| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
is not mandatory for exercise- and training-induced adaptive gene responses in mouse skeletal muscle1Centre of Inflammation and Metabolism and Copenhagen Muscle Research Centre, Department of Molecular Biology, and 2Copenhagen Muscle Research Centre, Department of Human Physiology, Institute of Exercise and Sport Sciences, University of Copenhagen, Copenhagen, Denmark; and 3Institute of Neurosciences and Department of Cellular Biology, Physiology, and Immunology, Autonomous University of Barcelona, Barcelona, Spain
Submitted 18 October 2007 ; accepted in final form 6 December 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
coactivator (PGC) 1
is required for exercise-induced adaptive gene responses in skeletal muscle. Whole body PGC-1 knockout (KO) and littermate wild-type (WT) mice performed a single treadmill-running exercise bout. Soleus and white gastrocnemius (WG) were obtained immediately, 2 h, or 6 h after exercise. Another group of PGC-1 KO and WT mice performed 5-wk exercise training. Soleus, WG, and quadriceps were obtained 
37 h after the last training session. Resting muscles of the PGC-1 KO mice had lower (20%) cytochrome c (cyt c), cytochrome oxidase (COX) I, and aminolevulinate synthase (ALAS) 1 mRNA and protein levels than WT, but similar levels of AMP-activated protein kinase (AMPK) 1, AMPK2, and hexokinase (HK) II compared with WT mice. A single exercise bout increased phosphorylation of AMPK and acetyl-CoA carboxylase-β and the level of HKII mRNA similarly in WG of KO and WT. In contrast, cyt c mRNA in soleus was upregulated in WT muscles only. Exercise training increased cyt c, COXI, ALAS1, and HKII mRNA and protein levels equally in WT and KO animals, but cyt c, COXI, and ALAS1 expression remained 20% lower in KO animals. In conclusion, lack of PGC-1 reduced resting expression of cyt c, COXI, and ALAS1 and exercise-induced cyt c mRNA expression. However, PGC-1 is not mandatory for training-induced increases in ALAS1, COXI, and cyt c expression, showing that factors other than PGC-1 can exert these adaptations.
peroxisome proliferator-activated receptor- coactivator; mitochondrial biogenesis; gene expression
Peroxisome proliferator activated receptor- coactivator (PGC) 1 is a transcriptional coactivator playing a key role in regulating mitochondrial biogenesis and in general in the transcriptional regulation of genes encoding proteins in oxidative metabolism (17). For example, overexpression of PGC-1 in myotubes (40) and in mouse skeletal muscle (15) induced mitochondrial biogenesis and increased expression of genes involved in oxidative phosphorylation such as cytochrome c (cyt c) and cytochrome oxidase (COX) in addition to aminolevulinate synthase (ALAS)1, providing heme for cytochromes in the respiratory chain. Overexpression of PGC-1 in skeletal muscle also markedly increased myoglobin protein expression, which was accompanied by transformation of otherwise white muscles to dark red in appearance (15). In accordance with this, PGC-1-deficient mice have been shown to have reduced mRNA content of genes encoding mitochondrial metabolic proteins (15, 30). Collectively, these findings provide evidence that PGC-1 is critical for the oxidative metabolism of a muscle. This is further in line with the relatively higher expression of endogenous PGC-1 in typical oxidative muscles like soleus (Sol) than in muscles with low oxidative capacity like the extensor digitorum longus (15).
Because mitochondrial DNA encodes 13 proteins in the respiratory chain (including COXI), whereas the remaining proteins (including cyt c) arise from nuclear-encoded genes, it is clear that adaptive responses of mitochondria require well-coordinated regulatory mechanisms. The role of PGC-1 in gene regulation lies in the ability to activate a broad range of transcription factors and thereby regulate transcription of their target genes, and interestingly, PGC-1 regulates the expression of both nuclear- and mitochondrial-encoded genes through effects on expression of various transcription factors. Hence PGC-1 regulates the expression of the nuclear-encoded mitochondrial transcription factor A (40), as well as expression and/or activity of many nuclear transcription factors like nuclear respiratory factors, forkhead box 01, and peroxisome proliferator-activated receptors (6, 34, 40). Thus PGC-1 has the potential to be important in coordinating adaptive cellular responses.
