- Access to this full-text is provided by Springer Nature.
- Learn more
Download available
Content available from Scientific Reports
This content is subject to copyright. Terms and conditions apply.
1
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
www.nature.com/scientificreports
Human Skeletal Muscle Possesses
an Epigenetic Memory of
Hypertrophy
Robert A. Seaborne1,2, Juliette Strauss2, Matthew Cocks2, Sam Shepherd2, Thomas D.
O’Brien2, Ken A. van Someren3, Phillip G. Bell3, Christopher Murgatroyd4, James P. Morton2,
Claire E. Stewart2 & Adam P. Sharples
1,2
It is unknown if adult human skeletal muscle has an epigenetic memory of earlier encounters with
growth. We report, for the rst time in humans, genome-wide DNA methylation (850,000 CpGs)
and gene expression analysis after muscle hypertrophy (loading), return of muscle mass to baseline
(unloading), followed by later hypertrophy (reloading). We discovered increased frequency of
hypomethylation across the genome after reloading (18,816 CpGs) versus earlier loading (9,153 CpG
sites). We also identied AXIN1, GRIK2, CAMK4, TRAF1 as hypomethylated genes with enhanced
expression after loading that maintained their hypomethylated status even during unloading
where muscle mass returned to control levels, indicating a memory of these genes methylation
signatures following earlier hypertrophy. Further, UBR5, RPL35a, HEG1, PLA2G16, SETD3 displayed
hypomethylation and enhanced gene expression following loading, and demonstrated the largest
increases in hypomethylation, gene expression and muscle mass after later reloading, indicating an
epigenetic memory in these genes. Finally, genes; GRIK2, TRAF1, BICC1, STAG1 were epigenetically
sensitive to acute exercise demonstrating hypomethylation after a single bout of resistance exercise
that was maintained 22 weeks later with the largest increase in gene expression and muscle mass after
reloading. Overall, we identify an important epigenetic role for a number of largely unstudied genes in
muscle hypertrophy/memory.
Numerous studies demonstrate that skeletal muscle can be programed, where early life exposure to environ-
mental stimuli lead to a sustained alteration of skeletal muscle phenotype in later life [reviewed in ref.1]. is has
been demonstrated in mammalian models in which reduced nutrient availability during gestation impairs skel-
etal muscle bre number, composition (fast/slow bre proportions) and size ofthe ospring1. Epidemiological
studies in human ageing cohorts also suggest that low birth weight and gestational malnutrition are strongly
associated with reduced skeletal muscle size, strength and gait speed in older age2,3. Driven by encounters with
the environment, foetal programming in skeletal muscle has been attributed in part to epigenetics4,5, which refers
to alterations in gene expression as a result of non-genetic structural modications of DNA and/or histones6.
Despite these compelling data, it is unknown if adult skeletal muscle possesses the capacity to respond dierently
to environmental stimuli in an adaptive or maladaptive manner if the stimuli have been encountered previously,
a concept recently dened as skeletal muscle memory1, or if this process is epigenetically regulated. Indeed, it is
known that skeletal muscle cells retain information or ‘remember’ the stem cell niche of the donor once derived
in-vitro from physically active7 obese8,9 and sarcopenic individuals [recently reviewed in ref.1]. Our group were
the rst to demonstrate this phenomenon, where human muscle stem cells derived from the skeletal muscle of
cancer patients exhibited overactive proliferation versus age matched control cells10. ese studies collectively
suggest that skeletal muscle cells could be epigenetically regulated, as they appear to not only retain information
from the environmental niche from which they originated, but also to pass this molecular ‘signature’ onto future
daughter cell progeny in-vitro. Furthermore, we have recently reported that mouse skeletal muscle cells (C2C12),
1Institute for Science and Technology in Medicine (ISTM), School of Medicine, Keele University, Staordshire, United
Kingdom. 2Research Institute for Sport and Exercise Sciences, Liverpool John Moores University, Liverpool, United
Kingdom. 3Department of Sport, Exercise and Rehabilitation, Northumbria University, Newcastle upon Tyne, United
Kingdom. 4School of Healthcare Science, Manchester Metropolitan University, Manchester, United Kingdom.
Correspondence and requests for materials should be addressed to A.P.S. (email: a.p.sharples@googlemail.com)
Received: 31 October 2017
Accepted: 16 January 2018
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
2
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
following an early-life inammatory stress, pass molecular information onto future generations (30 cellular divi-
sions), through a process of DNA methylation11. Importantly, the cells that encountered catabolic inammatory
stress in their earlier proliferative life had impaired dierentiation capacity when encountering the same inam-
matory stress in later proliferative life11. It has therefore been proposed that a memory and susceptibility of skele-
tal muscle to repeated encounters with inammation may be controlled by epigenetic modications such as DNA
methylation, a phenomenon we have termed skeletal muscle ‘epi-memory’1.
Mouse skeletal muscle in-vivo also appears to possess a memory from the anabolic growth steroid sex hor-
mone, testosterone. Where testosterone induced hypertrophy over a period of 3 months, resulted in enhanced
incorporation of myonuclei within muscle bres12,13. ese myonuclei were retained even following testosterone
withdrawal and the return of muscle mass to baseline12,13. Most notably the mice exposed to earlier life testoster-
one, exhibited a 31% increase in muscle cross-sectional area following mechanical loading versus control mice
that failed to grow in the same period of time12,13. is suggests an enhanced response to load induced mus-
cle hypertrophy when earlier life growth from testosterone had been encountered, and therefore corresponds
with the previously highlighted denition of muscle memory by Sharples et al.1. However, epigenetics has not
been studied in this model, and specically genome-wide DNA methylation has not been investigated aer adult
human skeletal muscle growth (hypertrophy) alone, or in skeletal muscle that has experienced later growth, to
investigate if skeletal muscle possesses an epigenetic memory from earlier life encounters with hypertrophy.
To provide parallel insights into the eect of the environment on genome-wide methylation changes in skeletal
muscle, recent studies have suggested that even an acute period of increased fat intake can alter the human DNA
methylome of CpGs in over 6,500 genes14. Like previous studies demonstrating rapid and dynamic alterations in
DNA methylation in skeletal muscle tissue aer acute metabolic stress (aerobic exercise)15,16 or disuse atrophy in
rats17, this study also suggested that large scale epigenetic modications can occur very rapidly in skeletal mus-
cle, aer only 5 days of high fat feeding. However, the authors also demonstrated a maintenance of methylation
following cessation of the high fat diet14. Where aer 8 weeks of returning to a normal diet, not all of the altered
methylation, particularly hypermethylation, was fully returned to baseline control levels14. is therefore suggests
that in response to an acute negative environmental stress, DNA methylation could be retained and accumulated
over time. Indeed, human skeletal muscle cells isolated from aged donors demonstrated a genome wide hyper-
methylated prole versus young adult tissue18. erefore, because DNA methylation, particularly within promoter
or enhancer regions of genes, generally leads to suppressed gene expression19, accumulation of high DNA methyl-
ation (hypermethylation) following a high fat diet and/or ageing could lead to universally suppressed gene expres-
sion. It may therefore be hypothesised that positive environmental encounters, such as muscle growth stimuli,
may induce a hypomethylated state (low DNA methylation) of important target transcripts or loci associated with
cellular growth and as a result, lead to enhanced gene expression when exposed to later life anabolic encounters.
To test this hypothesis, we aimed to investigate an epigenetic memory of earlier hypertrophy in adult human skel-
etal muscle using a within measures design, by investigating genome wide DNA methylation of over 850,000 CpG
sites aer: (1) Resistance exercise induced muscle growth (loading), followed by; (2) cessation of resistance exercise
to return muscle back towards baseline levels (unloading), and; (3) a subsequent later period of resistance exercise
induced muscle hypertrophy (reloading). is allowed us to assess the epigenetic regulation of skeletal muscle; (a)
hypertrophy, (b) a return of muscle back to baseline and, (c) memory of previous encounters with hypertrophy,
respectively. Importantly, these investigations for the rst time identied an increased frequency of hypomethyla-
tion across the genome during later reloading where lean muscle mass increases were enhanced compared to earlier
loading. We also detected genes; AXIN1, GRIK 2, CAMK4 and TRAF1 displayed increasing DNA hypomethylation
together with enhanced gene expression across loading, unloading and reloading. Where hypomethylation of these
genes was maintained even during unloading where muscle mass returned back to baseline, indicating an epigenetic
memory of earlier muscle growth. Furthermore, UBR5, RPL35a, HEG1 and PLA2G16 previously unstudied in skel-
etal muscle, together with SETD3 displayed hypomethylation and enhanced gene expression following loading ver-
sus baseline and displayed even larger increases in both hypomethylation and gene expression aer later reloading,
also indicating an epigenetically regulated memory leading to enhanced gene expression during reloading. Gene
expression of this cluster also strongly and positively correlated with increased muscle mass across all conditions,
conrming these transcripts to be novelresistance exercise induced- hypertrophy genes in skeletal muscle. Finally,
we identied genes GRIK2, TRAF1 (identied above), BICC1 and STAG1 were hypomethylated aer a single bout
of acute resistance exercise that were maintained as hypomethylated, and had enhanced gene expression aer later
reloading. Suggesting that these are epigenetically sensitive genes aer a single bout of resistance exercise and asso-
ciated with enhanced muscle hypertrophy 22 weeks later.
Methods
Human Participants and Ethical Approval. Eight healthy males gave written, informed consent to par-
ticipate in the study, following successful completion of a readiness to exercise questionnaire and a pre-biopsy
screening as approved by a physician. One participant withdrew from the study at experimental week 17 of 21,
for reasons unrelated to this investigation. However, consent allowed samples to be analysed prior to withdrawal,
therefore for this participant, this included all conditions excluding the nal reloading condition (for details see
below). Ethical approval was granted by the NHS West Midlands Black Country, UK, Research Ethics Committee
(NREC approval no. 16/WM/0103), all methods were performed in accordance with the relevant ethical guide-
lines and regulations.
Experimental Design. Using a within subject design, eight previously untrained male participants
(27.6 ± 2.4 yr, 82.5 ± 6.0 kg, 178.1 ± 2.8 cm, means ± SEM) completed an acute bout of resistance exercise (acute RE),
followed by 7 weeks (3d/week) of resistance exercise (loading), 7 weeks of exercise cessation (unloading) and a
further period of 7 weeks (3d/week) resistance exercise (re-loading). Graphical representation of experimental
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
3
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
design is provided in Fig.1A. Whole-body fan beam dual-energy x-ray absorptiometry (DEXA), strength of the
quadriceps via dynamometry and muscle biopsies from the vastus lateralis for RNA and DNA isolation were
obtained at baseline, aer 7 weeks loading (beginning of week 8), 7 weeks unloading (end of week 14) and 7 weeks
reloading (beginning of week 22). A muscle biopsy was also obtained 30 minutes aer acute RE prior to 7 weeks
loading. Genome-wide analysis of DNA methylation was performed via Illumina EPIC array (850,000 CpG sites-
detailed below) for participants across all conditions (n = 8 baseline, acute, loading, unloading, n = 7 reloading).
