Regulation of HDAC11 gene expression in early myogenic differentiation

Cell culture

C2C12 murine myoblasts (ATCC) were maintained in growth medium (GM), Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), in the incubator at 37 °C with 5% CO₂. Myogenic differentiation was induced with 80% confluent myoblasts cultures using differentiation medium (DM), DMEM supplemented with 2% horse serum. Bexarotene was purchased from the LC Laboratories.

Gene expression and histone acetylation profiles

The high degree correlation of the RNA-seq biological replicates have been described previously (Li et al., 2022; Khilji et al., 2021) with about 30 million mapped reads at 96% of mapping rate. The statistical power of the experimental design, calculated in RNASeqPower is 0.7. To determine differential gene expression between experimental conditions, a fold change greater than ±1.5-fold and below a false discover rate (FDR) of 5% were used to group the genes. The average enrichment of histone acetylation at the promoters of gene groups were visualized with ngs.plot (Shen et al., 2014). The ngs.plot calculates the coverage vectors for each query region based on specified alignment files, followed by normalization and transformation to generate an average profile with the number of reads normalized to the total number of mapped reads (in millions) in the dataset, in 20 bp bins within a 2 kb region, and centered at the transcription start sites (TSS). Integrative Genomics Viewer (IGV) was used for data browsing and representative snapshots. The heatmaps displaying differentially expressed genes were generated using the reshape2 and ggplot2 packages (Wickham, 2007, 2016), based on the FPKM values (Fragments Per Kilobase of transcript per Million mapped reads) or log2-transformed z-scores.

Reverse transcription qPCR analysis

As previously described, high-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Waltham, MA, USA) was used for reverse transcription and SYBR® Green PCR Master Mix and HotStarTaq DNA polymerase (Qiagen, Hilden, Germany) for the quantitative PCR on a CFX96 Touch Real-Time PCR Detection System (BioRad, Hercules, CA, USA) (Hamed et al., 2017). Quantification of the targets, normalized to endogenous reference and relative to a calibrator control, was calculated using the formula 2^–ΔΔCT as described (Hamed et al., 2017). Tbp primers for RT-qPCR have been described previously (Hamed, Chen & Li, 2022). Hdac11 primers for analysis were as follows:

Hdac11-F: 5′-TTACAACCGCCACATCTACC; R: 5′-GACATTCCTCTCCACCTTCTC.

Quantitative ChIP analysis

C2C12 myoblasts were crosslinked and sonicated followed by chromatin immunoprecipitation as previously described (Hamed et al., 2013). Antibodies against p300 and MyoD were purchased from Santa Cruz (sc-584x and sc-32758x). The immunoprecipitants were captured by Dynabeads protein-A, washed and eluted according to manufacturer’s protocol (Invitrogen, Waltham, MA, USA) (Hamed, Chen & Li, 2022). Chromatin DNA was reverse crosslinked at 65 °C for overnight, purified with the DNA purification kit (Qiagen, Hilden, Germany) and amplified with SYBR® Green and HotStarTaq DNA polymerase (Qiagen, Hilden, Germany) on a CFX96 or CFX384 Touch Real-Time PCR Detection System (BioRad, Hercules, CA, USA) (Hamed, Chen & Li, 2022). Each sample was amplified in triplicate PCR reactions. Purified input DNA was used to generate a standard curve for the PCR amplification of each immunoprecipitation (Hamed, Chen & Li, 2022). The abundance of immunoprecipitated DNA was quantified as the percentage relative to the input chromatin DNA (Hamed, Chen & Li, 2022). Primer pairs used for the amplification were as follows:

Hdac11: F5′-GGGTGTAGGGGGAATGGAGA; R5′-TGCCTTGAACCTGTTTCCCT. Chromosome 15 gene desert, a negative control locus: F5′-TCCTCCCCATCTGTGTCATC; R5′-GGATCCATCACCATCAATAACC.

Statistical analysis

All statistical analyses were performed using R Core Team (2022) or Microsoft Excel. Normally distributed data were analyzed by a two-tailed Student’s t-test and data are presented as the mean ± SEM. Each experiment was repeated at least three times. Non-normally distributed data were analyzed by non-parametric two-sided Wilcoxon rank sum tests. A p < 0.05 was considered statistically significant. Statistical details and number of replicates are indicated in the figures.

