Myogenic cell cultures derived from muscle biopsies are excellent models for human cell differentiation. We report the first comprehensive analysis of myogenesis-specific DNA hyper- and hypo-methylation throughout the genome for human muscle progenitor cells (both myoblasts and myotubes) and skeletal muscle tissue vs. 30 non-muscle samples using reduced representation bisulfite sequencing. We also focused on four genes with extensive hyper- or hypo-methylation in the muscle lineage (PAX3, TBX1, MYH7B/MIR499 and OBSCN) to compare DNA methylation, DNaseI hypersensitivity, histone modification, and CTCF binding profiles. We found that myogenic hypermethylation was strongly associated with homeobox or T-box genes and muscle hypomethylation with contractile fiber genes. Nonetheless, there was no simple relationship between differential gene expression and myogenic differential methylation, rather only for subsets of these genes, such as contractile fiber genes. Skeletal muscle retained ~30% of the hypomethylated sites but only ~3% of hypermethylated sites seen in myogenic progenitor cells. By enzymatic assays, skeletal muscle was 2-fold enriched globally in genomic 5-hydroxymethylcytosine (5-hmC) vs. myoblasts or myotubes and was the only sample type enriched in 5-hmC at tested myogenic hypermethylated sites in PAX3/CCDC140 andTBX1. TET1 and TET2 RNAs, which are involved in generation of 5-hmC and DNA demethylation, were strongly upregulated in myoblasts and myotubes. Our findings implicate de novo methylation predominantly before the myoblast stage and demethylation before and after the myotube stage in control of transcription and co-transcriptional RNA processing. They also suggest that, in muscle, TET1 or TET2 are involved in active demethylation and in formation of stable 5-hmC residues.

BackgroundFacioscapulohumeral muscular dystrophy (FSHD) is a dominant disease linked to contraction of an array of tandem 3.3-kb repeats (D4Z4) at 4q35. Within each repeat unit is a gene, DUX4, that can encode a protein containing two homeodomains. A DUX4 transcript derived from the last repeat unit in a contracted array is associated with pathogenesis but it is unclear how.MethodsUsing exon-based microarrays, the expression profiles of myogenic precursor cells were determined. Both undifferentiated myoblasts and myoblasts differentiated to myotubes derived from FSHD patients and controls were studied after immunocytochemical verification of the quality of the cultures. To further our understanding of FSHD and normal myogenesis, the expression profiles obtained were compared to those of 19 non-muscle cell types analyzed by identical methods.ResultsMany of the ~17,000 examined genes were differentially expressed (> 2-fold, p < 0.01) in control myoblasts or myotubes vs. non-muscle cells (2185 and 3006, respectively) or in FSHD vs. control myoblasts or myotubes (295 and 797, respectively). Surprisingly, despite the morphologically normal differentiation of FSHD myoblasts to myotubes, most of the disease-related dysregulation was seen as dampening of normal myogenesis-specific expression changes, including in genes for muscle structure, mitochondrial function, stress responses, and signal transduction. Other classes of genes, including those encoding extracellular matrix or pro-inflammatory proteins, were upregulated in FSHD myogenic cells independent of an inverse myogenesis association. Importantly, the disease-linked DUX4 RNA isoform was detected by RT-PCR in FSHD myoblast and myotube preparations only at extremely low levels. Unique insights into myogenesis-specific gene expression were also obtained. For example, all four Argonaute genes involved in RNA-silencing were significantly upregulated during normal (but not FSHD) myogenesis relative to non-muscle cell types.ConclusionsDUX4's pathogenic effect in FSHD may occur transiently at or before the stage of myoblast formation to establish a cascade of gene dysregulation. This contrasts with the current emphasis on toxic effects of experimentally upregulated DUX4 expression at the myoblast or myotube stages. Our model could explain why DUX4's inappropriate expression was barely detectable in myoblasts and myotubes but nonetheless linked to FSHD.

DNA methylation and chromatin DNaseI sensitivity were analyzed in and adjacent to D4Z4 repeat arrays, which consist of 1 to ∼100 tandem 3.3-kb units at subtelomeric 4q and 10q. D4Z4 displayed hypomethylation in some cancers and hypermethylation in others relative to normal tissues. Surprisingly, in cancers with extensive D4Z4 methylation there was a barrier to hypermethylation spreading to the beginning of this disease-associated array (facioscapulohumeral muscular dystrophy, FSHD) despite sequence conservation in repeat units throughout the array. We infer a different chromatin structure at the proximal end of the array than at interior repeats, consistent with results from chromatin DNaseI sensitivity assays indicating a boundary element near the beginning of the array. The relative chromatin DNaseI sensitivity in FSHD and control myoblasts and lymphoblasts was as follows: a non-genic D4Z4-adjacent sequence (p13E-11, array-proximal)> untranscribed gene standards > D4Z4 arrays> constitutive heterochromatin (satellite 2; P < 10−4 for all comparisons). Cancers displaying D4Z4 hypermethylation also exhibited a hypermethylation-resistant subregion within the 3.3-kb D4Z4 repeat units. This subregion contains runs of G that form G-quadruplexes in vitro. Unusual DNA structures might contribute to topological constraints that link short 4q D4Z4 arrays to FSHD and make long ones phenotypically neutral.