Previous studies have demonstrated that PGC-1 plays a role in the antioxidant defense of the cell by regulating the expression of scavenger enzymes, including superoxide dismutase (SOD) and uncoupling proteins (UCP) (30). The observation that genetically modified PGC-1 knockout (KO) cells lack a H2O2-induced increase in SOD and UCP mRNA expression demonstrates that PGC-1 mediates this effect (30). The data of St-Pierre et al. (30) also indicate that H2O2 is a direct stimulus to enhanced PGC-1 expression. In addition, exercise has been shown to increase the production of free radicals in skeletal muscle (9, 25), and training is known to increase muscle antioxidant defense systems (11), although the underlying signaling pathways for this effect remain largely unknown. Therefore, PGC-1 could be a possible mediator of training-induced improvements in the antioxidant defense of skeletal muscle.
The finding that skeletal muscle PGC-1 mRNA content is increased severalfold in skeletal muscle in response to an acute exercise bout (3, 7, 22) suggests that a rapid increase in PGC-1 protein content after exercise could play a role in initiating the transcriptional regulation of exercise-responsive genes in muscle and thereby contribute to training-induced adaptations in skeletal muscle. In addition, acute exercise results in activation of both p38 MAPK and AMP-activated protein kinase (AMPK) in skeletal muscle (1, 37). Because both of these have been shown to phosphorylate and activate PGC-1 (1, 10), a p38 MAPK/AMPK-mediated regulation of PGC-1 activity provides another potential mechanism by which PGC-1 could be involved in regulating exercise-induced gene expression.
The purpose of the present study was therefore to test the hypothesis that PGC-1 is required for exercise- and training-induced adaptive responses of mitochondrial genes/proteins in skeletal muscle and to examine the impact of PGC-1 in these processes in different muscle types.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
The generation and phenotype of PGC-1 KO black C57 mice have been described elsewhere (16). PGC-1 KO and wild-type (WT) mice were littermates produced by intercross breeding of heterozygote parents. The genotype of the offspring was assessed by determining the presence of either a WT- or KO-specific DNA fragment after extraction of DNA from a tail piece by the phenol-chloroform:isoamyl method, amplification of fragments by PCR using specific primers for both the WT and KO (16), and separation on an agarose gel. The survival of PGC-1 KO mice into adulthood was approximately half of the expected Mendelian ratio. All mice were kept on a 12:12-h light:dark cycle and received standard rodent chow (Altromin no. 1324, Chr. Pedersen, Ringsted, Denmark), and the mice were studied at 4–5 mo of age. Experiments were approved by the Danish Animal Experimental Inspectorate and complied with the European convention for the protection of vertebrate animals used for experiments and other scientific purposes (Council of Europe, no. 123, Strasbourg, France, 1985).
Exercise Performance Tests
Exercise performance of the PGC-1 KO and WT mice was examined by treadmill running. For the performance test, mice were placed on the treadmill and allowed to adapt to the surroundings for 3–5 min before starting. The treadmill speed was 8 m/min with a slope of 10% for 10 min followed by an increase in speed by 3 m/min every 3 min until exhaustion, defined by an inability to maintain running speed despite repeated encouragement with an air gun. In addition, after 15 min of running, each mouse was scored on a scale from 1 to 5, with 1 as a very poor runner and 5 as an excellent runner. The PGC-1 KO mice showed reduced running capacity, with average run-to-exhaustion of 23 ± 2.1 min compared with the WT mice, which ran 30 ± 1.8 min (n = 4–5), and a mean performance score of 2.5 for PGC-1 KO mice compared with 3.9 for WT mice (n = 4–5). These results were used to choose the running speed in the acute exercise and training studies described below.