Rt-qRT-PCR was used to investigate corresponding transcript expression of epigenetically altered genes identied
via the genome wide DNA methylation analysis.
Resistance exercise induced muscle hypertrophy: Loading, unloading and reloading.
Untrained male subjects initially performed an exercise familiarization week, in which participants performed
all exercises with no/low load to become familiar with the exercise type (detailed below). In the nal session of
the familiarzation week, the load that participants could perform 4 sets of 8–10 repetitions for each exercise was
assessed. Due to participants being uncustomized to resistance exercise, assessment was made on competence of
liing technique, range of exercise motion and verbal feedback (participant), and a starting load was set for each
participant on an individual basis (mean load for this starting load is included below). ree to four days later,
participants then undertook a single bout of lower limb resistance exercise (acute RE, exercises detailed below)
followed by biopsies 30 minutes post exercise. Following this single bout of acute RE they then began a chronic
resistance exercise program, completing 60-min training sessions (Monday-Wednesday-Friday), for 7 weeks,
Figure 1. (A) Schematic representation of experimental conditions and types of analysis undertaken across
the time-course. e image of a muscle represents the time point for analysis of muscle mass via (i) DEXA and
strength via (ii) isometric quadriceps muscle torque using an isokinetic dynamometer. e images of muscle
tissuealso represent the time point of skeletal muscle biopsy of the Vastus Lateralis, muscle sample preparation
for downstream analysis of (iii) Innium MethylationEPIC BeadChip arrays (850 K CpG sites) methylome wide
array (iv) and rt-qRT-PCR for gene expression analysis of important genes identied following methylome wide
analysis. (B) Weekly total volume of resistance exercise undertaken by human participants (n = 7) during the
rst 7-week resistance exercise period (loading, weeks 1–7), followed by a 7 week cessation of resistance exercise
(unloading, weeks 8–14) and the later second period of 7 weeks resistance exercise (reloading, weeks 15–21).
Data represents volume load as calculated by ((load (Kg) x reps) x sets)) averaged across 3 resistance exercise
sessions per week. Data presented mean ± SEM. (Ci) Lean lower limbmass changes in human subjects (n = 7)
aer a period of 7 weeks resistance exercise (loading), exercise cessation (unloading) and a subsequent second
period of 7 weeks resistance exercise (reloading). Total limb lean mass normalised to baseline (percentage
change). Signicant change compared to baseline represented by * and signicant dierence to all other
conditions represented by ** (Cii) Total lean mass percentage change when loading is normalised to baseline,
and reloading normalised to unloading to account for starting lean mass in both conditions. Pairwise t-test of
signicance indicated by *. All data presented as mean ± SEM (n = 7).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
4
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
with 2 sessions/week focusing on lower limb muscle groups (Monday and Friday) and the third session focusing
on upper body muscle groups (Wednesday). Lower limb exercises included, behind head squat, leg press, leg
extension, leg curl, Nordic curls, weighted lunges and calf raises. Upper limb exercises included, at barbell bench
press, shoulder press, latissimus pull down, dumbbell row and triceps cable extension. To ensure progression in
participants with no previous experience in resistance exercise, a progressive volume model was adopted20 in
which investigators regularly assessed competency of sets, reps and load of all exercises. Briey, exercises were
performed for 4 sets of 10 reps in each set, ~90–120 s in between sets and ~3 mins between exercises. When par-
ticipants could perform 3 sets of 10 repetitions without assistance and with the correct range of motion, load was
increased by ~5–10% in the subsequent set and participants continued on this new load until further modica-
tions wererequired, as similar to that previously described20. Where subjects failed to complete 10 full repetitions
(usually for their nal sets), they were instructed to reduce the load in order to complete a full repetition range
for the subsequent (usually nal)set. Total weekly volume load was calculated as the sum of all exercise loads;
..=∗∗Totalexercisevolumekgs Exercise load kgsNoofRepsNoSets()(())
e acute resistance exercise session resulted in a total load of 8,223 kg (±284kg). ereaer, the load-
ing and reloading phases resulted in a progressive increase in training volume (±SEM) of 2,257 ± 639 kg and
2,386 ± 222 kg respectively per week (Fig.1B), with the reloading phase displaying a signicant (P = 0.043)
increase in average load. Loading and reloading programs were conducted in an identical manner, with the same
exercises, program layout (same exercises on thesame day), sets and repetition pattern and rest between sets and
exercises. During the 7 week unloading phase, participants were instructed to return to habitual pre-intervention
exercise levels and not to perform any resistance training. Regular verbal communication between researcher and
participant ensured subjects followed these instructions. A trainer was present at all resistance exercise sessions
to enable continued monitoring, provide verbal encouragement and to ensure sucient progression. No injuries
were sustained throughout the exercise intervention.
Lean mass and strength of the lower limbs by dual-energy x-ray absorptiometry (DEXA) and
dynamometry. A whole-body fan beam dual-energy x-ray absorptiometry (DEXA; Hologic QDR Series,
Discovery A, Bedford, MA, USA) scan was performed after loading, unloading and reloading (depicted in
Fig.1A) to assess lower limb changes in lean mass. All scans were performed and analysed (QDR for Windows,
version 12:4:3) by the same trained operator, according to Hologic guidelines. e DEXA scan was automatically
analysed via the QDR soware and the operator conrmed areas of interest including lower limb positions. Lean
mass was calculated on absolute values for each condition, and presented as percentage change compared to
baseline. Furthermore, in addition, a separate analysis was undertakento assess whether later reloading altered
lean mass,where loading was normalised to baseline, and reloading was normalised to unloading to account for
anyresidual starting mass(even if non-signicant)following the earlier loading period. A pairwise t-test was
then used to analyse the percentage increase in lean mass as a consequence of reloading compared to loading.
To assess quadriceps muscle strength, in-vivo isometric knee extension maximal voluntary contractions (MVC)
were performed using an isokinetic dynamometer (IKD; Biodex, New York, USA) to measure peak joint torque.
Data presented as percentage increase to baseline (%) using absolute values (Nm), unless otherwise stated. A full
description of strength assessment can be found in Supplementary File1.
Muscle Biopsies and Sample Preparation. At baseline, 30 minutes post acute resistance exercise (RE)
and aer 7 weeks loading (beginning of week 8), 7 weeks unloading (end of week 14) and 7 weeks reloading
(beginning of week 22) (Fig.1A), a conchotome muscle biopsy was obtained from the vastus lateralis muscle of
the quadriceps from each participant, avoiding areas of immediate proximity to previous incisions, before being
carefully cleaned and dissected using a sterile scalpel on a sterile petri dish. In the unlikely event of any brous/fat
tissue, this was removed using a scapel, leaving only lean tissue. Separate samples were immediately snap frozen
in liquid nitrogen before being stored at −80 °C for RNA and DNA analysis.
DNA Isolation, Bisulte Conversion and Methylome Wide BeadChip Arrays. DNA was extracted
from frozen tissue samples using a commercially available DNA isolation kit (DNeasy Blood and Tissue Kit,
Qiagen, Manchester, UK) in accordance with manufacturer’s instructions, before being analysed (Nanodrop,
ThermoFisher Scientific, Paisley, UK) for yield (mean ± SDEV8.0 µg ± 4.2) and quality (260/280 ratio of
mean ± SDEV1.88 ± 0.09). Five-hundred ng of prepared DNA was bisulte converted using the EZ-96 DNA
Methylation Kit (Zymo Research Corp., CA, USA) following the manufacturer’s instructions for use of the DNA
in Illumina assays. Innium MethylationEPIC BeadChip array examined over 850,000 CpG sites of the human
epigenome (Innium MethylationEPIC BeadChip, Illumina, California, United States) and data was analysed
in Partek Genomics Suite V.6.6 (Partek Inc. Missouri, USA). Raw data les (.IDAT) were normalised via the
Subset-Quantile Within Array Normalisation (SWAN) method, as previously described21. Initial quality control
steps were undertaken to detect samples withinarrays that were identied as outliers. Principal component anal-
ysis (PCA) and normalisation histograms detected two observable outliers across all samples. ese samples were
removed from any further analysis (Supplementary Figure1A & B). While skeletal muscle tissue samples may
contain a small proportion of other non-muscle cells this analysis suggests sample homogeneity was consistent
in the experimental groups and therefore downstream analysis was representative of skeletal muscle tissue and
its niche. Data sets represent SWAN-normalised beta (β)-values which correspond to the percentage of methyl-
ation at each site and are calculated as a ratio of methylated to unmethylated probes22. Dierential methylation
was subsequently detected across all experimental conditions, and between conditions to identify statistically
dierentially regulated CpG sites. Fold change in CpG specic DNA methylation and statistical signicance
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
5
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
was performed using Partek Genomic Suite V.6.6 soware, where statistical signicance was obtained following
ANOVA (with bonferroni correction) analysis.
Hierarchical Clustering Dendogram. Unadjusted p-value signicance (P < 0.05) was used to create a
CpG site marker list of standardized beta-values. A standardized expression normalisation was performed to
shi CpG sites to mean of zero and scale to a standard deviation of one. Unsupervised hierarchical clustering
was performed and dendograms wereconstructed to represent dierentially methylated CpG loci and statistical
clustering of experimental samples. Heatmaps represent expression of CpG loci, where reduced methylation at
DNA sites (hypomethylated) are represented in green, increased methylation at DNA sites (hypermethylated) in
red, and unchanged sites are represented in black.
