GLP inhibits heterochromatin clustering and myogenic differentiation by repressing MeCP2

GLP inhibits heterochromatin clustering and myogenic differentiation by repressing MeCP2 Abstract Myogenic differentiation is accompanied by alterations in the chromatin states, which permit or restrict the transcriptional machinery and thus impact distinctive gene expression profiles. The mechanisms by which higher-order chromatin remodeling is associated with gene activation and silencing during differentiation is not fully understood. In this study, we provide evidence that the euchromatic lysine methyltransferase GLP regulates heterochromatin organization and myogenic differentiation. Interestingly, GLP represses expression of the methyl-binding protein MeCP2 that induces heterochromatin clustering during differentiation. Consequently, MeCP2 and HP1γ localization at major satellites are altered upon modulation of GLP expression. In GLP knockdown cells, depletion of MeCP2 restored both chromatin organization and myogenic differentiation. These results identify a novel regulatory axis between a histone methylation writer and DNA methylation reader, which is important for heterochromatin organization during differentiation. myogenic differentiation, chromatin, methylation Introduction Myogenic differentiation is accompanied by spatial reorganization of chromatin. In addition to epigenetic modifications that occur locally at promoters, heterochromatin is modified and gets progressively clustered in distinctive chromocenters. Concomitantly, major and minor satellite RNA transcripts accumulate during differentiation (Brero et al., 2005; Terranova et al., 2005; Agarwal et al., 2007; Luo et al., 2009; Sdek et al., 2011; Chen et al., 2017) DNA methylation at CpG dinucleotides and histone 3 lysine 9 methylation (H3K9me) are hallmarks of heterochromatin. Consequently, proteins that recognize methylation marks, such as methyl-CpG-binding domain (MBD) protein MeCP2, as well as heterochromatin protein 1 (HP1), are enriched in pericentric heterochromatin. Moreover, MeCP2 and HP1 result in heterochromatin clustering, which is thought to be essential for maintenance of the differentiated state in skeletal and cardiac myocytes (Brero et al., 2005; Agarwal et al., 2007; Sdek et al., 2011). The interplay between DNA and histone modifications in heterochromatin organization is complex and not fully understood. For instance, the absence of DNA methylation in DNMT1 or DNMT3a/3b deficient cells results in reduced MeCP2 localization but does not alter Suv39h-mediated H3K9 tri-methylation (H3K9me3) marks or HP1α localization at pericentric heterochromatin (Lehnertz et al., 2003). Furthermore, while MeCP2 interacts with Suv39h1/h2 (Fuks et al., 2003b), exogenous MeCP2 still induces pericentric heterochromatin clustering in Suv39h double-null cells, indicating that the MeCP2–Suv39h interaction is not essential (Brero et al., 2005). On the other hand, H3K9me is required for DNA methylation (Lehnertz et al., 2003). DNMT3a and 3b are directed to H3K9me marks via association with HP1α and HP1β (Fuks et al., 2003a). In Suv39h null cells, DNA methylation and MeCP2 localization is reduced at pericentric satellite repeats (Lehnertz et al., 2003). G9a and G9a-like protein (GLP) are the primary mediators of H3K9me2 in euchromatin (Tachibana et al., 2005; Trojer and Reinberg, 2007) and their auto-methylation acts as a docking site for HP1α and HP1γ (Chin et al., 2007; Fritsch et al., 2010). Depletion of GLP or G9a restricts localization of HP1γ to heterochromatin in mouse embryonic stem cells (Tachibana et al., 2005). In addition, G9a/GLP also associate with DNMT3a via MPP8 (Chang et al., 2011). Gene disruption studies in mice have shown that loss of G9a or GLP abrogates global H3K9me2 and results in embryonic lethality at E9.5 (Tachibana et al., 2005). Notably, the Ankyrin (ANK) repeats of GLP are required for efficient establishment and maintenance of H3K9me2 during embryonic development while the ANK repeats of G9a are dispensable (Liu et al., 2015), indicating independent functions of GLP and G9a during embryonic development. A few recent studies have suggested that GLP may be involved in skeletal myogenesis. For instance, GLP stabilizes PRDM16 and steers Myf5+ precursor cells towards a brown adipocyte lineage, in turn suppressing myogenic promoters (Ohno et al., 2013). In addition, conditional deletion of GLP in neurons results in the upregulation of myogenic genes (Schaefer et al., 2009). In cultured myoblasts, GLP overexpression was found to block myogenic differentiation (Battisti et al., 2016). However, the molecular mechanisms by which GLP inhibits muscle differentiation and its potential role in heterochromatin organization is not well understood. In this study, we have examined the role of GLP in skeletal myogenesis. We show that GLP overexpression inhibits terminal differentiation of muscle precursor cells. The defect in myogenic differentiation correlates with impaired heterochromatin clustering. HP1γ localization at heterochromatin and its occupancy at major satellites is reduced in GLP-overexpressing cells, and conversely is increased in GLP knockdown primary myoblasts. These changes occur in the absence of any changes in H3K9me3 levels or in DNA methylation. Interestingly however, MeCP2, which recruits HP1γ to heterochromatin, is repressed in GLP-overexpressing cells resulting in reduced occupancy at major satellites. We demonstrate that MeCP2 is a direct GLP target gene. Consequently, depletion of endogenous MeCP2 expression in the background of GLP knockdown reverts the enhanced myogenic differentiation and HP1γ localization to chromocenters seen in GLP knockdown cells. Taken together, our results provide evidence that GLP influences chromatin states and myogenic differentiation through repression of MeCP2. Results GLP inhibits skeletal muscle differentiation While recent studies have implicated GLP in skeletal muscle differentiation (Schaefer et al., 2009; Ohno et al., 2013; Battisti et al., 2016), the underlying mechanisms are not completely understood. To this end, we first examined endogenous GLP mRNA and protein expression in C2C12 cells. Similar to G9a (Ling et al., 2012; Jung et al., 2015; Ow et al., 2016), GLP expression declined during differentiation and inversely correlated with differentiation markers Myogenin and TroponinT (Figure 1A and B). To test its potential role in myogensis, we performed loss-of-function studies and examined proliferation and differentiation of GLP knockdown cells. C2C12 cells were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 48 h (Figure 1C). Cells were pulsed with BrdU and stained with anti-BrdU antibody. No significant changes in proliferation of siGLP cells were observed compared to control cells (Figure 1D). Interestingly however, siGLP cells exhibited a significant increase in myogenic differentiation as seen by immunofluorescence staining using anti-myosin heavy chain (MHC) antibody and myogenic index (Figure 1E). Consistently, Myogenin, TroponinT, and MyoD expression levels were elevated in siGLP cells upon differentiation, while G9a levels were not overtly altered (Figure 1F). To validate the impact of GLP on differentiation, siRNA-mediated depletion of GLP in mouse primary myoblasts was performed (Figure 1G). Similar to the effect in C2C12 cells, siGLP primary myoblasts expressed increased levels of Myogenin and myosin heavy chain 1 (Myh1) during differentiation (Figure 1G). Taken together, these results indicate that endogenous GLP inhibits myogenesis. Figure 1 View largeDownload slide Endogenous GLP inhibits skeletal muscle differentiation. (A) GLP mRNA levels were measured in undifferentiated (Day 0) and differentiating (Day 1 and Day 3) C2C12 cells by Q-PCR. (B) Expression of GLP, Myogenin, TroponinT, and MyoD in C2C12 cells was analyzed by western blot. β-actin was used as loading control. The molecular weights of proteins are indicated by numbers on the left. (C) C2C12 myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 48 h. GLP expression was analyzed by western blot. β-actin was used as loading control. (D) Control and siGLP C2C12 myoblasts were pulsed with BrdU and stained with anti-BrdU antibody (green). Nuclei were stained with DAPI (blue). For each sample, >1000 nuclei in 5 fields were counted and the percentage of BrdU+ cells was calculated. Scale bar, 60 μm. The results are representative of two independent experiments. Bars indicate the mean of BrdU+ cells in each experiment ±SD. n.s., P-value not significant. (E) Control and siGLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control and siGLP cells were scored. Scale bar, 15 μm. The results are representative of three independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (F) Expression of G9a, MyoD, Myogenin, and TroponinT was analyzed by western blot in Control and siGLP C2C12 cells. β-actin was used as a loading control. (G) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 3 days. control and siGLP cells were analyzed for expression of GLP, Myogenin, and Myh1 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Error bars indicate the mean of Q-PCR triplicates in each experiment ±SD. Figure 1 View largeDownload slide Endogenous GLP inhibits skeletal muscle differentiation. (A) GLP mRNA levels were measured in undifferentiated (Day 0) and differentiating (Day 1 and Day 3) C2C12 cells by Q-PCR. (B) Expression of GLP, Myogenin, TroponinT, and MyoD in C2C12 cells was analyzed by western blot. β-actin was used as loading control. The molecular weights of proteins are indicated by numbers on the left. (C) C2C12 myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 48 h. GLP expression was analyzed by western blot. β-actin was used as loading control. (D) Control and siGLP C2C12 myoblasts were pulsed with BrdU and stained with anti-BrdU antibody (green). Nuclei were stained with DAPI (blue). For each sample, >1000 nuclei in 5 fields were counted and the percentage of BrdU+ cells was calculated. Scale bar, 60 μm. The results are representative of two independent experiments. Bars indicate the mean of BrdU+ cells in each experiment ±SD. n.s., P-value not significant. (E) Control and siGLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control and siGLP cells were scored. Scale bar, 15 μm. The results are representative of three independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (F) Expression of G9a, MyoD, Myogenin, and TroponinT was analyzed by western blot in Control and siGLP C2C12 cells. β-actin was used as a loading control. (G) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 3 days. control and siGLP cells were analyzed for expression of GLP, Myogenin, and Myh1 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Error bars indicate the mean of Q-PCR triplicates in each experiment ±SD. The SET domain and ANK repeats of GLP are required to inhibit myogenic differentiation We next examined whether overexpression of GLP is sufficient to inhibit myogenic differentiation. Cells were transfected with an empty vector (pCS2), full-length GLP (FL-GLP) or a deletion mutant lacking both ANK repeats and SET domain (ΔANKΔSET) (Figure 2A). Both proteins were expressed at equivalent levels (Figure 2B). Overexpression of FL-GLP significantly inhibited myogenic differentiation compared to control cells. However, no significant change in differentiation of ΔANKΔSET cells was seen (Figure 2C). Correspondingly, Myogenin and TroponinT levels were lower only in FL-GLP expressing cells (Figure 2D). To confirm these findings, GLP was retrovirally overexpressed in C2C12 cells (Figure 2E). Compared to control cells, pMaRX-GLP cells showed reduced myogenic differentiation as observed by MHC staining, myogenic index (Figure 2F), and downregulation of myogenic markers (Figure 2G). No significant change of MyoD levels was seen in undifferentiated pMaRX-GLP cells, while a modest reduction was apparent during differentiation (Figure 2G). Altogether, these results confirm that GLP overexpression inhibits myogenesis. Figure 2 View largeDownload slide GLP overexpression inhibits myogenic differentiation through the SET domain and ANK repeats. (A) Schematic diagram depicting full-length GLP (FL-GLP) construct with conserved Glu-rich region (E), Cys-rich region (C), ANK repeats (A), and SET domain (S). The mutant GLP (ΔANKΔSET) construct lacks the ANK repeats and SET domain. (B) C2C12 cells expressing pCS2, FL-GLP, and ΔANKΔSET were analyzed for expression of GLP by western blot using anti-Flag (left) and anti-GLP (right) antibodies. β-actin was used as loading control. (C) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (green). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pCS2, FL-GLP, and ΔANKΔSET cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and analyzed for Myogenin and TroponinT levels. β-actin was used as loading control. (E) C2C12 cells were transfected with an empty vector control (pMaRX) or pMaRX-GLP. GLP and G9a expression levels were analyzed by western blot. (F) pMaRX and pMaRX-GLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pMaRX and pMaRX-GLP cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (G) pMaRX and pMaRX-GLP cells were differentiated for 2 days and analyzed for Myogenin, TroponinT, and MyoD levels. β-actin was used as loading control. Figure 2 View largeDownload slide GLP overexpression inhibits myogenic differentiation through the SET domain and ANK repeats. (A) Schematic diagram depicting full-length GLP (FL-GLP) construct with conserved Glu-rich region (E), Cys-rich region (C), ANK repeats (A), and SET domain (S). The mutant GLP (ΔANKΔSET) construct lacks the ANK repeats and SET domain. (B) C2C12 cells expressing pCS2, FL-GLP, and ΔANKΔSET were analyzed for expression of GLP by western blot using anti-Flag (left) and anti-GLP (right) antibodies. β-actin was used as loading control. (C) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (green). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pCS2, FL-GLP, and ΔANKΔSET cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and analyzed for Myogenin and TroponinT levels. β-actin was used as loading control. (E) C2C12 cells were transfected with an empty vector control (pMaRX) or pMaRX-GLP. GLP and G9a expression levels were analyzed by western blot. (F) pMaRX and pMaRX-GLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pMaRX and pMaRX-GLP cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (G) pMaRX and pMaRX-GLP cells were differentiated for 2 days and analyzed for Myogenin, TroponinT, and MyoD levels. β-actin was used as loading control. Chromatin organization is altered in GLP-overexpressing cells Previous studies have shown that GLP interacts with HP1α and HP1γ, and its depletion restricts HP1γ to heterochromatin, suggesting that it may have a role in heterochromatin organization (Tachibana et al., 2005; Chin et al., 2007). In myoblasts, HP1γ localizes in both euchromatin and heterochromatin but localizes exclusively in heterochromatin upon myogenic differentiation (Agarwal et al., 2007; Sdek et al., 2013). This association of HP1γ to heterochromatin in myotubes is attributed to MeCP2, which recruits HP1γ to the chromocenters upon the induction of heterochromatin clustering (Agarwal et al., 2007). We therefore examined whether HP1γ levels or sub-nuclear localization was altered upon GLP overexpression during differentiation. To analyze heterochromatin clustering, cells were differentiated for 6 days when distinctive chromocenter formation is observed (Terranova et al., 2005). HP1γ levels were unchanged in pMaRX-GLP cells, nor was there any change in H3K9me3, a heterochromatin mark (Figure 3A). Focal nuclear staining of heterochromatin can be visualized with DAPI staining of A/T rich satellite repeats. Heterochromatin clustered into distinct chromocenters in both pMaRX and pMaRX-GLP cells (Figure 3B). However, pMaRX-GLP cells showed a higher number, but smaller size of DAPI stained chromocenters, and reduced localization of HP1γ to chromocenters compared to control cells (Figure 3C), suggesting that heterochromatin clustering may be impaired. H3K9me3 marks are established docking sites for HP1γ, but H3K9me3 sub-nuclear localization was not altered upon GLP overexpression (Figure 3B and C). To validate these findings, control or siGLP primary myoblasts were differentiated and chromocenters were analyzed. Compared to controls, siGLP cells showed a lower number, larger size of DAPI stained chromocenters, and enhanced HP1γ localization to heterochromatin (Figure 3D and E). Figure 3 View largeDownload slide Chromatin organization during differentiation is altered in GLP-overexpressing cells. (A) pMaRX and pMaRX-GLP cells were differentiated for 0 and 6 days and were analyzed for HP1γ and H3K9me3 levels. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-HP1γ (red) and anti-H3K9me3 (green) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, Mander’s coefficient for HP1γ and DAPI, and Mander’s coefficient for H3K9me3 and DAPI in pMaRX and pMaRX-GLP myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. n.s., P-value not significant. (D) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-HP1γ antibody (red). Scale bar, 2 μm. (E) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siGLP primary myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Figure 3 View largeDownload slide Chromatin organization during differentiation is altered in GLP-overexpressing cells. (A) pMaRX and pMaRX-GLP cells were differentiated for 0 and 6 days and were analyzed for HP1γ and H3K9me3 levels. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-HP1γ (red) and anti-H3K9me3 (green) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, Mander’s coefficient for HP1γ and DAPI, and Mander’s coefficient for H3K9me3 and DAPI in pMaRX and pMaRX-GLP myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. n.s., P-value not significant. (D) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-HP1γ antibody (red). Scale bar, 2 μm. (E) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siGLP primary myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. MeCP2 and HP1γ are redistributed in GLP-overexpressing cells Since MeCP2 interacts with HP1γ to regulate heterochromatin clustering during myogenic differentiation (Brero et al., 2005; Agarwal et al., 2007), we examined its expression. Interestingly, MeCP2 expression was decreased in pMaRX-GLP cells compared to control cells (Figure 4A). Conversely, its expression was upregulated in siGLP primary myoblasts (Figure 4B) suggesting that MeCP2 de-regulation may contribute to modulation of HP1γ localization. Figure 4 View largeDownload slide GLP alters sub-nuclear distribution of MeCP2 and HP1γ without affecting DNA methylation at major satellites. (A) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). Scale bar, 2 μm. Box and whisker plot shows total corrected cell fluorescence (TCCF) of MeCP2. For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. (B) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). TCCF of MeCP2 in control and siGLP primary myotubes is shown. (C) pMaRX and pMARX-GLP cells were differentiated for 6 days and ChIP assays for occupancy of HP1γ, MeCP2, and H3K9me3 at major satellites were performed. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. n.s., P-value not significant. (D) pMaRX and pMARX-GLP cells were differentiated for 6 days and methylation status of major satellite repeat was analyzed. PCR product of bisulfite-treated DNA was digested with (+) or without (−) HpyCH4IV. Methylation percentage was scored. The results are representative of four independent experiments. Figure 4 View largeDownload slide GLP alters sub-nuclear distribution of MeCP2 and HP1γ without affecting DNA methylation at major satellites. (A) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). Scale bar, 2 μm. Box and whisker plot shows total corrected cell fluorescence (TCCF) of MeCP2. For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. (B) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). TCCF of MeCP2 in control and siGLP primary myotubes is shown. (C) pMaRX and pMARX-GLP cells were differentiated for 6 days and ChIP assays for occupancy of HP1γ, MeCP2, and H3K9me3 at major satellites were performed. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. n.s., P-value not significant. (D) pMaRX and pMARX-GLP cells were differentiated for 6 days and methylation status of major satellite repeat was analyzed. PCR product of bisulfite-treated DNA was digested with (+) or without (−) HpyCH4IV. Methylation percentage was scored. The results are representative of four independent experiments. To further ascertain these results, we tested for the presence of HP1γ, MeCP2, and H3K9me3 at heterochromatin by analyzing their occupancy at major satellites in pMaRX-GLP cells. Compared to control cells, pMaRX-GLP cells showed reduced HP1γ and MeCP2 occupancy at major satellites, although H3K9me3 enrichment remained unchanged (Figure 4C). Given that MeCP2 binds methylated CG dinucleotides, we tested whether change in DNA methylation at major satellites contributed to reduction of MeCP2 binding at heterochromatin. Combined bisulfite restriction analysis (COBRA) revealed that there was no significant change in DNA methylation at major satellites in pMaRX-GLP cells compared to control cells (Figure 4D). MeCP2 is a GLP target gene The lower MeCP2 levels in pMaRX-GLP cells suggested that it may be a downstream target of GLP. To test that possibility, we first examined endogenous MeCP2 expression during differentiation. Consistent with previous studies (Brero et al., 2005; Agarwal et al., 2007), MeCP2 expression dramatically increased upon terminal differentiation and inversely correlated with GLP levels (Figure 5A). In pMaRX-GLP cells, MeCP2 transcripts were downregulated suggesting that GLP may transcriptionally repress MeCP2 expression (Figure 5B). We also analyzed the expression of another MBD family member MBD2, which has also been shown to induce heterochromatin clustering in skeletal muscle (Brero et al., 2005). Similar to MeCP2, MBD2 mRNA levels were downregulated in pMaRX-GLP cells (Figure 5B). Correspondingly, MeCP2 and MBD2 were upregulated in siGLP primary myoblasts compared to controls (Figure 5C). The MeCP2 promoter has been reported to be repressed by H3K9me2 marks (Heard et al., 2001; Abuhatzira et al., 2011). We therefore examined whether MeCP2 is indeed a GLP target gene by assessing GLP occupancy and H3K9me2 marks at the MeCP2 promoter by ChIP assays. In control cells, GLP occupancy and H3K9me2 marks were higher in undifferentiated cells (Day 0) and reduced during differentiation (Day 3) (Figure 5D). A clear enrichment of GLP and H3K9me2 marks were evident at the MeCP2 promoter in pMaRX-GLP cells relative to control in both undifferentiated and differentiated cells (Figure 5D). No change in GLP occupancy was seen at the β-actin promoter in pMaRX-GLP cells, which was analyzed as a control (Figure 5D). Figure 5 View largeDownload slide GLP represses MeCP2 expression. (A) Expression of MeCP2, GLP, and TroponinT in C2C12 cells was analyzed by western blot after 0, 1, and 3 days of differentiation. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 0 and 3 days and analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (C) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 0 and 3 days. siRNA and siGLP cells were analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (D) ChIP assays for GLP occupancy (left) and H3K9me2 enrichment (middle) at the MeCP2 promoter were performed in undifferentiated (Day 0) and differentiated (Day 3) pMaRX and pMaRX-GLP cells. GLP occupancy at the β-actin promoter (right) was tested as negative control. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. Figure 5 View largeDownload slide GLP represses MeCP2 expression. (A) Expression of MeCP2, GLP, and TroponinT in C2C12 cells was analyzed by western blot after 0, 1, and 3 days of differentiation. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 0 and 3 days and analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (C) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 0 and 3 days. siRNA and siGLP cells were analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (D) ChIP assays for GLP occupancy (left) and H3K9me2 enrichment (middle) at the MeCP2 promoter were performed in undifferentiated (Day 0) and differentiated (Day 3) pMaRX and pMaRX-GLP cells. GLP occupancy at the β-actin promoter (right) was tested as negative control. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. Since MeCP2 is important for binding and aggregation of pericentric heterochromatin (Brero et al., 2005), we examined the impact of its depletion during myogenic differentiation. Cells were transfected with scrambled siRNA (Control) or MeCP2-specific siRNA (siMeCP2) (Figure 6A). Control and siMeCP2 cells were differentiated and immuno-stained with MeCP2 and HP1γ antibodies. Compared to control cells, siMeCP2 cells had a higher number and smaller size of chromocenters. In addition, similar to GLP-overexpressing cells, HP1γ localization to chromocenters was reduced indicating impaired heterochromatin clustering (Figure 6B and C). Given that MeCP2 is a GLP target gene, we investigated whether MeCP2 is downstream of GLP-mediated heterochromatin clustering by expressing exogenous MeCP2 in GLP-overexpressing cells (pMaRX-GLP + MeCP2) (Figure 6D). Control, pMaRX-GLP, and pMaRX-GLP + MeCP2 cells were differentiated and heterochromatin clustering was analyzed using anti-MeCP2 and anti-HP1γ antibodies. pMaRX-GLP cells showed a greater number and smaller size of chromocenters, and reduced HP1γ localization compared to controls (Figure 6E and F). Expression of exogenous MeCP2 in pMaRX-GLP + MeCP2 cells resulted in similar number and size of chromocenters and HP1γ localization as control cells (Figure 6E and F). Figure 6 View largeDownload slide MeCP2 regulates chromatin organization. (A) C2C12 myoblasts were transfected with scrambled siRNA (Control) or MeCP2-specific siRNA (siMeCP2) for 48 h. MeCP2 expression was analyzed by western blot. β-actin was used as loading control. (B) Control and siMeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siMeCP2 myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. (D) pMaRX-GLP cells were transfected with MeCP2 plasmid (pMaRX-GLP + MeCP2). pMaRX and pMaRX-GLP cells were transfected with pCS2 empty vector. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. (E) pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (F) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. Figure 6 View largeDownload slide MeCP2 regulates chromatin organization. (A) C2C12 myoblasts were transfected with scrambled siRNA (Control) or MeCP2-specific siRNA (siMeCP2) for 48 h. MeCP2 expression was analyzed by western blot. β-actin was used as loading control. (B) Control and siMeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siMeCP2 myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. (D) pMaRX-GLP cells were transfected with MeCP2 plasmid (pMaRX-GLP + MeCP2). pMaRX and pMaRX-GLP cells were transfected with pCS2 empty vector. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. (E) pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (F) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. Endogenous MeCP2 regulates heterochromatin organization and differentiation Given that exogenous MeCP2 restored heterochromatin clustering in GLP-overexpressing cells, we assessed whether inhibition of heterochromatin clustering by GLP is mediated via repression of endogenous MeCP2. Cells were transfected with scrambled siRNA (Control), siGLP, or both siGLP and siMeCP2 [double knockdown (DKD)]. Consistent with our previous data (Figure 3D and E), MeCP2 expression was elevated in siGLP cells compared to control cells (Figure 7A). We analyzed expression and localization of MeCP2 and HP1γ in differentiated control, siGLP and DKD cells with anti-MeCP2 and anti-HP1γ antibodies. Compared to control cells, siGLP cells showed lower number and larger size of chromocenters as seen by DAPI staining. HP1γ localization to heterochromatin was also increased (Figure 7B and C). On the other hand, while DKD cells exhibited no significant change in number of chromocenters compared to control cells (Figure 7C), there was a decrease in their size (Figure 7B) and HP1γ localization to heterochromatin was reduced (Figure 7C). Control, siGLP and DKD cells were also analyzed for myogenic differentiation. Interestingly, the enhanced myogenic differentiation in siGLP cells seen by MHC staining, myogenic index (Figure 7D), as well as Myogenin and TroponinT levels (Figure 7E), was reversed in DKD cells to a level similar to control cells. Figure 7 View largeDownload slide MeCP2 depletion restores chromatin organization in GLP knockdown cells. (A) C2C12 cells were transfected with siRNA (Control), siGLP, or both siGLP and siMeCP2 (DKD) for 48 h. