Treadmill running causes significant fiber damage in skeletal muscle of KATP channel-deficient mice
M. Thabet1,
T. Miki2,
S. Seino2 and
J.-M. Renaud1
1 Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario, Canada
2 Division of Cellular and Molecular Medicine, Kobe University Graduate School of Medicine, Kobe, Japan
 |
ABSTRACT
|
---|
Although it has been suggested that the ATP-sensitive K+ (KATP) channel protects muscle against function impairment, most studies have so far given little evidence for significant perturbation in the integrity and function of skeletal muscle fibers from inactive mice that lack KATP channel activity in their cell membrane. The objective was, therefore, to test the hypothesis that KATP channel-deficient skeletal muscle fibers become damaged when mice are subjected to stress. Wild-type and KATP channel-deficient mice (Kir6.2/ mice) were subjected to 45 wk of treadmill running at either 20 m/min with 0° inclination or at 24 m/min with 20° uphill inclination. Muscles of all wild-type mice and of nonexercised Kir6.2/ mice had very few fibers with internal nuclei. After 45 wk of treadmill running, there was little evidence for connective tissues and mononucleated cells in Kir6.2/ hindlimb muscles, whereas the number of fibers with internal nuclei, which appear when damaged fibers are regenerated by satellite cells, was significantly higher in Kir6.2/ than wild-type mice. Between 5% and 25% of the total number of fibers in Kir6.2/ extensor digitum longus, plantaris, and tibialis muscles had internal nuclei, and most of such fibers were type IIB fibers. Contrary to hindlimb muscles, diaphragms of Kir6.2/ mice that had run at 24 m/min had few fibers with internal nuclei, but mild to severe fiber damage was observed. In conclusion, the study provides for the first time evidence 1) that the KATP channels of skeletal muscle are essential to prevent fiber damage, and thus muscle dysfunction; and 2) that the extent of fiber damage is greater and the capacity of fiber regeneration is less in Kir6.2/ diaphragm muscles compared with hindlimb muscles.
Kir6.2 mice; exercise; internal nuclei
 |
INTRODUCTION
|
---|
THE ATP-SENSITIVE K+ (KATP) channel is a ligand-sensitive K+ channel. Its activity is modulated by changes in intracellular and extracellular metabolites that occur during metabolic stress, such as fatigue, hypoxia, and ischemia. In skeletal muscle, the metabolite changes include decreases in intracellular ATP (29) and pH (10) and increases in intracellular MgADP (33) and extracellular adenosine (3). Thus the KATP channel links the electrical activity of the cell membrane to the cell energy state. One postulated function is that the channel protects muscles against large ATP depletion to prevent dysfunction.
Studies using channel openers have elucidated one mechanism of action by which KATP channels carry out such function in skeletal muscle. Activating KATP channels with pinacidil during fatigue results in faster decrease in action potential amplitude (12) compared with control conditions. As a consequence of the pinacidil effect on action potentials, less Ca2+ is released by the sarcoplasmic reticulum (4, 11) and less force is developed by the contractile components (12, 19, 34). None of the pinacidil effects are observed in muscles with no KATP channel activity in the cell membrane (12), i.e., muscles from Kir6.2/ mice, which are null mice for the Kir6.2 gene, which encodes for the protein making the pore of the channel (15, 20, 28). These studies, therefore, suggest that KATP channels directly reduce action potential amplitude leading to smaller Ca2+ release and force. Lower Ca2+ release means less Ca2+ to pump back in the sarcoplasmic reticulum and less ATP hydrolysis by Ca2+-ATPase pumps. Fewer actomyosin links also reduce ATP hydrolysis by myosin ATPases. Finally, the capacity to recover force after fatigue is significantly improved in muscle exposed to pinacidil during fatigue (12, 19), supporting the notion that KATP channels prevent muscle dysfunction.
If the activation of KATP channels leads to faster decrease in force during fatigue, then abolishing their activity should result in a slower rate of fatigue. However, most studies have reported that blocking KATP channels of both amphibian and mammalian muscles with glibenclamide before fatigue does not affect the decrease in force during fatigue (8, 16, 19, 34). Furthermore, extensor digitum longus (EDL) and soleus muscles of Kir6.2/ mice have similar rates of fatigue as those from wild-type muscles (12, 13).
It is unlikely that the absence of the slower fatigue rate in Kir6.2/ and glibenclamide-exposed wild-type muscles is because KATP channels are not activated under control conditions. When glibenclamide is added during (as opposed to before) the fatigue period (11) or metabolic inhibition (14), the decrease in force is initially slower and then becomes faster than in control conditions (i.e., no glibenclamide). When dog diaphragms in situ are stimulated to fatigue, the plasma K+ concentration in the venous return increases under control conditions but decreases when glibenclamide is injected before fatigue (7). Resting tension develops during fatigue when the muscle fails to fully relax between contractions. Abolishing KATP channel activity with glibenclamide or by using Kir6.2/ muscle results in a greater generation of resting tension during fatigue (1214, 19). Abolishing KATP channel activity during fatigue also impairs the capacity of skeletal muscles to recover force after fatigue (8, 12, 13, 16, 19). Together, these studies provide indirect evidence for the activation of KATP channels during fatigue and about their importance in preventing muscle dysfunction as they reduce resting tension and improve force recovery.
