Mol. Cells, Vol. 22, No. 3, pp. 275-284
Molecules
and
Cells
©KSMCB 2006
Differential Effects of Two Period Genes on the Physiology and
Proteomic Profiles of Mouse Anterior Tibialis Muscles
Kiho Bae*, Kisoo Lee, Younguk Seo, Haesang Lee, Dongyong Kim, and Inho Choi*
Department of Life Science, Yonsei University, Wonju 220-710, Korea.
(Received June 12, 2006; Accepted September 10, 2006)
The molecular components that generate and maintain
circadian rhythms of physiology and behavior in
mammals are present both in the brain (suprachiasmatic nucleus; SCN) and in peripheral tissues. Examination of mice with targeted disruptions of either
mPer1 or mPer2 has shown that these two genes have
key roles in the SCN circadian clock. Here we show
that loss of the clock gene mPer2 affects forced locomotor performance in mice without altering muscle
contractility. A proteomic analysis revealed that the
anterior tibialis muscles of the mPer2 knockout mice
had higher levels of glycolytic enzymes such as triose
phosphate isomerase and enolase than those of either
the wild type or mPer1 knockout mice. In addition, the
level of expression of HSP90 in the mPer2 mutant mice
was also significantly higher than in wildtype mice.
These results suggest that the reduced locomotor endurance of the mPer2 knockout mice reflects a greater
dependence on anaerobic metabolism under stress
conditions, and that the two canonical clock genes,
mPer1 and mPer2, play distinct roles in the physiology
of skeletal muscle.
Keywords: Circadian Clock; mPer Mutant Mice; Muscle
Contraction; Treadmill Running; Two-dimensional Gel
Electrophoresis.
Introduction
Most animals have a circadian rhythm of locomotor activity with periods similar to the earth’s daily rotation about
its axis. Circadian rhythms in mammals are controlled by
free-running and self-sustaining pacemaker neurons located in the anterior hypothalamic region, the suprachi* To whom correspondences should be addressed.
Tel: 82-33-760-2280, 2244; Fax: 82-33-760-2183
E-mail: [email protected] (KB)/ [email protected] (IC)
asmatic nucleus (SCN). It is widely accepted that the cellular mechanism underlying circadian rhythms is composed of interacting transcriptional and translational feedback loops in which the expression of so-called “circadian
clock genes” is feedback regulated by their protein products (Lowrey and Takahashi, 2004; Reppert and Weaver,
2001).
In mammals, two basic helix-loop-helix (bHLH) domain-containing transcription factors called mCLOCK and
BMAL1 form dimers that bind to the promoter regions of
clock genes. They activate the expression of clock genes
including three Drosophila period (per) homologs named
mPer1, mPer2, and mPer3 and two Cryptochrome (Cry)
genes, mCry1 and mCry2. The mPERs and mCRYs form
multimeric protein complexes and enter the nucleus. Once
in the nucleus, they function as negative regulators of
mCLOCK-BMAL1, which results in the decline of their
own expression and lets mCLOCK-BMAL1, reinitiate a
new round of transcription (Gekakis et al., 1998; Shearman
et al., 2000; Ueda et al., 2005).
Peripheral tissues such as liver and skeletal muscles
also express clock genes. Many workers have suggested
that the rhythmicity in peripheral tissues is under the control of the central SCN clock. Neuronal synaptic circuits
and/or humoral factors might mediate communication
between central and peripheral clocks (Balsalobre et al.,
2000; Silver et al., 1997; Stokkan et al., 2001). Therefore,
circadian rhythms in locomotion may be a result of clock
gene activity in both central clocks and cell-specific regulation in muscle cells.
The physiological roles of genes in the circadian clock
are studied by generating mice with gene-specific knockouts (KO). The rhythmicity of locomotor activity is completely disrupted in Bmal1 single KO mice, mPer1 and
mPer2 double KO mice, and mCry1 and mCry2 double
KO mice as measured by wheel-running activity rhythms
Abbreviations: KO, knockouts; SCN, suprachiasmatic nucleus.
276
Skeletal Muscle Physiology in Clock Mutant Mice
in constant conditions (Bae et al., 2001; Bunger et al.,
2000; van der Horst et al., 1999; Zheng et al., 2001). Previously, one of the present authors generated mPer1 and
mPer2 single KO mice and showed that these two clock
genes have distinct roles in the SCN circadian clock. Altough the two KO mice had similar behavioral phenotypes
as measured by wheel-running activity, their running endurance was different. Although a number of factors can
influence locomotor performance (e.g., neural, circulorespiratory, bone properties, etc.), the properties of skeletal muscle are crucial (Dowson et al., 1998; Payne et al.,
2005). Muscle cells are packed with myofibrillar and
metabolic proteins, and the contractile capacity and subcellular attributes of the cells are good predictors of wholeorganism performance (Dawson et al., 1998; Hornberger
and Esser, 2004).
It is still unclear to what exent the two canonical clock
gene, mPer1 or mPer2, affect the functional properties of
muscles. One possibility is that an impairment of the SCN
clock gene significantly affects gene expression in skeletal muscle cells and that this causes changes in their
physiological and biochemical attributes that result in
altered locomotor capacity. However, our Northern blot
analysis of skeletal muscle after a day in constant darkness showed that the level and the phase of clock gene
expression were unaffected in either single KO mouse
(Lee and Bae, 2004). Secondly, local expression of these
two canonical clock genes may have effects unrelated to
the circadian clock. In that case the physiological and
biochemical attributes of the muscles could remain unchanged, or unergo moderate change as the tissues adjust
to altered physiological activities e.g., locomotion.
