Brain (2002), 125, 1510±1521
Change of chloride ion channel conductance is an
early event of slow-to-fast ®bre type transition
during unloading-induced muscle disuse
Sabata Pierno,1,* Jean-FrancËois Desaphy,1,* Antonella Liantonio,1 Michela De Bellis,1
Gianpatrizio Bianco,1 Annamaria De Luca,1 Antonio Frigeri,2 G. Paola Nicchia,2 Maria Svelto,2
Claude LeÂoty,3 Alfred L. George Jr4 and Diana Conte Camerino1
1Sezione
di Farmacologia, Dipartimento FarmacoBiologico, FacoltaÁ di Farmacia, 2Dipartimento di
Fisiologia Generale ed Ambientale, UniversitaÁ degli Studi
di Bari, Italy, 3Laboratoire de Physiologie GeÂneÂrale,
Faculte des Sciences et des Techniques, Universite de
Nantes, France and 4Division of Genetic Medicine,
Department of Pharmacology, Vanderbilt University
Medical Center, Nashville, TN, USA
Summary
Disuse of postural slow-twitch muscles, as it occurs in
hypogravity, induces a slow-to-fast myo®bre type transition. Nothing is known about the effects of weightlessness on the resting membrane chloride conductance
(gCl), which controls sarcolemma excitability and in¯uences ®bre type transition during development and
adult life. Using the current±clamp method, we
observed that rat hindlimb unloading (HU) for 1±3
weeks increased gCl in ®bres of the slow-twitch soleus
(Sol) muscle toward values found in fast muscle.
Northern blot analysis suggested that this effect resulted
from an increased ClC-1 chloride channel mRNA level.
In the meantime, a 4-fold increase in ®bres expressing
fast isoforms of the myosin heavy chain (MHC) was
Correspondence to: D. Conte Camerino, Sezione di
Farmacologia, Dipartimento Farmaco-Biologico,
UniversitaÁ degli Studi di Bari, Via Orabona 4 ± campus,
70125, Bari, Italy
E-mail: [email protected]
*These authors contributed equally to this work
observed by immunostaining of muscle sections. Also,
Sol muscle function evolved toward a fast phenotype
during HU, as demonstrated by the positive shift of the
threshold potential for contraction. After 3-days HU,
Sol muscle immunostaining and RT±PCR experiments
revealed no change in MHC protein and mRNA expression, whereas the gCl was already maximally increased,
due to a pharmacologically probed, increased activity of
ClC-1 channels. Thus the increase in gCl is an early
event in Sol muscle experiencing unloading, suggesting
that gCl may play a role in muscle adaptation to modi®ed use. Pharmacological modulation of ClC-1 channels
may help to prevent disuse-induced muscle impairment.
Keywords: chloride channel expression; fast and slow skeletal muscles; hindlimb unloading; mechanical threshold; muscle
plasticity; myosin heavy chain
Abbreviations: CPP = 2-(p-chlorophenoxy)propionic acid; EDL = extensor digitorum longus; gCl = resting membrane
chloride conductance; gK = resting membrane potassium conductance; gm = total membrane conductance; HU = hindlimb
unloading; k = rate constant to reach R; MHC = myosin heavy chain; MT = mechanical threshold; R = rheobase voltage;
Rm = membrane resistance; RMP = resting membrane potential; Sol = soleus; TA = tibialis anterior
Introduction
Skeletal muscles acquire speci®c function to respond to
speci®c needs during development, but can further adapt
in response to modi®ed functional request during adult
life. When exposed to hypogravity conditions as they
occur during space¯ights, the slow-twitch soleus (Sol)
muscle, normally dedicated to postural maintenance,
ã Guarantors of Brain 2002
undergoes a number of functional, morphological and
biochemical changes due to transcriptional and translational processes (Baldwin et al., 1990; Miu et al., 1990;
Baldwin, 1996). Skeletal muscle changes during hypogravity result in a reduction of strength and endurance of
the muscular apparatus, which makes postural mainten-
Chloride ion channel activity during modi®ed muscle use
ance more dif®cult and decreases the locomotion activity
upon return to Earth gravity. Similar symptoms can also
be observed in response to muscle disuse that may occur
during limb immobilization due to spinal cord injury and
acquired or inherited neurological disorders. With the aim
of understanding the mechanism underlying these effects,
several studies have been performed by using the animal
model of rat hindlimb unloading (HU), which closely
matches the microgravity conditions that occur during
space¯ight missions. After space¯ight or following HU,
Sol muscle shows a marked atrophy (Riley et al., 1990;
Booth and Criswell, 1997) and acquires properties typical
of fast-twitch muscles (Diffee et al., 1991; Talmadge,
2000). For instance, despite a decrease in Sol total protein
content, upregulation of the dihydropyridine receptor and
of the fast isoforms of myosin heavy chain (MHC) and
sarcoplasmic reticulum Ca-ATPase has been shown to
occur in response to HU. Also, the metabolic pro®le of
Sol is modi®ed according to a slow-to-fast transition, with
a reduction of oxidative enzymes and an increase of
glycolytic enzyme activity (Manchester et al., 1990). All
these modi®cations are accompanied by a shortening of
relaxation and contraction times, and easier fatigability of
HU slow-twitch muscle ®bres.
However, in order to obtain a complete slow-to-fast
transition of Sol muscle ®bres, changes in ion channel
expression in response to HU must occur. Indeed, ion
channels are deeply involved in the control of sarcolemma
excitability and consequently of muscle contraction, and their
pattern of expression depends on muscle ®bre type.
Nevertheless, very little is known about the modi®cations in
ion channel function induced by hypogravity and other
situations of muscle disuse. Recently, we reported that
expression of voltage-gated sodium channels (Desaphy
et al., 2001) and aquaporin-4 water channels (Frigeri et al.,
2001) is upregulated in the Sol muscles of HU rats in a way
parallel to the slow-to-fast transition of MHC proteins. Of
particular importance in the skeletal muscle are the chloride
channels that support the resting sarcolemmal chloride
conductance (gCl). The gCl plays a pivotal role in the
stabilization of membrane potential and in the control of
excitability (Bretag, 1987). In fact, a dramatic decrease of gCl
due to mutations in the ClC-1 chloride channel gene (CLCN1)
is responsible for the abnormal hyperexcitability typical of
hereditary myotonia (Lehmann-Horn and Jurkat-Rott, 1999).
