J Appl Physiol
90: 1900–1908, 2001.
Cardiac and skeletal muscle adaptations to voluntary
wheel running in the mouse
DAVID L. ALLEN,1 BROOKE C. HARRISON,2 ALEXANDER MAASS,1
MATTHEW L. BELL,2 WILLIAM C. BYRNES,2 AND LESLIE A. LEINWAND1
Departments of 2Kinesiology and 1Molecular, Cellular, and Developmental Biology,
University of Colorado, Boulder, Colorado 80309-0347
Received 28 September 2000; accepted in final form 8 December 2000
myosin heavy chain; endurance exercise; plasticity; cardiac
hypertrophy; transgenic mice
INCREASES IN MUSCULAR ACTIVITY,
such as those occurring
during endurance exercise, are associated with rapid
and dramatic adaptations of the cardiovascular and
musculoskeletal systems (29, 34). These physiological
adaptations result in more efficient delivery, uptake,
and utilization of energy substrates to produce sustained physical work. Endurance exercise adaptation
in the rat is characterized by an increase in relative
heart weight (2, 19, 26, 35, 43) as well as shifts in
skeletal muscle metabolic (2–5, 9, 18, 20, 23, 27, 32, 33,
37) and contractile (8, 14, 17, 22, 36, 39) protein expression toward a slower contracting, more oxidative
Address for reprint requests and other correspondence: L. Leinwand, Dept. of MCD Biology, Univ. of Colorado, Boulder, Boulder,
CO 80309-0347 (E-mail: [email protected]).
1900
fiber profile that is better suited for prolonged muscle
activation.
Studies on voluntary endurance exercise adaptation
in the mouse have been less numerous, confirming
some but not all of the results obtained from rats (21,
46). Specifically, evidence of cardiac hypertrophy (25)
and increased skeletal muscle oxidative capacity (21,
44) have been observed in mice after voluntary endurance exercise training. Less well characterized, however, are the changes in fiber contractile protein expression in response to endurance exercise training in
mice. Wernig and co-workers (45) demonstrated an
increase in histochemical type I (slow) fibers in the
soleus with endurance exercise but no change in fiber
composition of the extensor digitorum longus, a fast
flexor muscle. A recent study demonstrated no appreciable change in immunohistochemically typed fiber
percent in the medial gastrocnemius after 8 wk of
voluntary cage-wheel exercise in selectively bred house
mice (46). Others have also reported no significant
difference in gastrocnemius fiber-type composition between laboratory and wild house mice (15). Together
these studies suggest that the contractile element of
mouse skeletal muscle may be less adaptive than that
of rats. This contention is supported by interspecies
comparisons of the adaptive response to chronic electrical stimulation, which show that mouse tibialis anterior (TA) is much less responsive than rat TA to this
supraphysiological stimulus (38). Given the relative
dearth of studies on the exercise response of mice, as
well as the current position of the mouse as the genetic
model of choice, the purpose of this study was to systematically evaluate the time course of both cardiac
and skeletal muscle adaptations to voluntary wheelrunning exercise in the mouse. Our results demonstrate that mice will run distances on a cage wheel that
compare favorably to those achieved by rats undergoing involuntary treadmill or voluntary wheel-running
exercise. Voluntary wheel running in the mouse resulted in both cardiac hypertrophy as well as an increase in the expression of mRNAs associated with
cardiac hypertrophy. Moreover, voluntary running exThe costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
8750-7587/01 $5.00 Copyright © 2001 the American Physiological Society
http://www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Allen, David L., Brooke C. Harrison, Alexander
Maass, Matthew L. Bell, William C. Byrnes, and Leslie
A. Leinwand. Cardiac and skeletal muscle adaptations to
voluntary wheel running in the mouse. J Appl Physiol 90:
1900–1908, 2001.—In this paper, we describe the effects of
voluntary cage wheel exercise on mouse cardiac and skeletal
muscle. Inbred male C57/Bl6 mice (age 8–10 wk; n ⫽ 12) ran
an average of 4.3 h/24 h, for an average distance of 6.8 km/24
h, and at an average speed of 26.4 m/min. A significant
increase in the ratio of heart mass to body mass (mg/g) was
evident after 2 wk of voluntary exercise, and cardiac atrial
natriuretic factor and brain natriuretic peptide mRNA levels
were significantly increased in the ventricles after 4 wk of
voluntary exercise. A significant increase in the percentage of
fibers expressing myosin heavy chain (MHC) IIa was observed in both the gastrocnemius and the tibialis anterior
(TA) by 2 wk, and a significant decrease in the percentage of
fibers expressing IIb MHC was evident in both muscles after
4 wk of voluntary exercise. The TA muscle showed a greater
increase in the percentage of IIa MHC-expressing fibers than
did the gastrocnemius muscle (40 and 20%, respectively,
compared with 10% for nonexercised). Finally, the number of
oxidative fibers as revealed by NADH-tetrazolium reductase
histochemical staining was increased in the TA but not the
gastrocnemius after 4 wk of voluntary exercise. All results
are relative to age-matched mice housed without access to
running wheels. Together these data demonstrate that voluntary exercise in mice results in cardiac and skeletal muscle
adaptations consistent with endurance exercise.
VOLUNTARY RUNNING IN MICE
ercise produced a significant shift from IIb to IIa myosin heavy chain (MHC) expression in both the gastrocnemius and TA muscles. This shift in skeletal muscle
MHC expression is accompanied by an increase in the
percentage of NADH-positive fibers in the TA but not
the gastrocnemius, as well as hypertrophy of the more
oxidative type I and IIa fibers. Thus voluntary wheel
running produces changes in murine cardiac and skeletal muscle consistent with endurance exercise adaptation. We believe that this adaptive response, combined with the relative ease and affordability of the
voluntary wheel-running model, makes this an extremely useful model system for elucidating the molecular mechanisms underlying cardiac and skeletal muscle adaptation.