Tissue Homogenisation, RNA Isolation and rt-qRT-PCR. Skeletal muscle tissue (~30 mg) was
immersed in Tri-Reagent (Sigma-Aldrich, MO, United States) in MagNA Lyser 1.4 mm beaded tubes (MagNA
Lyser Green Beads, Roche, Germany) and homogenised for 40 secs at 6 m/s in a MagNA Lyser Homogeniser
(Roche, Germany), before being stored on ice for 5 mins. is step was repeated three times to ensure complete
disruption of muscle tissue sample. RNA was extracted using standard Tri-Reagent procedure via chloroform/iso-
propanol extractions and 75% ethanol washing as per manufacturer’s instructions. RNA pellets were resuspended
in 30 μl of RNA storage solution (Ambion, Paisley, UK) and analysed (Nanodrop, ermoFisher Scientic, Paisley,
UK) for quantity (mean ±SDEV ; 6671 ± 3986 ng) andan indication of quality (260/280 ratio of mean ± SDEV,
1.95 ± 0.09). For rt-qRT-PCR using QuantiFastTM SYBR® Green RT-PCR one-step kit on a Rotorgene 3000Q,
reactions were setup as follows; 9.5 μl experimental sample (5.26 ng/μl totaling 50 ng per reaction), 0.15 μl of both
forward and reverse primer of the gene of interest (100 μM), 0.2 μl of QuantiFast RT Mix (Qiagen, Manchester,
UK) and 10 μl of QuantiFast SYBR Green RT-PCR Master Mix (Qiagen, Manchester, UK). Reverse transcrip-
tionwas initiated with a hold at 50 °C for 10 minutes (cDNA synthesis), followed by a 5-minute hold at 95 °C
(transcriptase inactivation and initial denaturation), before 40–45 PCR cycles of; 95 °C for 10 sec (denaturation)
followed by 60 °C for 30 secs (annealing and extension). Primer sequences are provided in Supplementary File7.
Gene expression analysis was performed on at least n = 7 for all genes, unless otherwise stated. All relative gene
expression was quantied using the comparative Ct (∆∆Ct) method. Individual participants own baseline Ct val-
ues were used in ∆∆Ct equation as the calibrator using RPL13a as the reference gene. e average Ct value for the
reference gene was consistent across all participants and experimental conditions (20.48 ± 0.64, SDEV) with low
variation of 3.17%.
Statistical Analysis. Analysis of exercise volume load was performed via a T-test (MiniTab Version 17.2.1)
of average participant load during the loading vs. reloading phases. DEXA and isometric peak torque; for
n = 7, as well as correlation analysis was analysed via a statistical package for the social sciences soware for
Microso (SPSS, version 23.0, SPSS Inc, Chicago, IL) using a one-way repeated measures ANOVA, where appli-
cable. Pearson correlation of coecient analysis (two tailed) was conducted for gene expression vs. percentage
change of leg lean mass. Methylome wide array data sets (n = 8 for baseline, acute RE, loading, unloading, n = 7
for reloading) were analysed for signicant epigenetically modied CpG sites in Partek Genome Suite (ver-
sion 6.6). All gene ontology and KEGG signalling pathway23–25 analysis was performed in Partek Genomic Suite
and Partek Pathway, on generated CpG lists of statistical signicance (P < 0.05) across conditions (ANOVA) or
pairwise comparisonsbetween conditions. In MiniTab Statistical Soware (MiniTab Version 17.2.1) follow up
rt-qRT-PCR gene expression was analysed using both a MANOVA, to detect for signicant interactions across
time for identied clusters of genes, and an ANOVA for follow up of individual genes over time. A pairwise
t-test was used to analyse gene expression following acute RE vs. baseline. For follow up fold change in CpG
DNA methylation analysis was performedvia ANOVA in MiniTab Statistical Soware (MiniTab Version 17.2.1).
Statistical values were considered signicant at the level of P ≤ 0.05. All data represented as mean ± SEM unless
otherwise stated.
Results
Lean leg muscle mass is increased after loading, returns toward baseline during unloading and
is further increased after reloading. Analysis of lower limb lean mass via DEXA, identied a signicant
increase of 6.5% ( ± 1.0%; P = 0.013) in lean mass aer 7-wks of chronic loading compared to baseline (20.74 ± 1.11 kg
loading vs. 19.47 ± 1.01 kg baseline). Following 7-wks of unloading, lean mass signicantly reduced by 4.6% ± 0.6%
(P = 0.02) vs. the 7 weeks loading, back towards baseline levels (unloading, 19.83 ± 1.06 kg), conrmed by no signif-
icant dierence between unloading and baseline. Subsequently, a signicant increase in lean mass of the lower limbs
was accrued aer the reloading phase of 12.4 ± 1.3%, compared to baseline (reloading, 21.85 ± 2.78 kg, P = 0.001,
Fig.1Ci), resulting in an increase of 5.9 ± 1.0% compared to the earlier period of loading (P = 0.005). Pairwise t-test
analysis that corrected for any lean mass that was maintained during unloading demonstrated a signicant increase
in lean muscle mass in the reloading phase (unloading to reloading), compared to the loading phase (baseline to
loading) (P = 0.022; Fig.1Cii). Analysis of muscle strength suggested a similar trend. Isometric peak torque increased
by 9.3 ± 3.5% from 296.2 ± 22.1 Nm at baseline to 324.5 ± 27.3 Nm aer 7-wks of loading, this dierence was not
statistically signicant (Supplementary Figure2). Upon 7-wks of unloading, peak torque reduced by 8.3 ± 2.8% vs.
loading, back towards baseline levels. Upon subsequent reloading, a signicant 18 ± 3.6% increase in isometric peak
torque production (349.6 ± 27.7 Nm) was observed compared to baseline (P = 0.015; Supplementary Figure2A).
The largest DNA hypomethylation across the genome occurred following reloading. The
frequency of statistically (P < 0.05) differentially regulated CpGs in each condition was analysed (Fig.2A;
Supplementary File2B). 17,365 CpG sites were signicantly (P < 0.05) dierentially epigenetically modied fol-
lowing loading induced hypertrophy compared to baseline, with a larger number being hypomethylated (9,153)
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
6
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
compared to hypermethylated (8,212) (Fig.2A; Supplementary File2A & B). e frequency of hypomethylated
epigenetic modications was similar to loading aer unloading (8,891) (Fig.2A; Supplementary File2A & C),
where we reported lean muscle mass returned back towards baseline. Importantly, following reloading induced
muscle growth we observed an increase in the number of epigenetically modied sites (27,155) and an enhanced
number of hypomethylated DNA sites (18,816, Fig.2A; Supplementary File2A & D). is increase in hypometh-
ylation coincided with the largest increase in skeletal muscle mass in reloading. By contrast, hypermethylation
remained stable (8,339) versus unloading (8,638) and initial loading (8,212). To further analyse the reported
increased frequency of hypomethylated genes across the genome following reloading, gene ontologies were ana-
lysed for the frequency of hypo and hypermethylated CpG sites. In agreement with our above frequency analysis,
the most statistically signicant enriched GO terms identied an increased number of hypomethylated CpG sites
compared to baseline (Fig.2Bi–iii). Indeed, the most statistically signicantly (FDR < 0.05) enriched GO terms
were: 1) molecular function GO:0005488 encoding for genes related to ‘binding’, that displayed 9,577 (68.71%)
CpG sites that were hypomethylated following reloading and 4,361 (31.29%) sites as hypermethylated compared
to baseline (Fig.2Bi), and: 2) Biological process GO:0044699 encoding for genes related to ‘single-organism pro-
cesses’ that displayed 7,586 (68.57%) hypomethylated CpG sites compared to 3,493 (31.43%) sites proled as
hypermethylated aer reloading compared to baseline (Fig.2Bii). Finally, 3) cellular component, GO:004326
encoding for genes related to ‘organelle’ reported 7,301 hypomethylated CpG sites following reloading and 3,311
hypermethylated sites, compared to baseline, therefore favouring a majority 68.88% hypomethylated prole
(Fig.2Biii).
Following conrmation that the largest alteration in CpG DNA methylation occurred upon laterreload-
ingevoked hypertrophy, we sought to elucidate how the serine/threonine AKT signaling pathway, a critical path-
way involved in mammalian growth, proliferation and protein synthesis26,27, was dierentially regulated across
experimental conditions (Fig.3, Supplementary Figure3A and B). Intuitively, we report that the PI3K/AKT path-
way was signicantly enriched upon all pairwise comparisons of baseline vs. loading, unloading and reloading,
respectively (P < 0.022; Supplementary Figure3A,B and Fig.3A), suggesting that the pathway was signicantly
epigenetically modied following periods of skeletal muscle perturbation. Importantly, frequency analysis of
statistically dierentially regulated transcripts (Fig.3B) attributed to this pathway, reported an enhanced number
of dierentially regulated CpG sites (444 CpG sites) following reloading (Fig.3A), compared to loading ( 264 CpG
sites; Supplementary Figure3A) and unloading (283 CpG sites; Supplementary Figure3B) alone. In accordance
with our previous ndings, the enhanced number of statistically dierentially regulated CpG sites in this pathway
upon reloading is attributed to an enhanced number of hypomethylated (299 CpG sites, 67.3%) compared to
hypermethylated (145 sites, 32.7%) CpG sites (Raw data: Supplementary File3).
Figure 2. (A) Innium MethylationEPIC BeadChip arrays (850 K CpG sites) identied an enhanced frequency
of hypomethylated CpG sites upon reloading (n = 7). (B) Gene ontology analysis using forest plot schematics
conrmed an enhanced hypomethylated prole aer reloading across various (i) molecular function, (ii)
biological processes and (iii) cellular components. Functional groups with a fold enrichment >3 (as indicated
via shaded blue region) represents statistically ‘over expressed’ (in this case epigenetically modied) KEGG
pathways FDR < 0.05 (n = 8).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
7
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
Genome-wide DNA methylation analysis identied two clusters of temporal DNA methylation
patterns that provide initial evidence of an epigenetic memory. Changes in genome-wide DNA meth-
ylation were analysed following loading, unloading and reloading induced muscle adaptation. A dendogram of
the top 500 most statistically epigenetically modied CpG sites across each experimental condition compared to
baseline, identied large alterations in DNA methylation proles (Fig.4A; Supplementary File4A). A ranked unsu-
pervised hierarchical clustering analysis demonstrated signicant dierences between the initial loading (weeks
1–7) vs. all other conditions (Fig.4A). Closer analysis of the top 500 CpG sites across experimental conditions high-
lighted a clear temporal trend occurring within dierent gene clusters. e rst cluster (named Cluster, A) displayed
enhanced hypomethylation with earlier loading-induced hypertrophy. is cluster was methylated at baseline and
became hypomethylated aer loading, re-methylated with unloading (Fig.4A) and hypomethylated aer reloading.
e second temporal trend (named Cluster B) also displayed an enhanced hypomethylated state across the top
500 CpG sites as a result of load induced hypertrophy. As with Cluster A, Cluster B genes were methylated at baseline
and became hypomethylated aer initial loading. In contrast to Cluster A, Cluster B remained hypomethylated with
unloading, even when muscle returned to baseline levels, and this hypomethylation was also maintained/‘remem-
bered’ aer reload induced hypertrophy (Cluster B, depicted Fig.4A). e third temporal trend, named Cluster C,
revealed genes as hypomethylated at both baseline and aer initial loading, suggesting no epigenetic modication
aer the rst period of hypertrophy in these genes (Cluster C, Fig.4A). During unloading, genes were hypermeth-
ylated and remained in this state during reloading. e nal cluster (Cluster D) of genes, were hypomethylated at
baseline, became hypermethylated aer loading (Cluster D, Fig.4A), reverted back to a hypomethylated state with
Figure 3. (A) Representation of the DNA methylation modications that occurred within the PI3K/AKT
KEGG pathway following 7 weeks of reloading in human subjects. Signalling analysis performed on statistically
dierentially regulated CpG sites compared to baseline, with green indicating a hypomethylated fold change and
red indicating a hypermethylated change, with strength of colour representing the intensity of fold change23–25.
