J Appl Physiol 115: 1237–1244, 2013.
First published August 15, 2013; doi:10.1152/japplphysiol.00075.2013.
Changes in circulating microRNAs levels with exercise modality
Sébastien Banzet,1 Mounir Chennaoui,1 Olivier Girard,2 Sébastien Racinais,2 Catherine Drogou,1
Hakim Chalabi,2 and Nathalie Koulmann1,3
1
2
Institut de Recherche Biomédicale des Armées, Armed Forces Biomedical Research Institute, Brétigny sur Orge, France;
Aspetar, Qatar Othopaedic and Sports Medicine Hospital, Doha, Qatar; and 3Ecole du Val-de-Grâce, Paris, France
Submitted 22 January 2013; accepted in final form 12 August 2013
muscle miRNAs; cardiac miRNAs; muscle injury; biomarkers
are able to send
biological signals such as myokines or natriuretic peptides to
other tissues. In contrast, in some conditions, molecules released by muscles are not signals but markers of tissue damage.
Exercise-induced muscle damage is very frequent following
unaccustomed eccentric exercise, which involves the active
lengthening of muscle during tasks such as walking downhill,
and results in perturbation of myofiber membrane integrity
allowing cytosolic or contractile proteins to leak out in bloodstream.
Besides peptides and proteins, microRNAs (miRNA) are of
increasing interest as signaling molecules or injury biomarkers
but have currently been scarcely studied in exercise physiology. MicroRNAs are small noncoding RNA involved in the
posttranscriptional regulation of gene expression and play a
role in a wide range of biological and pathological processes.
DURING EXERCISE, SKELETAL AND CARDIAC MUSCLES
Address for reprint requests and other correspondence: S. Banzet, Institut de
Recherche Biomédicale des Armées, BP-73, 91223 Brétigny sur Orge, France
(e-mail: [email protected]).
http://www.jappl.org
Numerous miRNAs are highly tissue and cell specific (15, 18).
For instance, muscle-specific miRNAs such as miR-1, 133a,
133b, 206, 208a, 208b, and 499 have been described in skeletal
and/or cardiac muscles (10). They are involved in cardiac and
skeletal muscle development, myoblast proliferation or differentiation and tissue remodeling. Their functions and roles
have been recently reviewed (11, 19, 33). Changes in muscle
miRNAs expression have been described in the presence of
altered contractile activity. Hence, muscle unloading induced a
significant decrease in miR-206, 208b, and 499 (1, 21) within
muscle tissue. In a functional overload model, primary transcripts for miR-1, 206, and 133a were increased, with no changes
or a decrease in the corresponding mature miRNAs (20). Finally,
endurance exercise transiently up-regulated miR-1 expression in
mice, while both miR-1 and 133a in human skeletal muscle
increased immediately after exercise cessation (25, 30). In addition, other miRNAs, not restricted to muscle tissue, have been
shown to play important roles in muscle cells; e.g., mir-181 is
up-regulated in regenerating skeletal muscle (24) and is increased
in mice quadriceps after a running bout (30). Mir-214 is involved
in muscle cell differentiation (13) and in cardiomyocyte survival
during cardiac injury (2). Thus current literature shows that
miRNAs expression is altered within muscle in response to
changes in contractile activity or muscle damage, suggesting they
could be involved in muscle adaptation or repair.
Recently, noncellular miRNA have been found in various body
fluids including plasma or serum. There is growing interest in
using circulating miRNA as biomarkers of various pathologies,
especially in cancer detection (22), but also as biomarkers of
tissue injuries (17). In rats, systemic administration of muscle
toxic drugs induces an increase in plasma miR-133 (17) and in
rodents as well as in human, circulating cardiac miRNAs are
specific and early biomarkers of cardiac infarction (35). Additionally, accumulating experimental arguments suggest that secreted
miRNAs are mediators of intercellular communication [review in
(8)]. In a pioneer study, Baggish et al. (3) investigated circulating
levels of various miRNAs in healthy subjects and described a
variety of expression profiles at the end of an acute exhaustive
cycling exercise bout and/or after 90 days endurance training.
This work demonstrated that exercise can affect circulating
miRNA expression, and the authors proposed they could be
exercise biomarkers or possible physiological mediators of exercise-induced adaptations. While this work focused on angiogenesis and inflammation-related miRNA, studies investigating circulating miRNAs acute changes in response to exercise are scarce.
In the present work, our goal was to investigate if circulating
muscle-specific (hsa-mir-1, 133a, 133b, 208a, 208b, and 499)
or muscle related (hsa-mir-181 and 214) miRNAs are affected
by exercise in human, and whether exercise modality (i.e.,
damaging vs. nondamaging exercise modes) can differentially
influence their expression. To achieve this goal, we alterna-
8750-7587/13 Copyright © 2013 the American Physiological Society
1237
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
Banzet S, Chennaoui M, Girard O, Racinais S, Drogou C,
Chalabi H, Koulmann N. Changes in circulating microRNAs levels
with exercise modality. J Appl Physiol 115: 1237–1244, 2013. First
published August 15, 2013; doi:10.1152/japplphysiol.00075.2013.—
Here, we studied muscle-specific and muscle-related miRNAs in
plasma of exercising humans. Our aim was to determine whether they
are affected by eccentric and/or concentric exercise modes and could
be biomarkers of muscle injuries or possible signaling molecules. On
two separate days, nine healthy subjects randomly performed two
30-min walking exercises, one downhill (high eccentric component)
and one uphill (high concentric component). Perceived exertion and
heart rate were higher during the uphill exercise, while subjective pain
and ankle plantar flexor strength losses within the first 48-h were
higher following the downhill exercise. Both exercises increased
serum creatine kinase and myoglobin with no significant differences
between conditions. Plasma levels of circulating miRNAs assessed
before, immediately after, and at 2-, 6-, 24-, 48-, and 72-h recovery
showed that 1) hsa-mir-1, 133a, 133b, and 208b were not affected
by concentric exercise but significantly increased during early recovery of eccentric exercise (2 to 6 h); 2) hsa-mir-181b and 214 significantly and transiently increased immediately after the uphill, but not
downhill, exercise. The muscle-specific hsa-mir-206 was not reliably
quantified and cardiac-specific hsa-mir-208a remained undetectable.