Single Treadmill Exercise Bout
Starting 1 wk before the experimental day, mice were housed in individual cages and given standard rodent chow. PGC-1 KO and WT mice were on three separate days acclimatized to treadmill exercise for 10 min, and the last acclimatization treadmill exercise session was performed 37 h before the experimental trial. Each 10-min treadmill exercise bout consisted of 5 min at 8 m/min at 10% slope and 5 min at 14 m/min at 10% slope. An air gun was used to encourage running when necessary both during acclimatization and during the experimental run. On the experimental day, PGC-1 KO and WT mice performed a single exercise bout on a treadmill (60 min at 14 m/min, 10% slope). All mice were killed by a cervical dislocation in a 2-h fasted state between 4 and 6 p.m. Mice were killed immediately after running (0 h) or at 2 or 6 h of recovery. A control group that did not run acutely was also included (pre). Sol, white gastrocnemius (WG), and quadriceps (Quad) muscles were removed from the hind limbs and were quickly frozen in liquid nitrogen.
Exercise Training
PGC-1 KO and WT mice were randomly allocated to either a training group (8 males and 8 females of each genotype) or a control group (8 males and 8 females of each genotype), and all mice were placed in individual cages.
The training groups completed 5 wk of treadmill exercise training consisting of 60 min at 14 m/min at 10% slope 5 times per wk (Columbus Instruments, Exer 4 treadmill, Columbus, OH). In addition, the training-group mice had free access to an activity wheel in the cage (Minimitter Activitycage). Time spent in the activity wheel per day was measured by a computer (Sigma Sport, Neustadt, Germany). To ensure similar running volumes between WT and KO mice, the activity wheel of some WT mice was blocked 4 h into the dark period every evening during the last 4 wk. Therefore the total wheel-running duration per day was similar in WT and KO mice, with male WT 106 ± 10 min, male KO 98 ± 13 min, female WT 92 ± 8 min, and female KO 88 ± 13 min. Recordings indicated that PGC-1 KO mice voluntary ran at a slightly lower pace in the running wheel than WT mice, which is in agreement with a reduced running ability in PGC-1 KO mice, as shown in the present study and a previous study (14), as well as in muscle-specific PGC-1 KO mice (8).
Mice were killed 36–37 h after the last running bout at 8–10 a.m., and the Sol, WG, and Quad muscles were removed from both legs and immediately frozen in liquid nitrogen. Food was removed from the mice 2 h before death.
RNA Isolation and Reverse Transcription
Sol from both legs were used, whereas a piece of WG was cut off. Both Quad muscles (including both the white and the red portions) were crushed in liquid nitrogen to ensure homogeneity, and a part of this powdered muscle was used. RNA isolation was then performed on 15–20 mg muscle tissue with the guanidinium thiocyanate-phenol-choloroform method modified from Chomczynski and Sacchi (5), as described previously (21).
Reverse transcription (RT) was performed by using the Superscript II RNase H-system (Invitrogen) as previously described (21) and was diluted in nuclease-free H2O.
PCR
Real-time PCR was performed with an ABI 7900 sequence-detection system (Applied Biosystems, Foster City, CA). Primers and TaqMan probes for amplifying gene-specific fragments were designed by using a mouse-specific database (Ensemble) and Primer Express (Applied Biosystems). Primers and probes were obtained from TAG-Copenhagen (Copenhagen, Denmark), with all probes being 5'-FAM and 3'-TAMRA labeled, except GAPDH (Applied Biosystems), which was 5'-VIC and 3'-TAMRA labeled. Primers and probes used for real-time PCR are shown in Table 1.
|
Muscle Lysate Preparation
From WG and part of the pulverized Quad (including both the white and the red portions), muscle lysates were prepared by homogenization in 2-ml Eppendorf tubes using a Polytron (PT 1200, Kinematica). Homogenates were rotated end-over-end at 4°C for 1 h. Lysates were prepared from the homogenates by centrifugation for 30 min at 17,000 g and 4°C. Total protein content was determined by the bicinchoninic acid method (Pierce Chemical, Rockford, IL). Unless stated specifically, all chemicals were of analytic grade from Sigma-Aldrich (Denmark).
