Ten healthy Danish male volunteers, aged between 19 and 25 (22.6 ± 1.6) years old, were recruited to participate in the study. The participants were sedentary and had a body mass index between 19.5 and 28.5 (23.1 ± 2.8, more clinical characteristics can be found in Table S1 in Supplementary Material). All participants reported to the laboratory and completed a peak pulmonary oxygen uptake test under fasting conditions to determine their VO₂ max. This test consisted of an incremental exercise bout to volitional fatigue on an electromagnetically braked cycle ergometer (Monark Ergomedic 839E) during which pulmonary gas exchange was measured breath-by-breath with a gas analyzing system (Oxycon Pro, Jaeger). The participants then performed an intense (80% VO₂ max) 15-min exercise bout and skeletal muscle biopsies were collected under fasting conditions at rest (basal) and 4 h after cessation of exercise. The aerobic (rather than resistance) exercise was chosen since the literature reporting EPS model describe genes that respond to aerobic exercise (6, 12). The 4-h time point was chosen based on previous observations from our group and others, which peaked that mRNA expression is seen 3–5 h after exercise (1, 4). Biopsies were obtained from the Vastus lateralis muscle using the Bergström needle technique (14), and they were immediately snap-frozen in liquid nitrogen and stored at −80°C for further analysis.

This study was carried out in accordance with the recommendations of the Ethic Committee from the Capital Region of Denmark with written informed consent from all subjects. All subjects gave written informed consent in accordance with the Declaration of Helsinki. The protocol was approved by the Ethic Committee from the Capital Region of Denmark (reference H-1-2013-064).

Total RNA was purified from the skeletal muscle biopsy samples using an AllPrep DNA/RNA/miRNA Universal Kit (Qiagen) following manufacturer’s instructions. The quality of recovered RNA was assessed by using the Agilent RNA 6000 Nano Kit (Agilent Technologies). RNA concentration was determined by spectrophotometry using a NanoDrop 2000. After purification, RNA samples were stored at −80°C until further analysis.

Libraries were prepared after ribosomal RNA depletion with the TruSeq Stranded Total Sample Prep Kit with Ribo-Zero Gold (Illumina). Sequencing was done on a HiSeq 2500, single-read, 100 bp read length per lane. Reads were trimmed for adapters and low quality flanking ends using Trim Galore v0.3.7 and Cutadapt v1.4.2 (15). Pre-processed reads were mapped to the human reference genome (hg19) with tophat2 using the “--b2-very-sensitive” setting (16, 17), and gene coverages were computed with featureCounts (18) using the Gencode v19 annotation. Genes with a coverage larger than 2.5 RPKM in at least five samples were tested for differential expression with edgeR using the glmQLFit/glmQLFTest modeling framework (19).

Mouse skeletal muscle cell line C2C12 (ATCC CRL1772) were seeded into six-well plates at a density of 10⁵ cells/well, cultured in growth medium containing Dulbecco’s modified Eagle’s medium (25 mM glucose) supplemented with 10% fetal bovine serum (FBS), and 1% penicillin–streptomycin at 37°C, 5% CO₂. After reaching ~90% confluence, differentiation was induced by switching 10% FBS to 2% horse serum. Differentiation medium was changed every day until fully differentiated myotubes were visualized under a light microscope (5–6 days).

A six-well plate, containing differentiated C2C12 myotubes was connected to C-dish electrode (IonOptix). The electrical current was discharged from a carbon electrode that was immersed in 8 ml differentiation medium added 24 h before the EPS. EPS was applied by C-pace cell culture stimulator (C-pace EP, IonOptix). EPS-induced muscle contraction was validated using light microscopy. Voltage, frequency, pulse duration, and stimulation time were optimized to induce reproducible increases in mRNA expression of exercise-responsive genes (Figure 1). Cells were harvested immediately after EPS at 0, 30, 60, 120, 180, and 240 min.

Total RNA was isolated from cells with TRIzol (Invitrogen) or AllPrep DNA/RNA/miRNA universal kit (Qiagen) according to the manufacturer’s instructions. RNA was converted to cDNA by iScript cDNA synthesis kit (Bio-Rad). Gene expression was determined by quantitative real-time PCR (qPCR) using Brilliant III Ultra-Fast SYBR Green Master mix (Agilent Technologies). Sequences of qPCR primers used for Ppargc1a, Ppard, Cyclin D1 (Ccnd1), and Nr4a3 genes are provided (Table S2 in Supplementary Material). mRNA expression was quantified by normalizing Ct values to cDNA amounts.

Genomic DNA was isolated from the muscle cell cultures using an AllPrep DNA/RNA/miRNA Universal kit (Qiagen) according to the manufacturer’s instructions.

Genomic DNA was bisulfite-converted using EZ DNA Methylation-Lightning Kit (Zymo Research) according to the manufacturer’s instructions. Bisulfite primers were designed to span the promoter region of Ppard, Ccnd1, and Nr4a3 genes within −3,000 to +2,000 bp relative to transcriptional start site (TSS) (Table S2 in Supplementary Material) using MethPrimer program (20). All bisulfite primers were mixed at equimolar ratio in a multiplexed PCR reaction using 200 ng of DNA as starting material and the HotStarTaq Plus DNA Polymerase kit (Qiagen). The PCR program was run at 95°C for 5 min, 35 cycles of 94°C (1 min) 60°C (1 min) 72°C (1 min), followed by 72°C for 10 min. PCR products were purified using MinElute PCR purification Kit (Qiagen). DNA libraries were prepared using NEBNext Ultra DNA Library Prep Kit for Illumina (New England Biolabs) and sequenced on a Miseq instrument (Illumina) with 151 cycles paired-end run. After sequencing, reads were trimmed for adapters and low quality flanking ends using Trim Galore v0.3.7 and Cutadapt v1.4.2 (15). Pre-processed reads were mapped to the mm10 reference genome and DNA methylation levels quantified by Bismark v0.14.4 and Bowtie2 (21, 22).