Figure3b. Venn diagram analysis of the statistically dierentially regulated CpG sites attributed to the PI3K/
AKT pathway following loading, unloading and reloading, compared to be baseline. Ellipsis reports number of
commonly statistically dierentially regulated CpG sites across each condition. Analysis conrms an enhanced
number of dierentially regulated CpG sites upon reloading condition.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
8
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
unloading and then maintained the hypomethylated state aer reloading, reecting the prole of the baseline targets
in the same cluster (Cluster D, Fig.4A). ese two clusters (C&D) did report a maintenance of the DNA methylation
prole from unloading to reloading conditions. Cluster C also reported a hypermethylated prole aer unloading
following a period of loading, that may therefore identify important CpG sites that are hypermethylated when mus-
cle mass is reduced (we therefore include a full list from cluster C that includes the CpG sites signicantly modied
in loading vs. unloading, Supplementary File4G). However, both Cluster C&D suggest no retention of epigenetic
modications from the rst loading period to the later reloading phase.
Identification of gene expression clusters inversely associated with DNA methylation. To
assess whether the changes in DNA methylation aected gene expression, the 100 most signicantly dierentially
modied CpG sites across all conditions were identied and cross referenced with the most frequently occurring
(Supplementary File4B) CpG modications in pairwise comparisons of all conditions (Supplementary File4C to H).
is identied 48 genes that were then analysed by rt-qRT-PCR to assess gene expression. Forty-six percent of the
Figure 4. (A) Heat map depicting unsupervised hierarchical clustering of the top 500 statistically dierentially
regulated CpG loci (columns) and conditions (baseline, loading, unloading and reloading) in previously
untrained male participants (n = 8). e heat-map colours correspond to standardised expression normalised
β-values, with green representinghypomethylation, red hypermethylation and unchanged sites are represented
in black. (4B and C) Relative gene expression (i) and schematic representation of CpG DNA methylation and
gene expression relationship (ii) in two identied gene clusters from genome wide methylation analysis aer
a period of 7 weeks resistance exercise (loading), exercise cessation (unloading) and a subsequent secondary
period of 7 weeks resistance exercise (reloading). (Bi) Expression of genes that displayed a signicant increase
compared to baseline (represented by *) upon earlier loading, that returned to baseline during unloading, and
displayed enhanced expression aer reloading (signicantly dierent to all other conditions **). MANOVA
analysis reported a signicant eect over theentire time course of the experiment (P < 0.0001). (Bii)
Representative schematic displaying the inverse relationship between mean gene expression (solid black lines)
and CpG DNA methylation (dashed black lines) of grouped transcripts (RPL35a, C12orf50, BICC1, ZFP2,
UBR5, HEG1, PLA2G16, SETD3 and ODF2). Data represented as fold change for DNAmethylation (le y axis)
and gene/mRNA expression (right y axis). (Ci) Clustering of genes that portrayed an accumulative increase
in gene expression aer loading, unloading and reloading. With the largest increase in gene expression aer
reloading. Culminating in signicance in the unloading (baseline vs. unloading*), and reloading (reloading vs.
baseline**). (Cii) Representative schematic displaying the inverse relationship between mean gene expression
(solid black lines) and CpG DNA methylation (dashed black lines) of grouped transcripts (AXIN1, TRAF1,
GRIK2, CAMK4). Data represented as fold change for methylation (le y axis) and mRNA expression (right
y axis). All data represented as mean ± SEM for gene expression (n = 7 for UBR5, PLA2G16, AXIN1, GRIK2;
n = 8 for all others) and CpG DNA methylation (n = 8 for baseline, loading and unloading; n = 7 for reloading).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
9
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
top 100 CpG sites were within gene promotor regions with 18%residing in intergenic regions (Supplementary File5).
Interestingly, gene expression analysis identied two distinct clusters of genes that had dierent transcript proles.
is rst cluster included RPL35a, C12orf50, BICC1, ZFP2, UBR5, HEG1, PLA2G16, SETD3 and ODF2 genes that
displayed a signicant main eect for time (P < 0.0001) aer MANOVA analysis (Fig.4Bi). Chromosome loca-
tions, reference sequence numbers and generegion section details for these genes can be found in Supplementary
File5. Importantly, this rst cluster displayed a mirrored (inverse) temporal pattern to those identied previously in
Cluster A above for CpG methylation (in the top 500 dierentially regulated CpG sites, Fig.4A). Where, upon 7-wks
of loading, gene expression of this cluster signicantly increased (1.22 ± 0.09, P = 0.004) and CpG methylation of
the same genes was non-signicantly reduced (hypomethylated) (0.95 ± 0.04 Fig.4Bii). During unloading, meth-
ylation returned to baseline (1.03 ± 0.07), which was met by a return to baseline in gene expression (0.93 ± 0.05),
as indicated by both CpG methylation and gene expression displaying no signicant dierence compared to base-
line (Fig.4Bii). Importantly, upon reloading, both CpG methylation and gene expression displayed an enhanced
response compared to the baseline and loading time point, respectively. Indeed, upon reloading, this cluster became
hypomethylated (0.91 ± 0.03, P = 0.05, Fig.4Bii). is was met with a signicant enhancement (1.61 ± 0.06) in gene
expression of the same cluster compared to baseline and loading (P < 0.001, Fig.4Bii).
A second separate gene cluster was identied and included: AXIN1, GRIK2, CAMK4, TRAF1, NR2F6 and
RSU1. Although together there was no signicant eect of time via MANOVA analysis. ANOVA analysis reported
that this cluster displayed increased gene expression aer loading (1.19 ± 0.08) that then further increased during
unloading (1.58 ± 0.13) resulting in statistical signicance (P = 0.001) compared to baseline alone. Gene expression
was then even further enhanced (1.79 ± 0.09) upon reload induced hypertrophy (P < 0.0001; Fig.4Ci; Chromosome
locations, reference sequence numbers, region section details for this cluster of genes can be found in Supplementary
File5). In this cluster we identied an accumulative increase in gene expression, attaining signicance at unloading
condition (ANOVA; P = 0.001) compared to baseline, gene expression was subsequently further increased follow-
ing reloading conditions (ANOVA; P < 0.0001). is temporal gene expression pattern was inversely associated
to CpG methylation observed in Cluster B (identied previously in the top 500 dierentially regulated CpG sites,
Fig.4A). Closer fold-change analysis of CpG DNA methylation of this gene cluster, identied a distinct inverse
relationship with methylation and gene expression of 4 out of 6 of the targets (AXIN1, GRIK2, CAMK4, TRAF1).
Where, upon loading, these genes became signicantly hypomethylated (0.78 ± 0.09; P = 0.036) compared to base-
line, with this prole being maintained during unloading (0.84 ± 0.09) and reloading (0.83 ± 0.05) conditions, albeit
non-signicantly. Collectively, we report that a sustained hypomethylated state in 4 out of 6 of the genes in this
cluster that correspond to an increased transcript expression of the same genes (Fig.4Cii).
Identication of a number of novel genes at the expression level associated with skeletal mus-
cle hypertrophy. To ascertain the relationship between skeletal muscle hypertrophy and gene expression, fold
change in gene expression was plotted against percentage changes (to baseline) in leg lean mass. Interestingly, in
our rst cluster of genes identied above (RPL35a, C12orf50, BICC1, ZFP2, UBR5, HEG1, PLA2G16, SETD3 and
ODF2), a signicant correlation between gene expression and lean mass was observed for genes RPL35a, UBR5,
SETD3, PLA2G16 and HEG1 (Fig.5A & BI–V). Following exposure to 7-wks of load induced hypertrophy, RPL35a
gene expression displayed a non-signicant increase compared to baseline (1.13 ± 0.23; Fig.5AI), that upon unload-
ing returned back to the baseline levels (1.01 ± 0.21). Upon reloading, the expression of RPL35a increased to 1.70
( ± 0.44; Fig.5AI) compared to baseline (P = 0.05). is expression pattern across loading, unloading and reload-
ing conditions corresponded to a signicant correlation with percentage changes in skeletal muscle mass (R = 0.6,
P = 0.014; Fig.5BI), with RPL35a accounting for 36% of the variation in muscle across experimental conditions.
Both UBR5 and SETD3 displayed similar percentage accountability for the change in skeletal muscle mass across
conditions. Indeed, UBR5 and SETD3 accounted for 33.64% and 32.49% of the variability in skeletal muscle mass,
respectively, both portraying strong correlations between their gene expression and the percentage change in lean
leg mass (UBR5, R = 0.58, P = 0.018, Fig.5BII; SETD3, R = 0.57, P = 0.013, Fig.5BII, respectively). Additionally,
UBR5 (1.65 ± 0.4; Fig.5BII) and SETD3 (1.16 ± 0.2; Fig.5AIII) both demonstrated non-signicant increases in
gene expression aer 7-wks of loading (P > 0.05), with the expression of both genes, UBR5 (0.82 ± 0.27) and SETD3
(0.90 ± 0.15), returning to baseline levels upon 7-wks of unloading (Fig.5AII and AIII, respectively). Furthermore,
upon reloading UBR5 displayed its greatest increase in expression (1.84 ± 0.5; Fig.5AII), demonstrating a trend
for significance compared to baseline condition (P = 0.07), and a significant increase compared to unloading
(P = 0.035). Whereas, SETD3 demonstrated a fold increase of 1.48 ( ± 0.25; Fig.5AIII) approaching signicance
compared to baseline (P = 0.072) and achieving signicance compared to unloading (P = 0.036). PLA2G16 also
demonstrated a signicant correlation between its fold change in gene expression and the percentage change in
skeletal muscle mass (R = 0.55; P = 0.027; Fig.5BIV), with PLA2G16 accounting for 30.25% of the change in skeletal
muscle. Interestingly, across conditions, PLA2G16 demonstrated the greatest signicant changes in gene expres-
sion. Indeed, loading induced hypertrophy, PLA2G16 displayed a non-signicant increase compared to baseline
in expression (1.09 ± 0.17; Fig. 5AIV), that upon unloading returned back to the baseline levels (1.04 ± 0.25).