The cause for the lowered recovery capacity after fatigue in the absence of KATP channel activity is unknown but may involve damaged fibers that no longer generate maximum force. Furthermore, if in the absence of KATP channel activity muscles develop more resting tension, then running may lead to eccentric contractions as muscles contract while antagonist muscles have not fully relax. Eccentric contractions are a known cause of fiber damage in skeletal muscle (27). It was therefore the objective of this study to test the hypothesis that "fiber damage occurs during exercise leading to fatigue when there is no cell membrane KATP channel activity." To test this hypothesis, we subjected wild-type and Kir6.2/ mice to treadmill running for 45 wk. This approach was used because 1) contrary to in vitro conditions (1, 30), KATP channels are active at rest under in vivo conditions (17, 25) and are further activated during muscular activity leading to fatigue (7); and 2) the effects of no KATP channel could be tested under more physiological conditions, i.e., in situ versus in vitro experiments. The results show that treadmill running causes mild to severe fiber damage in EDL, tibialis, plantaris, and diaphragm muscles of Kir6.2/ mice but not in wild-type muscles.
 |
METHODS AND MATERIALS
|
---|
Animals and Treadmill Running
C57Bl6 mice were used as wild-type mice, and Kir6.2/ mice were generated as previously described by Miki et al. (21). Kir6.2/ mice were backcrossed with the wild-type mice over four generations before this study. Furthermore, in vitro studies have demonstrated that the kinetics of fatigue of Kir6.2/ EDL and soleus muscles (13) are similar to those observed in wild-type muscles (19) exposed to glibenclamide, a KATP channel blocker. Kir6.2/ mice did not require any special care. All mice were bred, fed ad libitum, and housed according to the guidelines of the Canadian Council for Animal Care. The Animal Care Committee of the University of Ottawa approved all experimental procedures.
A total of 14 wild-type male mice (2.0 ± 0.1 mo old, 26.3 ± 0.8 g initial body wt, means ± SE) and 14 Kir6.2/ male mice (2.6 ± 0.1 mo old, 27.1 ± 0.8 g initial body weight) were used in this study. Wild-type and Kir6.2/ mice were divided into three groups: 1) nonexercised, 2) 5 wk of treadmill running at 20 m/min with 0° inclination, and 3) 4 wk of treadmill running at 24 m/min with 20° uphill inclination. Mice were elicited to run by touching their back with a pencil. For the running speed of 20 m/min, mice ran until they could no longer maintain the set speed, i.e., until they were unable to stay away from the back wall of the treadmill despite stimulations. For the running speed of 24 m/min, mice also ran until they could no longer maintain the set speed during the first week. However, after 5 days, wild-type mice ran a distance that was more than threefold longer than Kir6.2/ mice; so, for the remaining 3 wk, wild-type and Kir6.2/ mice were elicit to run the same distance, i.e., 1.4 km (1 h).
Histology
After 45 wk of treadmill running, the plantaris, tibialis, EDL, soleus, and strips of diaphragm muscles were excised, embedded in OCT (Tissue Tek II), and frozen in isopentane precooled in liquid nitrogen. Cross sections (10 µm thick) were cut at 20°C using a cryostat. Slides always contained cross sections from wild-type and Kir6.2/ muscles to stain them simultaneously.
Hematoxylin and eosin staining.
Hematoxylin-eosin (H&E) staining was used to localize nuclei in muscle fibers. Cross sections were incubated for 7 min in hematoxylin solution (Shandon), washed 2 min with water, dipped 1 min in 1% acidic alcohol (20 drops of concentrated HCl in 200 ml of 70% alcohol), rinsed for 5 min in water, dipped 1 min in 1.0% lithium carbonate solution, washed 5 min in water, dipped 1 min in 75% ethanol, dipped three times in eosin solution (Shandon), incubated twice for 2 min in 95% alcohol, incubated twice for 2 min in 100% alcohol, incubated twice for 2 min in xylene, and mounted in Permount before the coverslip was placed over the sample.
Fiber typing.
Fiber type composition was determined using specific mouse monoclonal anti-embryonic myosin (MHCemb; anti-Bf-45, dilution 1:10) and anti-myosin type I (anti A484, dilution 1:50), IIA (anti Sc-71, dilution 1:50), and IIB (anti Bf-f3, dilution 1:100) as primary antibodies obtained from Dr. D. Parry (University of Ottawa). Horseradish peroxidase-labeled goat antimouse antibodies were used as secondary antibodies and for color development. Anti-IgG secondary antibodies (Jackson Immunoresearch) were used for Bf-45 (MHCemb, dilution 1:50) and SC-71 (type IIA, dilution 1:200). Anti-IgM secondary antibodies (Jackson Immunoresearch) were used for A484 (type I, dilution 1:200) and Bf-f3 (type IIB, dilution 1:50). All dilutions were in PBS containing 2% (wt/vol) dry milk.
Briefly, cross sections were exposed for 30 min to PBS with 2% (wt/vol) dry milk and then overnight at 4°C to primary antibodies (types I and IIA); the incubation period was reduced to 30 min for Bf-f3 (4°C, type IIB) and Bf-45 (room temperature, MHCemb) to reduce background staining. After being washed for 10 min for three times with PBS, cross sections were exposed 60 min at room temperature to secondary antibodies. After being washed for 10 min for three times, cross sections were placed 1 min in 50 mM diaminobenzadine and 0.1% hydrogen peroxide for color development. Cross sections were washed for 5 min in distilled water; dehydrated for 2 min in 95% alcohol, 2 min in 100% alcohol, 2 min in xylene; and mounted in Paramount before the coverslip was placed over the sample.
Morphological Analysis
Sections were viewed using a Sony digital camera (model DXC-950) attached to a Zeiss Axiophot fluorescent microscope and connected to a computer. The total numbers of fibers; type I, IIA, and IIB fibers; and fibers with internal nuclei were counted in all muscles. Antibody for type IIX myosin was not available, so the number of type IIX fibers was calculated as the difference between the total number of fibers and the total number of type I, IIA, and IIB fibers. Cross-sectional areas (CSAs) were determined using Northern Eclipse software. When possible, a total of 125 fibers for each fiber type was analyzed from five different areas (25 fibers each) chosen at random.