In conclusion, changes in circulating miRNAs were dependent on the
exercise mode. Circulating muscle-specific miRNAs primarily responded to a downhill exercise (high eccentric component) and could
potentially be alternative biomarkers of muscle damage. Two musclerelated miRNAs primarily responded to an uphill exercise (high
exercise intensity), suggesting they could be markers or mediators of
physiological adaptations.
1238
Circulating miRNAs and Exercise
tively measured plasma levels of miRNAs in response to an
uphill (mainly concentric exercise) or a downhill exercise (high
eccentric component), together with classical markers of exercise-induced muscle damage (muscle soreness, plasma markers, muscle strength) and exercise intensity (heart rate and
ratings of perceived exertion). We postulated that a higher
miRNA expression in response to concentric exercise would
suggest miRNAs are signaling molecules, whereas higher
miRNA in response to the eccentric exercise would suggest
muscle miRNA are released in response to muscle damage.
METHODS
Banzet et al.
Oy, Kempele, Finland). Ratings of perceived exertion (RPE) were
obtained every 5 min using the 6 –20 Borg scale.
Delayed onset muscle soreness assessment. A subjective evaluation
of the extent of DOMS in the plantar flexor muscles was performed
prior to assess muscle strength by using a 10-cm visual scale ranging
from 0 (“normal absence of soreness”) to 10 (“extremely painful”).
They were instructed to indicate the general soreness of the entire
muscle area when moving or using it.
Strength assessment. Isometric ankle plantar-flexion torque of the
right foot was measured using a dynamometric pedal (Captels, St
Mathieu de Treviers, France). The subject’s seating position was
standardized with pelvis, knee, and ankle angulations of 90°, the foot
securely strapped on the pedal by three straps. Subjects were instructed to perform three maximal isometric contractions for a 4- to
5-s duration with a rest interval of 30 s. They were vigorously
encouraged to perform maximally during voluntary efforts, and the
maximal voluntary contraction (MVC) torque was computed as mean
value recorded over all trials for a given experimental session. Prior to
strength assessment (except for postexercise session), subjects carried
out a standardized warm-up consisting of 10 isometric plantar flexions
(4-s duration interspaced with 10-s rest periods). Contraction intensities were self-adjusted by the subjects to be progressively increased up
to MVC level during the last three trials.
Blood sampling. Blood was drawn from the antecubital vein with a
new puncture at each time point. No catheter was used in order to
avoid local inflammation and alteration of miRNA expression. Blood
was collected into two separated tubes (Becton Dickinson, Franklin
Lakes, USA), one EDTA tube for plasma sample (5 ml), and one
serum separating tube for serum (SST) (7 ml). Plasma was separated
immediately by centrifugation (2,000 g, 4°C, 10 min) and aliquots
(250 ␮l) were frozen (⫺80°C). SST tubes were allowed to clot at
ambient temperature, 15 min, and centrifuged. Serum was analyzed
daily for creatine kinase (CK), aspartate aminotransferase (AST)
activities, myoglobin, and cardiac troponin I concentrations.
Serum markers measurement. CK, myoglobin AST, and cardiac
troponin I were measured in serum using Dimension RxL/XPan
clinical chemistry analyzer (Siemens Healthcare, Erlangen, Germany)
with corresponding Flex reagent cartridges. Specific performance
characteristics of the measures were as follow: coefficient of variation
0.5% to 1.9%, 1.9% to 2.3%, 0.02% to 0.08%, 0.8% to 2.7% (within
run) and 2.7% to 2.9%, 3.2% to 3.6%, 0.05% to 0.35% and 2.1% to
5.2% (total) for CK, myoglobin, troponin, and AST respectively.
miRNA extraction. Plasma aliquots were thawed at 4°C and centrifuged (10,000 g, 4°C, 10 min), 200 ␮l were used for total RNA
extraction. MirVana Paris isolation kit (Ambion, Austin, USA) was
used according to manufacturer protocol for liquid samples with an
additional precipitation step: column elution was performed with
180-␮l sterile water, 18-␮l sodium acetate 3M (Sigma), 400-␮l
ethanol (70%), and 1-␮l linear acrylamide (Ambion, Austin, USA)
were added. Tubes were vortexed and RNA were allowed to precipitate overnight at ⫺20°C. After centrifugation (12,000 g, 4°C, 10
min), supernatant was carefully removed and pellets were allowed to
dry 30 min. RNA were finally suspended in 12-␮l sterile water,
incubated at 50°C, 10 min, and frozen at ⫺80°C.