DNA (1 µg) was sheared into fragments (~200–300 bp) using a Bioruptor Plus sonicator (Diagenode). Hydroxymethylated DNA was captured from 220 ng of fragmented DNA using the hydroxymethylated DNA Immunoprecipitation (hMeDIP) ChIP Kit (Abcam) according to the manufacturer’s instructions. PCR primers were designed to cover a region of Nr4a3 promoter predetermined by DNA methylation analysis in response to EPS (Table S2 in Supplementary Material). DNA hydroxymethylation levels were determined by qPCR using Brilliant III Ultra-Fast SYBR Green Master mix (Agilent Technologies). Hydroxymethylated DNA was quantified by normalizing to total DNA input.

Data are presented as mean ± SE. Statistical analysis was performed using Graphpad Prism 6 (GraphPad Software). Two-way ANOVA was performed to determine the effect of EPS on gene expression, DNA methylation, and hydroxymethylation. When differences were statistically significant, the difference between control (non-stimulated) and EPS groups at corresponding time points was compared by an unpaired t-test.

We performed RNA sequencing on human skeletal muscle biopsies collected before and 4 h after an acute exercise bout to identify genes showing the highest transcriptional response. Analysis of differential mRNA expression showed 165 differentially expressed genes (out of 6,242 genes detected), where 72 were over-expressed and 93 were under-expressed after acute exercise (Table 1; Table S3 in Supplementary Material). The most upregulated gene was the Nuclear Receptor subfamily 4 group A member 3 (NR4A3) (P = 3.41E−08; false discovery rate = 2.51E − 05), which increased 26-fold after acute exercise. Thus, we selected NR4A3 to benchmark a model of muscle contraction in cultured myotubes.

Electrical pulse stimulation of cultured myotubes has been described (9, 11) and used to potentiate myotube differentiation (8) and increase the expression of specific genes, notably Ppargc1a (6). Here, we developed a novel EPS protocol to mimic the effects of acute exercise on the expression of Nr4a3. We tested several EPS conditions to stimulate C2C12 myotubes with contraction (Figure 1). While an EPS protocol applying 2 ms, 30 V, and 1 Hz had no effect on Ppargc1a mRNA expression (Figure 2A), it consistently induced the expression of Nr4a3 up to threefold at 60 min (Figure 2B) and also led to changes in mRNA expression of Ppard and Ccnd1 (Figures 2C,D). Thus, we selected this EPS protocol to study the epigenetic response of Nr4a3, Ccnd1, and Ppard after acute contraction.

We used targeted bisulfite sequencing to investigate the effect of EPS on Nr4a3, Ccnd1, and Ppard methylation. EPS altered the methylation status of several cytosines located in a region encompassing −2,201 to −2,081 bp relative to the TSS of the Nr4a3 gene (Figure 3A). Methylation of this region was inversely associated with Nr4a3 mRNA expression measured 60 min after EPS, at a time point where methylation was decreased (Figures 2B and 3B). Interestingly, Nr4a3 methylation was increased at 180 min, which marks the time point where Nr4a3 mRNA expression started to decrease (Figures 2B and 3B). EPS-induced transient methylation of the region +389 to +566 bp relative the TSS of the Ccnd1 gene at 30 and 60 min post-EPS (Figure 3C), which was associated with elevated mRNA expression at the 30-min time point (Figure 2D), suggesting hypermethylation is secondary to gene expression. Of potential interest, a wave of demethylation on the Ccnd1 promoter was detected at 120 min post-EPS, with mRNA expression tending to increase at the following two time points (Figure 2D). Similarly, methylation level of the Ppard promoter at a region of +706 to +852 bp relative to TSS was decreased at 60 min post-EPS (Figure 3D), which was followed by a progressive increase in Ppard mRNA expression in the following time points (Figure 2C). Our results provide evidence that EPS induces DNA methylation remodeling. Furthermore, we also show that contraction-induced DNA demethylation precedes transcriptional activation.

Given that EPS induced robust change in mRNA expression and demethylation of the Nr4a3 gene, we investigated the amount of the demethylation precursor hydroxymethylcytosine in the region −2,201 to −2,081 bp relative to the TSS of Nr4a3. We used hydroxymethylcytosine capture followed by quantitative PCR to gain further insight into the potential mechanisms by which EPS-induced myotube contraction induces Nr4a3 demethylation. We detected a threefold increase in hydroxymethylation levels of Nr4a3 immediately after EPS-induced contraction (0 min time point, Figure 4). This increase was transient, with the hydroxymethylation level restored to the baseline level by 30 min and further decreased at the 60 and 120 min time points (Figure 4). As the methylation level corresponding to this region was decreased at 60 and 120 min (Figure 3A), our results strongly suggest that contraction-induced Nr4a3 promoter demethylation occurs through intermediate hydroxymethylation.