In the present report, we examined the roles of mPer1
and mPer2 in peripheral tissue physiology. To this end, we
examined the effect of loss of either mPer1 or mPer2 at
three levels: whole-organism locomotor capacity, skeletal
muscle contractility, protein expression in the anterior tibialis (AT) muscles. Our results show that there are differences in the tissue-specific physiology and the level of protein expressions in skeletal muscles between the single KO
mice.
Eight to ten week-old male mice were housed in LD cycles at
least two weeks prior to analysis. All mice studied were of the
same isogenic 129/sv genetic background. This study was approved by the University Animal Care and Use Committee and
the animals were treated in accordance with the NIH guidelines
for the care and use of experimental animals.
Materials and Methods
Treadmill running To investigate inter-group differences in
locomotor capacity, we attempted treadmill running trials on the
mice. The general protocol used in this study was described
previously (Baker and Gleeson, 1999). Each animal was subjected to random-blind tests to avoid experimental bias. Locomotor capacity was gauged from running distance, a typical
endurance measure, conducted in the light cycle at ZT6-8 (Zeitgeber Time 6 [ZT6] means six hours after lights-on time in the
LD cycle) and in the dark cycle at ZT18-20 (ZT18 means six
hours after lights-off time in the LD cycle). The treadmill (l × w
× h =273 mm × 70 mm × 48 mm) was caged with 10 mm thick
plexiglass plates in an air-tight seal for oxygen consumption
measurements (Oxymax System, USA) (Lee et al., 2002). The
speed of the treadmill was adjusted precisely by a Panasonic
DVUX960Y digital speed controller (± 0.1 rpm).
Before the actual trial we let individuals accustomed to the
treadmill run for 10 min every morning and evening over seven
days. We chose the treadmill speed (30–32 m min–1) at which
the animals ran at about 70% of the maximum aerobic metabolic
rate. At the time of trial, a mouse was weighed and exposed to a
warm-up bout of 2 min during which the treadmill speed was
gradually increased to the experimental level. The actual trial
was then started, and video-recorded (30 frames per sec) for
accurate analysis. During the ZT18 treadmill test, a red backlight was placed right behind the treadmill for the video recording. Running duration (min) was determined from the
length of time over which the animal could sustain its speed on
the treadmill with a set number of pull-offs (Huang et al., 2002).
On the basis of preliminary experiments we defined a ‘pull-off’
as an incident in which the running animal pulled back its head
position from a half of the treadmill runway. The trial was completed as soon as the animals reached 20 pull-offs. Video playbacks demonstrated that the rate of pull-off increased with running time as the animals became more exhausted. We obtained
the duration of running of each individual from the video timer,
and calculated running distances (= duration × treadmill speed)
for the inter-group comparisons.
Subjects The animals used in the experiment have been previously described (Bae et al., 2001; Lee et al., 2004a). Breeding
stocks used to establish our colony were kindly provided by Dr.
Steven Reppert and Dr. David Weaver. Mice were maintained at
23° ± 1°C in 12-h light: 12-h dark (LD) cycles, with standard
Purina chow and water provided ad libitum. Genotypes were
determined by polymerase chain reaction (PCR) amplification
of genomic DNA extracted from tail biopsies, as previously
described (Bae et al., 2001). The animals were grouped as wildtype (W), mPer1 knockout (Per1) and mPer2 knockout (Per2).
Muscle contraction The experiment was performed using a
separate set of animals from those in the treadmill tests. Each
animal was anesthetized with sodium pentobarbital (i.p., 50 mg kg−1) either at ZT6 or ZT18, and the anterior tibialis muscles of
both right and left hindlimbs were quickly removed. The AT
muscle is one of the dorsiflexors at the ankle of the hind legs.
The left muscle was frozen immediately in liquid nitrogen for 2dimensional gel electrophoresis (2-DE) (see below). The right
tissue was soaked in 70 ml of cold oxygenated Ringer’s solution
(115 mM NaCl, 5 mM KCl, 4 mM CaCl2, 1 mM MgCl2, 1 mM
Kiho Bae et al.
NaH2PO4, 24 mM NaHCO3, 11 mM glucose, pH 7.39) containing 0.02 g L−1 tubocurarine chloride. One tendon was tied to
the 300B-LR servomotor arm (Aurora Scientific, Canada), and
the other tendon to a micrometer to adjust it to the optimal muscle length.
The servo system has the ability to make both force and
length steps in 1.3 ms and to shift from length control to force
control without any length transient. A custom software program
(DMC, Aurora Scientific) was used to control both the servo
system and a Grass S48 stimulator. The stimulator was connected to a pair of bright platinum electrodes, and supplied a
1.0-ms square wave pulse or pulse train. Data on force or length
changes were stored in a computer and analyzed by custom
software (DMA, Aurora Scientific).
All the experiments (n = 9 for each group) were conducted at
25°C. For each preparation we determined the optimum muscle
length and supramaximal voltage that generated maximum
twitch force. A stimulus frequency of 80 Hz was used to produce fully fused tetani for the three groups. Maximum tetanic
force (Fo) was stabilized, usually after the first three trials. A
rest period of 20 min was given between electrical stimulations.
The shortening velocity of the tissue was examined at 0.35 Fo
as a reference load at which a near maximum power (= shortening velocity x load) was generated, as in previous reports (Lee et
al., 2004b; Rome, 1998). To find this load, we first measured
the shortening velocity of the muscle from quick-release isotonic contractions, with 10 preset loads ranging from 0.1–0.8 Fo.
Length changes were measured over periods of 20–50 ms starting about 15 ms after the release. A hyperbola was fitted to the
data using Hill’s equation, and force-velocity and powervelocity relationships were established. From these preliminary
experiments (n = 3), we determined the load level, 0.35 Fo, that
best represented the force generating near-maximal power (see
Results). This load level was then used in the subsequent experiments to determine the shortening velocity of each muscle.
