(Received for publication, February 25, 1997, and in revised form, May 28, 1997)
From the Department of Molecular Biology and Research Center for
Cell Differentiation, Seoul National University, Seoul 151-742, Korea and the Department of Molecular Biology, Dankook
University, Seoul 140-714, Korea
The level of inwardly rectifying K+ channel 1 (IRK1) mRNA decreased upon denervation and increased during muscle differentiation in mouse skeletal muscle. To identify the mechanism(s) underlying the regulation of IRK1 mRNA expression, we examined its expression using the well differentiated C2C12 mouse skeletal muscle cell line as a model system. Since nerve-induced muscle activity results in contraction, it was questioned whether the changes in IRK1 expression might be relevant to the increased intracellular calcium that functions as a cytoplasmic messenger in excitation-contraction coupling. Indeed, activation of either L-type calcium channels or ryanodine receptors increased the level of IRK1 mRNA. More directly, ionomycin activated the IRK1 expression in time- and dose-dependent manners, which was abolished by treatment with EGTA. Genistein, a tyrosine kinase inhibitor, also abolished the stimulating effect of ionomycin. Meanwhile, activation of protein kinase C by 12-O-tetradecanoylphorbol acetate (TPA) markedly decreased the level of IRK1 mRNA, which required ongoing protein synthesis. Actinomycin D experiments revealed that ionomycin increased the half-life of IRK1 mRNA from 0.86 to 1.97 h, but TPA decreased it to 0.38 h. However, neither ionomycin nor TPA appreciably altered the rate of IRK1 gene transcription. Based on these observations, we conclude that intracellular calcium and protein kinase C are oppositely involved in the muscle activity-dependent regulation of IRK1 gene expression and that both act at the level of mRNA stability.
It has been known that denervation influences many biophysical and biochemical properties of skeletal muscle fibers (1-7). The mechanisms by which denervation initiates these changes are still unclear. They may be caused by the loss of neurotrophic factors normally released from the nerve terminals (8, 9). Alternatively, the electrical inactivity of the denervated muscle might be responsible, since direct electrical stimulation to the denervated muscle restores all passive electrical parameters of the membrane that were observed without denervation (10, 11). Considering the importance of calcium in the process of muscle contraction induced by neural activity, calcium may play a critical role in linking the biochemical and biophysical changes with muscle activity.
Calcium is known to be involved in many cellular events as a second
messenger. A regulatory role of calcium in the expression of sodium
channels, acetylcholinesterase, and nicotinic acetylcholine receptor
has been suggested (12-16). In addition, it has recently been shown
that calcium influx blocks the expression of nicotinic acetylcholine
receptor -subunit gene in chick skeletal muscle (17). There are
reports that coupling of the electrical activity with altered gene
expression is mediated by protein kinase C
(PKC)1 pathway (18, 19).
Gonoi and Hasegawa (20) demonstrate by using a patch clamp method that innervation of skeletal muscle fibers plays a key role in the induction and maintenance of inwardly rectifying K+ currents in mouse flexor digitorum longus muscle. The resting potential of many excitable cells, including skeletal muscle, is determined by resting potassium conductance of IRK that shows inward rectification, allowing potassium ions to move more readily inward the cell membrane than outward. Katz (21) first describes inward rectification of the resting K+ conductance of frog skeletal muscle. Since then, electrophysiological properties of this conductance have been studied by a number of investigators (22, 23), and subsequently the channel has been cloned from a mouse macrophage cell line (24).
The present work aims at elucidating the molecular mechanisms involved in the neural and developmental regulation of IRK1 expressions. Here we suggest that intracellular calcium and PKC oppositely regulate the expression of IRK1 mRNA in mouse skeletal muscle and both regulations are associated with mRNA stability.
[-32P]dCTP and
[
-32P]UTP and nylon membrane (Nytran) were purchased
from NEN Life Science Products and Schleicher & Schuell, respectively.
TPA, ryanodine, trifluoperazine, and Bay K 8644 were purchased from
Research Biochemical Inc., and 8-bromo-cyclic GMP, dibutyryl cyclic
AMP, genistein, and ionomycin were from Sigma. Culture dishes were
purchased from Corning Glass, and other culture reagents were obtained
from Life Technologies, Inc.
C2C12 cells were plated at a density of 3 × 104 cells/ml in growth medium (Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and 1% antibiotic-antimycotic solution) and cultured at 37 °C for 2 days. Differentiation from myoblasts to myotubes was induced by changing the growth medium with differentiation medium (Dulbecco's modified Eagle's medium with 2% horse serum and 1% antibiotic-antimycotic solution). Experiments were routinely done at 6 days after the medium change.