SDS-PAGE and Western Blotting
Muscle lysate proteins were separated by using 7.5–15% Tris·HCl gels (Bio-Rad) and were transferred (semidry) to PVDF membranes (Immobilion Transfer Membrane, Millipore). Standard Western-blotting procedures were used for detecting specific proteins as described previously (4, 38). Following detection and quantification with a charge-coupled device image sensor and 1-D software (Kodak Image Station, 2000MM), the protein content was expressed in arbitrary units relative to standard samples loaded in duplicate on each separate gel.
Primary antibodies were used for Western blotting: for AMPK subunit isoforms 1 and 2 were as described previously (4); anti--AMPK Thr172 phosphorylated (#2531, Cell Signaling Technology, Beverly, MA), anti-ACC Ser227 phosphorylated (Upstate Biotechnology), anti-HKII (Alpha Diagnostic International, San Antonio, TX), anti-cyt c (BD Biosciences-Pharmingen, San Diego, CA), anti-ALAS1 (Abnova, Taipai, Taiwan), and anti-COXI (Molecular Probes, Eugene, OR) were used. Secondary antibodies used were all species-specific horseradish peroxidase-conjugated immunoglobulins (DakoCytomation, Glostrup, Denmark).
Nucleotides and Glycogen Determination
Nucleotides were determined from 10 mg WG only. Muscle was freeze dried and extracted with perchloric acid and neutralized. Contents of ATP, ADP, AMP, and inosine monophosphate (IMP) were determined by reverse phase high-performance liquid chromatography according to previously described methods (32).
Muscle glycogen content was determined on 10 mg WG muscle as glycosyl units after acid hydrolysis (20).
Statistics
All data are presented as means ± SE. Two-way analysis of variance was applied to evaluate the effect of genotype and exercise/training intervention on mRNA content, protein content, glycogen, and nucleotides by using a Student-Newman-Keuls post hoc test to locate differences. Before statistical analysis, mRNA data was log transformed to approximate a normal distribution. Differences are considered significant at P < 0.05, and a tendency is reported with 0.1 < P < 0.05. Statistical calculations were performed by using Sigma Stat statistical software (Version 2.03).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Body weight.
The body weight of PGC-1 KO male (23.9 ± 0.3 g, n = 12) and KO female (19.4 ± 0.3 g, n = 16) mice was lower than WT male (27.7 ± 0.3 g, n = 12) and WT female (WT 21.0 ± 0.2g, n = 15) mice, respectively.
Muscle glycogen.
To confirm that the animals were metabolically challenged and the muscles recruited, WG muscle glycogen content was measured (Table 2). In PGC-1 KO mice, resting WG muscle glycogen concentration was 40% lower (P < 0.05) than in WT mice. The exercise bout reduced (P < 0.05) WG glycogen levels to 50% of the level in resting mice in both WT and KO animals. Whereas the WG glycogen level returned to the resting level already at 2 h of recovery in WT mice, the muscle glycogen concentration remained in KO mice at the level reached immediately after exercise (Table 2).
|
on the metabolic stress after exercise, ATP, ADP, AMP, and IMP were determined in WG at rest as well as immediately after exercise.
The single exercise bout did not induce detectable changes in ATP, ADP, AMP, or IMP either in PGC-1 KO or in WT mice, and no effect of genotype was apparent (Table 3).
|
.
PGC-1 mRNA was not detectable in PGC-1 KO muscle tissue. In WT mice, the exercise bout increased (P < 0.05) the PGC-1 mRNA content by five- to sevenfold in Sol and by approximately twofold in WG in recovery from exercise (Fig. 1, A and B).
|
. No genotype difference was detected for HKII mRNA in either Sol or WG muscles (Fig. 1, C and D). Sol HKII mRNA content was in both genotypes increased (P < 0.05) by two- to threefold at 2 and 6 h of recovery, whereas WG HKII mRNA content tended (main effect P = 0.06) to increase during recovery from exercise (Fig. 1, C and D).