We rechecked Fo after shortening velocity measurements, and
discarded the muscle and the data if this Fo was less than 93% of
the initial Fo.
After completion of each experiment we measured the optimal muscle length using a micrometer under a light microscope.
The tissue was frozen in place with liquid nitrogen, and its mass
was measured with a chemical balance after tendons and connective tissues were removed. The muscle was then sectioned (8
μm) in a Microm HM505E cryostat and stained by a routine
haematoxylin and eosin procedure. The cross-sectional area of
the section was determined as in a previous study. 25 Muscle
force was normalized with cross-sectional area, and shortening
velocity with muscle length. From the stored force curves, we
also measured tetanic rise time, half-relaxation time, and rate of
tetanic force production. Tetanic rise time was defined as the
time between 0.1Fo and 0.9Fo in the rising phase of the curve;
half-relaxation time was defined as the time between 0.9Fo and
0.5Fo in the falling phase; and the rate of tetanic force production was defined as the highest slope in the rising phase of the
curve.
277
Protein extraction and sample preparation To screen candidate proteins whose expression differed significantly between
the genotypes, we conducted a 2-DE analysis on the AT muscles
collected at ZT6 (n = 4 per group). Our preliminary experiments
showed no obvious differences in protein expression profiles at
two time points, ZT6 and ZT18. On the basis of this 2-DE outcome, we attempted immunoblotting to delineate the detailed
temporal responses of the proteins (see below). After tendons
and connective tissues were removed, each muscle sample was
crushed in liquid nitrogen using a mortar and pestle. The powdered sample was solubilized in 1 ml lysis buffer composed of 7
M urea, 2 M thiourea, 65 mM DTT, 4% CHAPS, 40 mM Tris,
0.5% immobilized pH gradients (IPG) buffer (pH 3–10), and
protein inhibitor Complete Mini (Roche, Penzberg, Germany).
The sample was then incubated for 1 hr at room temperature and
centrifuged for 30 min at 17,000 rpm. The supernatant was collected, and its protein concentration was determined using the
Bradford assay kit (Bio-Rad, USA). For each tissue, a protein
sample of 1.2 mg was used for 2-DE.
Separation of proteins by 2-DE The solubilized proteins were
incubated at 30°C for 30 min with endonuclease (Sigma Chemical, USA), and centrifuged for 15 min at 17,000 rpm. The protein suspension was mixed with a rehydration buffer [7 M urea,
2 M thiourea, 65 mM DTT, 4% CHAPS, 40 mM Tris, 0.5% IPG
buffer (pH 3–10)] to obtain a total volume of 450 μl. The first
dimension isoelectric separation was made on IPG of 3–10 L
(240 × 3 × 0.5 mm) dry strips (Amersham Pharmacia Biotech,
UK) using an Ettan IPGphor Isoelectric Focusing System (Amersham Pharmacia Biotech, UK). The strips were rehydrated at
60 V for 12 h, and focusing was run for 91,420 Vhr using a
gradually increasing voltage protocol implemented by a programmable high voltage power supply (final voltage 8000 V).
Following isoelectric focusing, the strips were equilibrated for
15 min in 10 mL tributyl phosphine (TBP) solution [6 M urea,
2% SDS, 50 mM Tris/HCl gel buffer, 30% (v/v) glycerol (87%
v/v), 5 mM TBP].
The second dimension separation was conducted with an Ettan DALT/six Electrophoresis Unit (Amersham Pharmacia Biotech, UK) on 12% SDS polyacrylamide gels. The strips were
embedded in a 0.5% (w/v) agarose stacking gel, and the proteins
were separated for 12 h at 30 mA per gel until the bromophenol
blue marker dye ran off the bottom of the gel.
After 2-DE, gels were fixed for 12 h in a mixture of 45%
methanol and 5% phosphoric acid, and stained for six hours in a
solution of 0.1% Coomassie brilliant blue G-250, 17% ammonium sulfate, 3.6% phosphoric acid, and 34% methanol. The
gels were then destained for 2 h by diffusion in a mixture of
15% methanol and 1% acetic acid, followed by washing for 1 h
in deionized distilled water.
Quantitative analysis The stained gel was scanned using a PowerLook 1100 scanner (UMAX Technology, USA). The scanned
images were processed with ImageMaster Analysis software
(Amersham Pharmacia Biotech, UK) to yield a list of position
278
Skeletal Muscle Physiology in Clock Mutant Mice
Table 1. Morphological properties of the three groups of mice studied.
Variables
Body mass (g)
Muscle mass (mg)
Muscle mass/body mass (× 10-3)
Muscle length (mm)
Muscle cross-sectional area (mm2)
W
20.96 ± 0.84
29.97 ± 0.11
01.43 ± 0.03
10.12 ± 0.18
02.96 ± 0.12
Per1
20.07 ± 0.600
27.01 ± 0.96*
01.35 ± 0.004
10.54 ± 0.250
2.56 ± 0.04
Per2
21.93 ± 0.380
32.82 ± 1.71*
1.50 ± 0.09
10.89 ± 0.250
3.04 ± 0.19
Data are means ± SE; n = 9.
* Significant differences between groups by one-way ANOVA and Scheffe’s post hoc tests (P < 0.05).
(pI/molecular weight), volume, and intensity information for
each detected spot. To quantify protein levels, spot volumes and
intensities were normalized by expressing the raw volume and
intensity of each individual spot on a gel as a percentage of the
total volume and intensity of protein present on that gel. This
approach offered good reproducibility because normalization
standardizes differences in the total amount of protein present
on each gel.