SurgeryFor denervation studies, 8-week-old ICR mice weighing about 25 g were anesthetized with avertin (0.014-0.018 ml of 2.5% avertin/g of body weight), and a 5-mm length of the right sciatic nerves at the upper thigh were cut out. The transection totally denervated the muscles of the lower leg. A sham operation was performed on the contralateral side of all denervated animals, and the contralateral innervated muscles were used as controls. At various times after the denervation, denervated and control muscles were isolated and prepared for RNA isolation.
Northern Blot AnalysisTotal RNA was isolated from mouse skeletal muscles and from cultured C2C12 muscle cells using the guanidinium thiocyanate-acidic phenol method (25). RNA was dissolved in 0.5% SDS and denatured in the presence of 50% formamide, 2.2 M formaldehyde, 20 mM MOPS, 4 mM sodium acetate, 0.5 mM EDTA at 60 °C for 5 min. Aliquots (15-30 µg) of RNA were size-fractionated by electrophoresis on a 1% (w/v) formaldehyde-denaturing-agarose gel and were transferred to Nytran membranes by capillary blotting.
Nytran membranes were placed in a polyethylene heat-sealable bag and
prehybridized with hybridization buffer (5 × saline/sodium/phosphate/EDTA, pH 7.4, 5 × Denhardt's solution,
0.5% SDS, 0.2 mg/ml fragmented, denatured salmon sperm DNA, and 50%
formamide) at 42 °C for 3 h. Hybridization was carried out at
42 °C for 2 days with heat-denatured IRK1 cDNA probe (5 × 106 cpm/ml). After hybridization, the membranes were washed
twice in 2 × SSC (1 × SSC: 150 mM NaCl, 15 mM citrate, pH 7.0), 0.1% SDS at room temperature for a
total of 20 min, then once in 0.1× SSC, 0.1% SDS at 42 °C for 20 min. Membranes were exposed to x-ray film at 70 °C for 1-7 days.
After autoradiography, the probe was stripped off the membrane by
incubation in distilled water at 100 °C for 10 min. Membranes were
then rehybridized with other control probes under the same conditions.
Northern blot experiments were repeated at least three times with
reproducible results.
Preparation of hybridization probes
was performed as described previously (26). The total RNA isolated from
mouse skeletal muscle was reverse-transcribed in the presence of random
hexamer (Boehringer Mannheim). For a polymerase chain reaction cloning of the IRK1 cDNA from mouse skeletal muscle, reverse-transcribed products were used as templates. The 5 and 3
primers were
5
-CGAGACCCAGACAACCAT-3
and 5
-TCCCCCATCACTATCGTT-3
, corresponding to
the 411-428 and 793-810 nucleotide sequences of IRK1 cDNA as
described previously (24). The fragments (400 base pairs) obtained were
ligated into the EcoRV site of pBluescript KS(+),
and sequences were analyzed. The sequences of the cDNA fragments
were identical to those of IRK1 in J774 mouse macrophage cell line
(24). 32P-labeled antisense DNA probes were synthesized
from linearized plasmids containing IRK1 fragments using Taq
polymerase (Promega) and 3
primer.
Nuclei isolation and nuclear run-on
transcription assays were performed as described by Greenberg and
Bender (27). After cells were harvested by centrifugation, the pellets
were resuspended in lysis buffer (10 mM Tris, pH 7.4, 3 mM CaCl2, and 2 mM
MgCl2), centrifuged for 5 min at 500 × g
and resuspended in the same lysis buffer containing 0.5% Nonidet P-40.
The cells were then broken in a Dounce homogenizer, and nuclei were
sedimented at 500 × g for 5 min. The nuclei were
resuspended in 50 mM Tris-HCl, pH 8.3, containing 40%
glycerol, 5 mM MgCl2, and 0.1 mM
EDTA and mixed with equal volume of 2 × reaction buffer
containing 10 mM Tris-HCl, pH 8.0, 5 mM
MgCl2, 0.3 M KCl, 5 mM
dithiothreitol, 1 mM each ATP, GTP, and CTP and 100 µCi
of [-32P]UTP and incubated at 37 °C for 30 min for
in vitro transcription. Radiolabeled mRNA was isolated
and hybridized for at least 36 h at 65 °C to slot-blotted
membranes containing 5 µg of linearized plasmid containing IRK1 or
GAPDH insert. After extensive washing with 2 × SSC, membranes
were exposed to x-ray film at
70 °C.