Importantly, upon reloading, the expression of PLA2G16 signicantly increased (1.60 ± 0.18; Fig.5AIV) compared
to baseline (P = 0.026) and unloading conditions (P = 0.046), as well as approaching a signicant increase compared
to the initial loading stimulus (P = 0.067 compared to load; Fig.5AIV). HEG 1 gene expression exhibited a signif-
icant correlation with skeletal muscle mass (R = 0.53, P = 0.05) with HEG 1 accounting for 28.09% of the changes
in muscle mass. However, HEG1 did not demonstrate any signicant fold changes in gene expression across the
experimental conditions. Furthermore, no signicant correlation was observed for the other identied cluster of
genes (AXIN1, GRIK2, CAMK4, TRAF1, NR2F6 and RSU1; P > 0.05; Data not shown). Collectively, these data
suggest that RPL35a, UBR5, SETD3 and PLA2G16 all display a signicantly enhanced gene expression upon reload-
ing induced hypertrophy. is suggests, that these genes portray a memory of earlier load induced hypertrophy, by
displaying the largest fold increases in gene expression aer reload induced growth.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
10
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
The E3 Ubiquitin Ligase, UBR5, has enhanced hypomethylation and the largest increase in
gene expression during reloading. e HECT E3 ubiquitin ligase gene UBR5 (Fig.6), for which the CpG
identied is located on chromosome 8 (start 103424372) in the promoter region 546 bp from the transcription
start site, was identied as being within the top 100 most statistically dierentially regulated CpG sites across all
pair-wise conditions (loading, unloading and reloading; Fig.6); but also the transcript that displayed the most
distinctive mirrored-inverse relationship with gene expression (Fig.5CII), aer every condition. Following the
initial period of 7-weeks of load induced hypertrophy, there was a non-signicant increase in UBR5 gene expres-
sion (1.65 ± 0.4) versus baseline, which was met with a concomitant (albeit non-signicant) reduction in CpG
DNA methylation (0.87 ± 0.03). Gene expression returned to baseline control levels aer unloading (0.82 ± 0.27)
demonstrated by a signicant reduction vs. loading (P = 0.05) and non-signicance versus baseline (P = N.S;
Fig.5CII). Aer the same unloading condition, we observed a signicant increase in CpG DNA methylation
compared to baseline (1.27 ± 0.02; P = 0.013; Fig.5CII). Importantly, upon reloading, UBR5 displayed its larg-
est increase in transcript expression, signicantly greater compared to unloading (1.84 ± 0.5 vs. 0.82 ± 0.27,
P = 0.035) and versus baseline levels to the level of P = 0.07. Concomitantly, aer the reloading condition, we
observed the largest statistically signicant reduction in CpG DNA methylation (0.78 ± 0.02) compared to base-
line (P = 0.039), and unloading (P ≤ 0.05; Fig .5CII).
Dynamic changes in DNA methylation after a single acute bout of resistance exercise precede
changes in gene expression after loading and reloading. We next wished to ascertain how dynamic and
transient DNA methylation of the identied genes were, aer a single acute bout of resistance exercise (acute RE).
Figure 5. Relative fold changes in: (A) gene expression; (B) correlation between gene expression and lower
limb leanmass across experimental conditions, and; (C) schematic representation of relationship between fold
changes in CpG DNA methylation (dashed black line; le y axis) and fold change in gene/mRNA expression
(solid black line; right y axis) for identied genes: RPL35a (I), UBR5 (II), SETD3 (III), PLA2G16 (IV) and
HEG1 (V). Statistical signicance compared to baseline and unloading represented by* and** respectively.
All signicance taken as p less than or equal to 0.05 unless otherwise state on graph. All data presented as
mean ± SEM (n = 7/8).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
11
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
We wanted to identify methylation sensitive genes (to single acute resistanceloading stimuli) that were still
aected at the DNA methylation and gene expression levels aer laterchronic load and reload induced hyper-
trophyconditions. We identied that acute loading evoked a greater hypomethylation compared to hypermeth-
ylation response of the human methylome (10,284 hypomethylated sites vs. 7,600 hypermethylated DNA sites;
Fig.7A) with hierarchical clustering analyses displaying distinct dierences between statistically signicant CpG
sites at baseline and acute RE conditions (P < 0.05; total of 17884 CpG sites, Fig.7A). is occurred with a sim-
ilar frequency versus loading where we previously reported 9,153 hypomethylated vs. 8,212 hypermethylated
(8,212) CpG sites (Fig.2A). Overlapping the top 100 signicantly dierentially identied targets from the loading,
unloading and reloading analysis (Supplementary File 4A) together with the 17,884 sites from acute stimulus
analysis (Supplementary File6), identied 27 CpG targets that were signicantly dierentially regulated across
comparisons (Fig.7B). We subsequently removed9 CpG sites that did not map to gene transcripts andwere there-
fore unable to analysefor corresponding gene expression. We identied that the fold change in DNA methylation
pattern of the remaining 18 CpG sites was virtually identical across these conditions (Fig.7C), displaying a signif-
icant correlation across acute RE to loading and reloading conditions (R = 0.94, P < 0.0001; Fig.7D), with follow
up broader hierarchical clustering analysis of the top 500 genes signicantly modied within these conditions
(Fig.7E) also conrming that the majority of sites in were hypomethylated. Suggesting that even aer a single
bout of acuteresistance exercise that the DNA methylation remained the sameaer later load and reload induced
hypertrophy. Interestingly, we identied 4 of the 18 CpG sites identied above (BICC1, GRIK2, ODF2, TRAF1)
that were also identied in our earlier analyses of loading, unloading and reloading conditions (Figs.7A and B).
is suggested that these genes were immediately altered following acute RE, and hypomethylation was retained
Figure 6. Representation and characterisation of the DNA methylation modications that occurred withinthe
ubiquitin mediated proteolysis pathway across all conditions of loading, unloading and reloading compared to
baseline (ANOVA). Signalling analysis performed on statistically dierentially regulated CpG sites compared
to baseline, with green indicating a hypomethylated fold change and red indicating a hypermethylated change,
with strength of colour representing the intensity of fold change23–25. Importantly, the novel HECT-type E3
ubiquitin ligase, UBR5, displays a signicantly hypomethylated state within this pathway.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
12
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
during chronic loading, unloading and subsequent reloading conditions. Finally, we analysed fold changes in
gene expression of a sub set of the 18 CpG sites identied as overlapping in both sets of methylome analysis exper-
iments (Supplementary Figure4a) and compared changes in gene expression to changes in CpG DNA methyla-
tion (Supplementary Figure4B). We identied that signicant hypomethylation upon acute resistance exercise
(Figure7Fi–vii) was not associated with signicant changes in gene expression (Figure7Fi–vii) in a sub-set of
analysed transcripts. However, upon continued loading (chronic loading and reloading conditions), changes in
CpG DNA methylation were associated with signicant changes in a number of these genes uponthe reloading
stimulus (Figure7Fi–vii).Suggesting that these newly identied epigenetically regulated genes (BICC1, GRIK2,
TRAF1 and STAG1) were acutely sensitive to hypomethylation aer a single bout of resistance exercise, that
enhanced gene expression 22 weeks aer a period of load induced hypertrophy, a return of muscle to baseline
and later reloading induced hypertrophy. erefore, the epigenetic regulation of these genes seems to be an early,
acute exercise biomarker of later muscle hypertrophy.
Discussion
Frequency of genome-wide hypomethylation is the largest after reloading induced hypertrophy
where lean muscle mass is enhanced. We aimed to investigate an epigenetic memory of earlier hyper-
trophy in adult human skeletal muscle using a within measures design, by undertaking: (1) resistance exercise
induced muscle growth (loading), followed by; (2) cessation of resistance exercise, to return muscle back towards
baseline levels (unloading), and; (3) a subsequent later period of resistance exercise induced muscle hypertrophy
(reloading). We rst conrmed that we were able to elicit an increase in lean mass of the lower limbs aer 7 weeks
loading, that returned back to baseline levels aer 7 weeks unloading, with 7 weeks reloading evoking the largest
Figure 7. Response of the methylome aer acute resistance loading stimulus compared to baseline, 7 weeks
loading and 7 weeks reloading: (A) Heat map depicting unsupervised hierarchical clustering of statistically
dierentially regulated (P = 0.05) CpG loci following exposure to acute RE compared to baseline; (B) a Venn
diagram depicting the number of CpG sites that were signicantly dierentially regulated in both methylome
analysis experiments (base, loading, unloading and reloading, blue circle; baseline and acute resistance
stimulus, red circle), and the amount of genes analysed for gene expression across acute, 7 weeks loading and 7
weeks reloading, respectively; (C) temporal pattern of fold change in DNA CpG methylation of the identied
overlapping CpG sites that mapped to relevant gene transcripts; (D) correlation of CpG DNA methylation of
acute RE vs. 7 weeks loading and reloading conditions, and(E)Heat map depicting unsupervised hierarchical
clustering of statistically dierentially regulated (P = 0.05) CpG loci following exposure to acute RE compared to
baseline,loading andreloading (F) representative schematic displaying the inverse relationship between mean
gene expression (solid black lines) and CpG DNA methylation (dashed black lines) of identied transcripts.
Signicance indicated in gene expression (*) and in CpG DNA methylation (§) when compared to baseline.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
13
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
increase in lean mass. Interestingly, aer DNA methylation analysis of over 850,000 CpG sites, we identied the
largest frequency of hypomethylation (18,816 CpG sites) occurred aer reloading where the largest lean mass
occurred. Previous studies have suggested that hypermethylation of over 6,500 genes are retained, aer an more
acute stress of high fat intake (for 5 days) 8 weeks later despite removal of the high fat diet14, and hypermethylation
occurs following early life inammatory stress in muscle cells and is maintained for over 30 cellular divisions11. e
present study also suggested hypomethylation was maintained during unloading (8,891 CpG sites) where muscle
mass returned to baseline having being subjected to an earlier period of load induced muscle growth (9,153 CpG
sites), then upon reloading the frequency of hypomethylation was enhanced in association with the largest
increases in lean mass. Furthermore, bioinformatic analysis of the PI3K/AKT pathway across loading, unloading
and reloading conditions, supports the ndings of an enhanced hypomethylated state upon secondary exposure to
resistance stimulus. Importantly, this pathway is identied as critical for cell proliferation/dierentiation, muscle
protein synthesis and therefore muscle hypertrophy27, and therefore, it is plausible that the enhanced hypometh-
ylated state of the genes in this pathways would lead to enhanced gene expression and protein levels. However,
further analysis is required to investigate the total protein or activity of these pathways in this model. Nonetheless,
collectively, these results provide initial evidence for a maintenance/memory of universal hypomethylation. e
only other study to demonstrate a memory of prior hypertrophy in skeletal muscle was in rodents following earlier
encounters with testosterone administration, where a retention of myonuclei occurred even during testosterone
withdrawal and a return of muscle to baseline levels13, suggesting a memory at the cellular level. However, these are
the rst studies to demonstrate that a memory occurs at the epigenetic level within skeletal muscle tissue.