cDNA synthesis. MicroRNA complementary DNA was synthesized
in a 10-␮l final volume with Universal cDNA Synthesis Kit (Exiqon,
Vedbaek, Denmark). Reaction mix contained 2-␮l 5X buffer, 2-␮l
nuclease free water and 1-␮l Enzyme mix. For each sample, 5-␮l
reaction mix and 5-␮l total RNA (diluted 1/6) were gently mixed and
incubated at 42°C, 60 min. Reaction was stopped with a final step at
95°C, 5 min, and cDNA were frozen (80°C).
qPCR. PCR were performed on a 96 wells LightCycler 480
instrument (Roche Applied Science, Mannheim, Germany). Each
sample was first controlled with an individual measurement of hsamiR-103 as a quality control. Pick & Mix microRNA PCR panels
(Exiqon, Vedbaek, Denmark) were designed with 22 miRNAs per
sample. Reaction was performed in a 10-␮l final volume per well,
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
Ethical approval. Experimental protocol was performed according
to the Helsinki Declaration of 1975 as revised in 2008 on ethical
principles for medical research involving human subjects. The project
was approved by the scientific committee of Aspetar and by the ethics committee of Shafallah Medical Genetics Center (Institutional
Review Board - Project number 2011-003). Participants gave their
informed, written consent prior to the commencement of the experiment, once the experimental procedures, associated risks, and potential benefits of participation had been explained.
Participants. Nine healthy male subjects (27 to 36 yr old) were
included in the study. Mean body mass index was 24.59 ⫾ 0.76
kg/m2. They were recreationally active, involved in individual (running, swimming, cycling) or collective sports (football). Participants
declared 1- to 4-h physical activity in the week, with no important
variation in the 3 mo preceding the experiment. They had no significant occupational physical activity. All subjects were unaccustomed
to eccentric exercise and were not involved in any regular resistance
training program for at least 3 mo before the experiment. Exclusion
criteria were chronic or recent (ⱕ2 wk) treatment with drugs acting on
skeletal muscle, recent (ⱕ2 wk) intramuscular injection, recent (ⱕ3
mo) history of traumatic muscle injury, history of heat-related illness,
contact sports practice (ⱖ2 h/wk).
Experimental design. Participant performed two randomly assigned, separate experimental sessions (downhill exercise or uphill
exercise) with at least a 6-wk washout period. During this period
subjects were asked to maintain their normal lifestyle (they were not
involved in resistance, aerobic, and/or flexibility training). They were
asked not to perform physical activity 48 h before and during the
experimental sessions. Blood samples, muscle strength, and delayed
onset muscle soreness (DOMS) were assessed preexercise and immediately postexercise as well as 24, 48, and 72 h after the recovery
period. Blood samples were additionally drawn 2 and 6 h after the end
of exercise.
Exercise protocol. Downhill exercise consisted of one bout of 30
min of backward downhill walking exercise on a motorized treadmill
(Cosmos, Nussdorf-Traunstein, Germany), at a constant velocity of 1
m/s, with a grade of 25%, with an additional load carriage of 12% of
body wt (loaded backpack), as described by Racinais (27). This
exercise protocol was previously found to induce significant damages
of the plantar flexor muscles.
The uphill exercise was performed under the same conditions as the
downhill session (i.e., 30 min, 1 m/s, grade of 25%, additional load
carriage of 12% of body wt), but the rotational direction of the
treadmill belt was inverted to make the participant walk uphill. This
exercise was therefore matched to downhill exercise in terms of
distance covered.
Exercise bouts were performed in a controlled thermoneutral environment (20°C/30% relative humidity). To prevent any potential
unknown daily variations in miRNAs that could interfere with our
results, exercise trials were performed at the same time of day
(between 8:30 and 10:00 am). Participants were allowed to drink ad
libidum during exercise and recovery. Heart rate (HR) was recorded
(5-s intervals) via a wireless Polar monitoring system (Polar Electro
•
Circulating miRNAs and Exercise
calculated as Cohen’s d (with d ⱕ 0.2 representing a trivial difference;
0.2– 0.5, a small difference; 0.5– 0.8, a moderate difference; and ⬎0.8,
a large difference). Analyses were performed with Statistica 7.0
software (Statsoft, Maisons-Alfort, France). Results were considered
significant for P ⬍ 0.05.
RESULTS
Physiological responses. Heart rate and RPE increased in
response to both forms of exercise and were significantly higher
during the uphill compared with the downhill session (P ⬍ 0.001)
(Fig. 1, A and B).
There was a significant effect of exercise (P ⬍ 0.001) and
condition (P ⬍ 0.001) on DOMS, with a significant interaction
between these factors (P ⬍ 0.001) (Fig. 1C). Uphill exercise
induced a modest and nonsignificant (P ⫽ 0.07) increase in
DOMS immediately postexercise, whereas downhill exercise
induced an increase in DOMS with peak values observed 48 h
postexercise. MVC torque was affected by exercise (P ⬍
0.001) and condition (P ⬍ 0.01), with a significant interaction
(P ⬍ 0.05) (Fig. 1D). There was a 7 ⫾ 6% decrease in MVC
torque at the end of the uphill exercise that was fully recovered
with 24 h. Immediately after the downhill exercise, MVC
torque dropped by 20 ⫾ 9% and remained significantly lower
than baseline until 72 h of recovery.