Protein identification From the scanned gel images, we identified 22 protein spots that appeared to differ in spot volume and
intensity between the three groups. These spots were excised from
the Coomassie-stained gels. Peptide mass fingerprinting was performed at Yonsei Proteome Research Center using a Perkin
Elmer/PerSeptive Biosystems Voyager-DE-PRO MALDI-TOF
mass spectrometer, operating in a delayed reflector mode at an
accelerating voltage of 20 kV.
Immunoblotting From each genotype, the AT muscles were collected at the indicated times during LD cycles and homogenized
at 4°C in three volumes of tissue lysis buffer [20 mM Hepes (pH
7.4), 75 mM NaCl, 2.5 mM MgCl2, 0.1 mM EDTA, 0.05% (v/v)
Triton X-100, 20 mM beta-Glycerophosphate, 1 mM Na3VO4, 10
mM NaF, 0.5 mM DTT, and protein inhibitor Complete Mini
(Roche, Germany)] using a minihomogenizer (Kontes). Homogenates were centrifuged twice (10 min at 12,000 × g) and supernatants saved to new tubes. Protein concentration was determined
using a Commassie protein assay kit according to the manufacturer’s instructions (Pierce). An equal volume of 2× SDS-sample
buffer was added to the supernatant and the mixture boiled at
95°C for 5 min. Equal amounts of total protein (~50 μg total)
were resolved on SDS-polyacrylamide gels, transferred to nitrocellulose membrane and the immunoblots exposed by chemiluminescence (ECL, Amersham) as previously described (Lee et
al., 2004a). 8% to 10% SDS-polyacrylamide gels were used to
visualize HSP90, whereas 10% to 12% gels were used to detect
enolase and triose phosphate isomerase (TPI). The antibodies
used in this study were purchased from Santa Cruz Biotechnology (catalog no. sc-7455 for enolase, sc-22031 for TPI, sc-2922
for HRP-conjugated rabbit anti-goat IgG; Santa Cruz, USA),
Stressgen (product # SPA-830 for HSP-90; Canada) or Lab
Frontier (catalog no. LF-PA0146 for alpha-tubulin; Korea) and
used at final concentrations of 1:3,000 to 1:10,000. Scanned
images of autoradiographs were prepared using Adobe Photoshop 7.0 software.
Data analysis All data are presented as mean ± 1SE, unless
otherwise noted. Differences in the means of morphological and
functional variables between groups were examined with oneway analysis of variance (ANOVA) and Scheffe’s post hoc multiple comparison tests. Statistical procedures were performed
using SPSS/PC+ (SPSS Inc.).
Results
Muscle Physiology
Morphology The three morphological variables (body
mass, muscle length and cross-sectional area) were not
significantly different between the W, Per1 and Per2
groups (one-way ANOVA, P > 0.05; Table 1), and specific mass (relative to body mass) was the same. However
muscle mass was greater in Per2 than in Per1 (P < 0.05).
Locomotion The running distances of the three groups
were compared to see whether diurnal rhythm affects locomotory capacity (Figs. 1A and 1B). Within each group,
running distance did not differ between the light and dark
phases (paired t-test, P > 0.05). In both phases, running
distance was 0.82- to 0.84-fold less in the Per2 group than
in the W group (one-way ANOVA and Scheffe’s tests, F2, 31
= 3.67, P = 0.04 for light phase; F2, 31 = 4.36, P = 0.02 for
dark phase). Running distance did not differ significantly
between the W and Per1 groups and between the Per1 and
Per2 groups (P > 0.05).
Muscle contractile function A typical tetanic force curve
and force-velocity and power-velocity curves are shown
in Figs. 2A and 2B for the AT muscle of wild-type mice.
The contractile properties of the muscle from the three
groups are summarized in Table 2. We found that the five
contractile variables, maximum tetanic tension, tetanic
rise time, half-relaxation time, rate of tension production,
and shortening velocity, did not differ between the three
groups (P > 0.05).
Kiho Bae et al.
279
Table 2. Contractile properties of the anterior tibialis muscle of the three groups.
Variables
Maximum tetanic tension (mN mm-2)
Tetanic rise time (msec)
Half-relaxation time (msec)
Rate of tetanic production (mN mm-2 msec-1)
Shortening velocity (ML sec-1)
W
185.02 ± 11.77
73.33 ± 1.90
12.76 ± 0.29
01.50 ± 0.10
02.15 ± 0.06
Per1
218.30 ± 16.39
76.13 ± 3.69
12.64 ± 0.62
01.77 ± 0.15
02.23 ± 0.05
Per2
206.93 ± 25.56
80.56 ± 4.57
12.66 ± 0.38
01.59 ± 0.21
02.26 ± 0.16
Data are means ± SE; n = 9.
A
B
Fig. 1. Running distance of the W, Per1, and Per2 groups examined in the light cycle (A) and the dark cycle (B). Treadmill speed
was set at 30–32 m min–1 depending on the aerobic capacity of
subjects, and video recording was used to accurately measure
running duration (min). Running distance was determined from
the product of treadmill speed × running duration. adenotes significant difference between W and Per2 for both light and dark
cycles (one-way ANOVA and Scheffe’s method, P < 0.05). Data
are means ± SE (n = 14 for W, 6 for Per1, and 13 for Per2).
Proteomic analysis We initially identified 22 candidate
proteins from the 2D gels and studied in detail four proteins that showed significant differences in spot volume
and intensity. Protein identities and the corresponding gel
spots are given in Table 3 and Fig. 3, respectively. Figure
4 presents comparisons of the expression levels of the
four proteins in the three groups. The results are grouped
as follows.
Contractile proteins We identified on the 2D gels the
seven proteins that are the major constituents of muscle
contraction: actin, tropomyosin α- and β-chains, myosin
essential light chain (skeletal isoform), myosin regulatory
light chain (skeletal isoform), troponin T (slow skeletal
isoform), and troponin I (fast skeletal isoform). None of
these proteins, however, showed significant differences in
expression level between the three groups (P > 0.05).