GENES ENCODING MITOCHONDRIAL PROTEINS.
In Sol and WG, the mRNA levels of cyt c and ALAS1 were lower (P < 0.05) in PGC-1 KO than in WT mice (Fig. 2, A, B, E, and F). In WG of PGC-1 KO mice, the COXI mRNA level was at all time points reduced to 50% of the level in WT (P < 0.05). However, no significant genotype difference in COXI mRNA level was detected in Sol (Fig. 2, C and D).
|
KO animals (Fig. 2A). No change was detected in WG cyt c mRNA after the single exercise bout (Fig. 2B). In addition, no change was detected in COXI or ALAS1 mRNA content in response to the single exercise bout in either of the muscles or genotypes (Fig. 2, C–F).
PRDM16.
To examine the expression and response of a transcriptional coactivator potentially able to compensate for PGC-1-mediated effects, the transcriptional coactivator PR-domain zinc finger protein 16 (PRDM16) mRNA was examined. No genotype difference or exercise-induced change was detected in PRDM16 mRNA content in either Sol or WG (data not shown).
Protein expression.
AMPK SIGNALING.
No effect of genotype or exercise was apparent in WG AMPK1 or AMPK2 protein content (Fig. 3, D and E). At rest, no genotype difference was detected in AMPK or ACC phosphorylation in WG muscle (Fig. 3, B and C). Immediately after the acute exercise bout, AMPK phosphorylation increased (P < 0.05) by 2-fold in both WT and PGC-1 KO mice and ACC phosphorylation increased (P < 0.05) by 1.5-fold in WT and 2-fold in KO animals. AMPK and ACC phosphorylation levels returned to pre-exercise levels at 2 h of recovery from exercise in both genotypes (Fig. 3, B and C).
|
Training Study
Body weight.
After training, the body weight of PGC-1 KO male (25.3 ± 0.7 g, n = 8) and KO female (22.1 ± 0.4 g, n = 7) mice was lower (P < 0.05) than WT male (29.3 ± 0.7 g, n = 8) and WT female (23.6 ± 0.3 g, n = 8) mice, respectively. Body weight was not affected by 5 wk of exercise training in either genotype.
Muscle glycogen.
The muscle glycogen content in WG and Quad was similar in PGC-1 KO and WT mice. Trained WT and PGC-1 KO mice tended to have higher (10%; 0.05 < P < 0.10) muscle glycogen concentration in WG muscle than the corresponding untrained animals (Table 4).
|
. No effect of genotype was observed on HKII mRNA content. Exercise training induced a 15% increase in HKII mRNA content in WT (P < 0.05) and PGC-1 KO animals (Fig. 4A).
|
KO than WT mice (Fig. 4, B and C). Exercise training induced a 15% increase in Quad cyt c mRNA and ALAS1 mRNA content in both WT and PGC-1 KO mice (Fig. 4, B and C). No effect of genotype or training was observed in COXI mRNA content either in WT or in PGC-1 KO mice (Fig. 4D).
MYOGLOBIN.
Myoglobin mRNA content was 20% lower (P < 0.05) in PGC-1 KO mice than in WT animals. Myoglobin mRNA content was 15% higher (P < 0.05) in trained WT and PGC-1 KO than in the corresponding untrained mice (Fig. 5C).
|
KO animals. No effect of exercise training was observed in SOD1 mRNA content in either genotype (Fig. 5A). However, the SOD2 mRNA content was 30% lower (P < 0.05) in PGC-1 KO than in WT mice, and the SOD2 mRNA content was 15% higher (P < 0.05) in trained WT and PGC-1 KO than in the corresponding untrained mice (Fig. 5B).