Plasma markers of muscle injuries. Plasma CK activity significantly increased in response to exercise (P ⬍ 0.001), peaking
A
B
C
D
*
*
*
1239
Banzet et al.
*
*
*
Fig. 1. Markers of exercise intensity and muscle damage. Heart rate (A) was recorded via a wireless monitoring system, and rate of perceived exertion (B) reported
by participants on a 6 –20 Borg scale during uphill (open circles) and downhill (filled circles) exercise bouts. Delayed onset muscle soreness (C) evaluated with
a 100-mm visual scale ranging from 0 (“normal absence of soreness”) to 10 (“extremely painful”) and isometric ankle plantar-flexion torque (D) measured using
a dynamometric pedal during maximal voluntary contraction. Values were recorded before, immediately after, and during recovery (24, 48, and 72 h) of uphill
(open bars) and downhill (filled bars) exercise bouts (n ⫽ 9).
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
with 5-␮l SYBR Green master mix, Universal RT (Exiqon, Vedbaek,
Denmark), and 5-␮l cDNA (diluted 1/40). PCR protocol consisted in
1 activation cycle (95°C, 10 min), 45 amplification cycles (denaturation 95°C, 10 s, and annealing/amplification 60°C, 1 min), and a
melting curve analysis step. Single measurements were performed
except for hsa-mir-21 that was measured in duplicate on each plate to
assess PCR repeatability: mean ⌬Cq was 0.13 (0.001 to 0.828)
corresponding to a mean coefficient of variation of 0.4% (0.001% to
2.2%).
Normalization. Quantification cycles (Cq) were determined using the
second derivative maximum method. An interplate calibration was performed according to manufacturer’s protocol. Quantification was performed according to the 2⫺⌬Cq method. Potent reference miRNAs were
chosen based on literature. The number of reference miRNAs and their
stability were examined using GeNorm software (34). Finally, five references (hsa-miR-20a, hsa-miR-103, hsa-miR-21, hsa-miR-192, and hsamiR-185) were necessary to validate the normalization. Final quantification was the geometric mean of the quantifications performed with each
reference.
Statistics. Data are presented as means ⫾ SE. Normal distribution
of the data was tested using the Kolmogorov-Smirnov test. Results
were analyzed using a two-way ANOVA for paired values. We first
assessed the effect of “exercise” (seven time points) to test the
hypothesis that circulating miRNAs could be affected by one bout of
exercise. We also tested the effect of “condition” (uphill or downhill
trials) and any potential interaction between “exercise” and “condition,” to test the hypothesis that exercise modality could differentially
affect circulating miRNAs. When appropriate, a Newman-Keuls post
hoc test was applied to compare groups. Effect sizes (ES) have been
•
1240
Circulating miRNAs and Exercise
A
•
Banzet et al.
B
C
A
*
Fig. 2. Serum creatine kinase, myoglobin, and aspartate amino transferase.
Creatine kinase (A), myoglobin (B), and aspartate amino transferase (C) were
measured in serum before, immediately after, and during recovery (2, 6, 24, 48,
and 72 h) of uphill (open bars) and downhill (filled bars) exercise bouts (n ⫽ 9).
after 6-h recovery (Fig. 2A). No significant difference between
uphill and downhill sessions (P ⫽ 0.12) and no interaction
between exercise and condition were found (P ⫽ 0.22). Plasma
concentrations of myoglobin increased in response to exercise
(P ⬍ 0.001), peaking after 2-h recovery (Fig. 2B), but no significant difference between conditions (P ⫽ 0.15) and no interaction
were found (P ⫽ 0.42). Exercise induced an increase in plasma
AST activity (P ⬍ 0.001) with peak values observed after 24 h of
recovery (Fig. 2C). AST values were significantly higher in
downhill condition (P ⬍ 0.05) without interaction (P ⫽ 0.80).
Finally, cardiac troponin I remained undetectable throughout the
experiment.
Plasma miRNAs. We first focused on muscle specific miRNAs.
Exercise significantly affected plasma levels of hsa-mir-1 (P ⬍
0.001), hsa-mir-133a (P ⬍ 0.001), and hsa-mir-133b (P ⬍
0.001) with peak values observed 6 h after exercise (Fig. 3).
Effect sizes comparing preexercise values to peak values were
hsa-mir-1 (1.11; CI95% -0.21 to 2.43), 133a (1.23; CI95%
-0.11 to 2.58), and 133b (1.31; CI95% 0.08 to 2.54). There was
a significant effect of condition for both hsa-mir-1 (P ⬍ 0.05)
and hsa-mir-133a (P ⬍ 0.05) with higher values observed in
B
*
*
C
*
Fig. 3. Plasma hsa-mir-1, hsa-mir-133a, and hsa-mir-133b. Hsa-mir-1 (A),
hsa-mir-133a (B), and hsa-mir-133b (C) were measured in plasma by RTqPCR before, immediately after, and during recovery (2, 6, 24, 48, and 72 h)
of uphill (open bars) and downhill (filled bars) exercise bouts (n ⫽ 9).