The slow-twitch muscle phenotype is characterized by a
typical low gCl (Bretag, 1987) that contributes to increasing
sarcolemmal excitability, while the concomitant lower
sodium channel density limits ®ring capability (Milton and
Behforouz, 1995). This allows the slow ®bres to be tonically
active and increases resistance to fatigue. Interestingly,
pharmacological block of gCl during the ®rst period of
postnatal life forces fast-twitch muscles to develop a slow
phenotype (Conte Camerino et al., 1989; De Luca et al.,
1990). In light of these observations it can be hypothesized
that ion channel activity can be a speci®c target of the
1511
hypogravity condition, and that the modi®cation of their
activity may be important for the functional impairment
observed in Sol muscle. If such is the case, pharmacological
modulation of ion channels may be a potential counteractive
measure that could be taken into consideration during muscle
disuse. The present study was aimed at evaluating the effects
of hypogravity on gCl and on the expression of ClC-1
chloride channels in rat Sol muscle ®bres after periods of HU
ranging from 3 days to 3 weeks. To compare the time-course
of these effects with the slow-to-fast transition occurring in
HU Sol muscle, the progression of muscle atrophy, the
transition in MHC proteins and the mechanical threshold
(MT, a measure of the excitation±contraction coupling
mechanism) were evaluated in parallel. Finally, similar
measures were performed in the fast-twitch extensor
digitorum longus (EDL) muscle to verify whether the
observed effects were speci®c to slow-twitch muscles. The
most striking result is that gCl dramatically increases in Sol
muscle ®bres of HU rats owing to an increase in ClC-1
channel expression. The maximum effect was obtained after 3
days in the HU model, before any change in MHC protein and
messenger could be detected, indicating that change in gCl is
an early event of the slow-to-fast transition of myo®bre type
occurring during muscle disuse.
Material and methods
Animal care and surgery
Experiments were conducted in accordance with the Italian
guidelines for the use of laboratory animals, which conform
with the European Community Directive published in 1986
(86/609/EEC). Male Wistar rats weighing 250±350 g
(Charles River Laboratories, Calco, Italy) were randomly
assigned to control or HU groups. To induce muscle
unloading, the animals were suspended individually in
special cages for 1±3 weeks by a harness linked to a trolley
by a lace (Desaphy et al., 2001). The trolley can move
horizontally on rails at the top of the cage, allowing free
movement of the animal. The lace length was adjusted to
allow the forelimbs of the rat to touch the bottom of the cage,
leaving the hindlimbs free and the animal's body inclined by
~45° from the horizontal plan. Control and suspended
animals had food and water ad libitum. At the end of the
period of suspension (0±3 weeks), muscles were removed
from the animal under deep anaesthesia induced by
intraperitoneal injection of urethane (1.2 g/kg body weight).
The slow-twitch Sol muscle and the fast-twitch EDL muscle
were promptly used for the electrophysiological experiments.
The Sol, EDL and tibialis anterior (TA) muscles isolated from
the contralateral limb of the same animals were immediately
frozen in liquid nitrogen and stored at ±80°C until RNA
isolation. At the end of the surgical intervention, animals
were killed either by an overdose of urethane or by
decapitation.
1512
S. Pierno et al.
Immunohistochemistry
Immuno¯uorescence staining was performed as described
previously (Frigeri et al., 1998, 2001). Brie¯y, Sol muscles
from control, 3 days, 1, 2 and 3 weeks HU rats were dissected
and rapidly frozen in isopentane cooled with liquid nitrogen.
Cryostat 5-mm cross sections were incubated with type IIa,
IIb (Schiaf®no et al., 1989) or I (1 : 500 dilution; Sigma, St
Louis, MO, USA) MHC mouse monoclonal antibodies for 1 h
at room temperature. After washing, sections were incubated
for 1 h with FITC-coupled goat anti-mouse antibodies (1 : 100
dilution). Sections were examined with a Leica DMRXA
photomicroscope equipped for epi¯uorescence, and digital
images were obtained with a cooled CCD camera (Princeton
Instruments, Princeton, NJ, USA). The percentage of type I,
IIa and IIb ®bres in Sol muscles was determined by the
immuno¯uorescence stain of cryostat sections.
Semi-quantitative RT±PCR
Sol muscle RNA was prepared using the TRIzol reagent
(Gibco-Life Technologies, Inc., Grand Island, NY, USA) and
cDNAs were prepared using random primers (Frigeri et al.,
2001). The cDNAs were used to amplify a 600-bp fragment
using speci®c primers for the type I MHC sequence (sense 5¢GAAGGCCAAGAAGGCCATC-3¢; antisense 5¢-GGTCTCAGGGCTTCACAGGC-3¢) and the type IIa MHC (sense 5¢GAAGGCCAAGAAGGCCATC-3¢; antisense 5¢-TCTACAGCATCAGAGCTGCC-3¢) (Wright et al., 1997). Relative
amounts of type I and type IIa MHC transcripts were
estimated by direct comparison between multiple samples
after standardization with co-ampli®ed 18S rRNA
(QuantumRNA kit; Ambion, Austin, Tex., USA).
Densitometric analysis was performed using Scion Image
software and statistical comparison between control and HU
rats was conducted by the Student's t-test for unpaired data.
Northern blot analysis of ClC-1 channel
expression
Total RNA was isolated from Sol, EDL and TA muscles of
control and HU rats using a modi®ed acid±phenol method
(Chomczynski and Sacchi, 1987). About 100±400 mg
(depending on muscle type: Sol muscles were smaller than
EDL or TA muscles) of each muscle was homogenized in
TRIzol reagent (1 ml/mg of tissue) and the homogenate was
centrifuged at 12 000 g for 10 min. The aqueous supernatant
containing the RNA was removed and precipitated with 2propanol, dissolved in diethylpyrocarbonate-treated water
and stored at ±80°C until analysis. RNA isolated with this
method was fully intact, as determined by agarose gel
electrophoresis. The amount of RNA recovered from the
tissue was determined by UV spectrophotometry at 260 nm.
Total RNA (10 mg) was size-fractionated on denaturing 1%
agarose/6% formaldehyde (v/v) gels and transferred to a
nylon membrane (Hybond-N; Amersham, Princeton, NJ,
USA) as described previously (Pierno et al., 1999a). Northern
blots were sequentially hybridized with a rat ClC-1 cDNA
probe (nucleotides 298±638) and with 18S ribosomal RNA
radiolabelled probe as internal reference in order to establish
the relative amount of RNA in each sample. The ClC-1 probe
was radiolabelled with [32P]dCTP using the random priming
method. Hybridization was performed at 42°C for 16 h in
50% formamide/53 SSPE (13 SSPE is 0.18 M NaCl/10 mM
Na2HPO4/1 mM EDTA)/1% sodium dodecyl sulphate (SDS)/
0.05 M Tris±HCl, pH 7.5/5 mM EDTA/1% bovine serum
albumin (BSA) and 1 3 10±6 c.p.m./ml 32P-labelled cDNA
probe. Blots were washed with a ®nal stringency of 65°C in
0.13 SSC/0.1% SDS following hybridization with cDNA
probes. A radiolabelled 18S RNA riboprobe was synthesized
by in vitro transcription using an antisense control template
(Ambion), T7 polymerase and [32P]CTP. Hybridization with
18S riboprobe was performed at 60°C with 0.5 3 10±6 c.p.m./
ml 32P-labelled probe in the same hybridization solution,
followed by washes in 0.13 SSC/0.1% SDS at 68°C.