Statistical Analysis
Split-plot ANOVA designs were used to determine significant differences between running and mice conditions (whole plot) and fiber type (split plot). ANOVA calculations were made using the general linear model procedures of Statistical Analysis Software (SAS Institute). The least-square difference (LSD) was used to locate significant differences (31). Statistical differences were for P < 0.05.
 |
RESULTS
|
---|
Running Distances
On the first day, wild-type mice ran an average of 2.2 km at 20 m/min with 0° inclination before they could no longer maintain the required speed (Fig. 1A). The mean distance ran by Kir6.2/ mice under the same conditions was only 0.9 km. Although the absolute difference in distance ran by wild-type and Kir6.2/ mice was not significant, it represented more than a twofold difference. The mean distance doubled on day 2 for wild-type mice and on day 3 for Kir6.2/ mice. Thereafter, the increases in mean distances were smaller. After 35 days, the mean running distance of wild-type mice had significantly increased by 3.4 km (from 2.2 to 5.6 km); for Kir6.2/ mice, the increase was significantly less, being only 1.2 km (from 0.9 to 2.1 km).
For the first day at 24 m/min (20° upward inclination), wild-type mice ran a mean distance of 1.9 km, whereas Kir6.2/ mice ran 0.6 km (Fig. 1B). Again, the difference in mean running distance between wild-type and Kir6.2/ mice was not significant for day 1 but represented more than a threefold difference between the mice. Wild-type mice ran significantly longer distances by day 3 and ran a mean distance of 5.2 km by day 5. For Kir6.2/ mice, the increase in mean running distance from day 1 to day 5 was not significant and was only 1.6 km on day 5, a value that was significantly shorter compared with wild-type mice. So, the increases in mean running distance over a 5-day period for wild-type and Kir6.2/ mice were, respectively, 3.3 and 1.0 km.
Histology
Total number of fibers and fiber type composition.
The total numbers of muscle fibers in plantaris, EDL, and soleus muscles were not different between wild-type and Kir6.2/ mice (Fig. 2). Kir6.2/ tibialis muscle, however, had on average 300 fewer fibers than wild-type muscle, a difference that was significant. Type IIB was the predominant fiber in tibialis, plantaris, and EDL muscles of nonexercised wild-type mice (Table 1). These muscles had fewer type IIA and IIX fibers, whereas type I fibers were almost nonexistent. Types I and IIA were the predominant fibers in the soleus muscle, and types IIA and IIX were the predominent fibers in the diaphragm. Significant differences in fiber type composition between nonexercised wild-type and Kir6.2/ mice were observed only in EDL and diaphragm muscles. Compared with wild-type muscles, the Kir6.2/ EDL muscle had more type IIB and fewer type IIX fibers, whereas the Kir6.2/ diaphragm had less type IIA and more type IIX.

View larger version (10K):
[in this window]
[in a new window]
|
Fig. 2. ATP-sensitive K+ (KATP) channel-deficient tibialis muscles had fewer fibers than WT tibialis muscles. There was no significant difference in the number of muscle fibers between nonexercised and exercised mice, so data were pooled. EDL, extensor digitum longus. Vertical bars represent the SEs of 14 muscles. *Mean total numbers of muscle fibers in Kir6.2/ mice (solid bars) was significantly different from those of WT mice (open bars; P < 0.05 by ANOVA and LSD test).
|
|
View this table:
[in this window]
[in a new window]
|
Table 1. ATP-sensitive K+ channel deficiency had small effects on fiber type composition of muscles of nonexercised mice
|
|
Treadmill running had few significant effects on fiber type composition in both wild type and Kir6.2/ muscles, as shown for two muscles in Fig. 3. Tibialis muscles of exercised Kir6.2/ mice had significantly fewer type IIA fibers than exercised wild-type mice. Furthermore, Kir6.2/ tibialis muscles had significantly more type IIB fibers (from mice that ran at 20 m/min) and more type IIX fibers (from mice that ran at 24 m/min) than wild-type tibialis muscles (Fig. 3A). For the plantaris muscle, treadmill running at 24 m/min significantly increased the number of type IIA fibers in both wild-type and Kir6.2/ mice, but the concomitant decreases in type IIB and IIX fibers were not significant (Fig. 3B). It is important to note that all significant differences between nonexercised and exercised mice as well as between wild-type and Kir6.2/ mice were <10%.

View larger version (32K):
[in this window]
[in a new window]
|
Fig. 3. Treadmill running had small effects on fiber type composition of WT and Kir6.2/ tibialis (A) and plantaris (B) muscles. Type IIA and IIB fibers were visualized using, respectively, anti-myosin IIA and IIB. The numbers of type IIX fibers were calculated as the difference between the total number of fibers and the total number of type I, IIA, and IIB fibers (the frequency of type I fibers is not shown because they are almost inexistent in those two muscles; see Table 1). Solid bars, nonexercised mice (n = 4); open bars, 5 wk of treadmill running at 20 m/min with 0° inclination (n = 4); hatched bars, 4 wk of treadmill running at 24 m/min with 20° uphill inclination (n = 6). Vertical bars represent the SEs. The frequency observed in muscles of exercised mice was significantly different from the frequency of nonexercised mice (P < 0.05 by ANOVA and LSD test); *the frequency observed in Kir6.2/ muscles was significantly different from the frequency for WT muscles under the same type of exercise (P < 0.05 by ANOVA and LSD test).
|
|
Internal nuclei and fiber damage.