In Quad muscle, the UCP2 mRNA content was similar in untrained WT and PGC-1 KO animals and the UCP2 mRNA content was 10% higher (P < 0.05) in trained than untrained mice for both genotypes (Fig. 5D).
Protein content.
AMPK SIGNALING.
The AMPK1 and AMPK2 protein content and phosphorylation of AMPK in WG were similar in PGC-1 KO and WT mice, and training did not affect the AMPK1 and AMPK2 protein level or phosphorylation in WG (data not shown).
METABOLIC PROTEINS.
The expression of COXI and cyt c protein was 15–30% lower (P < 0.05) in PGC-1 KO mice than in WT mice in both Sol, WG, and Quad muscles. This was evident both in untrained and in trained mice (Fig. 6). Cyt c protein content increased (P < 0.05) by 10% in Sol and WG and by 40% in Quad with exercise training in both WT and PGC-1 KO mice (Fig. 6, A, C, and D). In both genotypes, COXI protein content was 15% higher (P < 0.05) in WG and Quad muscles of trained mice than in untrained mice but did not change with exercise training in Sol muscle (Fig. 6, B, D, and F).
|
10% lower (P < 0.05) in Sol muscle of PGC-1 KO mice than WT mice in both untrained and trained mice (Fig. 7B). In WG and Quad, ALAS1 protein content was 10% lower (P < 0.05) in PGC-1 KO mice than in WT only in the trained animals (Fig. 7, D and F). Exercise training increased ALAS1 protein content by 15% in Sol and by 15–25% in WG and Quad muscles (Fig. 7, B, D, and E) in PGC-1 KO and WT mice.
|
130% in Quad (Fig. 7, A, C, and E). As expected, HKII protein content was similar in KO and WT mice in both untrained and trained animals. A representative blot is shown for each of the proteins cyt c, COXI, HKII, and ALAS1 in Fig. 8.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
KO mice had lower content of mitochondrial proteins than WT mice, the protein levels of AMPK1, AMPK2, and HKII were similar in muscles of the KO and WT mice. 2) PGC-1 KO abolished the exercise-induced upregulation of cyt c mRNA content but not HKII mRNA expression in skeletal muscle in response to a single exercise bout. 3) Exercise training induced a similar fold increase in cyt c, COXI, ALAS1, and HKII mRNA and protein levels in PGC-1 KO and WT mice.
The present results showing lower mRNA and protein expression of cyt c, COXI, and ALAS1 in skeletal muscles of PGC-1 KO mice are in agreement with previous studies showing reduced mRNA levels of these genes in the quadriceps muscle of PGC-1 KO mice and increased cyt c mRNA and protein cyt c expression in white vastus muscles of PGC-1-overexpressing mice (2, 14, 15, 16). Whereas previous studies with PGC-1 overexpression in mice and cell culture (15, 40) have demonstrated that increased levels of PGC-1 result in a general conversion of otherwise glycolytic, white muscles to very oxidative, red muscles (15), the present data show that lack of PGC-1 does lead to similar relative reductions in cyt c, COXI, and ALAS1 protein expression in the oxidative Sol, the glycolytic WG, and the metabolically mixed Quad muscle. This suggests that PGC-1-mediated regulation of mitochondrial proteins is independent of muscle type. In addition, PGC-1 is equally important in maintaining the levels of these mitochondrial proteins in muscles with very different metabolic enzyme profiles, which is likely related to the different endogenous expression levels of PGC-1 in these muscles (23). The present findings confirm the coordinating regulatory role of PGC-1 in skeletal muscle and show that this role is independent of metabolic muscle type. Hence the jointly reduced COXI, cyt c, and ALAS1 protein expression in all three muscles examined in PGC-1 KO mice underlines the existence of a PGC-1-mediated regulation of the respiration chain capacity of a muscle through concerted control of expression of nuclear- and mitochondrial-encoded cytochromes and production of the heme group needed for the cytochromes (15, 31, 34, 40) irrespective of the metabolic muscle type.