METHODS
value to gauge the shift from IIb to IId/x and IIa MHC
isoforms. Gels were dried between cellophane sheets, digitally scanned using a computer scanner, and densitometrically analyzed using NIH Image software. For each sample,
the IIb and IId/x⫹IIa bands were outlined and quantified;
background was eliminated by outlining a portion of the lane
not containing protein bands and subtracting this from each
band value.
Immunohistochemistry. Cryosections (10-␮m thickness)
for immunohistochemical analysis were cut from the midbelly of frozen muscle samples and placed on gelatin-coated
microscope slides. Slides were stored at ⫺20°C until use.
Slides were air dried for 20 min and then incubated in a
blocking solution (0.12% BSA, 0.12% nonfat dry milk, and
0.01% Triton X-100 in PBS) containing unlabeled Fab⬘ fragments (10 mg/ml) for 1 h to block nonspecific binding. After
being rinsed with PBS, sections were placed in primary
antibody solution overnight at 4°C or for 1 h at room temperature. The antibodies used were the following: SC-71,
which labels IIa MHC-containing fibers (16); MHCs (Novocastra), which labels type I MHC-containing fibers; and BFF3, which labels type IIb MHC-containing fibers (16). After
primary incubation, sections were rinsed three times in PBS
followed by three 5-min incubations in PBS. The secondary
antibody was goat anti-mouse IgG-peroxidase conjugate
(Vector Laboratories, Burlingame, CA) used at a 1:200 dilution in blocking solution. Sections were incubated in secondary antibody for 1 h at room temperature followed by numerous washes with PBS. Immunostaining was visualized using
a peroxidase 3,3⬘-diaminobenzidine reaction kit (Vector Laboratories) for 5 min followed by several rinses in distilled
H2O and mounting in Permount. Fiber percents were determined from counts of 500–1,000 fibers in 4–6 randomly
chosen fields. Because of the lack of an antibody specific to
type IId/x MHC, the percentage of type IId/x expressing
fibers was determined by subtracting the percentage of I, IIa,
and IIb fibers from 100%. In our hands, the antibodies used
yielded strong signals, which make interpretation quite easy.
However, to check this point, two independent observers
scored the same slides from the same animal, and we found
very little difference between the two interpretations (data
not shown).
Fiber cross-sectional area (CSA) was measured for type I
and type IIa fibers using a video camera (Videoscope International) mounted on a Zeiss bright-field microscope attached to a Macintosh compatible PowerBase 200 computer
(Power Computing, Austin, TX). Fiber CSA was determined
by using a computer mouse to trace the outline of positively
immunostained muscle fibers using NIH Image software.
The analysis software was calibrated by outlining defined
areas on a slide micrometer. A total of 50 fibers of each type
per muscle were analyzed from 5–10 randomly chosen areas
in each muscle. Histochemical staining for NADH-tetrazolium reductase (NADH-TR) was carried out by incubation for
30 min at 37°C in NADH-TR reaction solution (0.2 M Tris, 1.5
mM NADH, and 1.5 mM nitrotetrazolium blue) and then
dehydration through serial dilutions of acetone and mounting in Aquamount (Lerner Laboratories, Pittsburgh, PA).
The number of NADH-TR-positive fibers was counted for
approximately six fields per section (500–1,000 fibers per
muscle), and a mean was calculated for each mouse and each
time point. With NADH-TR staining, three types of fibers
were observed (see Fig. 4): unstained, moderately stained,
and darkly stained. Fibers that were moderately or darkly
stained were classified as NADH-TR positive.
Northern blotting. Northern blotting was carried out using
standard methods. RNA was isolated from individual mouse
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Voluntary cage-wheel exercise. Voluntary running was performed by inbred male C57/Bl6 mice, age 8–10 wk at the
start of exercise. Founder mice were originally obtained from
Jackson Laboratories; all mice used in the present study
were offspring of our breeding colony. Hamster-sized metal
cage wheels with a diameter of 11.5 cm (model no. 6208;
Petsmart, Phoenix, AZ) were fitted with digital magnetic
counters (model BC 600, Sigma Sport, Olney, IL) and placed
into 47 ⫻ 26 ⫻ 14.5-cm cages. The digital counters measure
maximum running speed, total distance run, and total time
run. The sampling frequency for maximum speed was based
on the amount of time required for one revolution of the
wheel, which was ⬃1–3 s. Each morning these data were
recorded from the computer and logged for each animal, and
the counter was reset. For a given litter, mice were randomly
assigned to either a particular exercise duration (1, 2, or 4
wk) or nonexercise littermate control. Control mice were
housed in cages without a cage wheel. All animals were
supplied with food and water ad libitum.
Body, heart, and muscle mass data. At different times after
the initiation of voluntary exercise (1, 2, and 4 wk), exercised
and nonexercised mice were euthanized by cervical dislocation under inhaled anesthesia. All nonexercised mice were
euthanized on the same day as exercised mice. Body, heart,
and muscle mass for selected muscles were all recorded.
Hearts and muscles were frozen either in liquid nitrogen (for
biochemical analysis) or in isopentane cooled in liquid nitrogen (for histochemical analysis).