It has been demonstrated using a patch clamp method
that innervation of skeletal muscle fibers plays a key role in the
induction and maintenance of inwardly rectifying currents in mouse
flexor digitorum longus muscle (20). To further clarify the role of innervation on IRK1 expression, the effect of chronic denervation on
the alteration in the IRK1 mRNA level was examined by Northern blot
analysis. Within 1 day after the denervation, the mRNA level in
slow twitch soleus muscle (soleus) was dramatically reduced, whereas
that in the contralateral muscle remained unchanged (Fig. 1, upper panel). Similar data
were obtained for denervated fast twitch extensor digitorum longus
muscle (EDL), although the reduction in the mRNA level was not as
obvious as that in denervated soleus. Interestingly, the IRK1 mRNA
level in the contralateral EDL was significantly higher than that seen
in the same muscle without denervation (Fig. 1, lower
panel). Although the reason for this change is not known at
present, it is possible that the increase in mRNA level in
contralateral EDL may be attributed to the increased use of the
contralateral leg. Analogous observations have been made in cat and
frog skeletal muscles. Steinbach (28) demonstrates the increase in
neuromuscular junctional size in fast twitch muscles contralateral to
denervated muscles of cats. In addition, denervation results in an
increase in multiple innervation of muscle fibers in contralateral
muscles of frog (29). Other works have also shown that muscle twitch
time was altered following contralateral denervation (30, 31).
We also examined whether IRK1 expression is regulated during muscle
development. When Northern blot analysis was carried out using total
RNAs in lower leg muscles obtained from mice at different ages, the
amounts of IRK1 mRNA increased with muscle development (Fig.
2A). Furthermore, the level of
IRK1 mRNA from C2C12 mouse muscle cell line also increased as the
myogenic development proceeded (Fig. 2B). These results
clearly demonstrate that the IRK1 expression is developmentally
regulated.
Calcium Is Responsible for Links between Muscle Activity and Increase of IRK1 mRNA Level
To characterize the factors
involved in the neural regulation of IRK1 expression, a series of
experiments were carried out in vitro using cultured C2C12
myotubes. Because nerve-induced muscle activity results in muscle
contraction, elements of ECC may be shared by the mechanism that links
the muscle activity with IRK1 gene regulation. The major components
involved in ECC are voltage-dependent calcium channels
(L-type calcium channels) in transverse tubule membrane and
ryanodine receptors in sarcoplasmic reticulum membrane (32). As shown
in Fig. 3, chronic depolarization of
C2C12 myotubes upon treatment with 40 mM extracellular
potassium stimulated the expression of IRK1. Activation of
L-type calcium channels by treatment with Bay K 8644 together with 40 mM potassium further increased the amount
of IRK1 mRNA. However, the increased expression of IRK1 by
depolarization could be reversed upon treatment with L-type
calcium channel blocker D600, although the mRNA level is still
higher than that seen without any treatment. Because both ryanodine and
caffeine have been well known to affect sarcoplasmic reticulum calcium
release, they have been used in the assessment of sarcoplasmic
reticulum function in controlling cytoplasmic calcium concentrations.
Ryanodine in nanomolar concentrations keeps the ryanodine receptor
channels to an open state (33), and caffeine activates the ryanodine
receptor channels by increasing channel opening probability (34). As
shown in Fig. 3, caffeine and ryanodine increased the IRK1 mRNA
level. These results suggest that elevation of intracellular calcium
level that is associated with ECC may mediate depolarization-IRK1 gene
activation coupling.
Consistent with this suggestion, calcium ionophore ionomycin was also
found to increase the expression of IRK1 mRNA in a
dose-dependent manner (Fig.
4A). Upon treatment of C2C12
myotubes with 0.5 µM ionomycin, the expression was
increased in a time-dependent manner up to a maximum level
at about 2 h after the treatment (Fig. 4B). Moreover,
the ionophore elevated the level of IRK1 mRNA only when free
calcium was present in the extracellular environment (Fig. 4C).
Genistein Abolishes the Ionomycin-induced Increase of IRK1 mRNA Level
In the nervous system, activity-induced increase in
cytoplasmic calcium activates various tyrosine kinase pathways (35). To
test whether a certain tyrosine kinase pathway is involved in the
increased IRK1 mRNA level by ionomycin, C2C12 cells were incubated
with various concentrations of genistein, a tyrosine kinase inhibitor,
in the presence of 0.5 µM ionomycin for 2 h. As
demonstrated in Fig. 5A,
genistein inhibited the increase in IRK1 mRNA by ionomycin in a
dose-dependent manner, whereas basal IRK1 mRNA level
still remained unchanged after the genistein treatment. Therefore, it
is likely that tyrosine kinase is somehow involved in
calcium-dependent IRK1 expression.