Hypomethylation is maintained from earlier load induced hypertrophy even during unload-
ing where muscle mass returns back towards baseline and is inversely associated with gene
expression. Following the frequencyanalysis of hypo/hypermethylated sites mentionedabove, closer analysis
of the top 500 most signicantly dierentially modied CpG sites across all conditions, identied two epigenet-
ically modied clusters of interest (named Cluster A&B). Cluster B supported the frequency analysis above and
demonstrated hypomethylation aer load induced hypertrophy that was then maintained following unloading
where muscle returned to baseline levels and this hypomethylation was then also maintained aer reload induced
hypertrophy. is maintenance of hypomethylation during unloading, suggested that the muscle ‘remembered’
the epigenetic modications that occurred aer an earlier period of load induced muscle hypertrophy. As reduced
DNA methylation of genes generally leads to enhanced gene expression due to the removal of methylation allow-
ing improved access of the transcriptional machinery and RNA polymerase that enable transcription, and also
creating permissive euchromatin19,28–30, this would be suggestive that the earlier period of hypertrophy leads
to increased gene expression of this cluster of genes that is then retained during unloading to enable enhanced
muscle growth in the later reloading period. To conrm this, in a separate analysis we identied the top 100 most
signicantly dierentially modied CpG sites across all conditions and cross referenced these with the most
frequently occurring CpG modications in all pairwise comparisons of experimental conditions. From this we
identied 48 genes that were frequently occurring in all pairwise comparisons and examined gene expression by
rt-qRT-PCR. Interestingly, we identied two clusters of genes with distinct temporal expression aer loading,
unloading and reloading. One of the clusters included AXIN1, GRIK2, CAMK4, TRAF1. Importantly, the major-
ity of these genes demonstrated a mirror/inverse relationship with DNA methylation of the CpG sites within the
same genes. Where DNA methylation reduced aer loading and remained low into unloading and reloading,
gene expression accumulated, demonstrating the highest expression aer reloading where the largest increase in
lean mass was also demonstrated. Overall, this suggested that these genes were hypomethylated and switched on
aer the earlier period of load induced hypertrophy, maintained during unloading due to methylation of these
genes remaining low, and then upon exposure to a later period of reload induced hypertrophy, these genes were
switched on to an even greater extent. Overall, this demonstrates that the methylation and collective responsive-
ness of these genes are important epigenetic regulators of skeletal muscle memory.
Interestingly, AXIN1 is a component of the beta-catenin destruction complex, where in skeletal muscle cells
AXIN1 has been shown to inhibit WNT/β-catenin signalling and enable differentiation31, where treatment
with the canonical WNT ligand suppresses dierentiation32. Other studies suggested that AXIN2 not AXIN1 is
increased aer dierentiation, however conrmed that the absence of AXIN1 reduced proliferation and myotube
formation32. erefore, together with the present data perhaps suggest an important epigenetic regulation of
AXIN1 involved in human skeletal muscle memory and hypertrophy at the tissue level, perhaps due to inhibition
of WNT/β-catenin signaling. GRIK2 (glutamate ionotropic receptor kainate type subunit 2, a.k.a. GluK2) belongs
to the kainate family of glutamate receptors, which are composed of four subunits and function as ligand-activated
ion channels33. Although reportedly expressed in skeletal muscle, its role in muscle growth or cellular function has
not been determined. CAMK4 is calcium/calmodulin-dependent protein kinase, that via phosphorylation, trig-
gers the CaMKK-CaMK4 signaling cascade and activates several transcription factors, such as MEF234. MEF2 has
been previously associated with a switch to slow bre types aer exercise35 and is hypomethylated aer 6 months
aerobic exercise36. While resistance exercise has been show to preferentially increase the size of type II faster bres,
chronic innervation even at higher loads can lead to an overall slowing in phenotype [reviewed in ref.37) and
therefore this epigenetically regulated gene, although not usallystudied during hypertrophy maybe important in
bre type changes in the present study. However, it is unknown how DNA methylation aects the protein levels
of CAMK4, and with its role in phosphorylation, would be important to ascertain in the future. Furthermore,
bre type properties were not analyzed in the present study and thereforerequire further investigation. TRAF1
is the TNF receptor-associated factor 1 and together with TRAF2 form the heterodimeric complex required for
TNF-α activation of MAPKs, JNK and NFκB38. In skeletal muscle, acute TNF exposure activates proliferation
via activation of MAPKs such as ERK and P38 MAPK39–41. erefore, acutely elevated systemic TNF-α following
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
14
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
damaging exercise such as resistance exercise correlates positively with satellite cell activation in-vivo aer dam-
aging exercise42,43, yet chronic administration in-vitro inhibits dierentiation, promotes myotube atrophy40,44 and
muscle wasting in-vivo44. Indeed, exposure to early life TNF-α during an early proliferative age in mouse C2C12s
results in maintenance of hypermethylation in the myoD promoter aer 30 divisions and an increased susceptibil-
ity to reduced dierentiation and myotube atrophy when muscle cells encounter TNF-α in later proliferative life11.
Suggesting a role for DNA methylation in retention of memory following earlier periods of high inammation.
Because resistance exercise evokes increases in TNF-α in the systemic circulation and has been shown increase
locally in muscle at the protein level (discussed above), these data collectively suggest an interesting epigenetic
role for TNF and TRAF1 in the epigenetic memory of earlier load induced muscle hypertrophy.
Identication of novel genes with the largest hypomethylation during reloading that are
associated with enhanced gene expression. e second DNA methylation cluster determined in the
top 500 dierentially modied CpG sites across all conditions, identied a cluster of genes (named Cluster A)
that was methylated at baseline and also became hypomethylated aer loading (similar to Cluster B above),
then, upon unloading, genes reverted back to a methylated state, and aer reloading switched back to hypo-
methylated. erefore, while not demonstrating an epigenetic memory per se, if hypomethylation was further
enhanced and was associated with enhanced gene expression in reloading versus loading would also sup-
port an epigenetic memory. Further gene expression analysis identied a cluster of genes that demonstrated
a mirror/inverse temporal pattern of gene expression versus their DNA methylation pattern. ese genes
included RPL35a, C12orf50, BICC1, ZFP2, UBR5, HEG1, PLA2G16, SETD3 and ODF2, that demonstrated
hypomethylation of DNA aer load induced growth and an increase in gene expression. Subsequently, then
both DNA methylation and gene expression returned back to baseline levels (in opposite directions) and aer
reload induced muscle growth DNA was hypomethylated again with an associated increase in gene expression.
Importantly, during reloading, gene expression was further enhanced versus loading, suggesting that an earlier
period of load induced growth was enough to produceenhanced gene expression when reload induced muscle
growth was encountered later, again suggesting a skeletal muscle memory at both the epigenetic andresultant
transcript level. Statistical analysis identied the genes RPL35a, UBR5, SETD3 and PLA2G16 as having signif-
icantly enhanced expression upon reloading. Importantly, these four genes, plus HEG1, displayed signicant
correlations between their gene expression and the percentage change in lean mass, suggesting for the rst time,
a role for these four genes in regulating adult human load induced skeletal muscle growth. Interestingly, SET
Domain Containing 3 (SETD3) is a H3K4/H3K36 methyltransferase, is abundant in skeletal muscle, and has
been shown to be recruited to the myogenin promoter, with MyoD, to promote its expression45. Furthermore,
overexpression of SETD3 in C2C12 murine myoblasts, evokes increases in myogenin, muscle creatine kinase,
and Myf6 (or MRF4) gene expression. Inhibition via shRNA in a myoblasts also impairs muscle cell dierentia-
tion45, suggesting a role for SETD3 in regulating skeletal muscle regeneration. However, less is known regarding
the role of PLA2G16 in skeletal muscle. PLA2G16 is a member of the superfamily of phospholipase A enzymes,
whose predominant localization is in adipose tissue. PLA2G16 is known to regulate adipocyte lipolysis in an
autocrine/paracrine manner, via interactions with prostaglandin and EP3 in a G-protein-mediated pathway46.
Indeed, ablation of PLA2G16 (referred to as Adpla), prevents obesity during periods high fat feeding in mouse
models, indicated via signicantly less adipose tissue and triglyceride content, compared to relevant controls46.
However, to date no known research has elucidated the role of PLA2G16 in skeletal muscle and therefore,
this requires future experimentation. Finally, HEG homology 1 (HEG1), initially reported as the heart of glass
gene, is recognised for its role in regulating the zebrash heart growth. HEG1 is a transmembrane receptor
that has been reported to be fundamental in the development of both the heart and blood vessels47. However,
a recent study reported a distinct role for HEG1 in regulating malignant cell growth48. Tsuji, et al.48 and col-
leagues reported that gene silencing of HEG1 in human MPM cell line, a cell linage that develop mesothelioma
tumours, signicantly reduced the survival and proliferation of mesothelioma cells, suggesting a role for HEG1
in regulating cellular growth. However, no known research has examined the role of HEG1 in regulating adult
skeletal muscle growth.