Gel electrophoresis. Muscle samples were prepared for
high-resolution gel electrophoresis by mild homogenization
by hand in 5 vol of myosin extraction buffer (0.3 M NaCl, 0.1
M NaH2PO4, 0.05 M Na2HPO4, 0.01 M sodium pyrophosphate, 1 mM MgCl2, 10 mM EDTA, and 1 mM dithiothreitol,
pH 6.5) and extraction on ice for 1 h as described by ButlerBrowne and Whalen (6). Samples were microfuged for 5 min
to remove cellular debris, mixed 1:1 in extraction buffer
containing 50% glycerol, and stored at ⫺20°C. Protein concentration was determined using the Bradford method, and
protein concentration for all samples was adjusted to 2.5
␮g/␮l.
Gel electrophoresis was carried out as described by Talmadge and Roy (41). Briefly, muscle samples were electrophoretically separated on gels containing 30% glycerol and
8% acrylamide for 40 h at 80 V at 4°C. Gels were stained in
Express Coomassie (R&D Systems, Mount Prospect, IL) to
visualize the bands. Using this system, consistent separation
of the IIa and IId/x protein bands could not be achieved; the
combined signal for both bands was therefore used as a single
1901
1902
VOLUNTARY RUNNING IN MICE
RESULTS
Wheel running. Mice ran an average of 4.3 h/24 h and
a distance of 6.8 km/24 h (Table 1). The range of
responses was 2.4–7.1 h/24 h and 3.1–16.2 km/24 h.
The average running speed was 26.4 m/min, and the
average maximum running speed was 44.8 m/min (Table 1). Across all groups, both distance ran and average
speed increased across the time examined; distance
was significantly higher during the second and third
weeks of exercise compared with the first week, and
average speed was significantly higher at 2, 3, and 4
wk of exercise compared with 1 wk (Table 1). Neither
total time run per night nor maximum speed differed
across the time period analyzed in this study (Table 1).
For comparison, we calculated the average distance
run per week for representative published studies employing involuntary treadmill running for rats (8, 14,
17, 31), voluntary wheel running for rats (22, 31, 33,
Table 1. Mean nightly wheel-running activity
in inbred C57/Bl6 mice
Week
No.
Time, h
Distance,
km
Average Speed,
m/min
Maximum Speed,
m/min
1
2
3
4
Overall
4.0 ⫾ 1.3
4.4 ⫾ 0.9
4.3 ⫾ 0.7
3.8 ⫾ 1.1
4.3 ⫾ 0.9
6.2 ⫾ 2.7
8.5 ⫾ 3.0*
8.4 ⫾ 2.2*
7.1 ⫾ 2.4
6.8 ⫾ 2.6
25.7 ⫾ 7.5
31.6 ⫾ 8.6*
32.8 ⫾ 8.3*
32.5 ⫾ 9.0*
26.4 ⫾ 8.6
42.7 ⫾ 5.6
43.9 ⫾ 4.3
46.2 ⫾ 4.9
45.9 ⫾ 4.3
44.8 ⫾ 5.5
Values are means ⫾ SD; n ⫽ 12 mice for week 1, n ⫽ 8 mice for
week 2, and n ⫽ 4 mice for weeks 3 and 4. * P ⬍ 0.0001 compared
with appropriate week 1 value.
Table 2. Body and heart mass with and without
voluntary free-wheel running
Nonexercised
1-wk Exercised
2-wk Exercised
4-wk Exercised
Body
Mass, g
Heart Mass,
mg
Heart/Body Mass,
mg/g
25.9 ⫾ 4.0
24.7 ⫾ 1.7
22.9 ⫾ 3.7
24.8 ⫾ 1.7
128.2 ⫾ 13.2
131.8 ⫾ 16.4
136.7 ⫾ 20.4
146.9 ⫾ 10.6*
5.0 ⫾ 0.5
5.3 ⫾ 0.3
6.0 ⫾ 1.0†
5.9 ⫾ 0.1†
Values are means ⫾ SD; n ⫽ 12 mice for nonexercised and n ⫽
4 for 1-, 2-, and 4-wk exercised. * P ⬍ 0.05 vs. nonexercised control.
† P ⬍ 0.01 vs. nonexercise control.
37), and voluntary wheel running in mice [observed in
the present study and in a recently published study
(21)]. The results demonstrate that C57/Bl6 mice run
significantly farther per week than rats undergoing
either involuntary or voluntary exercise (rat treadmill,
8.22 km/wk; rat wheel running, 24.9 km/wk; mouse
wheel running, 47.66 km/wk). Moreover, if these distances are normalized to the stride length and body
size of the animals, the differences are even more
dramatic (data not shown). Our distances are comparable to those reported by Houle-Leroy et al. (21) for
Hsd:ICR house mice (31.34 km/wk).
Body and muscle mass data. Voluntary running exercise resulted in no significant difference in body mass
through 4 wk of voluntary running (Table 2), consistent with the results of Swallow et al. (40). Relative
skeletal muscle weights were not significantly different
for the TA, gastrocnemius, or plantaris muscles at any
time (Table 3). However, a significant increase in absolute heart weight was observed in 4 wk exercised
mice compared with nonexercised mice (Table 2).
Heart weight relative to body mass was significantly
increased after 2 wk of voluntary exercise and persisted through 4 wk of exercise (Table 2). This demonstrates that mice were able to run sufficiently in a
voluntary model of exercise to induce physiological
cardiac hypertrophy.
Changes in cardiac mRNA expression. To determine
whether voluntary running exercise provided a sufficient stimulus to induce the expression of hypertrophic
markers such as ANF and BNP, Northern blotting
analysis was carried out. A significant increase in both
ANF and BNP mRNA levels was seen in mice run for 4
wk. This increase was first evident after 2 wk of voluntary running but only reached significance after 4
wk (Fig. 1).