Epigenetic feature associated with gene promoters in early myogenic differentiation

We have previously established a myoblast chromatin state model and defined residue-specific histone acetylation associated with active and poised enhancers in early myoblast differentiation (Khilji et al., 2018; Hamed et al., 2017). To survey the epigenetic features of myogenic promoters, we first profiled differential transcriptome which may contribute to myoblast fate decision by using our previously deposited RNA-seq data (GSE94560). Cuffdiff analysis discerned that 3,159 genes were significantly upregulated by over 1.5-fold in differentiating myoblasts in the first 24 h of differentiation, when compared to proliferating myoblasts, while 3,936 genes downregulated (Fig. 1A). As such, differential expression was observed on less than 30% of transcripts in differentiating myoblasts (Fig. 1B), revealing that a rapid change in myoblast fate transition arises from subsets of genes at onset of myoblast differentiation.

A previous study has reported that histone acetylation globally decreases at gene promoters during terminal myoblast differentiation (Blum et al., 2012). It is known that H3K9, H3K18, and H3K27 acetylation are found near transcription start sites (TSS), whereas H4K8 acetylation is associated to promoters and transcribed regions of active genes (Li, Carey & Workman, 2007; Di Cerbo & Schneider, 2013; Wang et al., 2008). We thus profiled residue-specific histone acetylation associated with promoters of differentially expressed genes in early myoblast differentiation by using our previously deposited histone acetylation ChIP-seq datasets (GSE94558) in matching conditions of the RNA-seq. As shown in Figs. 1C and 1D, at the promoters of downregulated genes, H4K8ac, H3K9ac and H3K18ac markedly decreased by 24 h of differentiation. Interestingly, change in H3K27ac signals was less evident, while the decrease in H3K9ac level was most pronounced (Figs. 1C and 1D). At the promoters of constitutively expressed genes, the levels of H4K8ac, H3K9ac, H3K18ac and H3K27ac were comparable between proliferating and differentiating myoblasts (Figs. 1C and 1D). However, at the promoters of upregulated genes, H3K18ac and H3K27ac levels were significantly augmented, whereas H4K8ac and H3K9ac signals remained similar in early differentiation (Figs. 1C and 1D). Taken together, our analysis indicates that changes in residue-specific histone acetylation at the promoters of differentially expressed genes may be intrinsic to myogenic activation at the onset of myoblast differentiation.

Regulation of Hdac11 by p300 in early myogenic differentiation

Histone deacetylases (HDACs) inhibit transcription by antagonizing HAT function through the removal of acetyl groups from the histones, which is essential for change of gene program in stem cell fate decision. Through RNA-seq analysis, we found that compared to other HDACs, the abundance of Hdac11 mRNA was very low in proliferating myoblasts (Fig. 2A). While some HDACs were differentially expressed in early myoblast differentiation, the upregulation of Hdac11 mRNA was most pronounced and significant, about 12-fold, in the first 24 h of differentiation compared to undifferentiated controls (Figs. 2A and 2B). To understand the control of HDAC11 function, we next sought to determine the molecular mechanisms by which Hdac11 gene expression is governed in early myoblast differentiation.

Since p300 is a critical HAT required for skeletal myogenesis (Polesskaya, 2001; Roth et al., 2003), we examined p300 occupancy and concurrence of histone acetylation signature at the Hdac11 locus following 24 h of myoblast differentiation, using our deposited p300 and histone acetylation ChIP-seq datasets (Khilji et al., 2018, 2020). We also utilized our established model of myoblast chromatin states which is defined by large-scale analyses of co-occurrence of different histone marks (Hamed et al., 2017). As shown in Fig. 2C, the Hdac11 promoter was classified as “active” in proliferating myoblasts based on the chromatin state model. Interestingly, the region spanning the active promoter was denoted as poised enhancer based on the presence of H3K4me1 signal but lack of H3K27ac (Creyghton et al., 2010; Hamed et al., 2017; Rada-Iglesias et al., 2011). At the Hdac11 promoter, a moderate enrichment of H4K8ac, H3K9ac, H3K18ac signals in proliferating myoblasts was observed on the genome browser view of ChIP-seq read signals (Fig. 2C). However, at the poised enhancer, about 400 base pair upstream of the promoter, a distinct p300 occupancy as well as H3K27 acetylation was detected in myoblasts differentiated for 24 h (Fig. 2C). Moreover, the H4K8ac, H3K9ac, H3K18ac peaks were augmented, broadened, and overlapped with the p300 peak in the differentiating myoblasts (Fig. 2C).