Metabolic proteins Among seven metabolic enzymes (9
spots) identified, two glycolytic enzymes (TPI and beta
enolase) showed significant inter-group differences in ex-
A
B
Fig. 2. A typical tetanic force curve (A), and force-velocity and
power-velocity curves (B) for the anterior tibialis muscle of a
wild-type mouse. Note that the force level that generated maximum power was 35.6% Fo for this muscle. Muscle mass = 26.4
mg, muscle length = 9.7 mm, and cross-sectional area = 2.72
mm2.
pression levels (Figs. 4A and 4B). The level of TPI, an enzyme catalyzing isomerization between glyceraldehyde 3phosphate and dihydroxyacetone phosphate, was significantly greater in Per2 than in W and Per1 (P < 0.05), but
was not different between W and Per1. The level of beta
enolase, a skeletal muscle enzyme catalyzing the transformation of 2-phosphoglycerate to phosphoenolpyruvate, was
also greater in Per2 than in W (Spot No. 27; P < 0.05). Another spot of beta enolase (No. 24), presumably an electrophoretic variant, showed a similar pattern although there
were no significant inter-group differences.
Regulatory and plasma proteins The level of carbonic
anhydrase III, which catalyzes the reversible hydration of
CO2 and HCO3-, was significantly lower in Per2 than in
W (P < 0.05), while there were no significant differences
in the other intergroup comparisons (Fig. 4C). The levels
of both creatine kinase M chain and serum albumin precursor did not differ.
Stress proteins The level of 90-kDa heat shock protein
(HSP90), which stabilizes proteins before folding or activation (Srikakulam and Winkelmann, 2004), was 0.6-fold
lower in Per1 and 1.7-fold higher in Per2 than in W (Fig.
4D) (P < 0.05). The level of 78-kDa glucose regulated
protein (GRP78), another chaperone protein (Lee et al.,
2002), did not differ among the three groups.
280
Skeletal Muscle Physiology in Clock Mutant Mice
Table 3. Summary of 22 identifiable proteins. The four proteins in boldfaced italic are those that showed significant differences in
expression between groups.
Theoretical pI/MW
using ExPASy*
Spot
No.
Protein identification
Contractile
proteins
4
5
6
7
9
14
43
Actin
Tropomyosin β-chain, skeletal muscle
Tropomyosin α-chain, skeletal muscle
Myosin essential light chain, skeletal isoform
Myosin regulatory light chain, skeletal isoform
Troponin T, slow skeletal muscle
Troponin I, fast skeletal muscle
Metabolic
proteins
3
12
13
24
27
32
33
34
Protein function
Estimated pI/MW
5.20/44.017
4.52/34.900
4.74/33.929
5.04/23.161
4.68/15.900
5.85/31.196
8.34/24.560
pI
5.23
4.72
4.71
5.08
4.70
5.46
8.86
MW
42.024
32.781
32.155
20.64
17.056
31.874
21.197
ATP synthase β-chain mitochondrial precursor
Pyruvate dehydrogenase
Lactate dehydrogenase B-chain
Beta enolase
Beta enolase
Phosphoglycerate kinase
Fructose bisphosphate aldolase A
Fructose bisphosphate aldolase A
4.99/50.947
5.9/38.848
5.7/36.573
6.68/45.750
6.52/46.003
7.57/42.000
8.03/39.728
7.69/39.728
5.04
5.41
5.35
6.63
6.52
7.64
8.00
7.67
50.527
30.053
31.979
45.882
46.131
44.600
39.352
39.331
36
Triose phosphate isomerase
6.87/25.937
6.70
25.775
Regulatory and
plasma
proteins
19
20
28
37
Serum albumin precursor
Serum albumin precursor
Creatine kinase M chain
Carbonic anhydrase III
5.41/66.352
5.49/66.176
6.75/40.352
6.9/27.220
5.44
5.49
6.60
6.90
67.860
67.798
43.019
27.163
Stress proteins
1
2
Heat shock protein 90-kDa
78-kDa glucose regulated protein
5.04/85.913
5.01/72.255
5.03
5.04
89.599
72.060
* The abbreviations pI, MW, and ExPASy represent isoelectric point, molecular weight, and the Expert Protein Analysis System proteomics
server (http://us.expasy.org), respectively.
Immunoblotting To determine the temporal expression
profiles of HSP90 and two glycolytic enzymes (TPI and
enolase), we performed immunoblotting using specific
antibodies against these proteins. Both of the metabolic
enzymes from the AT muscles were expressed constitutively in all three genetic backgrounds (data not shown).
The expression profiles of the two glycolytic enzymes in
W, Per1, and Per2 were similar overall throughout the
daily cycle, with a higher amount in Per2 than in W and
Per1 at ZT6, in accord with the 2-DE analysis. When the
gel runs of the protein samples collected at ZT6 were
compared side-by-side it was obvious that the levels of
HSP90 and TPI were higher in the mPer2 KO mice than
in wild-type mice (P < 0.05) (Figs. 5A and 5B).
Discussion
It is thought that the central SCN clock synchronizes peripheral tissues by means of hormones and the autonomic
nervous system, but very little is known about the role of
peripheral clocks in this process. The need for many pe-
A
B
C
Fig. 3. Representative 2-DE gel maps of the anterior tibialis
muscle for the wild type (A), and Per1 (B), and Per2 (C) mutants. Protein samples (1.2 mg) extracted from the tissue at ZT6
were separated on a wide-range (pH 3−10) IPG L strip in the
first dimension, followed by a 12% SDS-polyacrylamide gel in
the second. The gel was stained with Coomassie blue. Protein
names of the numbered spots are given in Table 3.
Kiho Bae et al.