Table 3. Normalized muscle mass with and without
voluntary free-wheel running
Nonexercised
1-wk Exercised
2-wk Exercised
4-wk Exercised
Tibialis Anterior
Plantaris
Gastrocnemius
1.49 ⫾ 0.16
1.61 ⫾ 0.03
1.57 ⫾ 0.13
1.57 ⫾ 0.18
0.56 ⫾ 0.07
0.64 ⫾ 0.05
0.61 ⫾ 0.04
0.61 ⫾ 0.13
4.59 ⫾ 0.11
5.16 ⫾ 0.28
4.63 ⫾ 0.37
4.72 ⫾ 0.25
Values are means ⫾ SD given in mg/g body wt; n ⫽ 4 mice per
condition.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
ventricles by using the RNA STAT-60 reagent (Tel-Test,
Friendswood, TX) according to the manufacturer’s recommendations, dissolved in sterile H2O, and stored at ⫺70°C
until use. Ten micrograms of total RNA were run on 1.2%
agarose-formaldehyde gels and then transferred by capillary
action to a nylon membrane where they remained overnight.
Membranes were cross-linked by ultraviolet radiation and
then rinsed in Perfect-Hyb prehybridization solution (Sigma
Chemical) for 3–4 h. Blots were hybridized overnight to
radiolabeled cDNA fragments specific to either atrial natriuretic factor (ANF; 650 bp) or brain natriuretic peptide (BNP;
380 bp) at 60°C and then washed two times with 2⫻ sodium
chloride-sodium citrate (SSC)-0.1% SDS and two times with
0.2⫻ SSC-0.1% SDS at 60°C. Membranes were visualized by
exposure to a phosphoimager, and signal intensity was analyzed using Image Quant software (Molecular Dynamics).
Membranes were then stripped and reprobed with a radiolabeled 250-bp cDNA fragment to glyceraldehyde-3-phosphate
dehydrogenase (GAPDH), and the ANF and BNP signals
were normalized to GAPDH signals. Results are reported as
means ⫾ SE.
Statistical analysis. All data are reported as means ⫾ SE.
One-way ANOVA with Fisher’s post hoc test for least squares
differences was used to evaluate the differences between
exercised and nonexercised animals at different time points.
To minimize variability, littermates were used for each time
point as either exercised or nonexercised. However, because
we observed no significant differences for any variable among
nonexercised animals at the different time points, these
animals were pooled for statistical comparison. Statistical
significance was taken as P ⬍ 0.05.
VOLUNTARY RUNNING IN MICE
1903
Skeletal muscle MHC adaptations. Voluntary running exercise produced a shift toward greater expression of type IIa⫹IId/x MHC as revealed by gel electrophoresis (Fig. 2). This shift was first evident after 1 wk
of running exercise and reached a peak after 2 wk of
exercise and did not change thereafter in the TA muscle; in the gastrocnemius muscle, these changes were
only significant at the 2-wk time point (Fig. 2). In
addition, immunohistochemistry followed by fiber percent analysis revealed that both the gastrocnemius and
the TA muscles experienced a shift in MHC expression
(Figs. 3 and 4). Specifically, an increase in the percentage of fibers expressing IIa MHC and a decrease in the
percentage of fibers expressing type IIb MHC were
observed in both muscles, and these changes were first
evident after 1 wk of exercise but increased throughout
the time points examined. After 4 wk of exercise, fibers
expressing IIa MHC comprised ⬃20 and 40% of the
total fibers in gastrocnemius and TA, respectively,
compared with ⬃10% in nonexercised muscles (Fig. 3).
No change in type I fiber percent was observed in
either muscle at any time point with exercise (Fig. 3).
Interestingly, we saw no change in the percentage of
fibers expressing IId/x MHC in the gastrocnemius muscle at any time point and only a small decrease in IId/x
fibers after 4 wk of voluntary exercise in the TA (Fig.
3B), but given the methodological limitations in analyzing IId/x MHC expression, specifically the lack of an
antibody specific to IId/x, it is difficult to assess the
relevance of this observation. Finally, we did not analyze coexpression of MHC isoforms, and it is therefore
possible that some of the fibers expressing IIa were
coexpressing IId/x and even IIb MHC.
We examined fiber size for type I and IIa fibers in
both the gastrocnemius and TA muscles to determine
whether voluntary running exercise induced fiber hypertrophy in the most oxidative muscle fibers. In the
gastrocnemius muscle, both type I and type IIa fibers
were significantly larger after 4 wk of voluntary exercise (Fig. 5). In addition, type IIa fibers were significantly larger after 1 and 2 wk of voluntary exercise
compared with those of nonexercised controls (Fig. 5).
In the TA, in contrast, there was no significant difference in mean fiber size for IIa fibers at any time point;
moreover, too few type I fibers could be found in either
the nonexercised or exercised TA to allow meaningful
statistical comparison.
Changes in oxidative fiber percents. Finally, we
stained transverse sections from unexercised and exercised mice to determine whether there was any
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Fig. 1. Northern blot analysis of atrial natriuretic factor (ANF) and brain natriuretic peptide (BNP) mRNA
from exercised and nonexercised mouse heart ventricles. A: gels. B: graphs. ANF and BNP values were
normalized to the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) signal to correct for loading differences. The gels show results from three nonexercised
(NE), four 2-wk exercised (2 week), and four 4-wk
exercised (4 week) mice. Values are means ⫾ SE for
nonexercised controls, 2-wk exercised, and 4-wk exercised mice. Voluntary wheel running results in a significant increase in both ANF and BNP mRNA levels
after 4 wk (* P ⬍ 0.05).
1904
VOLUNTARY RUNNING IN MICE
37). However, to date, there have been few studies on
exercise adaptation in mice. The present results demonstrate that voluntary wheel running in mice produces specific adaptations associated with endurance
training in both the cardiac and skeletal muscle systems.