Hybridizing bands were visualized by autoradiography and
quanti®ed by PhosphorImager analysis using ImageQuant
software (Molecular Dynamics, Sunnyvale, CA, USA).
Electrophysiological studies
Using a computer-assisted two intracellular microelectrodes
technique in current±clamp mode, the gCl was measured
in vitro from rat Sol and EDL muscle ®bres. Soon after the
removal from the rat, Sol or EDL muscle, tied at the end of
each tendon, was placed on a glass rod located in a 25-ml
muscle bath chamber maintained at 30°C and continuously
perfused with 95% O2/5% CO2-gassed normal and chloridefree physiological solutions (Bryant and Conte Camerino,
1991). The normal (chloride-containing) physiological solution had the following composition (in mM): 148 NaCl, 4.5
KCl, 2 CaCl2, 1 MgCl2, 12 NaHCO3, 0.44 NaH2PO4, 5.5
glucose, pH 7.3. The chloride-free solution was made by
equimolar substitution of methylsulphate salts for NaCl and
KCl and nitrate salts for CaCl2 and MgCl2 (Bryant and Conte
Camerino, 1991). The resting membrane potential (RMP)
was ®rst measured with a single standard microelectrode. The
membrane resistance (Rm) and the ®bre diameter were
calculated by injecting a hyperpolarizing constant square
current pulse into the muscle ®bre through one microelectrode and recording the resulting voltage de¯ection with a
second microelectrode inserted at two different distances
from the current electrode. The current pulse generation,
acquisition of the voltage records and calculation of the ®bre
constants were carried out under computer control, as detailed
elsewhere (Bryant and Conte Camerino, 1991). The total
membrane conductance (gm) was 1/Rm in the normal
physiological solution. The component membrane conductances to chloride and potassium ions were calculated from
Rm values in normal and chloride-free solutions. The
potassium conductance (gK) was gm in the chloride-free
solution. The mean chloride conductance gCl was calculated
Chloride ion channel activity during modi®ed muscle use
1513
Table 1 Effects of HU on cable parameters and macroscopic ionic conductances measured in muscle ®bres of adult rats
Muscle type
Nln
RMP (±mV)
Rm (W ´ cm2)
GK (mS/cm2)
GCl (mS/cm2
EDL
3 week
Sol
1 week
2 week
3 week
5/57
6/78
8/117
8/115
5/67
12/158
65
62
64
60
61
62
358
337
588
445
421
418
310
350
280
280
325
337
2728
2863
1577
2243
2257
2306
HU EDL
HU Sol
HU Sol
HU Sol
6
6
6
6
6
6
1.6
1.1
0.8
0.5²
0.9²
0.6*
6
6
6
6
6
6
15
12
15*
17²
17²
11*
6
6
6
6
6
6
24
27
25
15
30
21
6
6
6
6
6
6
81
74
49*
55²
76²
56*
Each parameter was measured in Sol muscle ®bres from rats after 1±3 weeks HU and compared with control EDL and Sol muscle ®bres.
N is the number of animals from which n ®bres were examined. All values are expressed as mean 6 SEM. *Signi®cantly different
compared with EDL (P < 0.005); ²signi®cantly different compared with Sol (P < 0.02).
as the mean gm minus the mean gK (Bryant and Conte
Camerino, 1991). The data are expressed as mean 6 standard
error of the mean (SEM) and were calculated by standard
methods from the variance of gm and gK, assuming no
covariance. Statistical differences between mean values were
evaluated using unpaired Student's t-test, considering P <
0.05 as signi®cant.
The MT for contraction was determined using a two
microelectrode point voltage clamp method in the presence of
3 mM tetrodotoxin, as described previously (Pierno et al.,
1999b). The holding potential was set at ±90 mV.
Depolarizing current pulses of increasing duration (5±500
ms) were given repetitively at a rate of 0.3 Hz, while the
impaled ®bres were viewed continuously with a stereomicroscope. The command voltage was increased until contraction
was just visible and the threshold membrane potential at this
point was read from a digital sample-and-hold voltmeter. The
mean threshold membrane potential V (mV) 6 SEM (n ®bres)
was plotted as a function of the pulse duration t (ms), and the
relationship was ®t using a non-linear least squares algorithm
using the equation, V(t) = [HR´exp(±k´t)]/[1 ± exp(±k´t)],
where H (mV) is the holding potential, R (mV) is the rheobase
voltage, and k (/ms) is the rate constant to reach R. The MT
values were expressed as the ®tted R parameter along with the
standard error that was determined from the variance±
covariance matrix in the non-linear least squares ®tting
algorithm. Statistical differences between ®t parameters were
evaluated using unpaired Student's t-test, considering P <
0.05 as signi®cant.
Results
HU increases the resting sarcolemmal chloride
conductance and ClC-1 chloride channel
expression in Sol muscle ®bres
The slow- and fast-twitch muscle ®bres showed signi®cant
differences in the passive cable properties (Table 1). The Rm
measured in Sol muscle ®bres was signi®cantly higher than in
EDL muscle ®bres, and this higher value was completely
attributable to a 43% lower resting gCl in Sol muscle
compared with that of EDL. Indeed, the value of gK was only
slightly, and not signi®cantly, lower in Sol muscle ®bres. In
addition, the RMP was quite similar in both muscles (Table
1). After HU for 1±3 weeks, a reduction of Rm toward that of
EDL muscle ®bres was observed in Sol muscle ®bres (Table
1). This reduction was due to a signi®cant 40% increase of
gCl in HU Sol muscle at each time point. The gK slightly
increased in both EDL and Sol muscles but this did not reach
statistical signi®cance. Such effects were constantly observed
with little variability between individual HU animals. The
changes in Rm and gCl were speci®c to the slow-twitch
muscle, because they were not observed in EDL muscle ®bres
after 3 weeks HU (Table 1). A slight but signi®cant reduction
of RMP was also measured after 1±3 weeks HU in both
muscle types (Table 1).