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
downhill condition, whereas statistical significance was not
reached for hsa-mir-133b (P ⫽ 0.053). Finally, a significant
interaction was found (P ⬍ 0.05 for hsa-mir-133a; P ⬍ 0.01
for both hsa-mir-1 and 133b). Post hoc analysis demonstrated
significantly higher values in downhill condition 2 h and 6 h
after exercise for hsa-mir-1 (P ⬍ 0.05 and P ⬍ 0.001) and 6 h
after exercise for hsa-mir-133a and 133b (P ⬍ 0.001). Individual
profiles for hsa-mir-1, 133a, and 133b are shown in Fig. 4. Plasma
levels of hsa-mir-206 could not be consistently measured because
of high Cq (⬎40).
Plasma levels of hsa-mir-499 –5p displayed a high variability between subjects. They were significantly affected by
exercise (P ⬍ 0.01) with a peak value 6 h after exercise, but no
effect of condition nor any interaction were found (Fig. 5A).
Hsa-mir-208b was significantly affected by exercise (P ⬍
0.001), and the effect of condition failed to reach significance
(P ⫽ 0.068) (Fig. 5B). However, a robust interaction was found
(P ⬍ 0.001), and post hoc test revealed a significant difference
6 h after exercise (P ⬍ 0.001). Finally, the cardiac specific
hsa-mir-208a remained undetectable throughout the experiment.
Plasma levels of hsa-mir-181b (Fig. 6A) and hsa-mir-214
(Fig. 6B) were not globally affected by exercise and condition,
but significant interactions were found (P ⬍ 0.05). Post hoc
analysis demonstrated a significantly increased value at the end
of uphill exercise compared with baseline (effect size 1.12;
Circulating miRNAs and Exercise
•
Banzet et al.
1241
CI95% 0.66 to 1.58) and other time points of the same
condition and after downhill exercise for hsa-mir-181b (P ⬍
0.05). Plasma hsa-mir-214 values were significantly higher at
the end of uphill exercise compared with baseline (effect size
0.74; CI95% ⫺0.13 to 1.60) and time points 2, 6, and 72 h after
exercise (P ⬍ 0.05). Individual profiles are shown in Fig. 7.
Finally, hsa-mir-181a plasma levels did not change throughout
the protocol (Fig. 6C).
DISCUSSION
Our main findings are 1) muscle-specific miRNAs (hsa-mir-1,
133a, 133b, and 208b) are up-regulated in the early recovery
period (i.e., 6 h) of the downhill exercise, and 2) muscle-related
miRNAs (hsa-mir-181b and 214) are transiently up-regulated at
the end of the uphill exercise. To our knowledge, we report for the
first time differential regulations of circulating miRNAs in response to eccentric vs. concentric exercise in humans.
As expected, uphill condition resulted in higher exercise
intensity than downhill exercise, which induced plantar flexors
muscles damage as shown by persistent decreases in MVC
torque. Serum CK and myoglobin increased following the
same pattern in both conditions. Uphill exercise was mainly
concentric, yet, because of the treadmill slope and load carrying, an eccentric component may have induced slight muscle
damages. Alternatively, alterations in CK, due to transient
alterations of myofibers membrane permeability, have been
reported in response to nondamaging exercise and may explain
the increases we found (5). Plasma CK is the most commonly
used blood marker of exercise-induced muscle damage; however, many limitations have been reported such as interindividual, gender, or ethnicity-related variations (5, 16). Only
serum AST displayed a significant effect of condition with
higher values for eccentric exercise. Unfortunately, AST is not
specific of muscle tissue. The present results further confirm
that in some situations, classical markers can hardly discriminate between situations with different levels of muscle damage.
We found that plasma hsa-mir-1, -133a, and -133b were
significantly increased 2 to 6 h after exercise, with a larger
response following the downhill exercise. Another musclespecific miRNA, hsa-mir-208b, had similar expression pattern,
but because it was not detectable in all subjects (i.e., 6 out of
9), we did not find a significant difference between conditions,
although a robust interaction was noted. Finally, hsa-mir499 –5p increased in both conditions, but displayed large interindividual differences that make any interpretation difficult.
Since all these miRNAs have a tissue-restricted expression,
muscle tissue is necessarily the main source for plasma variations. Nevertheless, the mechanism(s) by which muscle releases miRNAs in plasma has yet to be determined. An increase in mir-1 expression has been found in muscle of running
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
Fig. 4. Individual values of plasma hsa-mir-1, hsa-mir133a, and hsa-mir-133b. Hsa-mir-1, hsa-mir-133a, and
hsa-mir-133b were measured in plasma by RT-qPCR;
values are presented for time points before, immediately after, and during recovery (2, 6, 24, 48, and 72 h)
of uphill and downhill exercise bouts.
1242
Circulating miRNAs and Exercise
A
B
*
mice (30), and hsa-mir-1 and 133a (but not 133b nor 206)
increased within muscle tissue at the end of 1h cycling in
humans (25). However, because these miRNAs levels, displaying modest increases (20 –35%), returned to control values
within 3 h of the end of exercise, it is therefore unlikely that
such variations in miRNAs may account for the alterations we
report in plasma. Furthermore, a much larger muscle mass was
probably recruited in both legs in our uphill vs. downhill
condition, which would have resulted in higher miRNA levels.