A
B
281
A
B
C
D
Fig. 4. Inter-group comparisons of expression levels for four proteins: (A) triose phosphate isomerase, (B) beta enolase, (C) carbonic anhydrase III, and (D) 90-kDa heat shock protein. a indicates significant difference between W and Per2; b between Per1
and Per2; * between all the pairs compared (one-way ANOVA and
Scheffe’s method, P < 0.05). Data are means ± SE (n = 3−4).
ripheral clocks may ensure local tissue-specific control of
circadian functions. To decipher the individual roles of
per homologs in the peripheral tissues of mammals we
explored the locomotor and skeletal muscle responses of
mice that had lost one or other of two clock genes, mPer1
or mPer2.
Our data demonstrate that the locomotor capacities of
mPer2 KO mice were more affected than those of mPer1
KO mice; the running distance of the mPer2 mice was
about 20% lower than that of their wildtype littermates in
both light and dark phases (Fig. 1). The loss of mPer2
was not accompanied by significant alterations in the contractility and proteomic responses of the AT muscles except for the expression of a small number of proteins (Tables 2 and 3). There were essentially no significant differences between the tissue and subcellular responses of the
mPer1 mutants and those of wild type mice. These results
point to distinct roles of the two mPer genes in peripheral
tissues and also indicate that the peripheral clock finctions
independently of the SCN clock. It is well established that
the peripheral clock in the liver can be easily uncoupled
from the SCN clock by restricted daytime feeding (Damiola et al., 2000). Recently a master-slave clock hierarchy
was proposed and it was suggested that the SCN has the
role of an orchestrator of several peripheral clocks in different phases with respect to the organismic clock (Yoo et
al., 2004). It is noteworthy that our data indicate that the
Fig. 5. Immunoblotting of HSP90 and two metabolic enzymes,
enolase and triose phosphate isomerase (TPI). Representative gel
images are shown in (A). Protein samples from the anterior tibialis muscles of the W, Per1, and Per2 groups were prepared at
ZT6 and resolved on SDS-polyacrylamide gels, then analyzed
using specific antibodies to HSP90 (top gel), enolase (upper middle gel), or to TPI (lower middle gel). Protein loading was also
measured using antibodies to alpha-tubulin (bottom gel). Quantitation of gels and inter-group comparisons of protein expression
among the genotypes were performed in the same way as in Fig. 4
and are shown in (B). Data are means ± SE (n = 3).
individual clock genes have different effects on noncycling
gene expression in peripheral cells (Figs. 3 and 4). The
differences in the expression of four proteins in the mPer2
mutant mice reflect an adjustment of the myofibers to the
altered locomotor activity, but muscle contractile function
did not differ from the wildtype in either of the clock gene
mutants. This strongly suggests that the myofibrillar
properties of the AT muscle (e.g., numbers of crossbridges, size of sarcoplasmic reticulum) are unaffected by
changes in the circadian clock genes in the brain. Our
proteomic analysis is consistent with the contractile protein outcomes in that expression of most of the contractile
proteins was the same in all three groups.
The reduced running endurance of the mPer2 mutant
mice may be due to a greater dependence on energy production via glycolysis, as these mice expressed higher
levels of glycolytic enzymes (TPI, enolase) (Figs. 4 and
5). Feeding activates glycolysis, which generates pyruvates as end product, and the reduction of pyruvate to
lactate by the tissue-specific action of skeletal muscles
results in a net change of cellular redox state. This redox
change may entrain the peripheral clock. In this regard, it
is worth mentioning that the transcription factor NPAS2,
an mCLOCK homolog, interacts with BMAL1 to bind
DNA and induces circadian gene expression outside the
282
Skeletal Muscle Physiology in Clock Mutant Mice
SCN. The DNA binding activity of NPAS2:BMAL1
dimer is dependent on cellular redox state (Kaasik and
Lee, 2004; Rutter et al., 2001). Recently it has been
shown that the mCLOCK-BMAL1 dimer participates in
the regulation of glucose homeostasis (Rudic et al., 2004).
Moreover, Dallmann et al. (2006) report that their Per1Brd
mutant mice are lighter than WT and Per2Brd mice, and
that the metabolic rate of the Per1Brd mice is higher, with
impaired daily glucocorticoid rhythms. Therefore it is
likely that altered control in peripheral clocks, as in the
case of loss of mPer2, alters mCLOCK:BMAL1-mediated
gene expression, and this may be responsible for the
greater dependence of these mice on anaerobic glucose
metabolism.
The running endurance of mPer2 mice was further limited by the lower level of carbonic anhydrase III, because
cellular acidosis should be enhanced due to the decreased
CO2 and H+ buffering capacity during the 2–3 min of
treadmill running. Indeed, it is known that carbonic anhydrase, blood glucose, pH, and pCO2 display significant
diurnal changes in mammals (Challet et al., 2003) and
that disruption of D site-binding protein (DBP) results in
loss of circadian expression of other genes in the liver,
many of which take part in energy metabolism. There is
also evidence that clock genes regulate the expression of
non-cycling genes (Damiola et al., 2000; Sehgal, 2004).
Therefore, like DBP in liver, mPER2 may serve to perform tissue-specific functions in non-neural tissues including skeletal muscles by regulating the expression of
genes and the activity of their protein products. Furthermore, the expression of HSP90 in AT muscle at ZT6 was
significantly elevated in the mPer2 mice (Figs. 4 and 5).