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. Densitometric analysis of MeCP2 level is shown in the right panel. (B) Control, siGLP, and DKD cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control, siGLP, and DKD myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) Control, siGLP, and DKD cells were differentiated for 3 days and immuno-stained with anti-MHC antibody (red). Myogenic indices for Control, siGLP, and DKD myotubes were scored. Scale bar, 15 μm. For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control, siGLP, and DKD cells were scored. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (E) Control, siGLP, and DKD cells were differentiated for 3 days and expression of Myogenin and TroponinT was analyzed. β-actin was used as loading control. Figure 7 View largeDownload slide MeCP2 depletion restores chromatin organization in GLP knockdown cells. (A) C2C12 cells were transfected with siRNA (Control), siGLP, or both siGLP and siMeCP2 (DKD) for 48 h. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. Densitometric analysis of MeCP2 level is shown in the right panel. (B) Control, siGLP, and DKD cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control, siGLP, and DKD myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) Control, siGLP, and DKD cells were differentiated for 3 days and immuno-stained with anti-MHC antibody (red). Myogenic indices for Control, siGLP, and DKD myotubes were scored. Scale bar, 15 μm. For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control, siGLP, and DKD cells were scored. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (E) Control, siGLP, and DKD cells were differentiated for 3 days and expression of Myogenin and TroponinT was analyzed. β-actin was used as loading control. Discussion Chromatin organization is dynamically altered during myogenic differentiation. However, the mechanisms by which epigenetic alterations lead to distinct domains in chromatin that establish and maintain cell fates is unclear. In this study, we have identified a new role for GLP in transcriptional regulation of MeCP2. This in turn reduces HP1γ recruitment and heterochromatin remodeling during myogenic differentiation. We demonstrate that GLP inhibits skeletal muscle differentiation. Interestingly, the levels of G9a were unaltered in both siGLP and pMaRX-GLP cells indicating that GLP inhibits muscle differentiation independent of G9a as previously reported (Schaefer et al., 2009; Ohno et al., 2013). In line with our findings, a recent study also showed inhibition of myogenic differentiation upon GLP overexpression, although it promoted myogenic gene expression in myoblasts due to stabilization of MyoD (Battisti et al., 2016). GLP has been reported to interact with HP1γ and DNMT3a (Tachibana et al., 2005; Fritsch et al., 2010; Chang et al., 2011). However, its overexpression impairs rather than enhances heterochromatin aggregation during differentiation as evidenced by the number and size of DAPI stained chromocenters, as well as reduced HP1γ and MeCP2 localization at pericentric heterochromatin. H3K9me3 levels or DNA methylation at constitutive heterochromatin were not altered upon GLP overexpression. Since MeCP2 recruits HP1γ to heterochromatin during myogenic differentiation (Agarwal et al., 2007), our data suggests that reduced MeCP2 expression is central to the altered heterochromatin organization upon GLP overexpression. Indeed, knockdown of MeCP2 alone impacts HP1γ localization to heterochromatin mimicking the effect of GLP overexpression. Moreover, MeCP2 depletion in GLP-knockdown background reverted the enhanced HP1γ localization to pericentric heterochromatin, as well as myogenic differentiation. Our findings suggest that the previously established association of GLP with HP1γ (Tachibana et al., 2005; Chin et al., 2007; Fritsch et al., 2010) in euchromatin of undifferentiated myoblasts may be important to transcriptionally repress differentiation genes. Indeed, overexpression of HP1γ has been shown to repress Myogenin expression and curb MyoD transcriptional activity (Yahi et al., 2008). As GLP levels decline upon differentiation, MeCP2 expression is induced and recruits HP1γ to heterochromatin to repress proliferation genes in terminally differentiating myotubes. Consistent with this model, HP1γ is known to repress cell-cycle genes to maintain post-mitotic status of adult cardiac myocytes (Sdek et al., 2011) and cell-cycle genes are preferentially localized to pericentric heterochromatin upon skeletal muscle differentiation (Guasconi et al., 2010). Thus induction of MeCP2 during myogenic differentiation may be important for stabilization and maintenance of transcriptional programs during maturation of differentiated cells (Brero et al., 2005; Agarwal et al., 2007; Singleton et al., 2011; Becker et al., 2016). Consistent with previous studies showing that MeCP2 expression is transcriptionally repressed by H3K9me2 (Heard et al., 2001; Abuhatzira et al., 2011), we demonstrate that GLP represses MeCP2 expression. GLP occupancy and H3K9me2 enrichment is apparent at the MeCP2 promoter in proliferating myoblasts that correlate with inhibition of its expression. The downregulation of GLP during differentiation likely relieves this repression allowing MeCP2 to dramatically increase during myogenic differentiation (Brero et al., 2005; Agarwal et al., 2007). In line with this notion, in silico MeCP2 promoter analysis revealed that MyoD is able to bind to the MeCP2 promoter, indicating that MyoD could recruit GLP to the MeCP2 promoter and mediate repressive H3K9me2 marks to curb its expression (software Match 1.0, data not shown). Some studies have shown that MeCP2 does not impact myogenic differentiation likely due to redundancy with MBD2 (Brero et al., 2005; Becker et al., 2013), and has a non-cell autonomous role in skeletal muscle (Conti et al., 2015). Nonetheless, elevated MeCP2 expression has a significant detrimental impact on cardiac development in vivo (Alvarez-Saavedra et al., 2010), and peripheral MeCP2 knockout mice, which retain MeCP2 expression in neurons, exhibits reduced exercise capacity and aggravated fatigue (Ross et al., 2016). Moreover, a recent study using a Rett syndrome zebrafish model with MeCP2 loss-of-function showed severe downregulation of terminal myogenic differentiation genes, such as myosin heavy chain, TroponinT, and muscle-specific creatine kinase, at both embryonic and adult stages (Cortelazzo et al., 2017). Also, transcriptome analysis in mouse models with MeCP2 loss-of-function and gain-of-function in the hippocampus also indicates that TroponinT is positively regulated by MeCP2 (Chahrour et al., 2008). Given that depletion of MeCP2 in siGLP cells restored myogenic differentiation to control levels in our study, the difference between our analysis and MeCP2 knockout mice may be attributed to the different models employed in each study. Unlike analysis of terminally differentiated striated skeletal muscle tissue from adult mice, MeCP2 knockdown in myogenic precursor cells in vitro may recapitulate delayed differentiation during development. Analysis of missense Rett syndrome (RTT) mutations that lead to MeCP2 loss-of-function revealed that two-thirds of them show impaired heterochromatin clustering (Agarwal et al., 2011). It was also recently reported that a scaffolding protein SH2B1 also exerts pro-myogenic effect by modulating heterochromatin clustering during differentiation (Chen et al., 2017). Loss-of-function mutations in GLP gene underlie the neurological disorder Kleefstra syndrome (Kleefstra et al., 2006), whereas gain of MeCP2 leads to MeCP2 duplication syndrome (Amir et al., 1999; Ramocki et al., 2009). Both Kleefstra syndrome and MeCP2 duplication syndrome exhibit intellectual disability, muscle hypotonia, and dysmorphism as common symptoms. Notably, clinical screening of Kleefstra syndrome patients revealed a de novo missense mutation of MBD5, a methyl-binding protein similar to MeCP2 (Kleefstra et al., 2012). These studies demonstrate genetic interactions between GLP and MeCP2/MBD5. Thus, the GLP–MeCP2 regulatory axis identified in this study may be relevant in both disorders as well as several neuropathologies characterized by loss of MeCP2 expression. Materials and methods Cell culture and differentiation assays C2C12 cells were cultured in high glucose Dulbecco’s Modified Eagle’s Medium (DMEM; Sigma) supplemented with 20% fetal bovine serum (FBS; Hyclone). To induce differentiation, cells were cultured to 80%–90% confluency in growth media, and then cultured in differentiation medium [DMEM with 2% horse serum (Invitrogen)]. HEK293T and Phoenix cell lines were cultured in DMEM supplemented with 10% FBS. Isolation of primary myoblasts has been previously described (Sun et al., 2007; Rao et al., 2016). Differentiation was assessed by MHC staining as previously described (Ling et al., 2012). Briefly, differentiated cells were stained with mouse monoclonal anti-MHC (Sigma) antibody, and detected with secondary antibody conjugated to Alexa Fluor (Alexa Fluor 488; 568; Invitrogen). Nuclei were stained with the fluorochrome 4′,6-diamino-2-phenylindole (DAPI). Images were captured under Olympus DP72 fluorescence microscope. More than 1000 nuclei from five different fields were counted. Myogenic index was calculated by quantifying the ratio of nuclei in myosin heavy chain-positive myotubes to total nuclei. Plasmids, retroviral infection, and siRNA Plasmids FLAG-GLP and FLAG-ΔANKΔSET that lacks ANK repeats and SET domain of GLP have been previously described (Takahashi et al., 2012). To generate stable cell lines, C2C12 cells were transfected with FLAG-GLP or FLAG-ΔANKΔSET and pBABE vector in a 9:1 ratio using Lipofectamine 2000 reagent (Invitrogen). Mouse MeCP2 plasmid was purchased from ORIGENE (MR226839). Cells were selected with puromycin for 48 h and analyzed in differentiation experiments. Alternatively, pMaRX or pMaRX-GLP retroviral constructs were transfected into Phoenix packaging cells. Retroviral supernatants were used to transduce undifferentiated C2C12 cells with 8 μg/ml polybrene. Cells were subsequently subjected to selection with medium containing 2 μg/ml of puromycin for 48 h. Knockdown experiments were performed using 50 nM GLP-specific siRNA (siGLP; ON-TARGETplusSMARTpool; NM_001012518) or 50 nM MeCP2-specific siRNA (siMeCP2; ON-TARGETplusSMARTpool; NM_001081979) from Dharmacon using Lipofectamine RNAiMAX (Invitrogen) as described in the manufacturer’s protocol. Control cells were transfected with non-specific scrambled siRNA (Dharmacon, ON-TARGETplus, Non-Targeting Pool). Transfection of siRNA was done in growth medium for 48 h. RNA isolation and quantitative real-time polymerase chain reaction (Q-PCR) Total RNA was extracted using Trizol (Invitrogen). RNA was quantified using Nanodrop. Messenger RNA (mRNA) was converted to a single-stranded complementary DNA (cDNA) using iScript cDNA Synthesis Kit (Bio-Rad). Quantitative real-time polymerase chain reaction (Q-PCR) was performed using Lightcycler 480 SYBR Green 1 Master Kit (Roche). CT values of samples were normalized to internal control Gapdh to obtain delta CT (ΔCT). Relative expression was calculated by 2−ΔCT equation. Primer sequences for GLP (Ohno et al., 2013), Myogenin (Ling et al., 2012), MeCP2 (Murgatroyd et al., 2009), and MBD2 (Song et al., 2014) have been previously described. For Myh1, the following set of primers were used: 5′-AACAGCAGCGGCTGATCAAT-3′ and 5′-GCTGCCTCTTCAGCTCCTCA-3′. Chromatin immuno-precipitation (ChIP) ChIP assays were done as previously described (Ling et al., 2012). Briefly, 1 × 106 cells were cross-linked with 1% formaldehyde for 10 min at 37°C. Cells were sonicated using Bioruptor (Diagenode). ChIP was carried out according to the kit protocol (Millipore). Immuno-precipitates were reverse-cross-linked and DNA was extracted using phenol–chloroform–isoamylalcohol (Sigma). Q-PCR was performed as described above. DNA harvested from 10% input was used as control. Relative enrichment was calculated using 2−ΔCT equation. Following antibodies were used for ChIP assays: mouse monoclonal ChIP-grade anti-GLP (Abcam, ab41969), anti-H3K9me2 (Millipore, 17-681), anti-HP1γ (Millipore, MAB3450), anti-MeCP2 (Abcam, ab07013), anti-H3K9me3 (Abcam ab8898). Primers used for ChIP assays of MeCP2 promoters (Heard et al., 2001), major satellites (Terranova et al., 2005), and β-actin (Rao et al., 2016) have been described previously. Cell proliferation assay Proliferation was measured using BrdU incorporation assays (Azmi et al., 2004). Cells were pulsed with 10 μM BrdU for 30 min. Cells were fixed and stained with anti-BrdU antibody according to manufacturer’s protocol (Roche). More than 1000 nuclei from five different fields were counted. BrdU positivity was calculated. Immunofluorescence imaging and analysis For confocal imaging, samples were permeabilized with 0.1% Triton X-100 for 1 h at room temperature and blocked overnight in 10% horse serum in PBS. Samples were probed with primary antibodies overnight at 4°C in blocking buffer with 0.1 M glycine (Brocher et al., 2010). Primary antibodies used for immunofluorescence are anti-MeCP2 (Cell Signaling, 1:200), anti-HP1γ (Millipore, MAB3450, 1:350), and anti-H3K9me3 (Abcam, ab8988, 1:1000). Images were captured under FluoView FV1000 confocal fluorescence microscope (Olympus) with optical section thickness 1μm. More than 50 nuclei for each sample were counted for three independent experiments, unless otherwise stated. Number and size of DAPI positive chromocenters were quantified using ‘cell counter’ and ‘analyze particle’ tools in ImageJ software as described previously (Novo et al., 2016). Mander’s coefficient was used to measure the co-localization of DAPI and HP1γ using ‘JACoP’ tool in ImageJ as described previously (Bolte and Cordelières, 2006). Expression of MeCP2 was quantified using the total corrected cellular fluorescence (TCCF) equation as described previously (McCloy et al., 2014). Box and whisker plots were plotted using Microsoft Excel software. Combined bisulfite restriction assay (COBRA) COBRA assay was performed as previously described (Xiong and Laird, 1997). Briefly, genomic DNA (gDNA) was extracted using GeneJET Genomic DNA purification Kit (Thermo Fisher Scientific). gDNA was subject to bisulfite treatment using EZ DNA Methylation-Gold Kit (Zymo Research). Eluted DNA was PCR-amplified using the following set of primers: 5′-GGAATATGGCAAGAAAACTGAAAATCATGG-3′ and 5′-CCATATTCCAGGTCCTTCAGTGTGCATTTC-3′. The PCR reaction was carried out for 35 cycles of 95°C for 30 sec, 58°C for 60 sec, and 68°C for 20 sec. The purified PCR products were digested with restriction enzyme HpyCH4IV, which has a recognition sequence (ACGT) that should be destroyed by the bisulfite conversion. Restriction digestion reactions were performed at 37°C for 30 min. The digested PCR products were separated on 1% polyacrylamide gel and were visualized with ChemiDoc imaging system (Bio-rad). DNA methylation was calculated by quantifying ratio of methylated DNA and sum of un- and methylated DNA. Immunoblotting Protein lysate was harvested in radioimmunoprecipitation assay (RIPA) lysis buffer [50 mM NaCl, 50 mM Tris-HCl pH 7.0, 1 mM EDTA, 1% NP-40, 0.1% sodium deoxycholate, protease inhibitor cocktail (Roche)]. Histones were isolated using RIPA lysis buffer supplemented with 2% sodium dodecyl sulfate (SDS). The following primary antibodies were used for immunoblotting: mouse monoclonal anti-GLP (Abcam, ab41969, 1:2000), rabbit polyclonal anti-Myogenin (Santa Cruz Biotechnology, sc-576, 1:500), mouse monoclonal anti-Troponin-T (Sigma, T6277, 1:2000), rabbit polyclonal anti-MyoD (Santa Cruz Biotechnology, sc-304, 1:500), mouse monoclonal anti-Flag (Sigma, F3165, 1:500), rabbit polyclonal anti-MeCP2 (Cell Signaling, 3456, 1:2000), anti-HP1γ (Millipore, MAB3450, 1:5000), anti-H3K9me3 (Abcam, ab8988, 1:2000), and mouse monoclonal anti-β-actin (Sigma, A1978, 1:10000). Statistical analysis Significance was calculated using student’s t test (two-sided) and P values <0.05 were considered to be statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001). Bars represent mean ± standard deviation (SD) unless specified otherwise. For quantification significance for data with more than two groups, a one-way ANOVA followed by Bonferroni’s multiple comparison test was performed and P values <0.05 were considered to be statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001). Acknowledgements We thank Dr Eiji Hara (Osaka University) for the gift of various GLP plasmid constructs, Ms Lee Shu Yin (Confocal Microscopy Unit, National University of Singapore) for her help with confocal image analysis, and Dr Neerja Karnani (National University of Singapore) for valuable discussions. Funding This work was supported by a grant from the National Medical Research Council (NMRC/CBRG/0105/2016 to R.T.). M.H.C. and J.R.O. are supported by a NUS Graduate School for Integrative Sciences and Engineering Scholarship. Conflict of interest: none declared. References Abuhatzira, L., Shamir, A., Schones, D.E., et al.  . ( 2011). The chromatin-binding protein HMGN1 regulates the expression of methyl CpG-binding protein 2 (MECP2) and affects the behavior of mice. J. Biol. Chem.  286, 42051– 42062. Google Scholar CrossRef Search ADS PubMed  Agarwal, N., Becker, A., Jost, K.L., et al.  . ( 2011). MeCP2 Rett mutations affect large scale chromatin organization. Hum. Mol. Genet.  20, 4187– 4195. Google Scholar CrossRef Search ADS PubMed  Agarwal, N., Hardt, T., Brero, A., et al.  . ( 2007). MeCP2 interacts with HP1 and modulates its heterochromatin association during myogenic differentiation. Nucleic Acids Res.  35, 5402– 5408. Google Scholar CrossRef Search ADS PubMed  Alvarez-Saavedra, M., Carrasco, L., Sura-Trueba, S., et al.  . ( 2010). Elevated expression of MeCP2 in cardiac and skeletal tissues is detrimental for normal development. Hum. Mol. Genet.  19, 2177– 2190. Google Scholar CrossRef Search ADS PubMed  Amir, R.E., Van den Veyver, I.B., Wan, M., et al.  . ( 1999). Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpG-binding protein 2. Nat. Genet.  23, 185– 188. Google Scholar CrossRef Search ADS PubMed  Azmi, S., Ozog, A., and Taneja, R. ( 2004). Sharp-1/DEC2 inhibits skeletal muscle differentiation through repression of myogenic transcription factors. J. Biol. Chem.  279, 52643– 52652. Google Scholar CrossRef Search ADS PubMed  Battisti, V., Pontis, J., Boyarchuk, E., et al.  . ( 2016). Unexpected distinct roles of the related histone H3 lysine 9 methyltransferases G9a and G9a-like protein in myoblasts. J. Mol. Biol.  428, 2329– 2343. Google Scholar CrossRef Search ADS PubMed  Becker, A., Allmann, L., Hofstätter, M., et al.  . ( 2013). Direct homo- and hetero-interactions of MeCP2 and MBD2. PLoS One  8, e53730. Google Scholar CrossRef Search ADS PubMed  Becker, A., Zhang, P., Allmann, L., et al.  . ( 2016). Poly(ADP-ribosyl)ation of methyl CpG binding domain protein 2 regulates chromatin structure. J. Biol. Chem.  291, 4873– 4881. Google Scholar CrossRef Search ADS PubMed  Bolte, S., and Cordelières, F.P. ( 2006). A guided tour into subcellular colocalization analysis in light microscopy. J. Microsc.  224, 213– 232. Google Scholar CrossRef Search ADS PubMed  Brero, A., Easwaran, H.P., Nowak, D., et al.  . ( 2005). Methyl CpG-binding proteins induce large-scale chromatin reorganization during terminal differentiation. J. Cell Biol.  169, 733– 743. Google Scholar CrossRef Search ADS PubMed  Brocher, J., Vogel, B., and Hock, R. ( 2010). HMGA1 down-regulation is crucial for chromatin composition and a gene expression profile permitting myogenic differentiation. BMC Cell Biol.  11, 64. Google Scholar CrossRef Search ADS PubMed  Chahrour, M., Jung, S.Y., Shaw, C., et al.  . ( 2008). MeCP2, a key contributor to neurological disease, activates and represses transcription. Science  320, 1224– 1229. Google Scholar CrossRef Search ADS PubMed  Chang, Y., Sun, L., Kokura, K., et al.  . ( 2011). MPP8 mediates the interactions between DNA methyltransferase Dnmt3a and H3K9 methyltransferase GLP/G9a. Nat. Commun.  2, 533. Google Scholar CrossRef Search ADS PubMed  Chen, K.-W., Chang, Y.-J., Yeh, C.-M., et al.  . ( 2017). SH2B1 modulates chromatin state and MyoD occupancy to enhance expressions of myogenic genes. Biochim. Biophys. Acta  1860, 270– 281. Google Scholar CrossRef Search ADS PubMed  Chin, H.G., Estève, P.-O., Pradhan, M., et al.  . ( 2007). Automethylation of G9a and its implication in wider substrate specificity and HP1 binding. Nucleic Acids Res.  35, 7313– 7323. Google Scholar CrossRef Search ADS PubMed  Conti, V., Gandaglia, A., Galli, F., et al.  . ( 2015). MeCP2 affects skeletal muscle growth and morphology through non cell-autonomous mechanisms. PLoS One  10, e0130183. Google Scholar CrossRef Search ADS PubMed  Cortelazzo, A., Pietri, T., De Felice, C., et al.  . ( 2017). Proteomic analysis of the Rett syndrome experimental model mecp2(Q63X) mutant zebrafish. J. Proteomics  154, 128– 133. Google Scholar CrossRef Search ADS PubMed  Fritsch, L., Robin, P., Mathieu, J.R.R., et al.  . ( 2010). A subset of the histone H3 lysine 9 methyltransferases Suv39h1, G9a, GLP, and SETDB1 participate in a multimeric complex. Mol. Cell  37, 46– 56. Google Scholar CrossRef Search ADS PubMed  Fuks, F., Hurd, P.J., Deplus, R., et al.  . ( 2003a). The DNA methyltransferases associate with HP1 and the SUV39H1 histone methyltransferase. Nucleic Acids Res.  31, 2305– 2312. Google Scholar CrossRef Search ADS PubMed  Fuks, F., Hurd, P.J., Wolf, D., et al.  . ( 2003b). The methyl-CpG-binding protein MeCP2 links DNA methylation to histone methylation. J. Biol. Chem.  278, 4035– 4040. Google Scholar CrossRef Search ADS PubMed  Guasconi, V., Pritchard, L.-L., Fritsch, L., et al.  . ( 2010). Preferential association of irreversibly silenced E2F-target genes with pericentromeric heterochromatin in differentiated muscle cells. Epigenetics  5, 704– 709. Google Scholar CrossRef Search ADS PubMed  Heard, E., Rougeulle, C., Arnaud, D., et al.  . ( 2001). Methylation of histone H3 at Lys-9 is an early mark on the X chromosome during X inactivation. Cell  107, 727– 738. Google Scholar CrossRef Search ADS PubMed  Jung, E.-S., Sim, Y.-J., Jeong, H.-S., et al.  . ( 2015). Jmjd2C increases MyoD transcriptional activity through inhibiting G9a-dependent MyoD degradation. Biochim. Biophys. Acta  1849, 1081– 1094. Google Scholar CrossRef Search ADS PubMed  Kleefstra, T., Brunner, H.G., Amiel, J., et al.  . ( 2006). Loss-of-function mutations in euchromatin histone methyl transferase 1 (EHMT1) cause the 9q34 subtelomeric deletion syndrome. Am. J. Hum. Genet.  79, 370– 377. Google Scholar CrossRef Search ADS PubMed  Kleefstra, T., Kramer, J.M., Neveling, K., et al.  . ( 2012). Disruption of an EHMT1-associated chromatin-modification module causes intellectual disability. Am. J. Hum. Genet.  91, 73– 82. Google Scholar CrossRef Search ADS PubMed  Lehnertz, B., Ueda, Y., Derijck, A.A.H.A., et al.  . ( 2003). Suv39h-mediated histone H3 Lysine 9 methylation directs DNA methylation to major satellite repeats at pericentric heterochromatin. Curr. Biol.  13, 1192– 1200. Google Scholar CrossRef Search ADS PubMed  Ling, B.M.T., Bharathy, N., Chung, T.-K., et al.  . ( 2012). Lysine methyltransferase G9a methylates the transcription factor MyoD and regulates skeletal muscle differentiation. Proc. Natl Acad. Sci. USA  109, 841– 846. Google Scholar CrossRef Search ADS   Liu, N., Zhang, Z., Wu, H., et al.  . ( 2015). Recognition of H3K9 methylation by GLP is required for efficient establishment of H3K9 methylation, rapid target gene repression, and mouse viability. Genes Dev.  29, 379– 393. Google Scholar CrossRef Search ADS PubMed  Luo, S.-W., Zhang, C., Zhang, B., et al.  . ( 2009). Regulation of heterochromatin remodelling and myogenin expression during muscle differentiation by FAK interaction with MBD2. EMBO J.  28, 2568– 2582. Google Scholar CrossRef Search ADS PubMed  McCloy, R.A., Rogers, S., Caldon, C.E., et al.  . ( 2014). Partial inhibition of Cdk1 in G2 phase overrides the SAC and decouples mitotic events. Cell Cycle  13, 1400– 1412. Google Scholar CrossRef Search ADS PubMed  Murgatroyd, C., Patchev, A.V., Wu, Y., et al.  . ( 2009). Dynamic DNA methylation programs persistent adverse effects of early-life stress. Nat. Neurosci.  12, 1559– 1566. Google Scholar CrossRef Search ADS PubMed  Novo, C.L., Tang, C., Ahmed, K., et al.  . ( 2016). The pluripotency factor Nanog regulates pericentromeric heterochromatin organization in mouse embryonic stem cells. Genes Dev.  30, 1101– 1115. Google Scholar CrossRef Search ADS PubMed  Ohno, H., Shinoda, K., Ohyama, K., et al.  . ( 2013). EHMT1 controls brown adipose cell fate and thermogenesis through the PRDM16 complex. Nature  504, 163– 167. Google Scholar CrossRef Search ADS PubMed  Ow, J.R., Palanichamy Kala, M., Rao, V.K., et al.  . ( 2016). G9a inhibits MEF2C activity to control sarcomere assembly. Sci. Rep.  6, 34163. Google Scholar CrossRef Search ADS PubMed  Ramocki, M.B., Peters, S.U., Tavyev, Y.J., et al.  . ( 2009). Autism and other neuropsychiatric symptoms are prevalent in individuals with MeCP2 duplication syndrome. Ann. Neurol.  66, 771– 782. Google Scholar CrossRef Search ADS PubMed  Rao, V.K., Ow, J.R., Shankar, S.R., et al.  . ( 2016). G9a promotes proliferation and inhibits cell cycle exit during myogenic differentiation. Nucleic Acids Res.  44, 8129– 8143. Google Scholar CrossRef Search ADS PubMed  Ross, P.D., Guy, J., Selfridge, J., et al.  . ( 2016). Exclusive expression of MeCP2 in the nervous system distinguishes between brain and peripheral Rett syndrome-like phenotypes. Hum. Mol. Genet.  25, 4389– 4404. Google Scholar PubMed  Schaefer, A., Sampath, S.C., Intrator, A., et al.  . ( 2009). Control of cognition and adaptive behavior by the GLP/G9a epigenetic suppressor complex. Neuron  64, 678– 691. Google Scholar CrossRef Search ADS PubMed  Sdek, P., Oyama, K., Angelis, E., et al.  . ( 2013). Epigenetic regulation of myogenic gene expression by heterochromatin protein 1α. PLoS One  8, e58319. Google Scholar CrossRef Search ADS PubMed  Sdek, P., Zhao, P., Wang, Y., et al.  . ( 2011). Rb and p130 control cell cycle gene silencing to maintain the postmitotic phenotype in cardiac myocytes. J. Cell Biol.  194, 407– 423. Google Scholar CrossRef Search ADS PubMed  Singleton, M.K., Gonzales, M.L., Leung, K.N., et al.  . ( 2011). MeCP2 is required for global heterochromatic and nucleolar changes during activity-dependent neuronal maturation. Neurobiol. Dis.  43, 190– 200. Google Scholar CrossRef Search ADS PubMed  Song, C., Feodorova, Y., Guy, J., et al.  . ( 2014). DNA methylation reader MECP2: cell type- and differentiation stage-specific protein distribution. Epigenetics Chromatin  7, 17. Google Scholar CrossRef Search ADS PubMed  Sun, H., Li, L., Vercherat, C., et al.  . ( 2007). Stra13 regulates satellite cell activation by antagonizing Notch signaling. J. Cell Biol.  177, 647– 657. Google Scholar CrossRef Search ADS PubMed  Tachibana, M., Ueda, J., Fukuda, M., et al.  . ( 2005). Histone methyltransferases G9a and GLP form heteromeric complexes and are both crucial for methylation of euchromatin at H3-K9. Genes Dev.  19, 815– 826. Google Scholar CrossRef Search ADS PubMed  Takahashi, A., Imai, Y., Yamakoshi, K., et al.  . ( 2012). DNA damage signaling triggers degradation of histone methyltransferases through APC/C(Cdh1) in senescent cells. Mol. Cell  45, 123– 131. Google Scholar CrossRef Search ADS PubMed  Terranova, R., Sauer, S., Merkenschlager, M., et al.  . ( 2005). The reorganisation of constitutive heterochromatin in differentiating muscle requires HDAC activity. Exp. Cell Res.  310, 344– 356. Google Scholar CrossRef Search ADS PubMed  Trojer, P., and Reinberg, D. ( 2007). Facultative heterochromatin: is there a distinctive molecular signature? Mol. Cell  28, 1– 13. Google Scholar CrossRef Search ADS PubMed  Xiong, Z., and Laird, P.W. ( 1997). COBRA: a sensitive and quantitative DNA methylation assay. Nucleic Acids Res.  25, 2532– 2534. Google Scholar CrossRef Search ADS PubMed  Yahi, H., Fritsch, L., Philipot, O., et al.  . ( 2008). Differential cooperation between heterochromatin protein HP1 isoforms and MyoD in myoblasts. J. Biol. Chem.  283, 23692– 23700. Google Scholar CrossRef Search ADS PubMed  © The Author (2017). Published by Oxford University Press on behalf of Journal of Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Molecular Cell Biology Oxford University Press

GLP inhibits heterochromatin clustering and myogenic differentiation by repressing MeCP2

Loading next page...
 
/lp/ou_press/glp-inhibits-heterochromatin-clustering-and-myogenic-differentiation-EBNpo87yYG
Publisher
Oxford University Press
Copyright
© The Author (2017). Published by Oxford University Press on behalf of Journal of Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved.
ISSN
1674-2788
eISSN
1759-4685
D.O.I.