Otherwise, we propose that plasma levels of muscle-specific
miRNAs rise in response to myofibers damage. Skeletal muscle toxic injuries induced significant rises in plasma mir-133a
in rats (17). Recently, high plasma levels of mir-1, 206, and
133a have been shown in a mouse and a dog model of muscle
dystrophy, as well as in Duchenne muscular dystrophy (DMD)
patients. Rescuing dystrophin expression in mice substantially
decreased plasma miRNA levels, demonstrating they are reliable biomarkers of muscle dystrophy (6, 23). Here we report an
increase in muscle-specific miRNAs in response to the completion of a damaging walking exercise. Interestingly, the peak
values were found 6 h after the end of exercise, a time course
similar to what was observed for circulating mir-1, 133a, 499,
and 208a in experimental myocardial infarction in mice, another form of acute muscle injury (9, 35). In our study,
damages mainly occurred in a restricted muscle mass (plantar
flexors muscles), and despite the fact that loading paradigm
was high relative to every day experience, exercise-induced
muscle damages are expected to be very modest compared with
muscle dystrophy. Consistently, we report lower fold increases
in miRNA levels than in DMD patients (6); however, when
comparing peak with control values, we found large effect
sizes. Whether similar response would be found in trained
athletes is unknown. Taken together, the larger increase in
downhill condition, the time course, and the magnitude of
expression suggest that the alterations in plasma muscle-spe-
Banzet et al.
cific miRNA we found in exercising subjects could be biomarkers of exercise-induced skeletal muscle damages. This is
further supported by the lack of variation of plasma hsa-mir133a reported in response to one bout of cycling, an exercise
modality known to be almost exclusively concentric (3). This
hypothesis needs, however, to be extensively studied and
validated in various models of skeletal muscle damage.
Hsa-mir-1 and 133a and 208b are expressed in both skeletal
and cardiac muscles, and their plasma levels are increased after
acute myocardial infarction (35). They can therefore be considered as biomarkers of both skeletal and cardiac muscle
damages. However, it is very unlikely that the variations
observed come from heart tissue since higher levels were found
in the downhill condition, where cardiovascular demand was
rather modest. Moreover, both cardiac troponin I and the
cardiac-specific hsa-mir-208a remained undetectable throughout the experiment. Hsa-mir-133b and 206 are specific of
skeletal muscle and are not detectable in heart tissue. In our
experiment, hsa-mir-133b increased in response to eccentric
exercise, whereas hsa-mir-206 remained undetectable in most
of our samples. This may be explained by differences in the
abundance between the two miRNAs, as observed in rat
muscles (Banzet, unpublished data). Therefore, more severe
muscle damage may be necessary to induce a more noticeable
increase in plasma mir-206, as observed in DMD patients (6).
Taken together, our data suggest that circulating muscle-spe-
A
*
B
*
C
Fig. 6. Plasma hsa-mir-181b, hsa-mir-214, and hsa-mir-181a. Hsa-mir-181b (A),
hsa-mir-214 (B), and hsa-mir-181a (C) were measured in plasma by RT-qPCR
before, immediately after, and during recovery (2, 6, 24, 48, and 72 h) of uphill
(open bars) and downhill (filled bars) exercise bouts (n ⫽ 9).
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
Fig. 5. Plasma hsa-mir-499-5p and hsa-mir-208b. Hsa-mir-499-5p (A) and
hsa-mir-208b (B) were measured in plasma by RT-qPCR before, immediately
after, and during recovery (2, 6, 24, 48, and 72 h) of uphill (open bars) and
downhill (filled bars) exercise bouts (n ⫽ 9).
•
Circulating miRNAs and Exercise
•
Banzet et al.
1243
Fig. 7. Individual values of plasma hsa-mir-181b and hsamir-214. Hsa-mir-181b and hsa-mir-214 were measured in
plasma by RT-qPCR, and individual values are presented
for time points before and immediately after uphill and
downhill exercise bouts.
cells after one bout of cycling exercise (28, 29), the variations
reported here may reflect changes either in blood cell count or
miRNA expression. Whether circulating miRNAs could participate in the complex interactions reported between exercise
and immune system is a new question raised here.
Conclusions. Here we found that circulating muscle-specific
and muscle-related miRNAs are affected by exercise modality
according to two different patterns. Muscle specific miRNAs
respond to damaging exercise and increase during the recovery
period, raising the possibility that they could be alternative
biomarkers of muscle injury. Hsa-mir-181b and 214 respond
acutely to high-intensity exercise. In line with previous data,
our data show that acute exercise can alter circulating miRNAs
profiles. Further experiments are necessary to determine the
significance of these changes, unspecific alterations of plasma
miRNA levels in response to various stimuli, or specific
secretion/signaling intended to actively participate in exercise
responses or training adaptations. Now, a full profiling of
circulating miRNAs would provide an extensive overview of
the variations observed in response to various exercise modalities, thereby providing new orientations for future research.
GRANTS
This work was funded by Aspetar, Qatar Othopaedic and sports Medicine
Hospital and the French Délégation Générale pour l’Armement under contracts
07ca704 and 10ca701.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
AUTHOR CONTRIBUTIONS
Author contributions: S.B., M.C., O.G., S.R., C.D., H.C., and N.K. conception and design of research; S.B., M.C., O.G., S.R., C.D., H.C., and N.K.
performed experiments; S.B. and O.G. analyzed data; S.B., O.G., S.R., and
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
cific miRNA deserve more attention as a potentially interesting
and clinically relevant new class of plasma biomarkers to
assess skeletal muscle injuries in neuromuscular disorders, as
proposed by Cacchiarelli, but also in sports medicine and
research. Nevertheless, further experiments are required to
evaluate this alternative. Plasma hsa-mir-1, 133a, and 208b
have been proposed to be biomarkers of myocardial infarction
(32). Our results indicate that they should be carefully interpreted since concomitant skeletal muscle injuries, due to an
exercise bout or a fall, could significantly interfere with the
results. The cardiac specific hsa-mir-208a could help discriminating between muscle and cardiac injury.