One initial question was whether mice would run
sufficiently to produce an adaptational response. Most
of the studies on voluntary running in rats have suggested that the changes are qualitatively similar but
may often be quantitatively less robust than those
achieved using involuntary swimming or treadmill exercise. One reason this may be true is because animals
that exercise voluntarily do so at a percentage of exercise capacity insufficient to elicit the same magnitude
of changes that are seen with treadmill exercise, which
is typically done at a higher speed and/or inclination.
Indeed, most studies examining voluntary exercise in
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Fig. 2. Mean relative levels of IIa⫹IId/x myosin heavy chain (MHC)
protein in exercised and nonexercised mice. MHC isoforms were
separated by high-resolution SDS-PAGE, and individual bands were
densitometrically scanned for quantification. A: gastrocnemius muscle. B: tibialis anterior (TA) muscle. Values are means ⫾ SE; n ⫽ 2–3
animals per time point. Because the IIa and IId/x MHC isoforms run
extremely close together, we were unable to consistently individually
resolve both bands, and therefore we analyzed them as a single band.
Voluntary exercise resulted in a significant increase in IIa ⫹ IId/x
MHC protein at 2 wk for the gastrocnemius muscle and at 2 and 4 wk
for TA. *P ⬍ 0.05. **P ⬍ 0.001.
change in the percentage of NADH-TR-positive fibers.
In gastrocnemius muscle, there was no significant increase in the number of NADH-TR-positive fibers at
any time point (data not shown). In the TA, however,
there was a steady increase in the percentage of
NADH-TR-positive fibers, which was significant after 4
wk of voluntary exercise (Figs. 4 and 6).
DISCUSSION
The adaptational response of the cardiovascular and
musculoskeletal systems to the sustained muscle activation associated with endurance exercise has been
extensively documented. Studies on both involuntary
treadmill exercise training and voluntary wheel running in animals have consistently demonstrated that
endurance exercise is associated with cardiac hypertrophy (2, 19, 26, 43) and changes in skeletal muscle
protein expression, resulting in a slower contracting,
more oxidative muscle (2–5, 9, 18, 20, 22, 27, 32, 33,
Fig. 3. Fiber percents for the gastrocnemius (A) and TA (B) muscles
from exercised and nonexercised mice. Sections from the gastrocnemius and TA muscles were stained with antibodies to IIa, IIb, and
type I MHC. Counts were done on 500–1,000 fibers in 4–6 randomly
chosen fields to determine fiber percents. Values are means ⫾ SE for
3–4 animals per time point. Voluntary treadmill running resulted in
a decrease in IIb MHC-positive fibers and an increase in IIa MHCpositive fibers in both analyzed muscles. There was no change in the
percentage of IId/x fibers in the gastrocnemius, but in the TA there
was a significant decrease after 4 wk of voluntary running. There
was no change in the percentage of type I fibers in either muscle at
any time point. *P ⬍ 0.05. ** P ⬍ 0.01.
VOLUNTARY RUNNING IN MICE
1905
the rat have examined very long time courses (upward
of 1–2 mo) to provide a greater overall volume of work
for allowing adaptations to occur. In the present study,
we demonstrate that mice undergoing voluntary running exercise will run substantially farther per night
than has been reported in representative studies on
rats run on either a treadmill or voluntary wheel. This
is obviously an underestimate of the true difference,
given the dramatic difference in body size and there-
Fig. 5. Fiber size in exercised vs. nonexercised mice. Fiber crosssectional area (CSA) was measured using NIH Image software on
muscle cross sections immunohistochemically stained with antibodies specific to IIa (SC-71) and type I (MHCs) MHC. The area of 50
individual fibers was determined per muscle. Values are means ⫾
SE; n ⫽ 3–4 animals per time point. The CSA of fibers expressing IIa
MHC was significantly increased at 1, 2, and 4 wk of voluntary
exercise in gastrocnemius but not TA muscle; the CSA of fibers
expressing type I MHC was significantly increased at 4 wk of voluntary exercise in the gastrocnemius. *P ⬍ 0.05. **P ⬍ 0.01.
fore stride length between a mouse and a rat. In addition, the distances run per week reported in the
present study are quite similar to those recently reported for Hsd:ICR outbred mice (21). Thus, although
the intensity of voluntary exercise may be below that of
typical treadmill-based exercise, the substantial duration of the exercise stimulus appears to be sufficient to
produce fairly robust adaptational responses.
Significant cardiac hypertrophy was first evident after just 1 wk of voluntary running, and it reached a
peak after 2 wk of exercise that was maintained
through 4 wk of exercise (Table 2, Fig. 1). This hyper-
Fig. 6. The percentage of NADH-tetrazolium reductase-positive fibers (NADH ⫹) in the TA muscle in exercised and nonexercised mice.
Values are means ⫾ SE; n ⫽ 3–4 animals per time point. NADH ⫺,
NADH-tetrazolium reductase-negative fibers. Voluntary running exercise resulted in a significant increase in the number of NADHtetrazolium reductase positive fibers after 4 wk of voluntary running
exercise. * P ⬍ 0.05.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
Fig. 4. Representative micrograph of
immunostaining for type IIa MHC and
histochemical staining for NADH-tetrazolium reductase in nonexercised (A
and C) and 4-wk exercised (B and D)
mouse TA muscle. A and B: SC-71 immunofluorescence staining for type IIa
MHC. An increase in the number of
fibers positively stained for type IIa
MHC is evident at this time point. C
and D: NADH-tetrazolium reductase
histochemical staining. The number of
NADH-tetrazolium reductase-positive
fibers was increased after 4 wk of voluntary running exercise in mice.
