursor and alpha 1- securely tied to a fixed post with 4-0 suture at the knee. Soleus muscle antichymotrypsin in mouse skeletal muscle. Neuromuscul. Disord. 3, 477-481. maximum isometric twitch and tetanic force measurements were determined Akaaboune, M., Ma, J., Festoff, B. W., Greenberg, B. D. and Hantaï, D. (1994). by indirect nerve stimulation. See supplementary materials and methods for Neurotrophic regulation of mouse muscle beta-amyloid protein precursor and details. alpha 1-antichymotrypsin as revealed by axotomy. J. Neurobiol. 25, 503-514. Berghella, L., De Angelis, L., De Buysscher, T., Mortazavi, A., Biressi, S., Forcales, S. V., Sirabella, D., Cossu, G. and Wold, B. J. (2008). A highly Plasmids, siRNA and muscle electroporation conserved molecular switch binds MSY-3 to regulate myogenin repression in Plasmids and siRNAs were injected individually or together into innervated postnatal muscle. Genes Dev. 22, 2125-2138. soleus muscles. The molar ratio of stealth siRNA to enhanced green Berke, B., Wittnam, J., McNeill, E., Van Vactor, D. L. and Keshishian, H. (2013). + Retrograde BMP signaling at the synapse: a permissive signal for synapse fluorescent protein (EGFP) plasmid was >500-fold so as to ensure all GFP maturation and activity-dependent plasticity. J. Neurosci. 33, 17937-17950. fibers harbor siRNA. Nucleic acid uptake was facilitated by electroporation Buonanno, A., Apone, L., Morasso, M. I., Beers, R., Brenner, H. R. and Eftimie, as previously described (Tang et al., 2009). Following electroporation, the R. (1992). The MyoD family of myogenic factors is regulated by electrical activity: DEVELOPMENT

4047 STEM CELLS AND REGENERATION Development (2015) 142, 4038-4048 doi:10.1242/dev.125674

isolation and characterization of a mouse Myf-5 cDNA. Nucleic Acids Res. 20, Méjat, A., Ramond, F., Bassel-Duby, R., Khochbin, S., Olson, E. N. and 539-544. Schaeffer, L. (2005). Histone deacetylase 9 couples neuronal activity to muscle Cangiano, A., Lømo, T., Lutzemberger, L. and Sveen, O. (1980). Effects of chromatin acetylation and gene expression. Nat. Neurosci. 8, 313-321. chronic nerve conduction block on formation of neuromuscular junctions and Moresi, V., Williams, A. H., Meadows, E., Flynn, J. M., Potthoff, M. J., McAnally, junctional AChE in the rat. Acta Physiol. Scand. 109, 283-296. J., Shelton, J. M., Backs, J., Klein, W. H., Richardson, J. A. et al. (2010). Cao, Y., Kumar, R. M., Penn, B. H., Berkes, C. A., Kooperberg, C., Boyer, L. A., Myogenin and class II HDACs control neurogenic muscle atrophy by inducing E3 Young, R. A. and Tapscott, S. J. (2006). Global and gene-specific analyses ubiquitin ligases. Cell 143, 35-45. show distinct roles for Myod and Myog at a common set of promoters. EMBO J. 25, Nishitoh, H., Ichijo, H., Kimura, M., Matsumoto, T., Makishima, F., Yamaguchi, 502-511. A., Yamashita, H., Enomoto, S. and Miyazono, K. (1996). Identification of type I Carlson, B. M. and Faulkner, J. A. (1988). Reinnervation of long-term denervated and type II serine/threonine kinase receptors for growth/differentiation factor-5. rat muscle freely grafted into an innervated limb. Exp. Neurol. 