Several lines of experimental evidence support the suggestion that sarcolemmal gCl is mainly supported by the
skeletal muscle-speci®c voltage-gated ClC-1 channel
(Klocke et al., 1994; Pierno et al., 1999a). Thus, we looked
at ClC-1 transcript expression levels in whole fast- and slowtwitch muscles. As has already been described (Klocke et al.,
1994), northern blot analysis showed lower content of ClC-1
mRNA in Sol muscles as compared with EDL muscles (data
not shown), consistent with the lower gCl recorded in Sol
muscle ®bres. In Sol muscles, ClC-1 mRNA expression
increased with HU (Fig. 1A). Indeed the amount of ClC-1
mRNA (normalized to 18S rRNA) exhibited a 48% and 83%
increase in Sol muscles from 1 and 3 weeks HU rats,
respectively (Fig. 1B). A good linear correlation (P < 0.05)
was found between normalized ClC-1 mRNA levels and the
gCl measured in the contralateral Sol muscle (Fig. 1C). In
contrast, northern blot analysis of the EDL and TA fast-twitch
muscles of control and HU rats showed no change in ClC-1
mRNA content in agreement with the lack of modi®cation of
the electrical parameters, especially gCl, observed in EDL
muscles of HU rats (data not shown).
To determine whether the increase in gCl was homogeneous within the whole Sol muscle, we plotted the
distribution of single ®bres as a function of gCl, using a
sampling value of 300 mS/cm2 (Fig. 2). The histograms were
®tted with Gaussian functions using a Levenberg±Marquardt
least-squares ®tting routine (Pstat software, Pclamp 6.0
Package; Axon Instruments, Calif., USA). In both Sol and
EDL muscles of control rats (Fig. 2A and B), ®bres were
1514
S. Pierno et al.
Fig. 1 Increased ClC-1 mRNA expression in Sol muscles of HU
rats. (A) Representative northern blot shows ClC-1 mRNA and
18S rRNA signals in Sol muscles of 1 and 3 weeks HU rats and
control animals. Each lane contains 10 mg of total RNA extracted
from a Sol muscle of one animal. Size standards are indicated on
the left. (B) ClC-1 mRNA levels were quanti®ed by
PhosphorImager analysis and normalized with respect to the 18S
hybridization signals. Each bar represent the mean 6 SEM of at
least six muscles. (C) Normalized ClC-1 mRNA levels were
plotted against the value of gCl measured in the contralateral Sol
muscle of the same animal. Circles indicate the values found in
control animals. Squares and triangles indicate the values found in
1 and 3 weeks HU animals, respectively. The data are linearly
correlated (r = 0.73; P < 0.05).
Fig. 2 Fibre distribution as a function of gCl in rat muscles. (A)
gCl was measured in EDL ®bres (n = 57) and (B) Sol muscle
®bres (n = 117) of control rats compared with (C) gCl measured
in Sol muscle ®bres (n = 158) of HU rats. Histograms were
constructed using a sampling interval of 300 mS/cm2 and ®tted
with a Gaussian function. The ®bres of control Sol and EDL
muscles follow a monotone distribution, whereas ®bres of HU Sol
muscles are distributed among two distinct populations.
Chloride ion channel activity during modi®ed muscle use
1515
Fig. 3 Type IIa ®bre determination in Sol muscle after HU. Rat Sol muscle cryostat sections were assayed for MHC protein expression by
immuno¯uorescence. (A) Control Sol muscle sections stained with anti-type IIa MHC antibodies and (B) anti-type I MHC antibodies. (C±
F) Sections stained with anti-type IIa MHC antibodies from Sol muscles of rats after 3 days and 1, 2, 3 weeks HU in C, D, E and F,
respectively.
1516
S. Pierno et al.
normally distributed around one Gaussian mean value similar
to the arithmetic mean reported in Table 1. In contrast, after 1
week HU, the ®t of histogram was signi®cantly improved
using a two-Gaussian function, as evaluated by the F value (P
< 0.05), which compares the sum of squared errors of the
different ®tting models. Fibres pooled from Sol muscles of 1±
3 weeks HU rats clearly follow a bimodal distribution around
two mean values, 1689 and 2767 mS/cm2, similar to those
obtained in control Sol and EDL muscles, respectively (Fig.
2C). This suggests that after 1 week HU, about half of the
®bres of the Sol muscle maintained a gCl typical of the
control Sol muscle, whereas their counterparts acquired a gCl
typical of the fast EDL muscle.
HU modi®es MHC isoform distribution and
induces atrophy in Sol muscle
HU is well known to induce a slow-to-fast transition of Sol
muscles (Talmadge, 2000). We veri®ed such a change in ®bre
phenotype in our model by performing immuno¯uorescence
experiments in normal and suspended Sol muscles using
speci®c antibodies against type I, IIa and IIb MHC isoforms
(Fig. 3). No type IIb ®bres were detected in control or in HU
Sol muscles (data not shown). In contrast, as shown in Fig.
4A, the number of fast type IIa MHC ®bres was increased
dramatically from <15% of total ®bres in control Sol muscle
to ~35% in Sol muscles of 1±3 weeks HU rats. In parallel, the
number of type I MHC ®bres decreased from >85% in control
Sol muscle to 62% after 1 week HU, and to 55% after 2 and 3
weeks HU (Fig. 4A). From 1 week HU, some ®bres were not
recognized by any of the three antibodies, suggesting the
presence of type IIx MHC ®bres, as reported by other authors
(Stevens et al., 1999; Talmadge, 2000).
Together with slow-to-fast MHC transition, muscle atrophy is a main feature of HU-induced muscle change (Riley
et al., 1990; Booth and Criswell, 1997). Thus we evaluated
atrophy of Sol muscle by measuring the muscle-to-body
weight ratio and by calculating the ®bre diameter with the
program used for cable parameter evaluation (see Material
and methods). In control Sol muscle, the muscle-to-body
weight ratio was 0.47 6 0.04 mg/g (®ve rats) and the ®bre
diameter was 56.6 6 1.8 mm (117 ®bres from eight rats). A
marked and signi®cant decrease of these two parameters was
observed in HU Sol muscle compared with the control (Fig.
4B). The atrophy progressively developed along the HU
period. In contrast, atrophy was not observed in the fast EDL
muscle after 3 weeks of suspension. Indeed the muscle-tobody weight ratio for EDL muscles was 0.51 6 0.01 mg/g (n
= 5) and 0.49 6 0.03 mg/g (n = 7) in control and 3 weeks HU
rats, respectively (P > 0.05, unpaired Student's t-test). In
addition, only a slight and non-signi®cant change of ®bre
diameter was observed in EDL muscle, which decreased from
65.0 6 2.1 mm (57 ®bres from ®ve rats) in control rats to
59.6 6 1.9 mm (78 ®bres from six rats) after 3 weeks HU (P >
0.05, unpaired Student's t-test).