In the present study UBR5 displayed the most distinctive inverse relationship between DNA hypomethylation
and increased gene expressionfollowing loading and reloading. With the largest increase in hypomethylation and
gene expression aer reloading where the largest increase in lean mass was observed. UBR5 is a highly conserved
homologue of the drosophila tumour suppressor hyperplastic discs (HYD), and in the mammalian genome refers
to a protein that is a member of the HECT-domain E3 ubiquitin-ligase family49. E3 ubiquitin ligases play an
integral role in the ubiquitin - proteasome pathway, providing the majority of substrate recognition for the attach-
ment of ubiquitin molecules onto targeted proteins, preferentially modifying them for targeted autophagy/break-
down50. Indeed, extensive work has identied a distinct role of a number of E3 ubiquitin ligases such as MuRF1,
MAFbx and MUSA1 in muscle atrophy51,52. Furthermore, we have recently demonstrated that reduced DNA
methylation and increased gene expression of MuRF1 and MAFbx are associated with disuse atrophy in rats fol-
lowing nerve silencing of the hind limbs via tetrodotoxin exposure17. A process that is reversed upon a return to
habitual physical activity and a partial recovery of skeletal muscle mass17, suggesting a role for DNA methylation
in regulating the transcript behavior of a number of ubiquitinligases during periods of skeletal muscle atrophy
and recovery. However, there have been no studies that the authors are aware of, exmaining the role of UBR5 in
skeletal muscle atrophy or growth. Given the role of ubiquitin ligases in skeletal muscle, counterintuitively, we
report that the expression of the E3 ubiquitin ligase, UBR5, is increased during earlier periods of skeletal muscle
hypertrophy and are even further enhanced in later reload induced muscle growth. We further report that the
methylation prole of this E3 ubiquitin ligase portrays an inversed relationship with gene expression, support-
ing a role for DNA epigenetic modications in regulating its expression, as previously suggested17. However, in
support of its role in positively impacting on muscle, UBR5 has also been shown to promote smooth muscle
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
15
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
dierentiation through its ability to stabilize myocardin proteins53. While myocardin is only expressed in smooth
and cardiac muscle, it is considered the master regulator of smooth muscle gene expression54 and a known tran-
scription factor that upregulates smooth muscle myosin heavy chains (MYHCs), actin and desmin. It therefore
possesses a similar role to the myogenic regulatory factors during early dierentiation (Mrf5 and MyoD), dur-
ing fusion (myogenin) and during myotube hypertrophy (adult MYHC’s). Interestingly, it has previously been
observed that myocardin-related transcription factors (MRTF) interact with the myogenic regulatory factor,
MyoD, to activate skeletal muscle specic gene expression55, suggesting a potential cross-talk between muscle
specic regulatory factors, enabling skeletal muscle adaptations55,56. erefore, UBR5′s expression throughout
the time course of skeletal muscle cell dierentiation, its role in myotube hypertrophy are required in-vitro as well
as mammalian overexpression and knock-out of UBR5 to conrm its importance in-vivo. Further work is needed
to characterize UBR5, as well as other HECT-domain E3 ubiquitin ligase protein members identied in this work
via pathway analysis of the ubiquitin mediated proteolysis pathway, in the development of muscle growth to better
understand its role in facilitating skeletal muscle hypertrophy.
A single bout of acute resistance exercise evokes hypomethylation of genes that have
enhanced gene expression in later reload induced hypertrophy: Novel acutely exercise sensi-
tive DNA methylation biomarkers. Finally, we identied genes BICC1, STAG1, GRIK2 and TRAF1 were
hypomethylated aer a single bout of acute resistance exercise that were maintained as hypomethylated during
loading (as identied above) and reloading and demonstrated an enhanced gene expression aer later reloading.
Previous studies have suggested that acute aerobic exercise hypomethylates important genes in metabolic adapta-
tion and mitochondrial biogenesis such as PGC-1α, mitochondrial transcription factor A (TFAM) and pyruvate
dehydrogenase lipoamide kinase isozyme 4 (PDK4) post exercise, and reduces PPAR-δ methylation (hypometh-
ylates) 3 hours post exercise16, with corresponding increases in gene expression (3 hrs post exercise for PGC-1α,
PDK4 and PPAR-δ, immediately post for TFAM)16. Interestingly, hypermethylation of PGC 1α and reduced gene
expression, observed in skeletal muscle of the ospring of obese murine mothers, was reversed (hypomethylated)
by exercise in the mothers4. ese data support the role for aerobic exercise in hypomethylating candidate genes.
We also identify in the present study that hypomethylation (10,284 CpG sites) is favoured over hypermethylation
(7,600 CpG sites) across the genome 30 minutes post an acute bout of resistance exercise, yet without changes
in gene expression at this time point. Interestingly, however, hypomethylation of BICC1, STAG1, GRIK2 and
TRAF1 aer acute RE that was maintained aer 7 weeks loading and reloadinginduced hypertrophy, resulted in
signicantly enhanced gene expression 22 weeks later. is suggested that DNA methylation of these genes aer
a single bout of resistance exercise were more sensitive biomarkers than their acutely corresponding gene expres-
sion for later load induced hypertrophy. BICC1 is an RNA binding protein that has an undermined role in adult
skeletal muscle. It has been identied as dierentially expressed during prenatal muscle development between
two dierent pig breads57. RNA binding proteins in general are important in post transcriptional modications,
suggesting that perhaps reduced DNA methylation and increased gene expression may indicate an increase in
post-transcriptional modication aer reloading, however this requires further investigation to conrm. STAG1
(Cohesin subunit SA-1) isfundamental in cell division andpart of the cohesin complex, which is required for
the cohesion of sister chromatids aer DNA replication58. However, to the authors knowledge there is no spe-
cic rolefor STAG1 identied in adult skeletal muscle hypertrophy. GRIK2 and TRAF2 were also identied as
being hypomethylated aer loading and reloading together with enhanced gene expression. As suggested above,
GRIK2′s role in skeletal muscle is not well dened. However, TRAF1 has been widely implicated in skeletal muscle
cell proliferation and dierentiation, as discussed above, and hypomethylation of TRAF1 appears to be both sensi-
tive to acute RE, as well as maintained following repeated loading and reloading induced hypertrophy that resulted
in the largest increase in gene expression aer reloading, 22 weeks aer being detected as hypomethylated aer
acute RE. Overall, suggesting an important role for TRAF2 in skeletal muscles epigenetic memory of hypertrophy.
Conclusion
We identify that human skeletal muscle possesses an epigenetic memory of earlier acute and chronic anabolic
stimuli when encountering later muscle hypertrophy.
References
1. Sharples, A. P., Stewart, C. E. & Seaborne, . A. Does seletal muscle have an ‘epi’-memory? e role of epigenetics in nutritional
programming, metabolic disease, aging and exercise. Aging cell 15, 603–616, https://doi.org/10.1111/acel.12486 (2016).
2. Patel, H. P. et al. Developmental inuences, muscle morphology, and sarcopenia in community-dwelling older men. J Gerontol A
Biol Sci Med Sci 67, 82–87, https://doi.org/10.1093/gerona/glr020 (2012).
3. Patel, H. P. et al. Lean mass, muscle strength and gene expression in community dwelling older men: ndings from the Hertfordshire
Sarcopenia Study (HSS). Calcied tissue international 95, 308–316, https://doi.org/10.1007/s00223-014-9894-z (2014).
4. Laer, . C. et al. Exercise prevents maternal high-fat diet-induced hypermethylation of the Pgc-1alpha gene and age-dependent
metabolic dysfunction in the ospring. Diabetes 63, 1605–1611, https://doi.org/10.2337/db13-1614 (2014).
5. Zeng, Y., Gu, P., Liu, . & Huang, P. Maternal protein restriction in rats leads to reduced PGC-1alpha expression via altered DNA
methylation in seletal muscle. Molecular medicine reports 7, 306–312, https://doi.org/10.3892/mmr.2012.1134 (2013).
6. Jaenisch, . & Bird, A. Epigenetic regulation of gene expression: how the genome integrates intrinsic and environmental signals.
Nature genetics 33, Suppl, 245–254, https://doi.org/10.1038/ng1089 (2003).
7. Green, C. J., Bunprajun, T., Pedersen, B. . & Scheele, C. Physical activity is associated with retained muscle metabolism in human
myotubes challenged with palmitate. J Physiol 591, 4621–4635, https://doi.org/10.1113/jphysiol.2013.251421 (2013).
8. Aguer, C. et al. Intramyocellular lipid accumulation is associated with permanent relocation ex vivo and in vitro of fatty acid
translocase (FAT)/CD36 in obese patients. Diabetologia 53, 1151–1163, https://doi.org/10.1007/s00125-010-1708-x (2010).
9. Maples, J. M. et al. Lipid exposure elicits dierential responses in gene expression and DNA methylation in primary human seletal
muscle cells from severely obese women. Physiological genomics 47, 139–146, https://doi.org/10.1152/physiolgenomics.00065.2014
(2015).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
16
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
10. Foulstone, E. J., Savage, P. B., Crown, A. L., Holly, J. M. & Stewart, C. E. Adaptations of the IGF system during malignancy: human
seletal muscle versus the systemic environment. Horm Metab es 35, 667–674, https://doi.org/10.1055/s-2004-814159 (2003).
11. Sharples, A. P. et al. Seletal muscle cells possess a ‘memory’ of acute early life TNF-alpha exposure: role of epigenetic adaptation.
Biogerontology 17, 603–617, https://doi.org/10.1007/s10522-015-9604-x (2016).
12. Egner, I. M., Bruusgaard, J. C., Eestøl, E. & Gundersen, . A cellular memory mechanism aids overload hypertrophy in muscle long
after an episodic exposure to anabolic steroids. The Journal of Physiology 591, 6221–6230, https://doi.org/10.1113/
jphysiol.2013.264457 (2013).
13. Bruusgaard, J. C., Johansen, I. B., Egner, I. M., ana, Z. A. & Gundersen, . Myonuclei acquired by overload exercise precede
hypertrophy and are not lost on detraining. Proc Natl Acad Sci USA 107, 15111–15116, https://doi.org/10.1073/pnas.0913935107
(2010).
14. Jacobsen, S. C. et al. Eects of short-term high-fat overfeeding on genome-wide DNA methylation in the seletal muscle of healthy
young men. Diabetologia 55, 3341–3349, https://doi.org/10.1007/s00125-012-2717-8 (2012).
15. Barres, . et al. Non-CpG methylation of the PGC-1alpha promoter through DNMT3B controls mitochondrial density. Cell Metab
10, 189–198, https://doi.org/10.1016/j.cmet.2009.07.011 (2009).
16. B arres, . et al. Acute exercise remodels promoter methylation in human seletal muscle. Cell Metab 15, 405–411, https://doi.
org/10.1016/j.cmet.2012.01.001 (2012).
17. Fisher, A. et al. Transcriptomic and Epigenetic egulation of Disuse Atrophy and the eturn to Activity in Seletal Muscle. Faseb J
31, 5268–5282, https://doi.org/10.1096/.201700089 (2017).
18. Bigot, A. et al. Age-Associated Methylation Suppresses SPY1, Leading to a Failure of e-quiescence and Loss of the eserve Stem
Cell Pool in Elderly Muscle. Cell reports 13, 1172–1182, https://doi.org/10.1016/j.celrep.2015.09.067 (2015).
19. Bogdanovic, O. & Veenstra, G. J. DNA methylation and methyl-CpG binding proteins: developmental requirements and function.
Chromosoma 118, 549–565, https://doi.org/10.1007/s00412-009-0221-9 (2009).
20. Peterson, M. D., Pistilli, E., Ha, G. G., Homan, E. P. & Gordon, P. M. Progression of volume load and muscular adaptation during
resistance exercise. European journal of applied physiology 111, 1063–1071, https://doi.org/10.1007/s00421-010-1735-9 (2011).