Interestingly, the expression of HSP90 in both mPer mice
was higher than in wild type mice at ZT18 when measured by immunoblotting (data not shown). This could
reflect a differential level of time-of-day-specific responses to stress in the two groups of mPer mice. We
have obtained evidence for daily variation in the expression of HSP isoforms in the SCN of mPer mutant mice
(Kim and Bae, 2006). Many researchers, including us,
have noted that the expression of this protein is sensitive
to the amount of unloading-reloading stress impinging on
the myofibers (Kregel, 2002; Isfort et al., 2002; Seo et al.,
2006). Hence the mPer2 mutant mice with their higher
level of expression of this protein may be under greater
stress on account of their greater activity and the resulting
acidosis.
We also analyzed the detailed temporal expression profiles of two glycolytic enzymes, TPI and enolase, by immunoblotting. Neither enzyme was rhythmic but instead
constitutively expressed in all genetic backgrounds, ruling
out the possibility that we are simply detecting a phase
difference between the expression of various proteins in
the different genotypes (data not shown). Although the
overall level of enolase in the mPer2 mutant mice was
higher than in the mPer1 mutant and wild type mice, this
was not statistically significant (Fig. 5). The discrepancy
between the 2-DE and immunoblotting data may reflect
the sensitivity of experimental methods (Hirano et al.,
2006). Of the three isoenzymes of enolase, alpha, beta,
and gamma, ββ homodimers and αβ heterodimers are the
predominant forms in muscle tissues (Merkulova et al.,
2000). For immunoblotting, we used polyclonal antibody
raised against a peptide near the C-terminus of the alpha
enolase. The antibody detects both isoforms of enolase
that are abundantly expressed in muscle, and possibly a
variant of beta enolase (Spot No. 24 in 2-DE gel), all of
which makes it difficult to identify subtle differences between the genotypes. An increase in the level of the major
muscle enolase subunit β is correlated with the regeneration of myofibers after injury (Merkulova et al., 2000).
From the 2-DE analysis, the amount of another glycolytic
enzyme, TPI, was significantly higher in the mPer2 mutant mice than in wild type mice. All together the differential modulation of the levels of these proteins may reflect
the adjustment of peripheral effector muscles to altered
physiological needs, particularly in the mPer2 mutant
mice.
There is little other evidence pointing to distinct functions of the two mPER proteins in the circadian clock.
Normally, the two mPERs are expressed in peripheral tissues as well as in the SCN itself in response to a variety of
inducing signals (Balsalobre et al., 2000). Both mPer1 and
mPer2 in the SCN respond to light pulses in a time-of-day
specific manner (Yan and Silver, 2004). Others have shown
that mPer2 mutant mice are prone to develop certain types
of cancer (Fu et al., 2002). Our current results indicate that
there are clear differences in the locomotor capacity and
levels of protein expression in the AT muscles between
mPer1 and mPer2 mutant mice, strengthening the view that
the two mammalian per homologs have distinct functions
in peripheral tissue physiology.
Acknowledgments We are grateful to Drs. Isaac Edery and
David Weaver, and two anonymous reviewers for their valuable
comments on the manuscript. This work was supported by a
Korea Research Foundation Grant (KRF-2002-015-CP0348) to
K. Bae and I. Choi.
References
Bae, K., Jin, X., Maywood, E. S., Hastings, M. H., Reppert, S.
M., et al. (2001) Differential functions of mPer1, mPer2 and
mPer3 in the SCN circadian clock. Neuron 30, 525−536.
Baker, E. J. and Gleeson, T. T. (1999) The effects of intensity
on the energetics of brief locomotor activity. J. Exp. Biol.
202, 3081−3087.
Balsalobre, A., Marcacci, L., and Schibler, U. (2000) Multiple
signaling pathways elicit circadian gene expression in cul-
Kiho Bae et al.
tured Rat-1 fibroblasts. Curr. Biol. 10, 1291−1294.
Bunger, M. K., Wilsbacher, L. D., Moran, S. M., Clendenin, C.,
Radcliffe, L. A., et al. (2000) Mop3 is an essential component of the master circadian pacemaker in mammals. Cell
103, 1009−1017.
Challet, E., Malan, A., Turek, F. W., and van Reeth, O. (2003)
Daily variations of blood glucose, acid-base state and PCO2
in rats: effect of light exposure. Neurosci. Lett. 355, 131−135.
Dallmann, R., Touma, C., Palme, R., Albrecht, U., and Steinlechner,
S. (2006) Impaired daily glucocorticoid rhythm in Per1Brd mice.
J. Comp. Physiol. A. 192, 769−775.
Damiola, F., Minh, L., Preitner, N., Kornmann, B., Fleury-Olela,
F., et al. (2000) Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the
suprachiasmatic nucleus. Genes Dev. 14, 2950−2961.
Dawson, B., Fitzsimons, M., Green, S., Goodman, C., Carey, M.,
et al. (1998) Changes in performance, muscle metabolites,
enzymes and fibre types after short sprint training. Eur. J.
Appl. Physiol. Occup. Physiol. 78, 163−169.
Dowson, M. N., Nevill, M. E., Lakomy, H. K., Nevill, A. M.,
and Hazeldine, R. J. (1998) Modeling the relationship between isokinetic muscle strength and sprint running performance. J. Sports Sci. 16, 257−265.
Fu, L., Pelicano, H., Liu, J., Huang, P., and Lee, C. C. (2002)
The circadian gene Period2 plays an important role in tumor
suppression and DNA damage response in vivo. Cell 111,
41−50.
Gekakis, N., Staknis, D., Nguyen, H. B., Davis, F. C., Wilsbacher, L. D., et al. (1998) Role of the CLOCK protein in the
mammalian circadian mechanism. Science 280, 1564−1569.
Hirano, M., Rakwal, R., Shibato, J., Agrawal, G. K., Jwa, N-S.,
et al. (2006) New protein extraction/solubilization protocol
for gel-based proteomics of rat (female) whole brain and
brain regions. Mol. Cells 22, 119−125.