Since MyoD is a master regulator of skeletal myogenesis (Tapscott, 2005), and can recruit the HAT p300 to instigate histone acetylation (Puri et al., 1997; Hamed et al., 2013; Sartorelli et al., 1997), we also examined the binding of MyoD to the Hdac11 locus using a public available dataset (Yue et al., 2014). As shown in the genome browser, a prominent MyoD peak was found overlapping with the p300 read signal at the upstream of Hdac11 promoter in myoblast differentiated for 24 h (Fig. 2C). Taken together, our bioinformatic analyses suggest that p300 and MyoD may be involved in the regulation of Hdac11 gene expression in early myoblasts differentiation.

To exam the involvement of p300 and MyoD in the control of Hdac11 gene expression, we designed a set of primers spanning the summits of p300 and MyoD read signals at the poised enhancer for quantitative ChIP analysis. As shown in Figs. 2D and 2E, p300 occupancy and MyoD binding at the Hdac11 locus were both enriched by about 5-fold, following 24 h of differentiation as compared to proliferating myoblasts. On the other hand, the signals at a control locus were neglectable (Figs. 2D and 2E).

To further determine the requirement of p300 for the control of Hdac11 expression, we next used our established p300 shRNA knockdown myoblasts (Chen et al., 2015) to assess the levels of Hdac11 mRNA in early myoblast differentiation. As shown in Fig. 2F, knockdown of endogenous p300 negatively impacted Hdac11 gene expression, resulting in a significant reduction in Hdac11 mRNA level in myoblasts differentiated for 24 h, when compared to corresponding nonsilencing shRNA control as determined by quantitative RT-PCR analysis. Taken together, our data suggest that p300 function is important for the upregulation of Hdac11 expression at the onset of myogenic differentiation, possibly mediated by MyoD to activate the poised enhancer immediately upstream of the promoter.

Effect of RXR signaling on Hdac11 gene expression in early myoblast differentiation

RXR selective signaling enhances myogenic differentiation through a direct upregulation of MyoD gene expression as a transcription factor (Hamed et al., 2017), which may consequently impact Hdac11 gene expression. As such, we profiled in detail the expression of HDACs using our deposited RNA-seq datasets on myoblasts differentiated in the presence or absence of an agonist of RXR, bexarotene. As shown in the heatmap, Hdac11 mRNA was markedly upregulated as early as 12 h of differentiation (Figs. 2A and 3A). Interestingly, at 24 h, treatment with bexarotene significantly enhanced the expression of Hdac11, but not that of other HDACs (Fig. 3A). As shown in Fig. 3B, bexarotene augmented the level of Hdac11 mRNA by additional three-fold, compared to untreated myoblasts, determined by the quantitative RT-PCR analysis. Nevertheless, profiles of our deposited RXR ChIP-seq dataset (GSM2478303-2478305, at 24 h of C2C12 differentiation) did not exhibit specific RXR signal enrichment at the Hdac11 locus. Therefore, the positive effect of RXR-selective signaling on Hdac11 expression is possibly mediated through an indirect mechanism, directly enhancing MyoD gene expression (Hamed et al., 2017), which in turn positively regulates the Hdac11 locus.

Since rexinoid-responsive myogenic expression is mediated through MyoD that is known to interact with HAT p300 (Khilji et al., 2018), we examined p300 and histone acetylation signature at the Hdac11 locus in the context of RXR signaling. The Genome browser view of ChIP-seq read signals revealed that in differentiating myoblasts, the enrichment of p300, H4K8ac, H3K9ac, H3K18ac and H3K27ac read signals at the Hdac11 promoter and putative enhancer were further augmented following 24 h of differentiation in the presence of bexarotene (Fig. 3C). Next, we used quantitative ChIP analysis to ascertain the involvement of p300 and MyoD in signal enhanced Hdac11 gene expression. As shown in Figs. 3D and 3E, p300 occupancy and MyoD binding at the Hdac11 locus were both further augmented by an additional 2-fold in response to bexarotene treatment when compared to untreated controls. Taken together, our data suggests that p300 and MyoD are involved in rexinoid-augmented Hdac11 expression in early myogenic differentiation.

We also sought to determine the requirement of p300 function for Hdac11 gene expression enhanced by RXR-selective signaling. As shown in Fig. 3F, p300 knockdown had a negative impact on RXR signaling-mediated Hdac11 gene expression as exhibited by the attenuation of bexarotene-augmented Hdac11 mRNA level following 24 h of differentiation, compared to untreated control, shown by the quantitative RT-PCR analysis. Taken together, our data suggest that p300 function is important for the upregulation of Hdac11 gene expression, possibly through positive regulation of Hdac11 locus in signal-enhanced myogenic differentiation.