Muscle fibers of tibialis, plantaris, EDL, and soleus muscles of all nonexercised and exercised wild-type mice showed no histological signs of fiber damage (Fig. 4A for one wild-type tibialis muscle and Fig. 4B for one Kir6.2/ muscle). Small areas of connective tissues and mononucleated cells (Fig. 4B) were observed in only 2 of 56 Kir6.2/ tibialis, plantaris, EDL, and soleus muscles (from 4 nonexercised and 10 exercised mice). The incidence of internal nuclei, which appear when damaged fibers are regenerated by satellite cells (5, 9), was very small in all wild-type muscles and in muscles of nonexercised Kir6.2/ mice (Fig. 4C). However, this was not the case for muscles of exercised Kir6.2/ mice. After Kir6.2/ mice ran 5 wk at 20 m/min, more than 5% of Kir6.2/ tibialis and EDL muscle fibers had internal nuclei, a significant difference compared with wild-type muscles. After Kir6.2/ mice ran 4 wk at 24 m/min, about 12% of tibialis and plantaris fibers and 25% of EDL fibers had internal nuclei, values that were significantly different from wild-type mice. Of all the fibers with internal nuclei, 9899% were type IIB fibers and only 12% were type IIA and IIX fibers, whereas none of the type I fibers had internal nuclei (Fig. 4D). Cross sections of tibialis muscles were also stained for MHCemb, which appears in the early stage of fiber regeneration by satellite cells (9). Whereas neonatal muscle fibers (positive control) stained strongly for anti-MHCemb, none of the tibialis muscle fibers did (data not shown).

View larger version (64K):
[in this window]
[in a new window]
|
Fig. 4. KATP channel deficiency increased the incidence of internal nuclei during treadmill running. A: hematoxylin and eosin (H&E) staining of tibialis muscle from WT mice exercised at 24 m/min with 20° uphill inclination. B: H&E staining of tibialis muscle from Kir6.2/ mice. Open arrows, fibers with internal nuclei; solid arrows, areas with mononucleated cells and connective tissue infiltration. C: incidence of muscle fibers with internal nuclei. Open bars, nonexercised mice (n = 4); solid bars, 5 wk of treadmill running at 20 m/min with 0° inclination (n = 4); hatched bars, 4 wk of treadmill running at 24 m/min with 20° uphill inclination (n = 6). Vertical bars represent SEs. D: distribution of fibers with internal nuclei among different fiber types. Values are expressed as a percentage of the total number of fibers with internal nuclei (include all fibers from C). *Mean percent values were significantly different between WT and Kir6.2/ mice (P < 0.05 by ANOVA and LSD test).
|
|
Similar to hindlimb muscles, diaphragm muscles from nonexercised wild-type and Kir6.2/ mice had very few fibers (<1%) with internal nuclei (data not shown). Diaphragms of nonexercised and exercised wild-type mice were all 1213 fibers thick, and none of them showed signs of fiber damage as shown from one diaphragm muscle (Fig. 5A). Mild to severe fiber damage was observed only in the diaphragms of Kir6.2/ mice that had run at 24 m/min. In four of six diaphragm muscles, small areas of connective tissues and mononucleated cells were observed as well as areas with less than seven fibers in thickness (Fig. 5B). For the other two diaphragms, severe fiber damage and a reduction in the number of fibers (<6 fibers in thickness) were observed (Fig. 5C).

View larger version (171K):
[in this window]
[in a new window]
|
Fig. 5. Treadmill running resulted in mild to severe muscle fiber damages in Kir6.2/ diaphragm muscle. AC: H&E staining of diaphragm muscles from WT (A) and Kir6.2/ (B and C) diaphragms. Mice ran at 24 m/min with 20° uphill inclination for 4 wk. Solid arrows indicate areas with extra connective tissues and/or mononucleated cells. The diaphragm appears between pieces of liver (top and bottom).
|
|
Fiber CSA.
In most muscles, treadmill running and KATP channel deficiency had no significant effect on fiber CSA (data not shown) except for type IIB fibers in tibialis muscles and all fiber types in diaphragm muscle. In the tibialis muscle, the mean CSA of type IIB fibers, only in nonexercised muscles, was significantly larger in Kir6.2/ mice (2,903 µm2) compared with wild-type mice (2,453 µm2; Fig. 6A). The greater mean CSA of Kir6.2/ fibers was due to a shift in the frequency distribution toward higher CSA: in wild-type mice, 97% of type IIB fibers had CSAs between 1,500 and 3,500 µm2 compared with 86% of the fibers ranging between 2,000 and 4,500 µm2 for Kir6.2/ fibers (Fig. 6B). Treadmill running did not significantly affect the mean CSA in both wild-type and Kir6.2/ mice (Fig. 6, A and C), but it significantly increased the variability (as seen by the SE bar) for Kir6.2/ type IIB fibers. This variability was in part due to 1) a doubling in the number of fibers exceeding 4,000 µm2 from 6% to 12% and 2) a greater number of fibers with CSAs below 2,000 µm2, i.e., from 9% (nonexercise) to 24% (exercise; Fig. 6D).