10.1093/jmcb/mjx038
Publisher site
See Article on Publisher Site

Abstract

Abstract Myogenic differentiation is accompanied by alterations in the chromatin states, which permit or restrict the transcriptional machinery and thus impact distinctive gene expression profiles. The mechanisms by which higher-order chromatin remodeling is associated with gene activation and silencing during differentiation is not fully understood. In this study, we provide evidence that the euchromatic lysine methyltransferase GLP regulates heterochromatin organization and myogenic differentiation. Interestingly, GLP represses expression of the methyl-binding protein MeCP2 that induces heterochromatin clustering during differentiation. Consequently, MeCP2 and HP1γ localization at major satellites are altered upon modulation of GLP expression. In GLP knockdown cells, depletion of MeCP2 restored both chromatin organization and myogenic differentiation. These results identify a novel regulatory axis between a histone methylation writer and DNA methylation reader, which is important for heterochromatin organization during differentiation. myogenic differentiation, chromatin, methylation Introduction Myogenic differentiation is accompanied by spatial reorganization of chromatin. In addition to epigenetic modifications that occur locally at promoters, heterochromatin is modified and gets progressively clustered in distinctive chromocenters. Concomitantly, major and minor satellite RNA transcripts accumulate during differentiation (Brero et al., 2005; Terranova et al., 2005; Agarwal et al., 2007; Luo et al., 2009; Sdek et al., 2011; Chen et al., 2017) DNA methylation at CpG dinucleotides and histone 3 lysine 9 methylation (H3K9me) are hallmarks of heterochromatin. Consequently, proteins that recognize methylation marks, such as methyl-CpG-binding domain (MBD) protein MeCP2, as well as heterochromatin protein 1 (HP1), are enriched in pericentric heterochromatin. Moreover, MeCP2 and HP1 result in heterochromatin clustering, which is thought to be essential for maintenance of the differentiated state in skeletal and cardiac myocytes (Brero et al., 2005; Agarwal et al., 2007; Sdek et al., 2011). The interplay between DNA and histone modifications in heterochromatin organization is complex and not fully understood. For instance, the absence of DNA methylation in DNMT1 or DNMT3a/3b deficient cells results in reduced MeCP2 localization but does not alter Suv39h-mediated H3K9 tri-methylation (H3K9me3) marks or HP1α localization at pericentric heterochromatin (Lehnertz et al., 2003). Furthermore, while MeCP2 interacts with Suv39h1/h2 (Fuks et al., 2003b), exogenous MeCP2 still induces pericentric heterochromatin clustering in Suv39h double-null cells, indicating that the MeCP2–Suv39h interaction is not essential (Brero et al., 2005). On the other hand, H3K9me is required for DNA methylation (Lehnertz et al., 2003). DNMT3a and 3b are directed to H3K9me marks via association with HP1α and HP1β (Fuks et al., 2003a). In Suv39h null cells, DNA methylation and MeCP2 localization is reduced at pericentric satellite repeats (Lehnertz et al., 2003). G9a and G9a-like protein (GLP) are the primary mediators of H3K9me2 in euchromatin (Tachibana et al., 2005; Trojer and Reinberg, 2007) and their auto-methylation acts as a docking site for HP1α and HP1γ (Chin et al., 2007; Fritsch et al., 2010). Depletion of GLP or G9a restricts localization of HP1γ to heterochromatin in mouse embryonic stem cells (Tachibana et al., 2005). In addition, G9a/GLP also associate with DNMT3a via MPP8 (Chang et al., 2011). Gene disruption studies in mice have shown that loss of G9a or GLP abrogates global H3K9me2 and results in embryonic lethality at E9.5 (Tachibana et al., 2005). Notably, the Ankyrin (ANK) repeats of GLP are required for efficient establishment and maintenance of H3K9me2 during embryonic development while the ANK repeats of G9a are dispensable (Liu et al., 2015), indicating independent functions of GLP and G9a during embryonic development. A few recent studies have suggested that GLP may be involved in skeletal myogenesis. For instance, GLP stabilizes PRDM16 and steers Myf5+ precursor cells towards a brown adipocyte lineage, in turn suppressing myogenic promoters (Ohno et al., 2013). In addition, conditional deletion of GLP in neurons results in the upregulation of myogenic genes (Schaefer et al., 2009). In cultured myoblasts, GLP overexpression was found to block myogenic differentiation (Battisti et al., 2016). However, the molecular mechanisms by which GLP inhibits muscle differentiation and its potential role in heterochromatin organization is not well understood. In this study, we have examined the role of GLP in skeletal myogenesis. We show that GLP overexpression inhibits terminal differentiation of muscle precursor cells. The defect in myogenic differentiation correlates with impaired heterochromatin clustering. HP1γ localization at heterochromatin and its occupancy at major satellites is reduced in GLP-overexpressing cells, and conversely is increased in GLP knockdown primary myoblasts. These changes occur in the absence of any changes in H3K9me3 levels or in DNA methylation. Interestingly however, MeCP2, which recruits HP1γ to heterochromatin, is repressed in GLP-overexpressing cells resulting in reduced occupancy at major satellites. We demonstrate that MeCP2 is a direct GLP target gene. Consequently, depletion of endogenous MeCP2 expression in the background of GLP knockdown reverts the enhanced myogenic differentiation and HP1γ localization to chromocenters seen in GLP knockdown cells. Taken together, our results provide evidence that GLP influences chromatin states and myogenic differentiation through repression of MeCP2. Results GLP inhibits skeletal muscle differentiation While recent studies have implicated GLP in skeletal muscle differentiation (Schaefer et al., 2009; Ohno et al., 2013; Battisti et al., 2016), the underlying mechanisms are not completely understood. To this end, we first examined endogenous GLP mRNA and protein expression in C2C12 cells. Similar to G9a (Ling et al., 2012; Jung et al., 2015; Ow et al., 2016), GLP expression declined during differentiation and inversely correlated with differentiation markers Myogenin and TroponinT (Figure 1A and B). To test its potential role in myogensis, we performed loss-of-function studies and examined proliferation and differentiation of GLP knockdown cells. C2C12 cells were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 48 h (Figure 1C). Cells were pulsed with BrdU and stained with anti-BrdU antibody. No significant changes in proliferation of siGLP cells were observed compared to control cells (Figure 1D). Interestingly however, siGLP cells exhibited a significant increase in myogenic differentiation as seen by immunofluorescence staining using anti-myosin heavy chain (MHC) antibody and myogenic index (Figure 1E). Consistently, Myogenin, TroponinT, and MyoD expression levels were elevated in siGLP cells upon differentiation, while G9a levels were not overtly altered (Figure 1F). To validate the impact of GLP on differentiation, siRNA-mediated depletion of GLP in mouse primary myoblasts was performed (Figure 1G). Similar to the effect in C2C12 cells, siGLP primary myoblasts expressed increased levels of Myogenin and myosin heavy chain 1 (Myh1) during differentiation (Figure 1G). Taken together, these results indicate that endogenous GLP inhibits myogenesis. Figure 1 View largeDownload slide Endogenous GLP inhibits skeletal muscle differentiation. (A) GLP mRNA levels were measured in undifferentiated (Day 0) and differentiating (Day 1 and Day 3) C2C12 cells by Q-PCR. (B) Expression of GLP, Myogenin, TroponinT, and MyoD in C2C12 cells was analyzed by western blot. β-actin was used as loading control. The molecular weights of proteins are indicated by numbers on the left. (C) C2C12 myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 48 h. GLP expression was analyzed by western blot. β-actin was used as loading control. (D) Control and siGLP C2C12 myoblasts were pulsed with BrdU and stained with anti-BrdU antibody (green). Nuclei were stained with DAPI (blue). For each sample, >1000 nuclei in 5 fields were counted and the percentage of BrdU+ cells was calculated. Scale bar, 60 μm. The results are representative of two independent experiments. Bars indicate the mean of BrdU+ cells in each experiment ±SD. n.s., P-value not significant. (E) Control and siGLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control and siGLP cells were scored. Scale bar, 15 μm. The results are representative of three independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (F) Expression of G9a, MyoD, Myogenin, and TroponinT was analyzed by western blot in Control and siGLP C2C12 cells. β-actin was used as a loading control. (G) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 3 days. control and siGLP cells were analyzed for expression of GLP, Myogenin, and Myh1 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Error bars indicate the mean of Q-PCR triplicates in each experiment ±SD. Figure 1 View largeDownload slide Endogenous GLP inhibits skeletal muscle differentiation. (A) GLP mRNA levels were measured in undifferentiated (Day 0) and differentiating (Day 1 and Day 3) C2C12 cells by Q-PCR. (B) Expression of GLP, Myogenin, TroponinT, and MyoD in C2C12 cells was analyzed by western blot. β-actin was used as loading control. The molecular weights of proteins are indicated by numbers on the left. (C) C2C12 myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 48 h. GLP expression was analyzed by western blot. β-actin was used as loading control. (D) Control and siGLP C2C12 myoblasts were pulsed with BrdU and stained with anti-BrdU antibody (green). Nuclei were stained with DAPI (blue). For each sample, >1000 nuclei in 5 fields were counted and the percentage of BrdU+ cells was calculated. Scale bar, 60 μm. The results are representative of two independent experiments. Bars indicate the mean of BrdU+ cells in each experiment ±SD. n.s., P-value not significant. (E) Control and siGLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control and siGLP cells were scored. Scale bar, 15 μm. The results are representative of three independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (F) Expression of G9a, MyoD, Myogenin, and TroponinT was analyzed by western blot in Control and siGLP C2C12 cells. β-actin was used as a loading control. (G) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 3 days. control and siGLP cells were analyzed for expression of GLP, Myogenin, and Myh1 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Error bars indicate the mean of Q-PCR triplicates in each experiment ±SD. The SET domain and ANK repeats of GLP are required to inhibit myogenic differentiation We next examined whether overexpression of GLP is sufficient to inhibit myogenic differentiation. Cells were transfected with an empty vector (pCS2), full-length GLP (FL-GLP) or a deletion mutant lacking both ANK repeats and SET domain (ΔANKΔSET) (Figure 2A). Both proteins were expressed at equivalent levels (Figure 2B). Overexpression of FL-GLP significantly inhibited myogenic differentiation compared to control cells. However, no significant change in differentiation of ΔANKΔSET cells was seen (Figure 2C). Correspondingly, Myogenin and TroponinT levels were lower only in FL-GLP expressing cells (Figure 2D). To confirm these findings, GLP was retrovirally overexpressed in C2C12 cells (Figure 2E). Compared to control cells, pMaRX-GLP cells showed reduced myogenic differentiation as observed by MHC staining, myogenic index (Figure 2F), and downregulation of myogenic markers (Figure 2G). No significant change of MyoD levels was seen in undifferentiated pMaRX-GLP cells, while a modest reduction was apparent during differentiation (Figure 2G). Altogether, these results confirm that GLP overexpression inhibits myogenesis. Figure 2 View largeDownload slide GLP overexpression inhibits myogenic differentiation through the SET domain and ANK repeats. (A) Schematic diagram depicting full-length GLP (FL-GLP) construct with conserved Glu-rich region (E), Cys-rich region (C), ANK repeats (A), and SET domain (S). The mutant GLP (ΔANKΔSET) construct lacks the ANK repeats and SET domain. (B) C2C12 cells expressing pCS2, FL-GLP, and ΔANKΔSET were analyzed for expression of GLP by western blot using anti-Flag (left) and anti-GLP (right) antibodies. β-actin was used as loading control. (C) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (green). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pCS2, FL-GLP, and ΔANKΔSET cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and analyzed for Myogenin and TroponinT levels. β-actin was used as loading control. (E) C2C12 cells were transfected with an empty vector control (pMaRX) or pMaRX-GLP. GLP and G9a expression levels were analyzed by western blot. (F) pMaRX and pMaRX-GLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pMaRX and pMaRX-GLP cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (G) pMaRX and pMaRX-GLP cells were differentiated for 2 days and analyzed for Myogenin, TroponinT, and MyoD levels. β-actin was used as loading control. Figure 2 View largeDownload slide GLP overexpression inhibits myogenic differentiation through the SET domain and ANK repeats. (A) Schematic diagram depicting full-length GLP (FL-GLP) construct with conserved Glu-rich region (E), Cys-rich region (C), ANK repeats (A), and SET domain (S). The mutant GLP (ΔANKΔSET) construct lacks the ANK repeats and SET domain. (B) C2C12 cells expressing pCS2, FL-GLP, and ΔANKΔSET were analyzed for expression of GLP by western blot using anti-Flag (left) and anti-GLP (right) antibodies. β-actin was used as loading control. (C) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (green). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pCS2, FL-GLP, and ΔANKΔSET cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) pCS2, FL-GLP, and ΔANKΔSET cells were differentiated for 2 days and analyzed for Myogenin and TroponinT levels. β-actin was used as loading control. (E) C2C12 cells were transfected with an empty vector control (pMaRX) or pMaRX-GLP. GLP and G9a expression levels were analyzed by western blot. (F) pMaRX and pMaRX-GLP cells were differentiated for 2 days and immuno-stained with anti-MHC antibody (red). For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for pMaRX and pMaRX-GLP cells were scored. Scale bar, 15 μm. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. (G) pMaRX and pMaRX-GLP cells were differentiated for 2 days and analyzed for Myogenin, TroponinT, and MyoD levels. β-actin was used as loading control. Chromatin organization is altered in GLP-overexpressing cells Previous studies have shown that GLP interacts with HP1α and HP1γ, and its depletion restricts HP1γ to heterochromatin, suggesting that it may have a role in heterochromatin organization (Tachibana et al., 2005; Chin et al., 2007). In myoblasts, HP1γ localizes in both euchromatin and heterochromatin but localizes exclusively in heterochromatin upon myogenic differentiation (Agarwal et al., 2007; Sdek et al., 2013). This association of HP1γ to heterochromatin in myotubes is attributed to MeCP2, which recruits HP1γ to the chromocenters upon the induction of heterochromatin clustering (Agarwal et al., 2007). We therefore examined whether HP1γ levels or sub-nuclear localization was altered upon GLP overexpression during differentiation. To analyze heterochromatin clustering, cells were differentiated for 6 days when distinctive chromocenter formation is observed (Terranova et al., 2005). HP1γ levels were unchanged in pMaRX-GLP cells, nor was there any change in H3K9me3, a heterochromatin mark (Figure 3A). Focal nuclear staining of heterochromatin can be visualized with DAPI staining of A/T rich satellite repeats. Heterochromatin clustered into distinct chromocenters in both pMaRX and pMaRX-GLP cells (Figure 3B). However, pMaRX-GLP cells showed a higher number, but smaller size of DAPI stained chromocenters, and reduced localization of HP1γ to chromocenters compared to control cells (Figure 3C), suggesting that heterochromatin clustering may be impaired. H3K9me3 marks are established docking sites for HP1γ, but H3K9me3 sub-nuclear localization was not altered upon GLP overexpression (Figure 3B and C). To validate these findings, control or siGLP primary myoblasts were differentiated and chromocenters were analyzed. Compared to controls, siGLP cells showed a lower number, larger size of DAPI stained chromocenters, and enhanced HP1γ localization to heterochromatin (Figure 3D and E). Figure 3 View largeDownload slide Chromatin organization during differentiation is altered in GLP-overexpressing cells. (A) pMaRX and pMaRX-GLP cells were differentiated for 0 and 6 days and were analyzed for HP1γ and H3K9me3 levels. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-HP1γ (red) and anti-H3K9me3 (green) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, Mander’s coefficient for HP1γ and DAPI, and Mander’s coefficient for H3K9me3 and DAPI in pMaRX and pMaRX-GLP myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. n.s., P-value not significant. (D) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-HP1γ antibody (red). Scale bar, 2 μm. (E) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siGLP primary myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Figure 3 View largeDownload slide Chromatin organization during differentiation is altered in GLP-overexpressing cells. (A) pMaRX and pMaRX-GLP cells were differentiated for 0 and 6 days and were analyzed for HP1γ and H3K9me3 levels. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-HP1γ (red) and anti-H3K9me3 (green) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, Mander’s coefficient for HP1γ and DAPI, and Mander’s coefficient for H3K9me3 and DAPI in pMaRX and pMaRX-GLP myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. n.s., P-value not significant. (D) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-HP1γ antibody (red). Scale bar, 2 μm. (E) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siGLP primary myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. MeCP2 and HP1γ are redistributed in GLP-overexpressing cells Since MeCP2 interacts with HP1γ to regulate heterochromatin clustering during myogenic differentiation (Brero et al., 2005; Agarwal et al., 2007), we examined its expression. Interestingly, MeCP2 expression was decreased in pMaRX-GLP cells compared to control cells (Figure 4A). Conversely, its expression was upregulated in siGLP primary myoblasts (Figure 4B) suggesting that MeCP2 de-regulation may contribute to modulation of HP1γ localization. Figure 4 View largeDownload slide GLP alters sub-nuclear distribution of MeCP2 and HP1γ without affecting DNA methylation at major satellites. (A) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). Scale bar, 2 μm. Box and whisker plot shows total corrected cell fluorescence (TCCF) of MeCP2. For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. (B) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). TCCF of MeCP2 in control and siGLP primary myotubes is shown. (C) pMaRX and pMARX-GLP cells were differentiated for 6 days and ChIP assays for occupancy of HP1γ, MeCP2, and H3K9me3 at major satellites were performed. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. n.s., P-value not significant. (D) pMaRX and pMARX-GLP cells were differentiated for 6 days and methylation status of major satellite repeat was analyzed. PCR product of bisulfite-treated DNA was digested with (+) or without (−) HpyCH4IV. Methylation percentage was scored. The results are representative of four independent experiments. Figure 4 View largeDownload slide GLP alters sub-nuclear distribution of MeCP2 and HP1γ without affecting DNA methylation at major satellites. (A) pMaRX and pMaRX-GLP cells were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). Scale bar, 2 μm. Box and whisker plot shows total corrected cell fluorescence (TCCF) of MeCP2. For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of three independent experiments. (B) Control and siGLP primary myoblasts were differentiated for 6 days and immuno-stained with anti-MeCP2 antibody (green). TCCF of MeCP2 in control and siGLP primary myotubes is shown. (C) pMaRX and pMARX-GLP cells were differentiated for 6 days and ChIP assays for occupancy of HP1γ, MeCP2, and H3K9me3 at major satellites were performed. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. n.s., P-value not significant. (D) pMaRX and pMARX-GLP cells were differentiated for 6 days and methylation status of major satellite repeat was analyzed. PCR product of bisulfite-treated DNA was digested with (+) or without (−) HpyCH4IV. Methylation percentage was scored. The results are representative of four independent experiments. To further ascertain these results, we tested for the presence of HP1γ, MeCP2, and H3K9me3 at heterochromatin by analyzing their occupancy at major satellites in pMaRX-GLP cells. Compared to control cells, pMaRX-GLP cells showed reduced HP1γ and MeCP2 occupancy at major satellites, although H3K9me3 enrichment remained unchanged (Figure 4C). Given that MeCP2 binds methylated CG dinucleotides, we tested whether change in DNA methylation at major satellites contributed to reduction of MeCP2 binding at heterochromatin. Combined bisulfite restriction analysis (COBRA) revealed that there was no significant change in DNA methylation at major satellites in pMaRX-GLP cells compared to control cells (Figure 4D). MeCP2 is a GLP target gene The lower MeCP2 levels in pMaRX-GLP cells suggested that it may be a downstream target of GLP. To test that possibility, we first examined endogenous MeCP2 expression during differentiation. Consistent with previous studies (Brero et al., 2005; Agarwal et al., 2007), MeCP2 expression dramatically increased upon terminal differentiation and inversely correlated with GLP levels (Figure 5A). In pMaRX-GLP cells, MeCP2 transcripts were downregulated suggesting that GLP may transcriptionally repress MeCP2 expression (Figure 5B). We also analyzed the expression of another MBD family member MBD2, which has also been shown to induce heterochromatin clustering in skeletal muscle (Brero et al., 2005). Similar to MeCP2, MBD2 mRNA levels were downregulated in pMaRX-GLP cells (Figure 5B). Correspondingly, MeCP2 and MBD2 were upregulated in siGLP primary myoblasts compared to controls (Figure 5C). The MeCP2 promoter has been reported to be repressed by H3K9me2 marks (Heard et al., 2001; Abuhatzira et al., 2011). We therefore examined whether MeCP2 is indeed a GLP target gene by assessing GLP occupancy and H3K9me2 marks at the MeCP2 promoter by ChIP assays. In control cells, GLP occupancy and H3K9me2 marks were higher in undifferentiated cells (Day 0) and reduced during differentiation (Day 3) (Figure 5D). A clear enrichment of GLP and H3K9me2 marks were evident at the MeCP2 promoter in pMaRX-GLP cells relative to control in both undifferentiated and differentiated cells (Figure 5D). No change in GLP occupancy was seen at the β-actin promoter in pMaRX-GLP cells, which was analyzed as a control (Figure 5D). Figure 5 View largeDownload slide GLP represses MeCP2 expression. (A) Expression of MeCP2, GLP, and TroponinT in C2C12 cells was analyzed by western blot after 0, 1, and 3 days of differentiation. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 0 and 3 days and analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (C) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 0 and 3 days. siRNA and siGLP cells were analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (D) ChIP assays for GLP occupancy (left) and H3K9me2 enrichment (middle) at the MeCP2 promoter were performed in undifferentiated (Day 0) and differentiated (Day 3) pMaRX and pMaRX-GLP cells. GLP occupancy at the β-actin promoter (right) was tested as negative control. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. Figure 5 View largeDownload slide GLP represses MeCP2 expression. (A) Expression of MeCP2, GLP, and TroponinT in C2C12 cells was analyzed by western blot after 0, 1, and 3 days of differentiation. β-actin was used as loading control. (B) pMaRX and pMaRX-GLP cells were differentiated for 0 and 3 days and analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (C) Primary myoblasts were transfected with scrambled siRNA (Control) or GLP-specific siRNA (siGLP) for 72 h and differentiated for 0 and 3 days. siRNA and siGLP cells were analyzed for expression of MeCP2 and MBD2 by Q-PCR. Expression was normalized to Gapdh. The results are representative of two independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. (D) ChIP assays for GLP occupancy (left) and H3K9me2 enrichment (middle) at the MeCP2 promoter were performed in undifferentiated (Day 0) and differentiated (Day 3) pMaRX and pMaRX-GLP cells. GLP occupancy at the β-actin promoter (right) was tested as negative control. The results are representative of three independent experiments. Bars indicate the mean of Q-PCR triplicates in each experiment ±SD. Since MeCP2 is important for binding and aggregation of pericentric heterochromatin (Brero et al., 2005), we examined the impact of its depletion during myogenic differentiation. Cells were transfected with scrambled siRNA (Control) or MeCP2-specific siRNA (siMeCP2) (Figure 6A). Control and siMeCP2 cells were differentiated and immuno-stained with MeCP2 and HP1γ antibodies. Compared to control cells, siMeCP2 cells had a higher number and smaller size of chromocenters. In addition, similar to GLP-overexpressing cells, HP1γ localization to chromocenters was reduced indicating impaired heterochromatin clustering (Figure 6B and C). Given that MeCP2 is a GLP target gene, we investigated whether MeCP2 is downstream of GLP-mediated heterochromatin clustering by expressing exogenous MeCP2 in GLP-overexpressing cells (pMaRX-GLP + MeCP2) (Figure 6D). Control, pMaRX-GLP, and pMaRX-GLP + MeCP2 cells were differentiated and heterochromatin clustering was analyzed using anti-MeCP2 and anti-HP1γ antibodies. pMaRX-GLP cells showed a greater number and smaller size of chromocenters, and reduced HP1γ localization compared to controls (Figure 6E and F). Expression of exogenous MeCP2 in pMaRX-GLP + MeCP2 cells resulted in similar number and size of chromocenters and HP1γ localization as control cells (Figure 6E and F). Figure 6 View largeDownload slide MeCP2 regulates chromatin organization. (A) C2C12 myoblasts were transfected with scrambled siRNA (Control) or MeCP2-specific siRNA (siMeCP2) for 48 h. MeCP2 expression was analyzed by western blot. β-actin was used as loading control. (B) Control and siMeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siMeCP2 myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. (D) pMaRX-GLP cells were transfected with MeCP2 plasmid (pMaRX-GLP + MeCP2). pMaRX and pMaRX-GLP cells were transfected with pCS2 empty vector. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. (E) pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (F) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. Figure 6 View largeDownload slide MeCP2 regulates chromatin organization. (A) C2C12 myoblasts were transfected with scrambled siRNA (Control) or MeCP2-specific siRNA (siMeCP2) for 48 h. MeCP2 expression was analyzed by western blot. β-actin was used as loading control. (B) Control and siMeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control and siMeCP2 myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. (D) pMaRX-GLP cells were transfected with MeCP2 plasmid (pMaRX-GLP + MeCP2). pMaRX and pMaRX-GLP cells were transfected with pCS2 empty vector. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. (E) pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (F) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in pMaRX, pMaRX-GLP, and pMaRX-GLP + MeCP2 (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. Endogenous MeCP2 regulates heterochromatin organization and differentiation Given that exogenous MeCP2 restored heterochromatin clustering in GLP-overexpressing cells, we assessed whether inhibition of heterochromatin clustering by GLP is mediated via repression of endogenous MeCP2. Cells were transfected with scrambled siRNA (Control), siGLP, or both siGLP and siMeCP2 [double knockdown (DKD)]. Consistent with our previous data (Figure 3D and E), MeCP2 expression was elevated in siGLP cells compared to control cells (Figure 7A). We analyzed expression and localization of MeCP2 and HP1γ in differentiated control, siGLP and DKD cells with anti-MeCP2 and anti-HP1γ antibodies. Compared to control cells, siGLP cells showed lower number and larger size of chromocenters as seen by DAPI staining. HP1γ localization to heterochromatin was also increased (Figure 7B and C). On the other hand, while DKD cells exhibited no significant change in number of chromocenters compared to control cells (Figure 7C), there was a decrease in their size (Figure 7B) and HP1γ localization to heterochromatin was reduced (Figure 7C). Control, siGLP and DKD cells were also analyzed for myogenic differentiation. Interestingly, the enhanced myogenic differentiation in siGLP cells seen by MHC staining, myogenic index (Figure 7D), as well as Myogenin and TroponinT levels (Figure 7E), was reversed in DKD cells to a level similar to control cells. Figure 7 View largeDownload slide MeCP2 depletion restores chromatin organization in GLP knockdown cells. (A) C2C12 cells were transfected with siRNA (Control), siGLP, or both siGLP and siMeCP2 (DKD) for 48 h. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. Densitometric analysis of MeCP2 level is shown in the right panel. (B) Control, siGLP, and DKD cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control, siGLP, and DKD myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) Control, siGLP, and DKD cells were differentiated for 3 days and immuno-stained with anti-MHC antibody (red). Myogenic indices for Control, siGLP, and DKD myotubes were scored. Scale bar, 15 μm. For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control, siGLP, and DKD cells were scored. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (E) Control, siGLP, and DKD cells were differentiated for 3 days and expression of Myogenin and TroponinT was analyzed. β-actin was used as loading control. Figure 7 View largeDownload slide MeCP2 depletion restores chromatin organization in GLP knockdown cells. (A) C2C12 cells were transfected with siRNA (Control), siGLP, or both siGLP and siMeCP2 (DKD) for 48 h. Expression of GLP and MeCP2 was analyzed by western blot. β-actin was used as loading control. Densitometric analysis of MeCP2 level is shown in the right panel. (B) Control, siGLP, and DKD cells were differentiated for 6 days and immuno-stained with anti-MeCP2 (green) and anti-HP1γ (red) antibodies. Scale bar, 2 μm. (C) Box and whisker plots show the number of chromocenters per nucleus, average size of chromocenters, and Mander’s coefficient for HP1γ and DAPI in control, siGLP, and DKD myotubes (from left to right). For each experiment, >50 nuclei were counted in 10–20 random fields. The results are representative of two independent experiments. Error bars indicate the difference between lower quartile and lower extreme and difference between upper quartile and upper extreme, respectively. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (D) Control, siGLP, and DKD cells were differentiated for 3 days and immuno-stained with anti-MHC antibody (red). Myogenic indices for Control, siGLP, and DKD myotubes were scored. Scale bar, 15 μm. For each experiment, >1000 nuclei in 5 fields were counted and myogenic indices for Control, siGLP, and DKD cells were scored. The results are representative of two independent experiments. Bars indicate the mean of MHC+ cells in each experiment ±SD. Significance was calculated using a one-way ANOVA followed by Bonferroni’s multiple comparison test. (E) Control, siGLP, and DKD cells were differentiated for 3 days and expression of Myogenin and TroponinT was analyzed. β-actin was used as loading control. Discussion Chromatin organization is dynamically altered during myogenic differentiation. However, the mechanisms by which epigenetic alterations lead to distinct domains in chromatin that establish and maintain cell fates is unclear. In this study, we have identified a new role for GLP in transcriptional regulation of MeCP2. This in turn reduces HP1γ recruitment and heterochromatin remodeling during myogenic differentiation. We demonstrate that GLP inhibits skeletal muscle differentiation. Interestingly, the levels of G9a were unaltered in both siGLP and pMaRX-GLP cells indicating that GLP inhibits muscle differentiation independent of G9a as previously reported (Schaefer et al., 2009; Ohno et al., 2013). In line with our findings, a recent study also showed inhibition of myogenic differentiation upon GLP overexpression, although it promoted myogenic gene expression in myoblasts due to stabilization of MyoD (Battisti et al., 2016). GLP has been reported to interact with HP1γ and DNMT3a (Tachibana et al., 2005; Fritsch et al., 2010; Chang et al., 2011). However, its overexpression impairs rather than enhances heterochromatin aggregation during differentiation as evidenced by the number and size of DAPI stained chromocenters, as well as reduced HP1γ and MeCP2 localization at pericentric heterochromatin. H3K9me3 levels or DNA methylation at constitutive heterochromatin were not altered upon GLP overexpression. Since MeCP2 recruits HP1γ to heterochromatin during myogenic differentiation (Agarwal et al., 2007), our data suggests that reduced MeCP2 expression is central to the altered heterochromatin organization upon GLP overexpression. Indeed, knockdown of MeCP2 alone impacts HP1γ localization to heterochromatin mimicking the effect of GLP overexpression. Moreover, MeCP2 depletion in GLP-knockdown background reverted the enhanced HP1γ localization to pericentric heterochromatin, as well as myogenic differentiation. Our findings suggest that the previously established association of GLP with HP1γ (Tachibana et al., 2005; Chin et al., 2007; Fritsch et al., 2010) in euchromatin of undifferentiated myoblasts may be important to transcriptionally repress differentiation genes. Indeed, overexpression of HP1γ has been shown to repress Myogenin expression and curb MyoD transcriptional activity (Yahi et al., 2008). As GLP levels decline upon differentiation, MeCP2 expression is induced and recruits HP1γ to heterochromatin to repress proliferation genes in terminally differentiating myotubes. Consistent with this model, HP1γ is known to repress cell-cycle genes to maintain post-mitotic status of adult cardiac myocytes (Sdek et al., 2011) and cell-cycle genes are preferentially localized to pericentric heterochromatin upon skeletal muscle differentiation (Guasconi et al., 2010). Thus induction of MeCP2 during myogenic differentiation may be important for stabilization and maintenance of transcriptional programs during maturation of differentiated cells (Brero et al., 2005; Agarwal et al., 2007; Singleton et al., 2011; Becker et al., 2016). Consistent with previous studies showing that MeCP2 expression is transcriptionally repressed by H3K9me2 (Heard et al., 2001; Abuhatzira et al., 2011), we demonstrate that GLP represses MeCP2 expression. GLP occupancy and H3K9me2 enrichment is apparent at the MeCP2 promoter in proliferating myoblasts that correlate with inhibition of its expression. The downregulation of GLP during differentiation likely relieves this repression allowing MeCP2 to dramatically increase during myogenic differentiation (Brero et al., 2005; Agarwal et al., 2007). In line with this notion, in silico MeCP2 promoter analysis revealed that MyoD is able to bind to the MeCP2 promoter, indicating that MyoD could recruit GLP to the MeCP2 promoter and mediate repressive H3K9me2 marks to curb its expression (software Match 1.0, data not shown). Some studies have shown that MeCP2 does not impact myogenic differentiation likely due to redundancy with MBD2 (Brero et al., 2005; Becker et al., 2013), and has a non-cell autonomous role in skeletal muscle (Conti et al., 2015). Nonetheless, elevated MeCP2 expression has a significant detrimental impact on cardiac development in vivo (Alvarez-Saavedra et al., 2010), and peripheral MeCP2 knockout mice, which retain MeCP2 expression in neurons, exhibits reduced exercise capacity and aggravated fatigue (Ross et al., 2016). Moreover, a recent study using a Rett syndrome zebrafish model with MeCP2 loss-of-function showed severe downregulation of terminal myogenic differentiation genes, such as myosin heavy chain, TroponinT, and muscle-specific creatine kinase, at both embryonic and adult stages (Cortelazzo et al., 2017). Also, transcriptome analysis in mouse models with MeCP2 loss-of-function and gain-of-function in the hippocampus also indicates that TroponinT is positively regulated by MeCP2 (Chahrour et al., 2008). Given that depletion of MeCP2 in siGLP cells restored myogenic differentiation to control levels in our study, the difference between our analysis and MeCP2 knockout mice may be attributed to the different models employed in each study. Unlike analysis of terminally differentiated striated skeletal muscle tissue from adult mice, MeCP2 knockdown in myogenic precursor cells in vitro may recapitulate delayed differentiation during development. Analysis of missense Rett syndrome (RTT) mutations that lead to MeCP2 loss-of-function revealed that two-thirds of them show impaired heterochromatin clustering (Agarwal et al., 2011). It was also recently reported that a scaffolding protein SH2B1 also exerts pro-myogenic effect by modulating heterochromatin clustering during differentiation (Chen et al., 2017). Loss-of-function mutations in GLP gene underlie the neurological disorder Kleefstra syndrome (Kleefstra et al., 2006), whereas gain of MeCP2 leads to MeCP2 duplication syndrome (Amir et al., 1999; Ramocki et al., 2009). Both Kleefstra syndrome and MeCP2 duplication syndrome exhibit intellectual disability, muscle hypotonia, and dysmorphism as common symptoms. Notably, clinical screening of Kleefstra syndrome patients revealed a de novo missense mutation of MBD5, a methyl-binding protein similar to MeCP2 (Kleefstra et al., 2012). These studies demonstrate genetic interactions between GLP and MeCP2/MBD5. Thus, the GLP–MeCP2 regulatory axis identified in this study may be relevant in both disorders as well as several neuropathologies characterized by loss of MeCP2 expression. Materials and methods Cell culture and differentiation assays C2C12 cells were cultured in high glucose Dulbecco’s Modified Eagle’s Medium (DMEM; Sigma) supplemented with 20% fetal bovine serum (FBS; Hyclone). To induce differentiation, cells were cultured to 80%–90% confluency in growth media, and then cultured in differentiation medium [DMEM with 2% horse serum (Invitrogen)]. HEK293T and Phoenix cell lines were cultured in DMEM supplemented with 10% FBS. Isolation of primary myoblasts has been previously described (Sun et al., 2007; Rao et al., 2016). Differentiation was assessed by MHC staining as previously described (Ling et al., 2012). Briefly, differentiated cells were stained with mouse monoclonal anti-MHC (Sigma) antibody, and detected with secondary antibody conjugated to Alexa Fluor (Alexa Fluor 488; 568; Invitrogen). Nuclei were stained with the fluorochrome 4′,6-diamino-2-phenylindole (DAPI). Images were captured under Olympus DP72 fluorescence microscope. More than 1000 nuclei from five different fields were counted. Myogenic index was calculated by quantifying the ratio of nuclei in myosin heavy chain-positive myotubes to total nuclei. Plasmids, retroviral infection, and siRNA Plasmids FLAG-GLP and FLAG-ΔANKΔSET that lacks ANK repeats and SET domain of GLP have been previously described (Takahashi et al., 2012). To generate stable cell lines, C2C12 cells were transfected with FLAG-GLP or FLAG-ΔANKΔSET and pBABE vector in a 9:1 ratio using Lipofectamine 2000 reagent (Invitrogen). Mouse MeCP2 plasmid was purchased from ORIGENE (MR226839). Cells were selected with puromycin for 48 h and analyzed in differentiation experiments. Alternatively, pMaRX or pMaRX-GLP retroviral constructs were transfected into Phoenix packaging cells. Retroviral supernatants were used to transduce undifferentiated C2C12 cells with 8 μg/ml polybrene. Cells were subsequently subjected to selection with medium containing 2 μg/ml of puromycin for 48 h. Knockdown experiments were performed using 50 nM GLP-specific siRNA (siGLP; ON-TARGETplusSMARTpool; NM_001012518) or 50 nM MeCP2-specific siRNA (siMeCP2; ON-TARGETplusSMARTpool; NM_001081979) from Dharmacon using Lipofectamine RNAiMAX (Invitrogen) as described in the manufacturer’s protocol. Control cells were transfected with non-specific scrambled siRNA (Dharmacon, ON-TARGETplus, Non-Targeting Pool). Transfection of siRNA was done in growth medium for 48 h. RNA isolation and quantitative real-time polymerase chain reaction (Q-PCR) Total RNA was extracted using Trizol (Invitrogen). RNA was quantified using Nanodrop. Messenger RNA (mRNA) was converted to a single-stranded complementary DNA (cDNA) using iScript cDNA Synthesis Kit (Bio-Rad). Quantitative real-time polymerase chain reaction (Q-PCR) was performed using Lightcycler 480 SYBR Green 1 Master Kit (Roche). CT values of samples were normalized to internal control Gapdh to obtain delta CT (ΔCT). Relative expression was calculated by 2−ΔCT equation. Primer sequences for GLP (Ohno et al., 2013), Myogenin (Ling et al., 2012), MeCP2 (Murgatroyd et al., 2009), and MBD2 (Song et al., 2014) have been previously described. For Myh1, the following set of primers were used: 5′-AACAGCAGCGGCTGATCAAT-3′ and 5′-GCTGCCTCTTCAGCTCCTCA-3′. Chromatin immuno-precipitation (ChIP) ChIP assays were done as previously described (Ling et al., 2012). Briefly, 1 × 106 cells were cross-linked with 1% formaldehyde for 10 min at 37°C. Cells were sonicated using Bioruptor (Diagenode). ChIP was carried out according to the kit protocol (Millipore). Immuno-precipitates were reverse-cross-linked and DNA was extracted using phenol–chloroform–isoamylalcohol (Sigma). Q-PCR was performed as described above. DNA harvested from 10% input was used as control. Relative enrichment was calculated using 2−ΔCT equation. Following antibodies were used for ChIP assays: mouse monoclonal ChIP-grade anti-GLP (Abcam, ab41969), anti-H3K9me2 (Millipore, 17-681), anti-HP1γ (Millipore, MAB3450), anti-MeCP2 (Abcam, ab07013), anti-H3K9me3 (Abcam ab8898). Primers used for ChIP assays of MeCP2 promoters (Heard et al., 2001), major satellites (Terranova et al., 2005), and β-actin (Rao et al., 2016) have been described previously. Cell proliferation assay Proliferation was measured using BrdU incorporation assays (Azmi et al., 2004). Cells were pulsed with 10 μM BrdU for 30 min. Cells were fixed and stained with anti-BrdU antibody according to manufacturer’s protocol (Roche). More than 1000 nuclei from five different fields were counted. BrdU positivity was calculated. Immunofluorescence imaging and analysis For confocal imaging, samples were permeabilized with 0.1% Triton X-100 for 1 h at room temperature and blocked overnight in 10% horse serum in PBS. Samples were probed with primary antibodies overnight at 4°C in blocking buffer with 0.1 M glycine (Brocher et al., 2010). Primary antibodies used for immunofluorescence are anti-MeCP2 (Cell Signaling, 1:200), anti-HP1γ (Millipore, MAB3450, 1:350), and anti-H3K9me3 (Abcam, ab8988, 1:1000). Images were captured under FluoView FV1000 confocal fluorescence microscope (Olympus) with optical section thickness 1μm. More than 50 nuclei for each sample were counted for three independent experiments, unless otherwise stated. Number and size of DAPI positive chromocenters were quantified using ‘cell counter’ and ‘analyze particle’ tools in ImageJ software as described previously (Novo et al., 2016). Mander’s coefficient was used to measure the co-localization of DAPI and HP1γ using ‘JACoP’ tool in ImageJ as described previously (Bolte and Cordelières, 2006). Expression of