HU modi®es Sol muscle function
To verify in our study whether the change in MHC isoforms
in HU Sol muscle was accompanied by changes in muscle
function, especially in calcium handling and contractile
properties, we measured the MT for contraction in single
®bres from Sol and EDL muscles of control and HU rats, as
an evaluation of the excitation±contraction coupling mechanism (Fig. 5). The threshold potential needed to elicit
contraction was determined by applying pulse current of
various duration (5±500 ms) and showed the typical dependence on command pulse duration, being more negative the
longer the duration of the pulse. Slow- and fast-twitch muscle
®bres showed different MT for contraction in accordance
with the different contractile patterns that characterize each
muscle phenotype. Indeed, ®bres from Sol muscle needed
signi®cantly less depolarization to contract at each pulse
duration and, consequently, the threshold potential±duration
relationship was shifted toward more negative potentials in
Sol muscles as compared with EDL (Fig. 5). R, estimated
from the ®t of the experimental points, was ±72.8 6 1.0 mV
in Sol muscle ®bres and ±62.3 6 0.2 mV (P < 0.001) in EDL
muscle ®bres. The value of k to reach R was signi®cantly
lower in Sol (0.11 6 0.01/ms) compared with EDL muscle
®bres (0.15 6 0.01/ms, P < 0.001). After 3 weeks HU, the
strength±duration curve measured in Sol was shifted towards
that of control EDL muscle (Fig. 5), R (±69.5 6 0.7 mV) and
k (0.12 6 0.01/ms) reaching an intermediate value between
those of control Sol muscle (P < 0.001) and control EDL
muscle (P < 0.001). Similar results were obtained after just 1
week HU in Sol muscle; indeed R was ±67.8 6 0.8 mV (P <
0.001 versus control Sol) and k was 0.12 6 0.01/ms (P <
0.001 versus control Sol; data not shown).
Early effects of HU on Sol muscle properties
To determine whether the change in gCl precedes or follows
the slow-to-fast ®bre type transition, we evaluated these
parameters in Sol muscles after 3 days HU. Immunostaining
experiments failed to reveal any change in MHC protein
expression in cross-section of 3 days HU Sol muscles (Figs
3C and 4A). Because effects may have been initiated at the
messenger level while protein content remained constant, we
looked at MHC mRNA expression levels using semi-quantitative RT±PCR. No signi®cant change in mRNA levels for
type I and IIa MHC was detected in Sol muscles of 3 days HU
rats (Fig. 6A). As expected from such a short HU period, only
a slight but signi®cant atrophy of Sol muscles was measurable
in 3 days HU rats (Fig. 4B). In contrast, the gCl was largely
increased in Sol muscles ®bres of 3 days HU rats, reaching
2242 6 53 mS/cm2 (mean 6 SEM of 80 ®bres from ®ve rats;
P < 0.02 versus control), a value very similar to that found in
1±3 weeks HU rats (Fig. 6B and Table 1). We took advantage
of a selective pharmacological tool to verify whether the
increase in gCl was attributable to change in ClC-1 channels.
We tested the S(±)-isomer of 2-(p-chlorophenoxy)propionic
Chloride ion channel activity during modi®ed muscle use
1517
acid (CPP), a chiral clo®bric acid derivative that selectively
blocks ClC-1 channels over the other members of the ClC
family (Pusch et al., 2000). The IC50 for gCl block by S(±)CPP in 3 days HU Sol muscle ®bres was very similar to that
found in control Sol muscle ®bres (Fig. 6C), and corresponded to the IC50 found in EDL muscle ®bres of control
adult rats (14 6 1.5 mM; see Pusch et al., 2000). We are aware
of only two studies reporting the use of ClC-1 antibodies,
where western blot analysis of hClC-1-transfected cell
preparation and mouse skeletal muscle membrane-enriched
preparation were performed using an af®nity-puri®ed rabbit
polyclonal antibody directed against the carboxy-terminus of
rat ClC-1 (Gurnett et al., 1995; Fahlke et al., 1997).
Unfortunately, we obtained no satisfactory signal using the
same antibody on whole-muscle protein preparation.
Discussion
The present study shows that muscle unloading for 1±3 weeks
causes an increase in expression of ClC-1 chloride channel
mRNA that accounts for an increase in gCl in fully
differentiated slow-twitch muscle. During the same period,
these effects are paralleled by the slow-to-fast transition of
MHC protein expression and contractile function of Sol
muscle. However, measurements in Sol muscle after 3 days
unloading show that the change of gCl occurs before any
change in MHC protein and mRNA expression can be
detected. This result indicates that the increase in gCl is an
early event in Sol muscle adaptation to modi®ed use.
Fig. 4 Changes in MHC protein expression and atrophy of Sol
muscle during HU. (A) MHC protein expression was determined
on Sol muscle cryostat sections by immuno¯uorescence using
speci®c antibodies directed toward type I, IIa and IIb MHC
isoforms. No type IIb MHC was detected. For MHC type I
(MHCI) and MHC type IIa (MHCIIa), each point represents the
mean 6 SEM of at least two muscles. The circles (indicated as
`others') are for the residual unstained ®bres, corresponding most
probably to ®bres expressing type IIx MHC. (B) Atrophy of Sol
muscle that occurred during HU. The ®bre diameter calculated
from ®bre cable parameters was averaged from 67 to 158 ®bres in
each experimental condition and normalized with respect to the
mean value found in control rat Sol muscle ®bres. The mean of
muscle-to-body weight ratio was calculated from ®ve to 12 rats in
each experimental condition and normalized with respect to that
found in control rats. Statistical differences between the mean
values were evaluated using the unpaired Student's t-test (*P <
0.001 versus control rat).
Fig. 5 MT in EDL and Sol muscle ®bres of control rats and in Sol
muscle ®bres of 3 weeks HU rats. The strength±duration
relationships were obtained by plotting the values of threshold
potentials (V) needed to obtain muscle ®bre contraction as a
function of the voltage pulse duration (t). Each point represents
the mean value 6 SEM of ®ve to 40 ®bres from two or three rats.
The curves ®tting the experimental points were obtained using the
equation V(t) = [±90 ± R´exp(±k´t)]/[1 ± exp(±k´t)]. Fit parameter
values along with the standard error of the ®t are given in the text.