miRNAs can leak out an injured cell but can also be actively
secreted (8) thereby targeting neighboring cells to exert a
paracrine function. In addition to muscle-restricted miRNAs,
we selected miRNAs, which are reported to have an important
role in muscle biology and/or be regulated by exercise within
skeletal muscle. We measured both hsa-mir-181a and 181b,
encoded by two separate genes. Whereas hsa-mir-181a was not
affected by exercise, we found a transient increase in plasma
hsa-mir-181b with a large effect size immediately at the end of
the uphill exercise. A similar profile was found for plasma
hsa-mir-214 with a medium effect size. Whether mir-214 is
altered in response to muscle contraction is not known, but
interestingly, both mir-181b and 214 are up-regulated in response to hypoxia (14, 31), a condition that may occur within
myofibers during sustained exercise (4). It could be tempting to
suggest that contracting muscle could account for variations
observed in plasma. Nevertheless, it is currently not possible to
know the source of mir-181b and 214, since they are present in
many tissues (7, 12). Blood cells account for a substantial part
of circulating miRNAs (26). Since mir-181b expression is
up-regulated in neutrophils and peripheral blood mononuclear
1244
Circulating miRNAs and Exercise
N.K. interpreted results of experiments; S.B. prepared figures; S.B. drafted
manuscript; S.B., O.G., S.R., H.C., and N.K. edited and revised manuscript;
S.B., M.C., O.G., S.R., C.D., H.C., and N.K. approved final version of
manuscript.
REFERENCES
Banzet et al.
18. Liang Y, Ridzon D, Wong L, Chen C. Characterization of microRNA
expression profiles in normal human tissues. BMC Genomics 8: 166, 2007.
19. McCarthy JJ. The MyomiR network in skeletal muscle plasticity. Exerc
Sport Sci Rev 39: 150 –154, 2011.
20. McCarthy JJ, Esser KA. MicroRNA-1 and microRNA-133a expression
are decreased during skeletal muscle hypertrophy. J Appl Physiol 102:
306 –313, 2007.
21. McCarthy JJ, Esser KA, Peterson CA, Dupont-Versteegden EE.
Evidence of MyomiR network regulation of beta-myosin heavy chain gene
expression during skeletal muscle atrophy. Physiol Genomics 39: 219 –
226, 2009.
22. Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, PogosovaAgadjanyan EL, Peterson A, Noteboom J, O’Briant KC, Allen A, Lin
DW, Urban N, Drescher CW, Knudsen BS, Stirewalt DL, Gentleman
R, Vessella RL, Nelson PS, Martin DB, Tewari M. Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad
Sci U S A 105: 10513–10518, 2008.
23. Mizuno H, Nakamura A, Aoki Y, Ito N, Kishi S, Yamamoto K,
Sekiguchi M, Takeda S, Hashido K. Identification of muscle-specific
microRNAs in serum of muscular dystrophy animal models: promising
novel blood-based markers for muscular dystrophy. PLoS One 6: e18388,
2011.
24. Naguibneva I, Ameyar-Zazoua M, Polesskaya A, Ait-Si-Ali S, Groisman R, Souidi M, Cuvellier S, and Harel-Bellan A. The microRNA
miR-181 targets the homeobox protein Hox-A11 during mammalian
myoblast differentiation Nat Cell Biol 8: 278 –284, 2006.
25. Nielsen S, Scheele C, Yfanti C, Akerstrom T, Nielsen AR, Pedersen
BK, Laye MJ. Muscle specific microRNAs are regulated by endurance
exercise in human skeletal muscle. J Physiol 588: 4029 –4037, 2010.
26. Pritchard CC, Kroh E, Wood B, Arroyo JD, Dougherty KJ, Miyaji
MM, Tait JF, Tewari M. Blood cell origin of circulating microRNAs: a
cautionary note for cancer biomarker studies. Cancer Prev Res (Phila) 5:
492–497, 2012.
27. Racinais S, Bringard A, Puchaux K, Noakes TD, Perrey S. Modulation
in voluntary neural drive in relation to muscle soreness. Eur J Appl Physiol
102: 439 –446, 2008.
28. Radom-Aizik S, Zaldivar F Jr, Leu SY, Adams GR, Oliver S, Cooper
DM. Effects of exercise on microRNA expression in young males peripheral blood mononuclear cells Clin Transl Sci 5: 32–38, 2012.
29. Radom-Aizik S, Zaldivar F Jr, Oliver S, Galassetti P, Cooper DM.
Evidence for microRNA involvement in exercise-associated neutrophil
gene expression changes J Appl Physiol 109: 252–261, 2010.
30. Safdar A, Abadi A, Akhtar M, Hettinga BP, Tarnopolsky MA. miRNA
in the regulation of skeletal muscle adaptation to acute endurance exercise
in C57Bl/6J male mice. PLoS One 4: e5610, 2009.
31. Sarkar J, Gou D, Turaka P, Viktorova E, Ramchandran R, Raj JU.
MicroRNA-21 plays a role in hypoxia-mediated pulmonary artery smooth
muscle cell proliferation and migration. Am J Physiol Lung Cell Mol
Physiol 299: L861–L871, 2010.