Hornberger, T. A. and Esser, K. A. (2004) Mechanotransduction
and the regulation of protein synthesis in skeletal muscle.
Proc. Nutr. Soc. 63, 331−335.
Huang, T. H., Yang, R. S., Hsieh, S. S., and Liu, S. H. (2002)
Effects of caffeine and exercise on the development of bone:
a densitometric and histomorphometric study in young wistar
rats. Bone 30, 293−299.
Isfort, R. J., Wang, F., Greis, K. D., Sun, Y., Keough, T. W., et
al. (2002) Proteomic analysis of rat soleus muscle undergoing hindlimb suspension-induced atrophy and reweighting
hypertrophy. Proteomics 2, 543−550.
Kaasik, K. and Lee, C. C. (2004) Reciprocal regulation of haem
biosynthesis and the circadian clock in mammals. Nature 430,
467−477.
Kim, H.-G. and Bae, K. (2006) Effect of mPER1 on the expression of HSP105 gene in the mouse SCN. J. Exp. Biomed. Sci.
12, 53−56.
Kregel, K. C. (2002) Heat shock proteins: modifying factors in
physiological stress responses and acquired thermotolerance.
J. Appl. Physiol. 92, 2177−2186.
Lee, J. W. and Bae, K. (2004) Functional implications of the
mammalian clock gene, mPer expression in the peripheral
tissues. J. Nano Bio. Tech. 2, 94−99.
Lee, M., Choi, I., and Park, K. (2002) Activation of stress signaling molecules in bat brain during arousal from hibernation.
J. Neurochem. 82, 867−873.
283
Lee, C., Weaver, D. R., and Reppert, S. M. (2004a) Direct association between mouse PERIOD and CKIε is critical for a
functioning circadian clock. Mol. Cell. Biol. 24, 584−594.
Lee, K., Lee, Y. S., Lee, M., Yamashita, M., and Choi, I.
(2004b) Mechanics and fatigability of the rat soleus muscle
during early reloading. Yonsei Med. J. 45, 690−702.
Lowrey, P. L. and Takahashi, J. S. (2004) Mammalian circadian
biology: Elucidating genome-wide levels of temporal organization. Ann. Rev. Genomics Hum. Genet. 5, 407−441.
Merkulova, T., Dehaupas, M., Nevers, M., Creminon, C., Alameddine, H., et al. (2000) Differential modulation of α, β
and γ enolase isoforms in regenerating mouse skeletal muscle.
Eur. J. Biochem. 267, 3735−3743.
Payne, R. C., Veenman, P., and Wilson, A. M. (2005) The role
of the extrinsic thoracic limb muscles in equine locomotion.
J. Anat. 206, 193−204.
Reppert, S. M. and Weaver, D. R. (2001) Molecular analysis of
mammalian circadian rhythms. Ann. Rev. Physiol. 63, 647−
676.
Rome, L. C. (1998) Some advances in integrative muscle physiology. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 120,
51−72.
Rudic, R. D., McNamara, P., Curtis, A. M., Boston, R. C.,
Panda, S., et al. (2004) BMAL1 and CLOCK, two essential
components of the circadian clock, are involved in glucose
homeostasis. PLosbiology 2, 1893−1899.
Rutter, J., Reick, M., Wu, L. C., and McKnight, S. L. (2001)
Regulation of Clock and Npas2 DNA binding by the redox
state of NAD cofactors. Science 293, 510−514.
Sehgal, A. (2004) Molecular biology of circadian rhythms. pp.
131−140 and 221−229, John Wiley & Sons, Inc., Hoboken.
Seo, Y., Lee, K., Park, K., Bae, K., and Choi, I. (2006) A proteomic assessment of muscle contractile alterations during
unloading and reloading. J. Biochem. 139, 71−80.
Shearman, L. P., Sriram, S., Weaver, D. R., Maywood, E. S.,
Chaves, I., et al. (2000) Interacting molecular loops in the
mammalian circadian clock. Science 288, 1013−1019.
Silver, R., LeSauter, J., Tresco, P. A., and Lehman, M. N. (1996)
A diffusible coupling signal from the transplanted suprachiasmatic nucleus controlling circadian locomotor rhythms. Nature
382, 810−813.
Srikakulam, R. and Winkelmann, D. A. (2004) Chaperonemediated folding and assembly of myosin in striated muscle. J.
Cell Sci. 117, 641−652.
Stokkan, K., Yamazaki, S., Tei, H., Sakaki, Y., and Menaker, M.
(2001) Entrainment of the circadian clock in the liver by
feeding. Science 293, 490−493.
Ueda, H. R., Hayashi, S., Chen, W., Sano, M., Machida, M., et
al. (2005) System-level identification of transcriptional circuits underlying mammalian circadian clocks. Nat. Genetics
37, 187−192.
van der Horst, G. T. J., Muijtjens, M., Kobayashi, K., Takano,
R., Kanno, S., et al. (1999) Mammalian Cry1 and Cry2 are
essential for maintenance of circadian rhythms. Nature 398,
627−630.
Yan, L. and Silver, R. (2004) Resetting the brain clock: time
course and localization of mPer1 and mPer2 protein expression in suprachiasmatic nuclei during phase shifts. Eur. J.
Neurosci. 19, 1105−1109.
Yoo, S. H., Yamazaki, S., Lowrey, P. L., Shimomura, K., Ko, C.
284
Skeletal Muscle Physiology in Clock Mutant Mice
H., et al. (2004) PERIOD::LUCIFERASE real-time reporting
of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. Proc. Natl. Acad. Sci. USA
101, 5339−5346.
Zheng, B., Albrecht, U., Kaasik, K., Sage, M., Lu, W., et al.
(2001) Nonredundant roles of the mPer1 and mPer2 genes in
the mammalian circadian clock. Cell 105, 683−694.