1518
S. Pierno et al.
Slow-to-fast transition of myo®bre type in Sol
muscle after hindlimb suspension
Owing to its plasticity, adult skeletal muscle has the capacity
to adapt by modifying its contractile and metabolic properties
in response to the in¯uence of different patterns of electrical
activity or to environmental stimuli such as muscle loading
(Buonanno and Fields, 1999; Pette and VrbovaÁ, 1999;
Talmadge, 2000). Indeed the removal of the mechanical
load normally imposed to the hindlimb muscles was shown to
modify the expression of genes encoding speci®c fast and
slow protein isoforms responsible for the determination of the
phenotype-speci®c contractile and metabolic properties of
muscle (Talmadge, 2000). These modi®cations were mainly
observed in the slow-twitch muscles involved in the postural
control and are consistent with the idea that HU induces
expression of the fast genes over a slow program.
Accordingly, using the rat HU model, we observed a slowto-fast myo®bre type transition in the antigravity slow-twitch
Sol muscle, characterized by an increase in type II MHCcontaining fast ®bres together with a decrease in type I MHCcontaining slow ®bres. This effect was detectable after 1
week HU but not after 3 days HU, suggesting that effects on
MHC protein contents initiate within this period. This is
consistent with recent data describing only slight effects on
Fig. 6 Effects of 3 days HU on Sol muscle properties. (A) Semi-quantitative RT±PCR measurements of type I and IIa MHC mRNA
levels. The 488-bp band corresponds to the 18S rRNA internal standard whereas the ~600-bp band corresponds to the MHC-speci®c PCR
product. The bar graph shows the averaged results 6 SEM obtained from three control and three HU animals. No signi®cant difference
was found using unpaired Student's t-test (P > 0.3). (B) Values of gCl found in control and 3 days HU rat Sol ®bres. (C) Concentration±
response relationships for the blocking effect of S(±)-CPP on gCl in Sol muscle ®bres. Each data point was calculated from the mean gCl
measured in 15±36 ®bres from a minimum of two rats. The relationships were ®tted using the equation gClCPP/gClcontrol = 1/{1 + ([CPP]/
IC50)h}, where IC50 is the half-maximal inhibitory concentration and h is the logistic slope factor. These parameters along with the
standard error of the ®t were IC50 = 11.7 6 0.3 mM and h = ±1.05 6 0.03 for control Sol muscles, and IC50 = 9.1 6 0.4 mM and
h = ±1.0 6 0.05 for HU Sol muscles.
Chloride ion channel activity during modi®ed muscle use
MHC protein expression after 4 days HU (Stevens et al.,
1999).
These changes were accompanied by a slow-to-fast
transition of HU Sol functional properties, as demonstrated
by the shift of the MT toward values typical of a fast muscle.
MT is an integrative measure of the excitation±contraction
coupling mechanism and can be considered as an index of the
functional state of skeletal muscle. Accordingly to Dulhunty
and Gage (1983), ®bres of Sol muscle had a more negative R
value with respect to those of EDL muscle, which means that
they required minor depolarization to contract. In response to
HU, the MT of Sol muscle ®bres was shifted towards that
observed in EDL muscle, and this effect was observable after
1 week HU. This observation is in line with the modi®ed
expression of various proteins involved in the calcium
homeostasis that occurs in response to HU (Kandarian et al.,
1992, 1996; Schulte et al., 1993; Huchet-Cadiou et al., 1996;
Peters et al., 1999; Stevens et al., 1999; Bastide et al., 2000;
Talmadge, 2000). The fact that the MT measured in HU Sol
muscles did not completely reach that of the fast EDL muscle
most probably re¯ects the presence of a mixture of slow and
fast ®bres in the HU Sol muscle.
Change in gCl is an early event of slow-to-fast
®bre type transition
A novel and important ®nding of this study was that HU
signi®cantly increased gCl in Sol muscle ®bres toward values
that are typical of fast-twitch muscle ®bres. The modi®cation
of gCl was recurrent in the Sol muscles of all the HU animals
examined but did not occur in EDL muscles, indicating that
this parameter is speci®cally involved in the pathogenesis of
HU-induced muscle alterations. It is important to mention
that unloading-induced atrophy is speci®c to type I myo®bres
and that it may be more dif®cult to measure membrane
conductance in very small atrophied ®bres (McMinn and
VrbovaÁ, 1964). Yet, although we can not fully exclude an
involuntary selection of type II ®bres when measuring gCl in
HU Sol muscle, a number of observations described below
argue against a potential bias in our experimental procedure.
Indeed, the level of atrophy regularly increases from 3 days to
3 weeks HU, whereas the gCl is maximally activated after 3
days HU and remains constant throughout the HU period. If
selective atrophy of type I myo®bres had introduced a bias in
the measure of gCl, this latter should have increased parallel
to atrophy extent. Moreover, the increase of gCl measured in
individual ®bres is correlated to the increase of ClC-1 channel
mRNA level in the whole Sol muscle, which strongly
suggests that gCl is a positively regulated target of unloading.
In addition, the number of ®bres with a high or low gCl value
is very similar to the number of ®bres expressing fast or slow
MHC proteins. Thus we believe that the proportion of ®bres
used for microelectrode measurement is representative of the
entire ®bre population in the control muscle as well as in the
HU muscle.
1519
The increase of gCl was sustained by an upregulation of
transcription of the ClC-1 chloride channel gene in Sol
muscles. This was stressed by the linear correlation between
the value of gCl measured in control and HU Sol muscles and
the level of ClC-1 mRNA found in the contralateral Sol
muscles. ClC-1 channel expression appeared to be a speci®c
target of HU, since we found a reduction of the total mRNA
content in the same HU muscles. Yet, the relation between
gCl and ClC-1 mRNA was less pronounced after 3 weeks
HU; in fact, both gCl and ClC-1 mRNA increased by ~45%
after 1 week HU, whereas ClC-1 mRNA still increased by up
to 80%, but gCl remained constant after 3 weeks HU. A
similar behaviour has been reported for expression of the
sarcoplasmic reticulum calcium pump (Schulte et al., 1993)
and of the type IIa MHC (Stevens et al., 1999), and is
consistent with the idea that a decreased protein translation
followed by an increased protein degradation takes place
during long chronic unloading (Booth and Criswell, 1997).
Another possibility may be the negative control of ClC-1
channel function by a factor activated during prolonged HU.
Importantly, preliminary results indicate that gCl progressively recovers toward control Sol value upon reloading after
both 3 days and 2 weeks HU (unpublished data).