21. Masimovic, J., Gordon, L. & Oshlac, A. SWAN: Subset-quantile within array normalization for illumina infinium
HumanMethylation450 BeadChips. Genome biology 13, 44, https://doi.org/10.1186/gb-2012-13-6-r44 (2012).
22. Pidsley, . et al. Critical evaluation of the Illumina MethylationEPIC BeadChip microarray for whole-genome DNA methylation
proling. Genome biology 17, 208, https://doi.org/10.1186/s13059-016-1066-1 (2016).
23. anehisa, M. & Goto, S. EGG: yoto encyclopedia of genes and genomes. Nucleic acids research 28, 27–30 (2000).
24. anehisa, M., Furumichi, M., Tanabe, M., Sato, Y. & Morishima, . EGG: new perspectives on genomes, pathways, diseases and
drugs. Nucleic acids research 45, D353–d361, https://doi.org/10.1093/nar/gw1092 (2017).
25. anehisa, M., Sato, Y., awashima, M., Furumichi, M. & Tanabe, M. EGG as a reference resource for gene and protein annotation.
Nucleic acids research 44, D457–462, https://doi.org/10.1093/nar/gv1070 (2016).
26. Egerman, M. A. & Glass, D. J. Signaling pathways controlling seletal muscle mass. Critical eviews in Biochemistry and Molecular
Biology 49, 59–68, https://doi.org/10.3109/10409238.2013.857291 (2014).
27. Schiano, S. & Mammucari, C. egulation of seletal muscle growth by the IGF1-At/PB pathway: insights from genetic models.
Seletal Muscle 1, 4, https://doi.org/10.1186/2044-5040-1-4 (2011).
28. Fus, F. et al. e methyl-CpG-binding protein MeCP2 lins DNA methylation to histone methylation. e Journal of biological
chemistry 278, 4035–4040, https://doi.org/10.1074/jbc.M210256200 (2003).
29. Lunya, V. V. et al. Corepressor-dependent silencing of chromosomal regions encoding neuronal genes. Science (New Yor, N.Y.)
298, 1747–1752, https://doi.org/10.1126/science.1076469 (2002).
30. ountree, M. . & Seler, E. U. DNA methylation inhibits elongation but not initiation of transcription in Neurospora crassa. Genes
& development 11, 2383–2395 (1997).
31. Figeac, N. & Zammit, P. S. Coordinated action of Axin1 and Axin2 suppresses beta-catenin to regulate muscle stem cell function.
Cellular signalling 27, 1652–1665, https://doi.org/10.1016/j.cellsig.2015.03.025 (2015).
32. Hurasin, D. et al. Wnt/beta-catenin signaling via Axin2 is required for myogenesis and, together with YAP/Taz and Tead1, active
in IIa/IIx muscle bers. Development 143, 3128–3142, https://doi.org/10.1242/dev.139907 (2016).
33. Han, Y., Wang, C., Par, J. S. & Niu, L. Channel-opening inetic mechanism for human wild-type Glu2 and the M867I mutant
ainate receptor. Biochemistry 49, 9207–9216, https://doi.org/10.1021/bi100819v (2010).
34. Blaeser, F., Ho, N., Prywes, . & Chatila, T. A. Ca2+-dependent Gene Expression Mediated by MEF2 Transcription Factors. Journal
of Biological Chemistry 275, 197–209, https://doi.org/10.1074/jbc.275.1.197 (2000).
35. Wu, H. et al. MEF2 responds to multiple calcium-regulated signals in the control of seletal muscle ber type. Embo j 19, 1963–1973,
https://doi.org/10.1093/emboj/19.9.1963 (2000).
36. Nitert, M. D. et al. Impact of an exercise intervention on DNA methylation in seletal muscle from rst-degree relatives of patients
with type 2 diabetes. Diabetes 61, 3322–3332, https://doi.org/10.2337/db11-1653 (2012).
37. Fry, A. C. The ole of esistance Exercise Intensity on Muscle Fibre Adaptations. Sports Medicine 34, 663–679, https://doi.
org/10.2165/00007256-200434100-00004 (2004).
38. Pomerantz, J. L. & Baltimore, D. NF-appaB activation by a signaling complex containing TAF2, TAN and TB1, a novel I-
related inase. Embo j 18, 6694–6704, https://doi.org/10.1093/emboj/18.23.6694 (1999).
39. Foulstone, E. J., Huser, C., Crown, A. L., Holly, J. M. & Stewart, C. E. Dierential signalling mechanisms predisposing primary
human seletal muscle cells to altered proliferation and dierentiation: roles of IGF-I and TNFalpha. Exp Cell es 294, 223–235,
https://doi.org/10.1016/j.yexcr.2003.10.034 (2004).
40. Girven, M. et al. l-glutamine Improves Seletal Muscle Cell Dierentiation and Prevents Myotube Atrophy Aer Cytoine (TNF-alpha)
Stress Via educed p38 MAP Signal Transduction. Journal of cellular physiology 231, 2720–2732, https://doi.org/10.1002/jcp.25380
(2016).
41. Li, Y. P. TNF-alpha is a mitogen in seletal muscle. Am J Physiol Cell Physiol 285, C370–376, https://doi.org/10.1152/
ajpcell.00453.2002 (2003).
42. Macey, A. L. et al. e inuence of anti-inammatory medication on exercise-induced myogenic precursor cell responses in
humans. J Appl Physiol 103, 425–431, https://doi.org/10.1152/japplphysiol.00157.2007 (2007).
43. van de Vyver, M. & Myburgh, . H. Cytoine and satellite cell responses to muscle damage: interpretation and possible confounding
factors in human studies. J Muscle es Cell Motil 33, 177–185, https://doi.org/10.1007/s10974-012-9303-z (2012).
44. Li, Y. P. et al. TNF-alpha acts via p38 MAP to stimulate expression of the ubiquitin ligase atrogin1/MAFbx in seletal muscle. Faseb
J 19, 362–370, https://doi.org/10.1096/.04-2364com (2005).
45. Eom, G. H. et al. Histone Methyltransferase SETD3 egulates Muscle Differentiation. Journal of Biological Chemistry 286,
34733–34742, https://doi.org/10.1074/jbc.M110.203307 (2011).
46. Jaworsi, . et al. AdPLA ablation increases lipolysis and prevents obesity induced by high-fat feeding or leptin deciency. Nature
medicine 15, 159–168, https://doi.org/10.1038/nm.1904 (2009).
47. leaveland, B. et al. egulation of cardiovascular development and integrity by the heart of glass-cerebral cavernous malformation
protein pathway. Nature medicine 15, 169–176, https://doi.org/10.1038/nm.1918 (2009).
48. Tsuji, S. et al. HEG1 is a novel mucin-lie membrane protein that serves as a diagnostic and therapeutic target for malignant
mesothelioma. Scientic reports 7, 45768, https://doi.org/10.1038/srep45768 (2017).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
www.nature.com/scientificreports/
17
SCIENTIfIC REPORtS | (2018) 8:1898 | DOI:10.1038/s41598-018-20287-3
49. Callaghan, M. J. et al. Identication of a human HECT family protein with homology to the Drosophila tumor suppressor gene
hyperplastic discs. Oncogene 17, 3479–3491, https://doi.org/10.1038/sj.onc.1202249 (1998).
50. Buetow, L. & Huang, D. T. Structural insights into the catalysis and regulation of E3 ubiquitin ligases. Nature reviews. Molecular cell
biology 17, 626–642, https://doi.org/10.1038/nrm.2016.91 (2016).
51. Bodine, S. C. et al. Identication of ubiquitin ligases required for seletal muscle atrophy. Science 294, 1704–1708, https://doi.
org/10.1126/science.1065874 (2001).
52. Sartori, . et al. BMP signaling controls muscle mass. Nature genetics 45, 1309–1318, https://doi.org/10.1038/ng.2772 (2013).
53. Hu, G. et al. Modulation of myocardin function by the ubiquitin E3 ligase UB5. The Journal of biological chemistry 285,
11800–11809, https://doi.org/10.1074/jbc.M109.079384 (2010).
54. Wang, Z., Wang, D. Z., Pipes, G. C. & Olson, E. N. Myocardin is a master regulator of smooth muscle gene expression. Proceedings
of the National Academy of Sciences of the United States of America 100, 7129–7134, https://doi.org/10.1073/pnas.1232341100 (2003).
55. Meadows, S. M., Warman, A. S., Salanga, M. C., Small, E. M. & rieg, P. A. e myocardin-related transcription factor, MAST,
cooperates with MyoD to activate seletal muscle gene expression. Proceedings of the National Academy of Sciences of the United
States of America 105, 1545–1550, https://doi.org/10.1073/pnas.0703918105 (2008).
56. Long, X., Creemers, E. E., Wang, D. Z., Olson, E. N. & Miano, J. M. Myocardin is a bifunctional switch for smooth versus seletal
muscle dierentiation. Proceedings of the National Academy of Sciences of the United States of America 104, 16570–16575, https://doi.
org/10.1073/pnas.0708253104 (2007).
57. Muráni, E., Murániová, M., Ponsusili, S., Schellander, . & Wimmers, . Identication of genes dierentially expressed during
prenatal development of seletal muscle in two pig breeds diering in muscularity. BMC Developmental Biology 7, 109, https://doi.
org/10.1186/1471-213x-7-109 (2007).
58. ong, X. et al. Distinct functions of human cohesin-SA1 and cohesin-SA2 in double-strand brea repair. Mol Cell Biol 34, 685–698,
https://doi.org/10.1128/mcb.01503-13 (2014).
Acknowledgements
is work was funded by a PhD studentship for Robert A. Seaborne by the Doctoral Training Alliance UK/LJMU/
Keele University awarded via Adam P. Sharples (PI). Genome-wide methylation/gene expression studies were
funded by a GlaxoSmithKline grant awarded to Adam P. Sharples (PI).
Author Contributions
Sharples concieved experiments,Sharples and Seaborne designed experiments and research methodology,
performed the research and data collection, analysed all the data and wrote the manuscript. Sharples, Seaborne,
Strauss, Cocks, Shepherd, O’Brien, van Someren, Bell, Murgatroyd, Morton, Stewart provided expertise for
sample, data collection and analysis. All authors reviewed the manuscript drafts and inputted corrections,
amendments and their expertise.
Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-20287-3.
Competing Interests: e authors declare that they have no competing interests.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2018
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
Content uploaded by Adam Philip Sharples
Author content
All content in this area was uploaded by Adam Philip Sharples on Jan 30, 2018
Content may be subject to copyright.