32. Tijsen AJ, Pinto YM, Creemers EE. Circulating microRNAs as diagnostic biomarkers for cardiovascular diseases. Am J Physiol Heart Circ
Physiol 303: H1085–H1095, 2012.
33. van Rooij E, Liu N, Olson EN. MicroRNAs flex their muscles. Trends
Genet 24: 159 –166, 2008.
34. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De
Paepe A, Speleman F. Accurate normalization of real-time quantitative
RT-PCR data by geometric averaging of multiple internal control genes.
Genome Biol 3: RESEARCH0034, 2002.
35. Wang GK, Zhu JQ, Zhang JT, Li Q, Li Y, He J, Qin YW, Jing Q.
Circulating microRNA: a novel potential biomarker for early diagnosis of
acute myocardial infarction in humans. Eur Heart J 31: 659 –666, 2010.
J Appl Physiol • doi:10.1152/japplphysiol.00075.2013 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.33.1 on July 28, 2017
1. Allen DL, Bandstra ER, Harrison BC, Thorng S, Stodieck LS, Kostenuik PJ, Morony S, Lacey DL, Hammond TG, Leinwand LL,
Argraves WS, Bateman TA, Barth JL. Effects of spaceflight on murine
skeletal muscle gene expression. J Appl Physiol 106: 582–595, 2009.
2. Aurora AB, Mahmoud AI, Luo X, Johnson BA, van Rooij E, Matsuzaki S, Humphries KM, Hill JA, Bassel-Duby R, Sadek HA, Olson EN.
MicroRNA-214 protects the mouse heart from ischemic injury by controlling Ca(2)(⫹) overload and cell death. J Clin Invest 122: 1222–1232,
2012.
3. Baggish AL, Hale A, Weiner RB, Lewis GD, Systrom D, Wang F,
Wang TJ, Chan SY. Dynamic regulation of circulating microRNA during
acute exhaustive exercise and sustained aerobic exercise training. J
Physiol 589: 3983–3994, 2011.
4. Birot OJ, Peinnequin A, Simler N, van Cuyck-Gandre H, Hamel R,
Bigard XA. Vascular endothelial growth factor expression in heart of rats
exposed to hypobaric hypoxia: differential response between mRNA and
protein. J Cell Physiol 200: 107–115, 2004.
5. Brancaccio P, Maffulli N, Limongelli FM. Creatine kinase monitoring in
sport medicine. Br Med Bull 81– 82: 209 –230, 2007.
6. Cacchiarelli D, Legnini I, Martone J, Cazzella V, D’Amico A, Bertini
E, Bozzoni I. miRNAs as serum biomarkers for Duchenne muscular
dystrophy. EMBO Mol Med 3: 258 –265, 2011.
7. Chen CZ, Li L, Lodish HF, Bartel DP. MicroRNAs modulate hematopoietic lineage differentiation. Science 303: 83–86, 2004.
8. Chen X, Liang H, Zhang J, Zen K, Zhang CY. Secreted microRNAs: a
new form of intercellular communication. Trends Cell Biol 22: 125–132,
2012.
9. Cheng Y, Tan N, Yang J, Liu X, Cao X, He P, Dong X, Qin S, Zhang
C. A translational study of circulating cell-free microRNA-1 in acute
myocardial infarction. Clin Sci (Lond) 119: 87–95, 2010.
10. Eisenberg I, Alexander MS, Kunkel LM. miRNAS in normal and
diseased skeletal muscle. J Cell Mol Med 13: 2–11, 2009.
11. Guller I, Russell AP. MicroRNAs in skeletal muscle: their role and
regulation in development, disease and function. J Physiol 588: 4075–
4087, 2010.
12. Hsu SD, Lin FM, Wu WY, Liang C, Huang WC, Chan WL, Tsai WT,
Chen GZ, Lee CJ, Chiu CM, Chien CH, Wu MC, Huang CY, Tsou
AP, Huang HD. miRTarBase: a database curates experimentally validated
microRNA-target interactions. Nucleic Acids Res 39: D163–169, 2011.
13. Juan AH, Kumar RM, Marx JG, Young RA, Sartorelli V. Mir-214dependent regulation of the polycomb protein Ezh2 in skeletal muscle and
embryonic stem cells. Mol Cell 36: 61–74, 2009.
14. Kulshreshtha R, Ferracin M, Wojcik SE, Garzon R, Alder H, AgostoPerez FJ, Davuluri R, Liu CG, Croce CM, Negrini M, Calin GA, Ivan
M. A microRNA signature of hypoxia. Mol Cell Biol 27: 1859 –1867,
2007.
15. Lagos-Quintana M, Rauhut R, Yalcin A, Meyer J, Lendeckel W,
Tuschl T. Identification of tissue-specific microRNAs from mouse. Curr
Biol 12: 735–739, 2002.
16. Landau ME, Kenney K, Deuster P, Campbell W. Exertional rhabdomyolysis: a clinical review with a focus on genetic influences. J Clin
Neuromuscul Dis 13: 122–136, 2012.
17. Laterza OF, Lim L, Garrett-Engele PW, Vlasakova K, Muniappa N,
Tanaka WK, Johnson JM, Sina JF, Fare TL, Sistare FD, Glaab WE.
Plasma MicroRNAs as sensitive and specific biomarkers of tissue injury.
Clin Chem 55: 1977–1983, 2009.
•