Very little is known about the mechanisms that modulate
ClC-1 channel expression. During the postnatal development
of rodent muscle, ClC-1 mRNA levels and gCl increase
steeply in parallel (Conte Camerino et al., 1989; Steinmeyer
et al., 1991), and both are downregulated upon denervation
(Camerino and Bryant, 1976; Klocke et al., 1994; Rich et al.,
1999), suggesting that ClC-1 channel expression is mainly
under the control of nerves. Two nerve-dependent regulatory
intracellular pathways of MHC isoform expression have been
proposed, which comprise the MyoD family of basic helix±
loop±helix transcription factors and the calcium-dependent
calcineurin:NF-AT pathway (Buonanno and Fields, 1999;
Talmadge, 2000). To date, only the former has been clearly
shown to be involved during HU, the mRNA levels of MyoD
being signi®cantly increased as soon as after 1 day HU
(Wheeler et al., 1999; Seward et al., 2001). The presence of
several E-boxes in the upstream sequence of the ClC-1 gene
(Klocke et al., 1994; Lorenz et al., 1994) therefore suggests
that ClC-1 expression may be regulated by MyoD during HU.
By analysing the distribution of ®bres as a function of gCl,
a bimodal repartition of ®bres between two mean values of
gCl was observed after 1 week HU in parallel to the half-andhalf content in type I and II MHC ®bres. Indeed, half of the
HU Sol muscle ®bres had a gCl overlapping that of control
Sol muscle ®bres, and the others had a gCl typical of control
fast-twitch EDL muscle ®bres. Nevertheless, the increase in
gCl was measurable after 3 days HU, whereas no change in
MHC protein and mRNA expression was detected. This
increase in gCl was attributable to an increase in ClC-1
channel activity, since gCl in 3 days HU Sol muscle ®bres
maintained the same sensitivity to the speci®c ClC-1 channel
blocker, S(±)-CPP, as compared with gCl in control Sol and
EDL muscle ®bres. This indicates that the increase in gCl due
1520
S. Pierno et al.
to ClC-1 channels constitutes an early event during the
adaptation of the slow muscle to unloading, which may be
necessary to attain complete MHC transition. In this regard, it
is important to bear in mind that the pharmacological block of
ClC-1 channels during development affects the maturation of
the EDL muscle, which retains slow contractile properties
(De Luca et al., 1990). Also, in adult animals, the reduction of
gCl, either due to natural mutations or induced by drugs, has
been shown to decrease the expression of `very fast' type IIb
MHC and to increase the expression of `less fast' type IIx and
IIa MHCs in fast-twitch muscles (Salviati et al., 1986; Goblet
and Whalen, 1995). In accordance with previous studies
suggesting that the decrease of gCl may trigger a fast-to-slow
myo®bre type transition, the present study suggests that the
increase of gCl may play a role in the slow-to-fast transition
occurring in unloaded slow-twitch muscles. Since the gCl
controls the electrical threshold of the sarcolemma, variation
of gCl amplitude modulates excitability and, in turn, may
modify the response of the myo®bre to speci®c nerve
stimulus. If this is the case, the pharmacological modulation
of gCl in Sol muscle ®bres might be useful to counteract the
slow-to-fast myo®bre type transition observed in pathophysiological situations of muscle disuse, and may help to prevent
muscle impairment and reduction of motor capacity of
humans after a space¯ight, or after long periods of limb
immobilization.
Acknowledgements
activity during neural and skeletal muscle development. [Review].
Curr Opin Neurobiol 1999; 9: 110±20.
Camerino D, Bryant SH. Effects of denervation and colchicine
treatment on the chloride conductance of rat skeletal muscle ®bers.
J Neurobiol 1976; 7: 221±8.
Chomczynski P, Sacchi N. Single-step method of RNA isolation by
acid guanidinium thiocyanate-phenol-chloroform extraction. Anal
Biochem 1987; 162: 156±9.
Conte Camerino D, De Luca A, Mambrini M, VrbovaÁ G. Membrane
ionic conductances in normal and denervated skeletal muscle of the
rat during development. P¯uÈgers Arch 1989; 413: 568±70.
De Luca A, Conte Camerino D, Connold A, VrbovaÁ G.
Pharmacological block of chloride channels of developing rat
skeletal muscle affects the differentiation of speci®c contractile
properties. P¯uÈgers Arch 1990; 416: 17±21.
Desaphy J-F, Pierno S, LeÂoty C, George AL Jr, De Luca A, Conte
Camerino D. Skeletal muscle disuse induces ®bre type-dependent
enhancement of Na+ channel expression. Brain 2001; 124: 1100±13.
Diffee GM, Caiozzo VJ, Herrick RE, Baldwin KM. Contractile and
biochemical properties of rat soleus and plantaris after hindlimb
suspension. Am J Physiol 1991; 260: C528±34.
Dulhunty AF, Gage PW. Asymmetrical charge movement in slowand fast-twitch mammalian muscle ®bres in normal and paraplegic
rats. J Physiol (Lond) 1983; 341: 213±31.
Fahlke C, Knittle T, Gurnett CA, Campbell KP, George AL Jr.
Subunit stoichiometry of human muscle chloride channels. J Gen
Physiol 1997; 109: 93±104.
This work was supported by grants from the Italian Space
Agency to D.C.C. (grants ASI ARS-98-71, ASI ARS-99-33,
ASI I/R/114/00) and the Italian Telethon Foundation to A.F.
(grant 983).
Frigeri A, Nicchia GP, Verbavatz JM, Valenti G, Svelto M.
Expression of aquaporin-4 in fast-twitch ®bers of mammalian
skeletal muscle. J Clin Invest 1998; 102: 695±703.
References
Baldwin KM. Effect of space¯ight on the functional, biochemical,
and metabolic properties of skeletal muscle. [Review]. Med Sci
Sports Exerc 1996; 28: 983±7.
Goblet C, Whalen RG. Modi®cations of gene expression in
myotonic murine skeletal muscle are associated with abnormal
expression of myogenic regulatory factors. Dev Biol (NY 1985)
1995; 170: 262±73.
Baldwin KM, Herrick RE, Ilyina-Kakueva E, Oganov VS. Effects
of zero gravity on myo®bril content and isomyosin distribution in
rodent skeletal muscle. FASEB J 1990; 4: 79±83.
Gurnett CA, Kahl SD, Anderson RD, Campbell KP. Absence of the
skeletal muscle sarcolemma chloride channel ClC-1 in myotonic
mice. J Biol Chem 1995; 270: 9035±8.
Bastide B, Conti A, Sorrentino V, Mounier Y. Properties of
ryanodine receptor in rat muscles submitted to unloaded conditions.
Biochem Biophys Res Commun 2000; 270: 442±7.
Huchet-Cadiou C, Bonnet V, MeÃme W, LeÂoty C. Hypogravity
increases cyclopiazonic acid sensitivity of rat soleus muscle. J Appl
Physiol 1996; 80: 1100±4.