Calcium activates a variety of cellular processes. For example, the activation of calcium/calmodulin-dependent protein kinase is a common mechanism mediating the effects of increase in intracellular calcium concentration (36). The calmodulin antagonists, trifluoperazine, however, had no effect on the expression of IRK1 in both control and ionomycin-treated myotubes (Fig. 5B). Several second messenger pathways associated with muscle electrical activity have been demonstrated. Nestler et al. (37) demonstrate that the cyclic GMP level increases upon electrical stimulation. In addition, the suppression of expression of embryonic-type nicotinic acetylcholine receptor genes by muscle activity can be reversed by increasing intracellular cAMP (38). However, neither dibutyryl cAMP nor 8-bromo-cGMP up to 0.5 mM showed any effect on the level of IRK1 mRNA in both control and ionomycin-treated cells (Fig. 5B). These findings exclude the possibility of involvement in the signaling pathway of cyclic nucleotides or calmodulin for the regulation of IRK1 expression.
Activation of PKC Decreases the Level of IRK1 mRNAPKC
activity has been reported to increase in active muscle fibers (39).
Huang et al. (18, 19) show that a calcium-requiring nuclear
PKC mediates depolarization-acetylcholine receptor gene inactivation
coupling. To test if the IRK1 expression is also regulated by PKC,
C2C12 cells were treated with a phorbol ester, TPA, that is an
activator of PKC. Against our expectation, TPA decreased the IRK1
expression in time- and dose-dependent manners (Fig.
6, A and B), and
this effect could be prevented upon co-treatment with staurosporine or
calphostin C, PKC inhibitors (Fig. 6C). Furthermore, TPA
reduced the stimulating effect of ionomycin on the IRK1 expression
(Fig. 6C). These results clearly demonstrate that PKC is
involved in the decrease in IRK1 mRNA level.
Ionomycin and TPA, both, Act at the Level of IRK1 mRNA Stability
Studies were performed to assess whether ionomycin and
TPA alter transcription rate or stability of IRK1 mRNA. First,
transcriptional regulation was tested by nuclear run-on analysis. As
shown in Fig. 7, treatment of 0.5 µM ionomycin or 0.5 µM TPA did not
appreciably influence the rate of transcription of IRK1 gene. Then, to
determine whether ionomycin or TPA alters the stability of IRK1
mRNA, we examined the decay of mRNA levels when the cells were
incubated with actinomycin D. As shown in Fig.
8, the level of IRK1 mRNA decayed
rapidly in control cells, the half-life being 0.86 ± 0.12 h.
Treatment of ionomycin increased the half-life of IRK1 mRNA to
1.97 ± 0.21 h, whereas TPA decreased it to 0.38 ± 0.18 h. The results imply that IRK1 mRNA level is regulated at
posttranscriptional level.
Down-regulation by TPA of IRK1 mRNA Level Requires Ongoing Protein Synthesis
To determine whether the regulation of IRK1
mRNA expression requires protein synthesis, C2C12 cells were
incubated with 250 µM cycloheximide followed by treatment
with TPA or ionomycin. As shown in Fig.
9, treatment with cycloheximide abolished
the inhibitory effect of TPA on the IRK1 expression. Moreover, the block of protein synthesis further stimulated the increase of ionomycin-mediated IRK1 mRNA, although basal IRK1 mRNA level
remained unchanged. Thus, only the PKC-dependent
down-regulation of IRK1 mRNA level appears to require de
novo protein synthesis.
The expression of IRK in skeletal muscle seems to be subjected to developmental and neural regulations (20, 26). To our knowledge, however, nothing is known about the mechanism underlying these phenomena. In this regard, the present work was undertaken to identify molecular mechanism(s) that might be involved in these regulations. One of the most important findings in the present studies is that calcium mediates muscle activity-IRK1 gene activation coupling through mRNA stabilization. In addition, tyrosine kinase-mediated signaling pathway is somehow involved in this regulation. In contrast, it appears that the increase of IRK1 mRNA induced by intracellular calcium does not seem to involve calmodulin- or cyclic nucleotide-dependent pathways since inhibitors or agonists of these cellular components had no effect. Instead, PKC-dependent pathway appears to decrease the level of IRK1 mRNA, and the PKC-mediated down-regulation of IRK1 expression is also modulated at the level of posttranscription.
