| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
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 |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Kir6.2– mice; exercise; internal nuclei
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
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 (12–14, 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 4–5 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 |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
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 4–5 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 A4–84, 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 A4–84 (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 |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
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.
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
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, 4–5 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 4–5 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 |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
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 |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODS AND MATERIALSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
deficiency affects resting tension not contractile force during fatigue in skeletal muscle. Am J Physiol Cell Physiol 279: C1351–C1358, 2000.This article has been cited by other articles:
|
|
 |
|
  S. P. Cairns, D. M. Robinson, and D. S. Loiselle Double-sigmoid model for fitting fatigue profiles in mouse fast- and slow-twitch muscle Exp Physiol, July 1, 2008; 93(7): 851 - 862. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Philip-Couderc, N. I. Tavares, A. Roatti, R. Lerch, C. Montessuit, and A. J. Baertschi Forkhead Transcription Factors Coordinate Expression of Myocardial KATP Channel Subunits and Energy Metabolism Circ. Res., February 1, 2008; 102(2): e20 - e35. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cifelli, F. Bourassa, L. Gariepy, K. Banas, M. Benkhalti, and J.-M. Renaud KATP channel deficiency in mouse flexor digitorum brevis causes fibre damage and impairs Ca2+ release and force development during fatigue in vitro J. Physiol., July 15, 2007; 582(2): 843 - 857. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. A. Macdonald, N. Ortenblad, and O. B. Nielsen Energy conservation attenuates the loss of skeletal muscle excitability during intense contractions Am J Physiol Endocrinol Metab, March 1, 2007; 292(3): E771 - E778. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Liang and B. Ventura Physiological genomics in PG and beyond: July to September 2005 Physiol Genomics, October 17, 2005; 23(2): 119 - 124. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
) fatigued faster than wild-type (WT) mice (•) during treadmill running. Mice were subjected to run at 20 m/min with 0° inclination (A) or at 24 m/min with 20° uphill inclination (B). For the running speed of 20 m/min, mice ran until they could no longer maintained the required speed throughout the 5-wk period (for clarity, the data shown in A represent the distance run for the first 5 days and the last day of the 5-wk running period). For the running speed of 24 m/min, mice also ran again until they could no longer maintained the required speed only during 1 wk; thereafter, all mice ran 1.4 km or 1 h (data in B show the mean running distances during the first 5 days). Distance was calculated as the product of running speed and the time mice ran. Vertical bars represent the SEs of 4 mice at 20 m/min and 6 mice at 24 m/min. 
Mean distance was significantly different from the mean distance on day 1 [P < 0.05 by ANOVA and least-square difference (LSD) test]; *mean distance for Kir6.2–/– mice was significantly less than the mean distance of WT mice (P < 0.05 by ANOVA and LSD test).