Booth FW, Criswell DS. Molecular events underlying skeletal
muscle atrophy and the development of effective countermeasures.
[Review]. Int J Sports Med 1997; 18 Suppl 4: S265±9.
Kandarian S, O'Brien K, Thomas K, Schulte L, Navarro J.
Regulation of skeletal muscle dihydropyridine receptor gene
expression by biomechanical unloading. J Appl Physiol 1992; 72:
2510±4.
Bretag AH. Muscle chloride channels. [Review]. Physiol Rev 1987;
67: 618±724.
Bryant SH, Conte-Camerino D. Chloride channel regulation in the
skeletal muscle of normal and myotonic goats. P¯uÈgers Arch 1991;
417: 605±10.
Buonanno A, Fields RD. Gene regulation by patterned electrical
Frigeri A, Nicchia GP, Desaphy J-F, Pierno S, De Luca A, Conte
Camerino D, et al. Muscle loading modulates aquaporin-4
expression in skeletal muscle. FASEB J 2001; 15: 1282±4.
Kandarian SC, Peters DG, Favero TG, Ward CW, Williams JH.
Adaptation of the skeletal muscle calcium-release mechanism to
weight-bearing condition. Am J Physiol 1996; 270: C1588±94.
Klocke R, Steinmeyer K, Jentsch TJ, Jockusch H. Role of
innervation, excitability, and myogenic factors in the expression
Chloride ion channel activity during modi®ed muscle use
of the muscular chloride channel ClC-1. J Biol Chem 1994; 269:
27635±9.
Lehmann-Horn F, Jurkat-Rott K. Voltage-gated ion channels and
hereditary disease. [Review]. Physiol Rev 1999; 79: 1317±72.
Lorenz C, Meyer-Kleine C, Steinmeyer K, Koch MC, Jentsch TJ.
Genomic organization of the human muscle chloride channel ClC-1
and analysis of novel mutations leading to Becker-type myotonia.
Hum Mol Genet 1994; 3: 941±6.
Manchester JK, Chi MM, Norris B, Ferrier B, Krasnov I, Nemeth
PM, et al. Effect of microgravity on metabolic enzymes of
individual muscle ®bers. FASEB J 1990; 4: 55±63.
McMinn RM, Vrbova G. The effect of tenotomy on the structure of
fast and slow muscle in the rabbit. Q J Exp Physiol 1964; 49: 424±9.
Milton RL, Behforouz MA. Na channel density in extrajunctional
sarcolemma of fast and slow twitch mouse skeletal muscle ®bres:
functional implications and plasticity after fast motoneuron
transplantation on to a slow muscle. J Muscle Res Cell Motil
1995; 16: 430±9.
Miu B, Martin TP, Roy RR, Oganov V, Ilyina-Kakueva E, Marini JF, et al. Metabolic and morphologic properties of single muscle
®bers in the rat after space¯ight, Cosmos 1887. FASEB J 1990; 4:
64±72.
Peters DG, Mitchell-Felton H, Kandarian SC. Unloading induces
transcriptional activation of the sarco(endo)plasmic reticulum Ca2+ATPase 1 gene in muscle. Am J Physiol 1999; 276: C1218±25.
Pette D, Vrbova G. What does chronic electrical stimulation teach
us about muscle plasticity? [Review]. Muscle Nerve 1999; 22: 666±
77.
Pierno S, De Luca A, Beck CL, George AL Jr, Conte Camerino D.
Aging-associated down-regulation of ClC-1 expression in skeletal
muscle: phenotypic-independent relation to the decrease of chloride
conductance. FEBS Lett 1999a; 449: 12±6.
Pierno S, De Luca A, Liantonio A, Camerino C, Conte Camerino D.
Effects of HMG-CoA reductase inhibitors on excitation-contraction
coupling of rat skeletal muscle. Eur J Pharmacol 1999b; 364: 43±8.
Pusch M, Liantonio A, Bertorello L, Accardi A, De Luca A, Pierno
S, et al. Pharmacological characterization of chloride channels
belonging to the ClC family by the use of chiral clo®bric acid
derivatives. Mol Pharmacol 2000; 58: 498±507.
1521
Rich MM, Kraner SD, Barchi RL. Altered gene expression in
steroid-treated denervated muscle. Neurobiol Dis 1999; 6: 515±
22.
Riley DA, Ilyina-Kakueva EI, Ellis S, Bain JL, Slocum GR, Sedlak
FR. Skeletal muscle ®ber, nerve, and blood vessel breakdown in
space-¯own rats. FASEB J 1990; 4: 84±91.
Salviati G, Biasia E, Betto R, Betto DD. Fast to slow transition
induced by experimental myotonia in rat EDL muscle. P¯uÈgers
Arch 1986; 406: 266±72.
Schiaf®no S, Gorza L, Sartore S, Saggin L, Ausoni S, Vianello M,
et al. Three myosin heavy chain isoforms in type 2 skeletal muscle
®bres. J Muscle Res Cell Motil 1989; 10: 197±205.
Schulte LM, Navarro J, Kandarian SC. Regulation of sarcoplasmic
reticulum calcium pump gene expression by hindlimb unweighting.
Am J Physiol 1993; 264: C1308±15.
Seward DJ, Haney JC, Rudnicki MA, Swoap SJ. bHLH
transcription factor MyoD affects myosin heavy chain expression
pattern in a muscle-speci®c fashion. Am J Physiol Cell Physiol
2001; 280: C408±13.
Steinmeyer K, Ortland C, Jentsch TJ. Primary structure and
functional expression of a developmentally regulated skeletal
muscle chloride channel. Nature 1991; 354: 301±4.
Stevens L, Sultan KR, Peuker H, Gohlsch B, Mounier Y, Pette D.
Time-dependent changes in myosin heavy chain mRNA and protein
isoforms in unloaded soleus muscle of rat. Am J Physiol 1999; 277:
C1044±9.
Talmadge RJ. Myosin heavy chain isoform expression following
reduced neuromuscular activity: potential regulatory mechanisms.
[Review]. Muscle Nerve 2000; 23: 661±79.
Wheeler MT, Snyder EC, Patterson MN, Swoap SJ. An E-box
within the MHC IIB gene is bound by MyoD and is required for
gene expression in fast muscle. Am J Physiol 1999; 276: C1069±78.
Wright C, Haddad F, Qin AX, Baldwin KM. Analysis of myosin
heavy chain mRNA expression by RT-PCR. J Appl Physiol 1997;
83: 1389±96.
Received July 25, 2001. Revised February 13, 2002.
Accepted February 18, 2002