Of special interest is the role of PGC-1 in adaptive responses in skeletal muscle to regular physical activity. Because previous cell culture (40) and mice (15) studies convincingly have demonstrated PGC-1 to play a key role in regulating mitochondrial biogenesis, it was anticipated that training-induced adaptive responses in mitochondrial proteins would be reduced or abolished when PGC-1 was not present. However, the present findings that exercise training indeed can induce increased mRNA and protein content of mitochondrial COXI, cyt c, and ALAS1 in addition to nonmitochondrial metabolic proteins like HKII in muscles of PGC-1 KO mice show that PGC-1 is not required for these changes. The observation that the fold changes in training-induced mRNA and protein responses even were very similar in the two genotypes only underlines this further. In addition, these adaptations in the mitochondrial proteins cyt c, COXI, ALAS1, and the nonmitochondrial HKII occurred in Sol, WG, and Quad, indicating that this adaptability in the absence of PGC-1 is independent of metabolic muscle type. Thus these findings show that PGC-1 is not mandatory for exercise training-induced adaptations in skeletal muscle mitochondrial proteins and myoglobin mRNA, despite the fact that PGC-1 is needed to maintain normal expression levels of the same mRNAs/proteins in skeletal muscle as shown in the present and previous studies (14, 15). We interpret this to mean that either PGC-1 is simply not involved in such training-induced adaptive responses in murine skeletal muscle or that compensatory mechanisms take over when PGC-1 is absent. Although the current results do not allow discrimination between these possibilities, the present findings open the search for factors that mediate such responses as seen in PGC-1 KO mice.
As such, this may involve rather small changes in many factors exerting coactivator and corepressor functions but could potentially also be a single factor with similar function as PGC-1. Because the transcriptional coactivator PRDM16 recently was reported to elicit similar responses as PGC-1 in brown adipose tissue (28), we hypothesized that changes in PRDM16 expression are involved in regulating training-induced adaptations in skeletal muscle and that PRDM16 can compensate for PGC-1 in such responses. Because no difference was observed in skeletal muscle PRDM16 mRNA expression between KO and WT mice or in response to exercise in the present study, the current results do not support such PRDM16-mediated compensatory mechanisms when PGC-1 is absent. Still, it cannot be ruled out that regulation of PRDM16 activity or protein expression is involved in adaptive responses in skeletal muscle even in WT mice.
The similar HKII mRNA response in PGC-1 KO and WT mice to a single exercise bout is in agreement with the comparable adaptations in HKII mRNA and protein with training independent of genotype and the expectation that HKII gene regulation is independent of PGC-1. However, although the 5-wk exercise-training protocol also induced adaptations in mitochondrial proteins in the absence of PGC-1, the observation that the acute cyt c mRNA response was abolished in Sol in PGC-1 KO mice does support a role of PGC-1 in adaptive responses to exercise. It should be noted that the acute responses were only examined until 6 h of recovery, and it is thus possible that the lack of PGC-1 only resulted in a delayed cyt c mRNA response and that a cumulative effect of transient cyt c gene responses still exists without PGC1 in the muscles. This would be in accordance with the increased cyt c mRNA and protein content in muscles of PGC-1 KO mice after a period of exercise training.
Although the single exercise bout elicited acute responses of HKII and cyt c mRNA, no changes were detected at the protein level after the single exercise bout. The similar lack of change in ALAS1 protein expression after the single running bout is in contrast to the increased ALAS1 protein content reported in rat skeletal muscle 3, 12, and 18 h after a single 6-h swimming bout (39). However, this difference between studies may well be due to the combination of exercise duration and time of sampling after exercise, where the last time point in the present study was at 6 h after 1 h of running exercise because the focus of the acute study was on mRNA responses. A delayed protein response in ALAS1, cyt c, and HKII beyond 6 h of recovery from the present exercise protocol would be in line with the idea that cellular adaptations in skeletal muscles can stem from cumulative effects of repeated exercise bouts (36).