View larger version (27K):
[in this window]
[in a new window]
|
Fig. 6. KATP channel deficiency caused hypertrophy in type IIB fibers of tibialis muscle. A: mean cross-sectional areas (CSAs) of type IIA, IIB, and IIX fibers. Open bars, nonexercised mice (n = 4); solid bars, 5 wk of treadmill running at 20 m/min with no inclination (n = 4); hatched bars, 4 wk of treadmill running at 24 m/min with 20° uphill inclination (n = 6). Vertical bars represent the SEs. BD: frequency distributions of CSA of tibialis type IIB fibers from nonexercised WT (solid lines; B) and Kir6.2/ mice (shaded lines; B) and from nonexercised (solid lines) and exercised (shaded lines; 24 m/min, 20° uphill inclination) WT (C) and Kir6.2/ mice (D). CSA were divided in a range of 500 µm2. *Means CSA of Kir6.2/ fibers were significantly different from those of WT fibers (P < 0.05 by ANOVA and LSD test).
|
|
For muscles of nonexercised mice, CSAs of type IIA and IIB fibers were smaller in Kir6.2/ than in wild-type diaphragm muscles, whereas no significant difference was observed for type I and IIX fibers (Fig. 7). After mice ran for 4 wk of treadmill running at 24 m/min, type I fibers of wild-type diaphragms were larger and type IIB fibers were smaller compared with diaphragm fibers from nonexercised mice. In the Kir6.2/ diaphragm, treadmill running resulted in large hypertrophy for all fiber types. The increases in CSAs for type I, IIA, IIB, and IIX fibers were, respectively, 42%, 50%, 60%, and 65%.

View larger version (33K):
[in this window]
[in a new window]
|
Fig. 7. KATP channel deficiency caused hypertrophy in Kir6.2/ diaphragm muscle fibers during treadmill running. Open bars, nonexercised WT mice (n = 4); hatched bars, exercised WT mice; solid bars, nonexercised Kir6.2/ mice; crosshatched bars, exercised Kir6.2/ mice. Exercise consisted of 4 wk treadmill running at 24 m/min with 20° inclination. A: type I, IIA, and IIX fibers; B: type IIB fibers. Vertical bars represent the SEs of 4 nonexercised and 6 exercised mice. Mean CSAs of fibers from exercised mice were significantly different from those of nonexercised mice (P < 0.05 by ANOVA and LSD test); *mean CSAs of fibers were significantly different between WT and Kir6.2 mice (P < 0.05 by ANOVA and LSD test)..
|
|
 |
DISCUSSION
|
---|
The major findings of this study are 1) during treadmill running, Kir6.2/ mice fatigued faster than wild-type mice; 2) wild-type and Kir6.2/ EDL, soleus, tibialis, and plantaris muscles had similar fiber type composition; 3) 45 wk of treadmill running resulted in a higher incidence of internal nuclei in Kir6.2/ hindlimb muscle fibers and mild to severe fiber damage in the Kir6.2/ diaphragm, whereas a similar situation was not observed in muscles of wild-type mice; and 4) significant hypertrophy was observed in type IIB fibers of the tibialis muscle and in all fibers of diaphragm muscles of Kir6.2/ mice compared with wild-type mice.
KATP Channel Deficiency Diminishes Running Performance
When treadmill speed and inclination are increased stepwise, the tolerated workload was three times less in Kir6.2/ mice than wild-type mice, i.e., Kir6.2/ mice fatigued faster than wild-type mice during running exercise (35). This study now demonstrates that at constant speed and inclination, Kir6.2/ mice also fatigue faster. For the entire running period at either 20 m/min (no inclination) or 24 m/min with 20° upward inclination, the distance ran before the required speed could not be maintained was threefold shorter for Kir6.2/ than for wild-type mice (Fig. 1). Another important difference between wild-type and Kir6.2/ mice is their capacity to increase running endurance. As expected from several studies, running endurance or the distance ran by wild-type mice increased significantly over time. However, the increases for Kir6.2/ mice were significantly smaller than those for wild-type mice.
The causes for the lower fatigue resistance and reduced capacity to improve running endurance of Kir6.2/ mice is most likely multifactorial because the Kir6.2 gene is expressed in several tissues along with either the SUR1 or SUR2 regulatory subunit (28). Although KATP channels are expressed in vascular smooth muscle cells, the evidence so far suggests that the Kir6.1 subunit, and not the Kir6.2 subunit, plays a major role in the regulation of blood pressure (22, 32). However, Zingman et al. (35) have shown that the enhancement of cardiac performance by sympathetic stimulation, which occurs during running, is compromised in Kir6.2/ mice. Furthermore, dog diaphragm muscles in situ fatigued faster in the absence of KATP channel activity (7), and, after 4 wk of treadmill running at 24 m/min, mild to severe fiber damages were observed in diaphragm muscles (Fig. 6). Thus lower cardiac performance and impaired diaphragm muscle contraction, especially at 24 m/min, are expected to contribute to the lower fatigue resistance in Kir6.2/ mice because they most likely cause insufficient blood flow to and oxygenation of active skeletal muscles.
In vitro studies have also provided evidence for dysfunction of skeletal muscle during fatigue development when there is no KATP channel activity in the cell membrane. Abolishing KATP channel activity, pharmacologically or by using Kir6.2/ mice, results in greater development of resting tension, which develops when muscles fail to completely relax between contractions (12, 13, 19). Consequently, during treadmill running, Kir6.2/ muscles must work harder than wild-type muscles because they have to overcome the resistance of antagonist muscles that develop more resting tension. The extra work undoubtedly results in lower fatigue resistance in Kir6.2/ mice. Thus this study and the one of Zingman et al. (35) demonstrated that KATP channels are essential in optimizing running performance.
KATP Channel Deficiency Has No Effect on Fiber Type Composition
Very few significant differences in fiber type composition of hindlimb and diaphragm muscles were observed between nonexercised wild-type and Kir6.2/ mice (Table 1). However, the significant differences were <9% and perhaps of little physiological consequence. Four weeks of voluntary wheel running have little effect on the fiber type composition of skeletal muscle of wild-type mice (2). In this study, 45 wk of treadmill running also had little effect on fiber type composition in both wild-type and Kir6.2/ mice (data not shown). We therefore suggest that the lack of sarcolemmal KATP channel activity had no major effect on fiber type composition of mouse skeletal muscle.