102, 50-56. J. Biol. Chem. 271, 21345-21352. Caroni, P. (1993). Activity-sensitive signaling by muscle-derived insulin-like growth Payne, S. H., Jr., and Brushart, T. M. (1997). Neurotization of the rat soleus muscle: factors in the developing and regenerating neuromuscular system. Ann. N. Y. a quantitative analysis of reinnervation. J. Hand Surg. 22, 640-643. Acad. Sci. 692, 209-222. Piccioli, Z. D. and Littleton, J. T. (2014). Retrograde BMP signaling modulates Davis, R. J., Pesah, Y. I., Harding, M., Paylor, R. and Mardon, G. (2006). rapid activity-dependent synaptic growth via presynaptic LIM kinase regulation of MouseDach2 mutants do not exhibit gross defects in eye development or brain cofilin. J. Neurosci. 34, 4371-4381. function. Genesis 44, 84-92. Rao, P. K., Kumar, R. M., Farkhondeh, M., Baskerville, S. and Lodish, H. F. Everman, D. B., Bartels, C. F., Yang, Y., Yanamandra, N., Goodman, F. R., (2006). Myogenic factors that regulate expression of muscle-specific microRNAs. Mendoza-Londono, J. R., Savarirayan, R., White, S. M., Graham, J. M., Jr, Proc. Natl. Acad. Sci. USA 103, 8721-8726. Gale, R. P. et al. (2002). The mutational spectrum of brachydactyly type C. Rassendren, F. A., Bloch-Gallego, E., Tanaka, H. and Henderson, C. E. (1992). Am. J. Med. Genet. 112, 291-296. Levels of mRNA coding for motoneuron growth-promoting factors are increased in Forsyth, S., Horvath, A. and Coughlin, P. (2003). A review and comparison of the denervated muscle. Proc. Natl. Acad. Sci. USA 89, 7194-7198. murine alpha1-antitrypsin and alpha1-antichymotrypsin multigene clusters with Sadasivam, G., Willmann, R., Lin, S., Erb-Vögtli, S., Kong, X. C., Rüegg, M. A. the human clade A serpins. Genomics 81, 336-345. and Fuhrer, C. (2005). Src-family kinases stabilize the neuromuscular synapse in Funakoshi, H., Belluardo, N., Arenas, E., Yamamoto, Y., Casabona, A., vivo via protein interactions, phosphorylation, and cytoskeletal linkage of Persson, H. and Ibanez, C. F. (1995). Muscle-derived neurotrophin-4 as an acetylcholine receptors. J. Neurosci. 25, 10479-10493. activity-dependent trophic signal for adult motor neurons. Science 268, Sartori, R., Schirwis, E., Blaauw, B., Bortolanza, S., Zhao, J., Enzo, E., 1495-1499. Stantzou, A., Mouisel, E., Toniolo, L., Ferry, A. et al. (2013). BMP signaling Goldman, D. and Staple, J. (1989). Spatial and temporal expression of controls muscle mass. Nat. Genet. 45, 1309-1318. acetylcholine receptor RNAs in innervated and denervated rat soleus muscle. Smith, R. G., McManaman, J. and Appel, S. H. (1985). Trophic effects of skeletal Neuron 3, 219-228. muscle extracts on ventral spinal cord neurons in vitro: separation of a protein with Goldman, D., Brenner, H. R. and Heinemann, S. (1988). Acetylcholine receptor morphologic activity from proteins with cholinergic activity. J. Cell Biol. 101, alpha-, beta-, gamma-, and delta-subunit mRNA levels are regulated by muscle 1608-1621. activity. Neuron 1, 329-333. Storm, E. E., Huynh, T. V., Copeland, N. G., Jenkins, N. A., Kingsley, D. M. and Gould, T. W., Yonemura, S., Oppenheim, R. W., Ohmori, S. and Enomoto, H. Lee, S.-J. (1994). Limb alterations in brachypodism mice due to mutations in a (2008). The neurotrophic effects of glial cell line-derived neurotrophic factor on new member of the TGF beta-superfamily. Nature 368, 639-643. spinal motoneurons are restricted to fusimotor subtypes. J. Neurosci. 