1906
VOLUNTARY RUNNING IN MICE
sion systems in mice are sensitive to endurance exercise training as has been previously characterized in
rats (14, 17, 34, 36, 39).
Few studies to date have examined the time course
of the adaptations in response to voluntary running
exercise (2). Indeed, most studies have used voluntary
running regimens lasting ⬎1 mo to achieve a steady
state of adaptation (4, 5, 18, 26, 45, 46). We sought to
determine the time course of adaptations in response
to voluntary wheel running and examined 1, 2, and 4
wk of voluntary running exercise. Our results demonstrate that cardiac and skeletal muscle adaptations
first become evident as early as 1 wk after initiation of
voluntary exercise, although most of these adaptations
did not reach statistical significance until 2 wk or in
some cases 4 wk of exercise. This suggests that the
adaptational response is initiated fairly rapidly on
initiation of exercise.
Much of the knowledge regarding the adaptation of
physiological systems has come from involuntary
treadmill exercise paradigms. This approach has several obvious advantages, such as the ability to standardize the exercise stimulus for all animals and the
ability to quantify the amount of work done and increase it in a specific manner. In addition, a treadmill
exercise paradigm can be used to diagnose abnormalities in cardiovascular or respiratory function (13).
However, treadmill training also results in increased
stress to animals (12, 30, 42). Voluntary wheel-running
paradigms appear to produce relatively less stress to
animals (31), presumably by allowing animals to exercise when they want, and to the extent that they want,
with no handling or shocks. The voluntary wheel-running paradigm has been fairly well characterized and
has been demonstrated to lead to qualitatively similar
adaptations as the treadmill paradigms, i.e., a shift
toward slower, more oxidative fiber types (22, 32, 33,
39). To date, most studies on voluntary wheel running
in the mouse have not focused on the extent of muscle
adaptation. Thus the present work is one of the first to
systematically characterize both cardiac and skeletal
muscle adaptations occurring with voluntary exercise
in the mouse.
One of the most powerful tools for analyzing the
molecular mechanisms underlying physiological function is the generation of genetically altered lines of
animals, and to date the overwhelming majority of
genetically altered animals have been mice, because
the techniques for undertaking genetic manipulation
of other animals remain less well established (7). Mice
containing alterations to cardiovascular or musculoskeletal genes could be used to analyze the importance
of a specific gene in the exercise adaptation process
using the voluntary exercise system described here (10,
13). Indeed, we have recently observed that a number
of different null and transgenic mouse lines will engage
in voluntary exercise to an extent comparable to wildtype, nontransgenic mice (unpublished observations),
making studies of this type feasible. This “candidate”
gene approach is an excellent complement to genetic
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
trophy was specific to cardiac muscle because there
was no change in either body mas or in mass of the
hindlimb muscles analyzed (Tables 2 and 3). In addition, we observed an increase in ventricular ANF and
BNP mRNA levels after 2 wk of voluntary running that
increased further after 4 wk of voluntary running (Fig.
1). Increased ANF expression is a hallmark of a number of cardiac hypertrophic states (1, 11, 24), although
the relationship among exercise, cardiac hypertrophy,
and ANF expression is more equivocal. An increase in
ANF mRNA was observed in the atria but not ventricles of endurance-trained rats (1), whereas a modest
increase in ANF mRNA but no cardiac hypertrophy
was observed in response to exercise in another study
(28). The present data support the contention that
voluntary running exercise provides a sufficient stimulus to induce both cardiac hypertrophy as well as
alterations in cardiac gene expression in mice.
Similarly, running exercise is associated with distinct metabolic adaptations to skeletal muscle (2–5, 9,
18, 20, 23, 27, 32, 33, 37). We observed a significant
increase in the percentage of NADH-positive muscle
fibers, although this was evident only in the TA and not
the gastrocnemius muscle (Figs. 4 and 6). The increase
in oxidative fibers in the TA is consistent with the
increase in the activity of a number of enzymes of
oxidative metabolism observed in mouse hindlimb
muscle in response to voluntary running (21). It is not
known why an increase in NADH-positive fibers was
observed in the TA but not the gastrocnemius muscle,
although the fact that the TA muscle also underwent a
greater shift in MHC expression (see below) suggests
that this muscle may have a greater adaptive capability in the mouse.
Less well established is the plasticity of the sarcomeric components of mouse skeletal muscle in response to endurance exercise. A study by Wernig et al.
(45) demonstrated histochemical changes in muscle
fiber ATPase in the mouse soleus but not extensor
digitorum longus muscle in response to long-term voluntary wheel running, whereas a recent study on the
medial gastrocnemius muscle of wild house mice revealed only negligible changes in immunohistochemically typed fiber percentages with voluntary wheel
running (46). We observed a decrease in type IIb MHC
and a corresponding increase in type IIa MHC that was
first evident after 1 wk of exercise and continued
through 4 wk of exercise (Figs. 3 and 4). This fiber-type
shift was observed in both an ankle extensor (gastrocnemius) and an ankle flexor (TA), although the magnitude of change was greater for the TA than for the
gastrocnemius. Moreover, we observed no increase in
the percentage of type I fibers in either muscle at any
time point up to 28 days of voluntary running, suggesting that the stimulus from voluntary running was
insufficient to result in a complete shift from fast to
slow MHC. This is consistent with the lack of “complete” fiber transformation to type I MHC expression in
rats and mice even with supraphysiological models
such as chronic electrical stimulation (38). The present
data support the idea that contractile protein expres-
VOLUNTARY RUNNING IN MICE
selection and screening studies, which have already
shown tremendous potential (15, 21, 46).