KATP Channel Deficiency Results in Muscle Fiber Damage
Hindlimb muscles.
The incidence of fibers with internal nuclei, which appears after damaged fibers have been regenerated by satellite cells (5, 9), was low in hindlimb muscles of nonexercised wild-type and Kir6.2/ mice as well as of exercised wild-type mice. However, exercised Kir6.2/ mice had several muscles with fibers containing internal nuclei (Fig. 4). Among the hindlimb muscles tested in this study, the Kir6.2/ soleus muscle was the only muscle with no increased incidence of internal nuclei after 45 wk of treadmill running, whereas EDL muscles had the largest incidences (Fig. 4C). Interestingly, among fibers with central nuclei in plantaris, tibialis, and EDL muscles, 98% were of type IIB fibers (Fig. 4D). Thus it appears that the lack of increased incidence of fibers with internal nuclei in the soleus muscle is because it has no or very few type IIB fibers (Table 1).
The cause for the fiber damage in type IIB fibers is unlikely related to a deficiency in blood flow or oxygenation as discussed above for the lower fatigue resistance of Kir6.2/ mice. First, type IIB fibers are the most glycolytic and the least oxidative and vascularized fibers. Second, under in vitro conditions for which there is no blood flow, the extent of force recovery is little affected by the level of KATP channel activity during fatigue in the soleus muscle, a muscle primarily composed of the most oxidative and vascularized fibers, i.e., type I and IIA fibers (Table 1). In the EDL muscle, a muscle primarily composed of type IIB fibers (Table 1), the extent of force recovery depends largely on KATP channel activity: it is significantly reduced in the absence of channel activity and significantly improved upon activation with pinacidil (12, 13, 19). Thus, in the absence of blood flow, force recovery is affected by KATP channel activity in muscle containing primarily type IIB fibers, and this fact correlates well with the observation in this study that the incidence of internal nuclei after treadmill running was basically all from type IIB fibers. Third, Kir6.2/ muscles develop more resting tension during fatigue in vitro than wild-type mice. Resting tension during treadmill running is expected to increase the extent of eccentric contractions, a well known cause of fiber damage (27), because antagonist muscles stretch each others when they fail to fully relax. Considering that the difference in resting tension between wild-type and Kir6.2/ muscle is more marked in the EDL muscle than in the soleus muscle (12, 13, 19), it is then not surprising to see more evidence of fiber damage in the EDL than soleus muscle. Therefore, we suggest that KATP channels are essential in preventing fiber damage during a stress, such as treadmill running, in type IIB fibers of hindlimb muscles.
The question as to why type IIB fibers are the only fibers affected during treadmill running cannot be answered from the results of this study. One possible reason may be due to the fact that type I, IIA and IIX fibers contain more mitochondria than type IIB fibers, and mitochondria have mechanisms that protect muscle against fiber damage during metabolic stress, at least in cardiac muscle (26). So perhaps type I, IIA, and IIX fibers rely less on cell membrane KATP channels to prevent fiber damage than type IIB fibers.
Diaphragm muscles.
Contrary to hindlimb muscles, the Kir6.2/ diaphragm muscle had very few fibers with internal nuclei, whereas it contained a large amount of mononucleated cells and connective tissues and had evidence for loss of fibers after mice had run at 24 m/min. Furthermore, all fiber types appeared to be affected, not just type IIB fibers as observed in hindlimb muscles. Interestingly, the order in the extent of fiber damage observed in this study (i.e., diaphragm > EDL > soleus) is similar to the order observed in mdx mice, which are deficient in dystrophin (18, 23, 24). Another interesting observation is that in hindlimb muscles, very few mononucleated cells and connective tissues were observed and there was no evidence for fibers expressing MHCemb. Because such a situation is normally observed during the first week of muscle regeneration (9), it thus appears that in hindlimb muscles fiber damage occurred early during the running period and thereafter fibers became resistant to further damage, whereas diaphragm muscle fibers fail to regenerate damaged fibers for unknown reasons. In fact, hindlimb muscles of mdx mice also have a greater capacity than diaphragm muscles to regenerate fibers and to become more resistant to further damage (18, 24).
KATP Channel Deficiency Results in Hypertrophy in Tibialis and Diaphragm Muscle Fibers
In most cases, the mean and distribution of muscle fiber CSAs were the same between wild-type and Kir6.2/ mice and between nonexercised and exercised mice (data not shown). There were only two exceptions. The first exception was an increased proportion of small CSA for the Kir6.2/ tibialis type IIB fibers after treadmill running, a situation not observed in wild-type muscles (Fig. 6, C and D). The net effect was an increased variability in CSA for Kir6.2 tibialis type IIB fibers after exercise (Fig. 6A). The increase in CSA variability may have been the result of the regeneration of damaged fibers as previously reported in mdx muscles (9).
The second exception was a significant hypertrophy of some fibers. In nonexercised mice, Kir6.2/ tibialis type IIB fibers had on average greater CSAs than those of wild-type muscle fibers (Fig. 5, A and B). Significant hypertrophy was also observed for all fiber types in diaphragm muscles of exercised (24 m/min, 20° inclination) Kir6.2/ mice compared with those from nonexercised Kir6.2/ mice (Fig. 7). In both cases, the hypertrophy was observed when there was evidence for a difference in the number of fibers. The Kir6.2/ tibialis muscle was the only muscle that had fewer fibers than its wild-type counterpart (Fig. 2). Diaphragm muscles with hypertrophied fibers were those for which there was a tremendous loss of fibers (Fig. 5). It is therefore proposed that the hypertrophy observed under those conditions was probably part of a compensation for the loss of fibers. Thus the lack of KATP channel activity causes hypertrophy in muscle, but only when there is a loss of fiber.