28, Tang, H. and Goldman, D. (2006). Activity-dependent gene regulation in skeletal 2131-2146. Griesbeck, O., Parsadanian, A. S., Sendtner, M. and Thoenen, H. (1995). muscle is mediated by a histone deacetylase (HDAC)-Dach2-myogenin signal Expression of neurotrophins in skeletal muscle: quantitative comparison and transduction cascade. Proc. Natl. Acad. Sci. USA 103, 16977-16982. significance for motoneuron survival and maintenance of function. J. Neurosci. Tang, H., Veldman, M. B. and Goldman, D. (2006). Characterization of a muscle- 42, 21-33. specific enhancer in human MuSK promoter reveals the essential role of Hennig, R. (1987). Late reinnervation of the rat soleus muscle is differentially myogenin in controlling activity-dependent gene regulation. J. Biol. Chem. 281, suppressed by chronic stimulation and by ectopic innervation. Acta Physiol. 3943-3953. Scand. 130, 153-160. Tang, H., Macpherson, P., Marvin, M., Meadows, E., Klein, W. H., Yang, X.-J. and Jansen, J. K. S., Lomo, T., Nicolaysen, K. and Westgaard, R. H. (1973). Goldman, D. (2009). A histone deacetylase 4/myogenin positive feedback loop Hyperinnervation of skeletal muscle fibers: dependence on muscle activity. coordinates denervation-dependent gene induction and suppression. Mol. Biol. Science 181, 559-561. Cell 20, 1120-1131. Jin, L. and Li, X. (2013). Growth differentiation factor 5 regulation in bone Veldman, M. B., Bemben, M. A., Thompson, R. C. and Goldman, D. (2007). Gene regeneration. Curr. Pharm. Des. 19, 3364-3373. expression analysis of zebrafish retinal ganglion cells during optic nerve Knapp, J. R., Davie, J. K., Myer, A., Meadows, E., Olson, E. N. and Klein, W. H. regeneration identifies KLF6a and KLF7a as important regulators of axon (2006). Loss of myogenin in postnatal life leads to normal skeletal muscle but regeneration. Dev. Biol. 312, 596-612. reduced body size. Development 133, 601-610. Wang, T., Xie, K. and Lu, B. (1995). Neurotrophins promote maturation of Koliatsos, V. E., Clatterbuck, R. E., Winslow, J. W., Cayouette, M. H. and Prices, developing neuromuscular synapses. J. Neurosci. 15, 4796-4805. D. L. (1993). Evidence that brain-derived neurotrophic factor is a trophic factor for Wigston, D. J. and English, A. W. (1992). Fiber-type proportions in mammalian motor neurons in vivo. Neuron 10, 359-367. soleus muscle during postnatal development. J. Neurobiol. 23, 61-70. Kong, X. C., Barzaghi, P. and Ruegg, M. A. (2004). Inhibition of synapse assembly Williams, A. H., Valdez, G., Moresi, V., Qi, X., McAnally, J., Elliott, J. L., Bassel- in mammalian muscle in vivo by RNA interference. EMBO Rep. 5, 183-188. Duby, R., Sanes, J. R. and Olson, E. N. (2009). MicroRNA-206 delays ALS Macpherson, P. C. D., Cieslak, D. and Goldman, D. (2006). Myogenin-dependent progression and promotes regeneration of neuromuscular synapses in mice. nAChR clustering in aneural myotubes. Mol. Cell. Neurosci. 31, 649-660. Science 326, 1549-1554. Macpherson, P. C. D., Wang, X. and Goldman, D. (2011). Myogenin regulates Zhang, C. L., McKinsey, T. A., Chang, S., Antos, C. L., Hill, J. A. and Olson, E. N. denervation-dependent muscle atrophy in mouse soleus muscle. J. Cell. (2002). Class II histone deacetylases act as signal-responsive repressors of Biochem. 112, 2149-2159. cardiac hypertrophy. Cell 110, 479-488. DEVELOPMENT

4048