MeCP2 was quantified using the total corrected cellular fluorescence (TCCF) equation as described previously (McCloy et al., 2014). Box and whisker plots were plotted using Microsoft Excel software. Combined bisulfite restriction assay (COBRA) COBRA assay was performed as previously described (Xiong and Laird, 1997). Briefly, genomic DNA (gDNA) was extracted using GeneJET Genomic DNA purification Kit (Thermo Fisher Scientific). gDNA was subject to bisulfite treatment using EZ DNA Methylation-Gold Kit (Zymo Research). Eluted DNA was PCR-amplified using the following set of primers: 5′-GGAATATGGCAAGAAAACTGAAAATCATGG-3′ and 5′-CCATATTCCAGGTCCTTCAGTGTGCATTTC-3′. The PCR reaction was carried out for 35 cycles of 95°C for 30 sec, 58°C for 60 sec, and 68°C for 20 sec. The purified PCR products were digested with restriction enzyme HpyCH4IV, which has a recognition sequence (ACGT) that should be destroyed by the bisulfite conversion. Restriction digestion reactions were performed at 37°C for 30 min. The digested PCR products were separated on 1% polyacrylamide gel and were visualized with ChemiDoc imaging system (Bio-rad). DNA methylation was calculated by quantifying ratio of methylated DNA and sum of un- and methylated DNA. Immunoblotting Protein lysate was harvested in radioimmunoprecipitation assay (RIPA) lysis buffer [50 mM NaCl, 50 mM Tris-HCl pH 7.0, 1 mM EDTA, 1% NP-40, 0.1% sodium deoxycholate, protease inhibitor cocktail (Roche)]. Histones were isolated using RIPA lysis buffer supplemented with 2% sodium dodecyl sulfate (SDS). The following primary antibodies were used for immunoblotting: mouse monoclonal anti-GLP (Abcam, ab41969, 1:2000), rabbit polyclonal anti-Myogenin (Santa Cruz Biotechnology, sc-576, 1:500), mouse monoclonal anti-Troponin-T (Sigma, T6277, 1:2000), rabbit polyclonal anti-MyoD (Santa Cruz Biotechnology, sc-304, 1:500), mouse monoclonal anti-Flag (Sigma, F3165, 1:500), rabbit polyclonal anti-MeCP2 (Cell Signaling, 3456, 1:2000), anti-HP1γ (Millipore, MAB3450, 1:5000), anti-H3K9me3 (Abcam, ab8988, 1:2000), and mouse monoclonal anti-β-actin (Sigma, A1978, 1:10000). Statistical analysis Significance was calculated using student’s t test (two-sided) and P values <0.05 were considered to be statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001). Bars represent mean ± standard deviation (SD) unless specified otherwise. For quantification significance for data with more than two groups, a one-way ANOVA followed by Bonferroni’s multiple comparison test was performed and P values <0.05 were considered to be statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001). Acknowledgements We thank Dr Eiji Hara (Osaka University) for the gift of various GLP plasmid constructs, Ms Lee Shu Yin (Confocal Microscopy Unit, National University of Singapore) for her help with confocal image analysis, and Dr Neerja Karnani (National University of Singapore) for valuable discussions. Funding This work was supported by a grant from the National Medical Research Council (NMRC/CBRG/0105/2016 to R.T.). M.H.C. and J.R.O. are supported by a NUS Graduate School for Integrative Sciences and Engineering Scholarship. Conflict of interest: none declared. References Abuhatzira, L., Shamir, A., Schones, D.E., et al.  . ( 2011). The chromatin-binding protein HMGN1 regulates the expression of methyl CpG-binding protein 2 (MECP2) and affects the behavior of mice. J. Biol. Chem.  286, 42051– 42062. Google Scholar CrossRef Search ADS PubMed  Agarwal, N., Becker, A., Jost, K.L., et al.  . ( 2011). MeCP2 Rett mutations affect large scale chromatin organization. Hum. Mol. Genet.  20, 4187– 4195. Google Scholar CrossRef Search ADS PubMed  Agarwal, N., Hardt, T., Brero, A., et al.  . ( 2007). MeCP2 interacts with HP1 and modulates its heterochromatin association during myogenic differentiation. Nucleic Acids Res.  35, 5402– 5408. Google Scholar CrossRef Search ADS PubMed  Alvarez-Saavedra, M., Carrasco, L., Sura-Trueba, S., et al.  . ( 2010). Elevated expression of MeCP2 in cardiac and skeletal tissues is detrimental for normal development. Hum. Mol. Genet.  19, 2177– 2190. Google Scholar CrossRef Search ADS PubMed  Amir, R.E., Van den Veyver, I.B., Wan, M., et al.  . ( 1999). Rett syndrome is caused by mutations in X-linked MECP2, encoding methyl-CpG-binding protein 2. Nat. Genet.  23, 185– 188. Google Scholar CrossRef Search ADS PubMed  Azmi, S., Ozog, A., and Taneja, R. ( 2004). Sharp-1/DEC2 inhibits skeletal muscle differentiation through repression of myogenic transcription factors. J. Biol. Chem.  279, 52643– 52652. Google Scholar CrossRef Search ADS PubMed  Battisti, V., Pontis, J., Boyarchuk, E., et al.  . ( 2016). Unexpected distinct roles of the related histone H3 lysine 9 methyltransferases G9a and G9a-like protein in myoblasts. J. Mol. Biol.  428, 2329– 2343. Google Scholar CrossRef Search ADS PubMed  Becker, A., Allmann, L., Hofstätter, M., et al.  . ( 2013). Direct homo- and hetero-interactions of MeCP2 and MBD2. PLoS One  8, e53730. Google Scholar CrossRef Search ADS PubMed  Becker, A., Zhang, P., Allmann, L., et al.  . ( 2016). Poly(ADP-ribosyl)ation of methyl CpG binding domain protein 2 regulates chromatin structure. J. Biol. Chem.  291, 4873– 4881. Google Scholar CrossRef Search ADS PubMed  Bolte, S., and Cordelières, F.P. ( 2006). A guided tour into subcellular colocalization analysis in light microscopy. J. Microsc.  224, 213– 232. Google Scholar CrossRef Search ADS PubMed  Brero, A., Easwaran, H.P., Nowak, D., et al.  . ( 2005). Methyl CpG-binding proteins induce large-scale chromatin reorganization during terminal differentiation. J. Cell Biol.  169, 733– 743. Google Scholar CrossRef Search ADS PubMed  Brocher, J., Vogel, B., and Hock, R. ( 2010). HMGA1 down-regulation is crucial for chromatin composition and a gene expression profile permitting myogenic differentiation. BMC Cell Biol.  11, 64. Google Scholar CrossRef Search ADS PubMed  Chahrour, M., Jung, S.Y., Shaw, C., et al.  . ( 2008). MeCP2, a key contributor to neurological disease, activates and represses transcription. Science  320, 1224– 1229. Google Scholar CrossRef Search ADS PubMed  Chang, Y., Sun, L., Kokura, K., et al.  . ( 2011). MPP8 mediates the interactions between DNA methyltransferase Dnmt3a and H3K9 methyltransferase GLP/G9a. Nat. Commun.  2, 533. Google Scholar CrossRef Search ADS PubMed  Chen, K.-W., Chang, Y.-J., Yeh, C.-M., et al.  . ( 2017). SH2B1 modulates chromatin state and MyoD occupancy to enhance expressions of myogenic genes. Biochim. Biophys. Acta  1860, 270– 281. Google Scholar CrossRef Search ADS PubMed  Chin, H.G., Estève, P.-O., Pradhan, M., et al.  . ( 2007). Automethylation of G9a and its implication in wider substrate specificity and HP1 binding. Nucleic Acids Res.  35, 7313– 7323. Google Scholar CrossRef Search ADS PubMed  Conti, V., Gandaglia, A., Galli, F., et al.  . ( 2015). MeCP2 affects skeletal muscle growth and morphology through non cell-autonomous mechanisms. PLoS One  10, e0130183. Google Scholar CrossRef Search ADS PubMed  Cortelazzo, A., Pietri, T., De Felice, C., et al.  . ( 2017). Proteomic analysis of the Rett syndrome experimental model mecp2(Q63X) mutant zebrafish. J. Proteomics  154, 128– 133. Google Scholar CrossRef Search ADS PubMed  Fritsch, L., Robin, P., Mathieu, J.R.R., et al.  . ( 2010). A subset of the histone H3 lysine 9 methyltransferases Suv39h1, G9a, GLP, and SETDB1 participate in a multimeric complex. Mol. Cell  37, 46– 56. Google Scholar CrossRef Search ADS PubMed  Fuks, F., Hurd, P.J., Deplus, R., et al.  . ( 2003a). The DNA methyltransferases associate with HP1 and the SUV39H1 histone methyltransferase. Nucleic Acids Res.  31, 2305– 2312. Google Scholar CrossRef Search ADS PubMed  Fuks, F., Hurd, P.J., Wolf, D., et al.  . ( 2003b). The methyl-CpG-binding protein MeCP2 links DNA methylation to histone methylation. J. Biol. Chem.  278, 4035– 4040. Google Scholar CrossRef Search ADS PubMed  Guasconi, V., Pritchard, L.-L., Fritsch, L., et al.  . ( 2010). Preferential association of irreversibly silenced E2F-target genes with pericentromeric heterochromatin in differentiated muscle cells. Epigenetics  5, 704– 709. Google Scholar CrossRef Search ADS PubMed  Heard, E., Rougeulle, C., Arnaud, D., et al.  . ( 2001). Methylation of histone H3 at Lys-9 is an early mark on the X chromosome during X inactivation. Cell  107, 727– 738. Google Scholar CrossRef Search ADS PubMed  Jung, E.-S., Sim, Y.-J., Jeong, H.-S., et al.  . ( 2015). Jmjd2C increases MyoD transcriptional activity through inhibiting G9a-dependent MyoD degradation. Biochim. Biophys. Acta  1849, 1081– 1094. Google Scholar CrossRef Search ADS PubMed  Kleefstra, T., Brunner, H.G., Amiel, J., et al.  . ( 2006). Loss-of-function mutations in euchromatin histone methyl transferase 1 (EHMT1) cause the 9q34 subtelomeric deletion syndrome. Am. J. Hum. Genet.  79, 370– 377. Google Scholar CrossRef Search ADS PubMed  Kleefstra, T., Kramer, J.M., Neveling, K., et al.  . ( 2012). Disruption of an EHMT1-associated chromatin-modification module causes intellectual disability. Am. J. Hum. Genet.  91, 73– 82. Google Scholar CrossRef Search ADS PubMed  Lehnertz, B., Ueda, Y., Derijck, A.A.H.A., et al.  . ( 2003). Suv39h-mediated histone H3 Lysine 9 methylation directs DNA methylation to major satellite repeats at pericentric heterochromatin. Curr. Biol.  13, 1192– 1200. Google Scholar CrossRef Search ADS PubMed  Ling, B.M.T., Bharathy, N., Chung, T.-K., et al.  . ( 2012). Lysine methyltransferase G9a methylates the transcription factor MyoD and regulates skeletal muscle differentiation. Proc. Natl Acad. Sci. USA  109, 841– 846. Google Scholar CrossRef Search ADS   Liu, N., Zhang, Z., Wu, H., et al.  . ( 2015). Recognition of H3K9 methylation by GLP is required for efficient establishment of H3K9 methylation, rapid target gene repression, and mouse viability. Genes Dev.  29, 379– 393. Google Scholar CrossRef Search ADS PubMed  Luo, S.-W., Zhang, C., Zhang, B., et al.  . ( 2009). Regulation of heterochromatin remodelling and myogenin expression during muscle differentiation by FAK interaction with MBD2. EMBO J.  28, 2568– 2582. Google Scholar CrossRef Search ADS PubMed  McCloy, R.A., Rogers, S., Caldon, C.E., et al.  . ( 2014). Partial inhibition of Cdk1 in G2 phase overrides the SAC and decouples mitotic events. Cell Cycle  13, 1400– 1412. Google Scholar CrossRef Search ADS PubMed  Murgatroyd, C., Patchev, A.V., Wu, Y., et al.  . ( 2009). Dynamic DNA methylation programs persistent adverse effects of early-life stress. Nat. Neurosci.  12, 1559– 1566. Google Scholar CrossRef Search ADS PubMed  Novo, C.L., Tang, C., Ahmed, K., et al.  . ( 2016). The pluripotency factor Nanog regulates pericentromeric heterochromatin organization in mouse embryonic stem cells. Genes Dev.  30, 1101– 1115. Google Scholar CrossRef Search ADS PubMed  Ohno, H., Shinoda, K., Ohyama, K., et al.  . ( 2013). EHMT1 controls brown adipose cell fate and thermogenesis through the PRDM16 complex. Nature  504, 163– 167. Google Scholar CrossRef Search ADS PubMed  Ow, J.R., Palanichamy Kala, M., Rao, V.K., et al.  . ( 2016). G9a inhibits MEF2C activity to control sarcomere assembly. Sci. Rep.  6, 34163. Google Scholar CrossRef Search ADS PubMed  Ramocki, M.B., Peters, S.U., Tavyev, Y.J., et al.  . ( 2009). Autism and other neuropsychiatric symptoms are prevalent in individuals with MeCP2 duplication syndrome. Ann. Neurol.  66, 771– 782. Google Scholar CrossRef Search ADS PubMed  Rao, V.K., Ow, J.R., Shankar, S.R., et al.  . ( 2016). G9a promotes proliferation and inhibits cell cycle exit during myogenic differentiation. Nucleic Acids Res.  44, 8129– 8143. Google Scholar CrossRef Search ADS PubMed  Ross, P.D., Guy, J., Selfridge, J., et al.  . ( 2016). Exclusive expression of MeCP2 in the nervous system distinguishes between brain and peripheral Rett syndrome-like phenotypes. Hum. Mol. Genet.  25, 4389– 4404. Google Scholar PubMed  Schaefer, A., Sampath, S.C., Intrator, A., et al.  . ( 2009). Control of cognition and adaptive behavior by the GLP/G9a epigenetic suppressor complex. Neuron  64, 678– 691. Google Scholar CrossRef Search ADS PubMed  Sdek, P., Oyama, K., Angelis, E., et al.  . ( 2013). Epigenetic regulation of myogenic gene expression by heterochromatin protein 1α. PLoS One  8, e58319. Google Scholar CrossRef Search ADS PubMed  Sdek, P., Zhao, P., Wang, Y., et al.  . ( 2011). Rb and p130 control cell cycle gene silencing to maintain the postmitotic phenotype in cardiac myocytes. J. Cell Biol.  194, 407– 423. Google Scholar CrossRef Search ADS PubMed  Singleton, M.K., Gonzales, M.L., Leung, K.N., et al.  . ( 2011). MeCP2 is required for global heterochromatic and nucleolar changes during activity-dependent neuronal maturation. Neurobiol. Dis.  43, 190– 200. Google Scholar CrossRef Search ADS PubMed  Song, C., Feodorova, Y., Guy, J., et al.  . ( 2014). DNA methylation reader MECP2: cell type- and differentiation stage-specific protein distribution. Epigenetics Chromatin  7, 17. Google Scholar CrossRef Search ADS PubMed  Sun, H., Li, L., Vercherat, C., et al.  . ( 2007). Stra13 regulates satellite cell activation by antagonizing Notch signaling. J. Cell Biol.  177, 647– 657. Google Scholar CrossRef Search ADS PubMed  Tachibana, M., Ueda, J., Fukuda, M., et al.  . ( 2005). Histone methyltransferases G9a and GLP form heteromeric complexes and are both crucial for methylation of euchromatin at H3-K9. Genes Dev.  19, 815– 826. Google Scholar CrossRef Search ADS PubMed  Takahashi, A., Imai, Y., Yamakoshi, K., et al.  . ( 2012). DNA damage signaling triggers degradation of histone methyltransferases through APC/C(Cdh1) in senescent cells. Mol. Cell  45, 123– 131. Google Scholar CrossRef Search ADS PubMed  Terranova, R., Sauer, S., Merkenschlager, M., et al.  . ( 2005). The reorganisation of constitutive heterochromatin in differentiating muscle requires HDAC activity. Exp. Cell Res.  310, 344– 356. Google Scholar CrossRef Search ADS PubMed  Trojer, P., and Reinberg, D. ( 2007). Facultative heterochromatin: is there a distinctive molecular signature? Mol. Cell  28, 1– 13. Google Scholar CrossRef Search ADS PubMed  Xiong, Z., and Laird, P.W. ( 1997). COBRA: a sensitive and quantitative DNA methylation assay. Nucleic Acids Res.  25, 2532– 2534. Google Scholar CrossRef Search ADS PubMed  Yahi, H., Fritsch, L., Philipot, O., et al.  . ( 2008). Differential cooperation between heterochromatin protein HP1 isoforms and MyoD in myoblasts. J. Biol. Chem.  283, 23692– 23700. Google Scholar CrossRef Search ADS PubMed  © The Author (2017). Published by Oxford University Press on behalf of Journal of Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Molecular Cell BiologyOxford University Press

Published: Oct 3, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off