Role of KATP Channels in Skeletal Muscle: a New Perspective
As discussed in the Introduction, KATP channel openers clearly demonstrate that the channel directly reduces action potential amplitude, and, as a consequence of this effect, it reduces Ca2+ release from the sarcoplasmic reticulum and the force developed by the contractile components, i.e., it suggests that KATP channels can contribute to the decrease in force during fatigue. However, despite evidence that KATP channels are active under control conditions, most studies fail to demonstrate a slower fatigue rate when KATP channel activity is abolished using pharmacological or knockout approaches. However, this study now demonstrates that the absence of KATP channel activity significantly affects muscle integrity, which leads to fiber damage and thus a reduction in the capacity of muscle to generate force.
On the basis of this study and others, we are now proposing the following roles for the KATP channel: their activation during fatigue contributes to the decrease in force by reducing action potential amplitude, and they are essential to prevent muscle dysfunction, that is, they prevent fiber damage (this study), membrane depolarization (14), resting tension, and improve force recovery after fatigue (12, 13, 19). Finally, we suggest that the apparent lack of effect on fatigue rate when KATP channel activity is abolished is because the resulting muscle dysfunction counteracts the expected slower fatigue rate.
 |
GRANTS
|
---|
This study was supported by a grant from the National Science and Engineering Research Council of Canada (to J.-M. Renaud) and by grants-in-aid for Scientific Research from the Ministry of Education, Science, Sports and Culture, Japan (to S. Seino).
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: J.-M. Renaud, Dept. of Cellular and Molecular Medicine, Univ. of Ottawa, 451 Smyth Rd., Ottawa, Ontario, Canada K1H 8M5 (E-mail: jmrenaud{at}uottawa.ca).
10.1152/physiolgenomics.00064.2005.
 |
REFERENCES
|
---|
- Allard B, Lazdunski M, and Rougier O. Activation of ATP-sensitive K+ channels by metabolic poisoning in adult mouse skeletal muscle: role of intracellular Mg2+ and pH. J Physiol 485: 283296, 1995.[Abstract]
- Allen DL, Harrison BC, Maass A, Bell ML, Byrnes WC, and Leinwand LA. Cardiac and skeletal muscle adaptations to voluntary wheel running in the mouse. J Appl Physiol 90: 19001908, 2001.[Abstract/Free Full Text]
- Barrett-Jolley R, Comtois A, Davies NW, Stanfield PR, and Standen NB. Effect of adenosine and intracellular GTP on KATP channels of mammalian skeletal muscle. J Membr Biol 152: 111116, 1996.[CrossRef][ISI][Medline]
- Burton FL and Smith GL. The effect of cromakalim on intracellular [Ca2+] in isolated rat skeletal muscle during fatigue and metabolic blockade. Exp Physiol 82: 469483, 1997.[Abstract]
- Carnwath JW and Shotton DM. Muscle dystrophy in the mdx mouse: histopathology of the soleus and extensor digitorum longus muscles. J Neurol Sci 80: 3954, 1987.[CrossRef][ISI][Medline]
- Chakkalakal JV, Harrison MA, Carbonetto S, Chin ER, Michel RB, and Jasmin BJ. Stimulation of calcineurin signaling attenuates the dystrophic pathology in mdx mice. Hum Mol Genet 13: 379388, 2004.[Abstract/Free Full Text]
- Comtois A, Sinderby C, Comtois N, Grassino A, and Renaud JM. An ATP-sensitive potassium channel blocker decreases diaphragmatic circulation in anesthetized dogs. J Appl Physiol 77: 127134, 1994.[Abstract/Free Full Text]
- Comtois AS, Light PE, and Renaud JM. Effect of tolbutamide on the rate of fatigue and recovery in frog sartorius muscle. J Pharmacol Exp Ther 274: 10611066, 1995.[Abstract]
- D'Albis A, Couteaux R, Mira JC, Janmot C, and Roulet A. Isoformes de la myosine synthétisées au cours de la régénération de muscles squelettiques à contraction rapide, en présence du nerf moteur et après dénervation. Etude chez le rat et la souris adultes. Reprod Nutr Dev 28: 753756, 1988.[ISI][Medline]
- Davies NW, Standen NB, and Stanfield PR. The effect of intracellular pH on ATP-dependent potassium channels of frog skeletal muscle. J Physiol 445: 549568, 1992.[Abstract]
- Duty S and Allen DG. The effects of glibenclamide on tetanic force and intracellular calcium in normal and fatigued mouse skeletal muscle. Exp Physiol 80: 520541, 1995.
- Gong B, Legault D, Miki T, Seino S, and Renaud JM. KATP channels depress force by reducing action potential amplitude in mouse EDL and soleus. Am J Physiol Cell Physiol 285: C1464C1474, 2003.[Abstract/Free Full Text]
- Gong B, Miki T, Seino S, and Renaud JM. A K
deficiency affects resting tension not contractile force during fatigue in skeletal muscle. Am J Physiol Cell Physiol 279: C1351C1358, 2000.[Abstract/Free Full Text]
- Gramolini A and Renaud JM. Blocking ATP-sensitive K+ channel during metabolic inhibition impairs muscle contractility. Am J Physiol Cell Physiol 272: C936C946, 1997.