In summary, we have demonstrated that a voluntary
wheel-running model of endurance exercise for mice
produces distinct adaptational responses in both cardiac and skeletal muscle. This approach complements
the existing model of involuntary treadmill running
and provides another system for examining the molecular and cellular mechanisms governing the adaptation of physiological systems to increased activational
states.
We thank Michelle Laird for help in analyzing muscle fiber size,
Silke Oberdorf-Maass for help with the Northern blotting, and Laura
Sycuro for critical discussions regarding the manuscript.
This research was funded by National Institute of General Medical Sciences Grant GM-29090 (to L. A. Leinwand). D. L. Allen is
supported by a research fellowship grant from the Muscular Dystrophy Association. A. Maass is supported by Deutsche Forschungsgemeinschaft Grant MA 2185/1-1.
1. Azizi C, Bouissou P, Galen FX, Lattion AL, Lartigue M,
and Carayon A. Alterations in atrial natriuretic peptide gene
expression during endurance training in rats. Eur J Endocrinol
133: 361–365, 1995.
2. Baldwin KM, Cooke DA, and Cheadle WG. Time course
adaptations in cardiac and skeletal muscle to different running
programs. J Appl Physiol 42: 267–272, 1977.
3. Baldwin KM, Klinkerfuss GH, Terjung RL, Mole PA, and
Holloszy JO. Respiratory capacity of white, red, and intermediate muscle: adaptive response to exercise. Am J Physiol 222:
373–378, 1972.
4. Beyer RE, Morales-Corral PG, Ramp BJ, Kreitman KR,
Falzon MJ, Rhee SY, Kuhn TW, Stein M, Rosenwasser MJ,
and Cartwright KJ. Elevation of tissue coenzyme Q (ubiquinone) and cytochrome C concentrations by endurance exercise in
the rat. Arch Biochem Biophys 234: 323–329, 1984.
5. Brown M, Ross TP, and Holloszy JO. Effects of ageing and
exercise on soleus and extensor digitorum longus muscles of
female rats. Mech Ageing Dev 63: 69–77, 1992.
6. Butler-Brown GS and Whalen RG. Myosin isozyme transitions occurring during the postnatal development of the rat
soleus muscle. Dev Biol 102: 324–334, 1984.
7. Charreau B, Tesson L, Soulillou JP, Pourcel C, and Anegon I. Transgenesis in rats: technical aspects and models.
Transgenic Res 5: 223–234, 1996.
8. Demirel HA, Powers SK, Naito H, Hughes M, and Coombes
JS. Exercise-induced alterations in skeletal muscle myosin
heavy chain phenotype: dose-response relationship. J Appl
Physiol 86: 1002–1008, 1999.
9. Dudley GA, Abraham WM, and Terjung RL. Influence of
exercise intensity and duration on biochemical adaptations in
skeletal muscle. J Appl Physiol 53: 844–850, 1982.
10. Dupont-Versteegden EE, McCarter RJ, and Katz MS. Voluntary exercise decreases the progression of muscular dystrophy
in diaphragm of mdx mice. J Appl Physiol 77: 1736–1741, 1994.
11. Feldman AM, Weinberg EO, Ray PE, and Lorell BH. Selective changes in cardiac gene expression during compensated
hypertrophy and the transition to cardiac decompensation in
rats with chronic aortic banding. Circ Res 73: 184–192, 1993.
12. Ferry A, Weill BL, and Rieu M. Immunomodulations induced
in rats by exercise on a treadmill. J Appl Physiol 69: 1912–1915,
1990.
13. Fewell JG, Osinska H, Klevitsky R, Ng W, Syfris G, Bahrehmand F, and Robbins J. A treadmill exercise regimen for
identifying cardiovascular phenotypes in transgenic mice. Am J
Physiol Heart Circ Physiol 273: H1595–H1605, 1997.
14. Fitzsimons DP, Diffee GM, Herrick RE, and Baldwin KM.
Effects of endurance exercise on isomyosin patterns in fast- and
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
slow-twitch skeletal muscles. J Appl Physiol 68: 1950–1955,
1990.
Garland T Jr, Gleeson TT, Aronovitz BA, Richardson CS,
and Dohm MR. Maximal sprint speeds and muscle fiber composition of wild and laboratory house mice. Physiol Behav 58:
869–876, 1995.
Gorza L. Identification of a novel type 2 fiber population in
mammalian skeletal muscle by combined use of histochemical
myosin ATPase and anti-myosin monoclonal antibodies. J Histochem Cytochem 38: 257–265, 1990.
Guezennec CY, Gilson E, and Serrurier B. Comparative
effects of hindlimb suspension and exercise on skeletal muscle
myosin isozymes in rats. Eur J Appl Physiol 60: 430–435, 1990.
Gulve EA, Rodnick KJ, Henriksen EJ, and Holloszy JO.
Effects of wheel running on glucose transporter (GLUT4) concentration in skeletal muscle of young adult and old rats. Mech
Ageing Dev 67: 187–200, 1993.
Hickson RC, Galassi TM, and Dougherty KA. Repeated
development and regression of exercise-induced cardiac hypertrophy in rats. J Appl Physiol 54: 794–797, 1983.
Holloszy JO, Oscai LB, Don IJ, and Mole PA. Mitochondrial
citric acid cycle and related enzymes: adaptive response to exercise. Biochem Biophys Res Commun 40: 1368–1373, 1970.
Houle-Leroy P, Garland T Jr, Swallow JG, and Guderley
H. Effects of voluntary activity and genetic selection on muscle
metabolic capacities in house mice Mus domesticus. J Appl
Physiol 89: 1608–1616, 2000.