Since depolarization triggered by neural activity at the neuromuscular junction results in skeletal muscle contraction, it seems possible that elements of the ECC pathway may be shared by the signaling pathway by which muscle activity is conveyed to regulate the IRK1 expression. Contraction of skeletal muscle has been shown to depend directly on sarcoplasmic reticulum calcium release (40). Therefore, calcium may act as a link between muscle activity and IRK1 gene activation. Consistent with this notion, activation of either L-type calcium channels or ryanodine receptors, the major elements of the ECC pathway, was found to increase the level of IRK1 mRNA (Fig. 3). In addition, the involvement of intracellular calcium was further confirmed by the observation that ionomycin induces the IRK1 expression in time- and dose-dependent manners (Fig. 4). Furthermore, the stimulatory effect of the increase in intracellular calcium on IRK1 expression is likely to mediate the tyrosine kinase pathway since a well known tyrosine kinase inhibitor genistein prevented the increase of IRK1 mRNA level by ionomycin (Fig. 5).
Intracellular calcium is an agonist of PKC, and hence elevated calcium levels either by influx or by release should lead to activation of the enzyme. Therefore, it was suspected that the intracellular calcium might increase the IRK1 expression through a PKC-dependent pathway. Nevertheless, treatment with TPA alone significantly reduced the level of IRK1 mRNA (Fig. 6). Yet, it is a well known fact that PKC also depends on intracellular calcium. It is thus no wonder that down-regulation of PKC with long term treatment of TPA or selective PKC inhibitor staurosporine further potentiates the ionomycin-mediated increase in IRK1 mRNA level (data not shown). It seems quite obvious that intracellular calcium can act on up-regulation of IRK1 expression by a mechanism that is distinct from the signaling pathway involving PKC, which instead is involved in down-regulation mechanism.
Stability of mRNA in eukaryotic cells, although not as widely and thoroughly studied as transcriptional control, is a regulated property that can determine the level of expression of a gene (41, 42). Elevated intracellular calcium increases IRK1 mRNA level through the stabilization of mRNA as evidenced by the experiments for both nuclear run-on (Fig. 7) and actinomycin D pulse-chase (Fig. 8). PKC also seems to down-regulate IRK1 mRNA level by reducing the mRNA stability. Additionally, it is likely that labile destabilizing factors are involved in regulating IRK1 mRNA stability since inhibition of protein synthesis by cycloheximide is able to prevent inhibitory effect of TPA (Fig. 9). Although the precise mechanism for IRK1 mRNA stabilization and the involved cis elements are not characterized, it is clear that intracellular calcium and PKC oppositely regulate IRK1 expression by modulating the mRNA stability.
The mechanism involving calcium of muscle activity-dependent expression offers some suggestions concerning the developmental change in the IRK1 expression. During the differentiation of skeletal muscle, mononucleate myoblasts align along their bipolar axes and fuse to form multinucleate myotubes (43). David et al. (44) demonstrate that calcium entry is necessary for the onset of myoblast fusion. Therefore, increase in the IRK1 mRNA level concurrent with muscle differentiation is relevant to intracellular calcium increase that is a prerequisite for myoblast fusion. In addition, the increased expression of IRK1 in EDL contralateral to denervated muscle may be due to the increased cytoplasmic calcium caused by increased use of contralateral leg after denervation.
The functional implication for the up-regulation of the IRK1 expression by the increase in intracellular calcium is not obvious at present. One of the speculative roles of IRK1 is its involvement in a pathway that facilitates potassium ion reentry from potassium-loaded transverse tubules after each action potential. Potassium ions tend to accumulate in the lumen of the transverse tubules even under normal conditions, and this accumulation is exaggerated with the prolonged action potentials and thus tends to partially depolarize the muscle fibers and increase their excitability (45). Adrian and Peachey (46) calculate that a single action potential alters the luminal potassium concentration by about +0.3 mM. Moreover, the extracellular potassium concentration elevated physiologically to 8-9 mM in the vicinity of stimulated skeletal muscles, causing hyperkalemic periodic paralysis (47). The potassium accumulation in the lumen of the transverse tubules can only be dissipated relatively slowly by diffusion out of the mouth of the transverse tubules and by active pumping back into the myoplasm across the transverse tubule wall. Our present findings suggest that IRK1, which is up-regulated by muscle activity through the mechanism involving intracellular calcium, contributes to the uptake of accumulated luminal potassium, thereby preventing the hyperexcitability of stimulated muscle fibers.
We thank Professor Kyungjin Kim and Dr. Woong Sun for helpful discussions throughout this study.