- Inagaki N, Gonoi T, Clement JP, Namba N, Inazawa J, Gonzalez G, Aguilar-Bryan L, Seino S, and Bryan J. Reconstitution of IKATP: an inward rectifier subunit plus the sulfonylurea receptor. Science 270: 11661170, 1995.[Abstract]
- Light PE, Comtois AS, and Renaud JM. The effect of glibenclamide on frog skeletal muscle: evidence for KATP+ channel activation during fatigue. J Physiol 475: 495507, 1994.[Abstract]
- Lindinger MI, Hawke TJ, Vickery L, Bradford L, and Lipskie SL. An integrative, in situ approach to examining K+ flux in resting skeletal muscle. Can J Physiol Pharmacol 79: 9961006, 2001.[CrossRef][ISI][Medline]
- Louboutin JP, Fichter-Gagnepain V, Thaon E, and Fardeau M. Morphometric analysis of mdx diaphragm muscle fibers. Comparison with hindlimb muscles. Neuromuscul Disord 3: 463469, 1993.[CrossRef][Medline]
- Matar W, Nosek TM, Wong D, and Renaud JM. Pinacidil suppresses contractility and preserves energy but glibenclamide has no effect during fatigue in skeletal muscle. Am J Physiol Cell Physiol 278: C404C416, 2000.[Abstract/Free Full Text]
- Miki T, Minami K, Zhang L, Morita M, Gonoi T, Shiuchi T, Minokoshi Y, Renaud JM, and Seino S. ATP-sensitive potassium channels participate in glucose uptake in skeletal muscle and adipose tissue. Am J Physiol Endocrinol Metab 283: E1178E1184, 2002.[Abstract/Free Full Text]
- Miki T, Nagashima H, Tashiro F, Kotake K, Yoshitomi H, Tamamoto A, Gonoi T, Iwanaga T, Miyazaki JI, and Seino S. Defective insulin secretion and enhanced insulin action in KATP channel-deficient mice. Proc Natl Acad Sci USA 95: 1040210406, 1998.[Abstract/Free Full Text]
- Miki T, Nagashima K, Tashiro F, Kotake K, Yoshitomi H, Tamamoto A, Gonoi T, Iwanaga T, Miyazaki JI, and Seino S. Mouse model of Prinzmetal angina by disruption of the inward rectifier Kir6.1. Nat Med 8: 466472, 1998.
- Moens P, Baatsen PHWW, and Maréchal G. Increased susceptibility of EDL muscles from mdx mice to damage induced by contractions with stretch. J Muscle Res Cell Motil 14: 446451, 1993.[CrossRef][ISI][Medline]
- Muller J, Vayssiere N, Royuela M, Leger ME, Muller A, Baxcou F, Pons F, Hugon G, and Mornet D. Comparative evolution of muscular dystrophy in diaphragm, gastrocnemius and masseter muscles from old male and mdx mice. J Muscle Res Cell Motil 22: 133139, 2001.[CrossRef][ISI][Medline]
- Nielsen JJ, Kristensen M, Hellsten Y, Bangsbo J, and Juel C. Localization and function of ATP-sensitive potassium channels in human skeletal muscle. Am J Physiol Regul Integr Comp Physiol 284: R558R563, 2003.[Abstract/Free Full Text]
- Oldenburg O, Cohen MV, Yellon DM, and Downey JM. Mitochondrial KATP channels: role in cardioprotection. Cardiovasc Res 55: 429437, 2002.[CrossRef][ISI][Medline]
- Proske U and Morgan DL. Muscle damage from eccentric exercise: mechanism, mechanical signs, adaptation and clinical applications. J Physiol 537: 333345, 2001.[Abstract/Free Full Text]
- Seino S. ATP-sensitive potassium channels: a model of heteromultimeric potassium channel/receptor assemblies. Annu Rev Physiol 61: 637662, 1999.
- Spruce AE, Standen NB, and Stanfield PR. Studies of the unitary properties of adenosine-5'-triphosphate-regulated potassium channels of frog skeletal muscle. J Physiol 382: 213236, 1987.[Abstract]
- Standen NB, Pettit AI, Davies NW, and Stanfield PR. Activation of ATP-dependent K+ currents in intact skeletal muscle fibers by reduced intracellular pH. Proc R Soc Lond B Biol Sci 247: 195198, 1992.[ISI][Medline]
- Steel RGD and Torrie JH. Principles and Procedures of Statistics. A Biometrical Approach. New York: McGraw-Hill, 1980.
- Suzuki M, Li RA, Miki T, Uemura H, Sakamoto N, Ohmoto-Sekine Y, Tamagawa M, Ogura T, Seino S, Marban E, and Nakaya H. Functional roles of cardiac and vascular ATP-sensitive potassium channels clarified by Kir6.2-knockout mice. Circ Res 88: 570577, 2001.[Abstract/Free Full Text]
- Vivaudou MB, Arnoult C, and Villaz M. Skeletal muscle ATP-sensitive K+ channels recorded from sarcolemmal blebs of split fibers: ATP inhibition is reduced by magnesium and ADP. J Membr Biol 122: 165175, 1991.[CrossRef][ISI][Medline]
- Weselcouch EO, Sargent C, Wilde MW, and Smith MA. ATP-sensitive potassium channels and skeletal muscle function in vitro. J Pharmacol Exp Ther 267: 410416, 1993.[Abstract]
- Zingman LV, Hodgson DM, Bast PH, Kane GC, Perez-Terzic C, Gumina RJ, Pucar D, Bienengraeber M, Dzeja PP, Miki T, Seino S, Alekseev AE, and Trezic A. Kir6.2 is required for adaptation to stress. Proc Natl Acad Sci USA 99: 1327813283, 2002.[Abstract/Free Full Text]