Ishihara A, Inoue N, and Katsuta S. The relationship of
voluntary running to fibre type composition, fibre area, and
capillary supply in rat soleus and plantaris muscles. Eur J Appl
Physiol 62: 211–215, 1991.
Ivy JL, Young JC, McLane JA, Fell RD, and Holloszy JO.
Exercise training and glucose uptake by skeletal muscle in rats.
J Appl Physiol 55: 1393–1396, 1983.
Izumo S, Nadal-Ginard B, and Mahdavi V. Protooncogene
induction and reprogramming of cardiac gene expression produced by pressure overload. Proc Natl Acad Sci USA 85: 339–
343, 1988.
Kaplan ML, Cheslow Y, Vikstrom K, Malhotra A, Greenen
DL, Nakouzi A, Leinwand LA, and Buttrick PM. Cardiac
adaptations to chronic exercise in mice. Am J Physiol Heart Circ
Physiol 267: H1167–H1173, 1994.
Kingwell BA, Arnold PJ, Jennings GL, and Dart AM. The
effects of voluntary running on cardiac mass and aortic compliance in Wistar-Kyoto and spontaneously hypertensive rats.
J Hypertens 16: 181–185, 1998.
Kriketos AD, Pan DA, Sutton JR, Hoh JF, Baur LA,
Clooney GJ, Jenkins AB, and Storlien LH. Relationships
between muscle membrane lipids, fiber type, and enzyme activities in sedentary and exercised rats. Am J Physiol Regulatory
Integrative Comp Physiol 269: R1154–R1162, 1995.
Mantymaa P, Arokoski J, Porsti I, Perhonen M, Arvola P,
Helminen HJ, Takala TE, Leppaluoto J, and Ruskoaho H.
Effect of endurance training on atrial natriuretic peptide gene
expression in normal and hypertrophied hearts. J Appl Physiol
76: 1184–1194, 1994.
Moore RL. Cellular adaptations of the heart muscle to exercise
training. Ann Med 30: 46–53, 1998.
Moraska A, Deak T, Spencer RL, Roth D, and Fleshner M.
Treadmill running produces both positive and negative physiological adaptations in Sprague-Dawley rats. Am J Physiol Regulatory Integrative Comp Physiol 279: R1321–R1329, 2000.
Noble EG, Moraska A, Mazzeo RS, Roth DA, Charlotte
Olsson M, Moore RL, and Fleshner M. Differential expression of stress proteins in rat myocardium after free wheel or
treadmill run training. J Appl Physiol 86: 1696–1701, 1999.
Powers SK, Grinton S, Lawler J, Criswell D, and Dodd S.
High intensity exercise training-induced metabolic alterations in
respiratory muscles. Respir Physiol 89: 169–177, 1992.
Rodnick KJ, Reaven GM, Haskell WL, Sims CR, and Mondon CE. Variations in running activity and enzymatic adaptations in voluntary running rats. J Appl Physiol 66: 1250–1257,
1989.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017
REFERENCES
15.
1907
1908
VOLUNTARY RUNNING IN MICE
34. Salmons S and Henriksson J. The adaptive response of skeletal muscle to increased use. Muscle Nerve 4: 94–105, 1981.
35. Samorajski T, Rolsten C, Przykorska A, and Davis CM.
Voluntary wheel running exercise and monoamine levels in
brain, heart, and adrenal glands of aging mice. Exp Gerontol 22:
421–431, 1987.
36. Seburn KL and Gardiner P. Adaptations of rat lateral gastrocnemius motor units in response to voluntary running. J Appl
Physiol 78: 1673–1678, 1995.
37. Sexton WL. Vascular adaptations in rat hindlimb skeletal muscle after voluntary running-wheel exercise. J Appl Physiol 79:
287–296, 1995.
38. Simoneau JA and Pette D. Species-specific effects of chronic
nerve stimulation upon tibialis anterior muscle in mouse, rat,
guinea pig, and rabbit. Pflügers Arch 412: 86–92, 1988.
39. Sullivan VK, Powers SK, Criswell DS, Tumer N, Larochelle JS, and Lowenthal D. Myosin heavy chain composition
in young and old rat skeletal muscle: effects of endurance exercise. J Appl Physiol 78: 2115–2120, 1995.
40. Swallow JG, Koteja P, Carter PA, and Garland T Jr. Artificial selection for increased wheel-running activity in house
41.
42.
43.
44.
45.
46.
mice results in decreased body mass at maturity. J Exp Biol 202:
2513–2520, 1999.
Talmadge RJ and Roy RR. Electrophoretic separation of rat
skeletal muscle myosin heavy chain isoforms. J Appl Physiol 75:
2337–2340, 1993.
Tharp GD and Pruess TL. Mitogenic response of T-lymphocytes to exercise training and stress. J Appl Physiol 70: 2535–
2538, 1991.
Tibbits G, Koziol BJ, Roberts NK, Baldwin KM, and Barnard RJ. Adaptation of the rat myocardium to endurance training. J Appl Physiol 44: 85–89, 1978.
Vihko V, Salminen A, and Rantamaki J. Exhaustive exercise, endurance training, and acid hydrolase activity in skeletal
muscle. J Appl Physiol 47: 43–50, 1979.
Wernig A, Irintchev A, and Weisshaupt P. Muscle injury,
cross-sectional area and fibre type distribution in mouse soleus
after intermittent wheel-running. J Physiol (Lond) 428: 639–
652, 1990.
Zhan WZ, Swallow JG, Garland T Jr, Proctor DN, Carter
PA, and Sieck GC. Effects of genetic selection and voluntary
activity on the medial gastrocnemius muscle in house mice.
J Appl Physiol 87: 2326–2333, 1999.
Downloaded from http://jap.physiology.org/ by 10.220.33.2 on June 18, 2017