A major role of PGC-1 in the antioxidant defense of the cell (30) through PGC-1-mediated regulation of expression of scavenger enzymes (13) and uncoupling proteins (29, 33) has been illustrated by the increased reactive oxygen species formation and damage in the brain of PGC-1 KO mice (16, 30). Whereas the present result of markedly lower SOD2 mRNA expression in the Quad muscle of PGC-1 KO mice is in line with previously reported microarray data (2) and findings in fibroblasts from PGC-1 KO mice (30), the lack of effect of PGC-1 KO on SOD1 and UCP2 mRNA expression in Quad muscle is in contrast to previous results in fibroblasts (30). The clearly different dependency of PGC-1 in regulating SOD1 and SOD2 mRNA expression in Quad in the present study may be related to the cellular localization of the two isoforms, with SOD1 located in the cytoplasm (19) and SOD2 in the mitochondria (35). Interestingly, the present demonstration of training-induced increases in SOD2 and UCP2 mRNA expression indicates a mechanism for improved antioxidant defense in trained muscles. Moreover, the similar fold increases in SOD2 and UCP2 mRNA content in PGC-1 KO and WT mice with training suggest that PGC-1 is either simply not involved in regulating these adaptive responses or that compensatory mechanisms come into play in training-induced regulation of this scavenger and uncoupling protein in mouse skeletal muscle when PGC-1 is not present.
Surprisingly, despite reduced oxidative capacity (17) and reduced running ability in whole body PGC-1 KO mice, as shown in the present and a previous study (14), as well as in muscle-specific PGC-1 KO mice (8), exercise-induced activation of the energy sensor AMPK seemed unaffected by the lack of PGC-1. Thus even when PGC-1 KO mice and WT mice ran at the same speed and slope, exercise-induced AMPK and ACC phosphorylation were similar in the two genotypes and exercise did not elicit changes in nucleotide content in either genotype. Although these findings do not as such support that the PGC-1 KO mice were more metabolically challenged, the lower muscle glycogen content in muscles of PGC-1 KO mice at rest and less net glycogen usage during exercise in KO muscles than in muscles from WT mice may suggest that PGC-1 KO and WT mice used different substrates during running.
Also, in the present study PGC-1 KO mice killed late in the light period (single running study) had reduced muscle glycogen levels compared with their WT littermates, whereas no difference was apparent between genotypes for mice killed early in the light period (training study). This may be due to mice mainly eating during the dark period (26), and the observed difference in muscle glycogen levels late in the afternoon may be due to impaired maintenance of glucose homeostasis in PGC-1 KO mice. This is in line with the fact that PGC-1 has been shown to regulate the expression of several liver gluconeogenesis enzymes (24, 41).
Of special interest, in line with the present findings regarding PGC-1, mice lacking AMPK1 or AMPK2 have been shown to have reduced resting content of skeletal muscle mitochondrial proteins. Furthermore, lack of AMPK1 or AMPK2 was dispensable for exercise training-induced responses in muscle (12). Therefore, it seems that although AMPK and PGC-1 signaling pathways are important in regulating resting gene/protein expression in muscles, other factors can take over for training-induced responses in muscles when either AMPK or PGC-1 is lacking.
In conclusion, although lack of PGC-1 reduces resting mRNA and protein expression of cyt c, COXI, and ALAS1 and exercise-induced cyt c mRNA expression in skeletal muscle, PGC-1 is not mandatory for training-induced increases in skeletal muscle ALAS1, COXI, and cyt c expression. Although this does not rule out that PGC-1, when present, can serve such a role, this study shows that other factors than PGC-1 can mediate training-induced adaptations in skeletal muscle and apparently even exert coordinated regulation of expression of nuclear- and mitochondrial-encoded genes in response to exercise training.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. M. Robinson, K. L. Hamilton, and B. F. Miller The interactions of some commonly consumed drugs with mitochondrial adaptations to exercise J Appl Physiol, July 1, 2009; 107(1): 8 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Leick, Y. Hellsten, J. Fentz, S. S. Lyngby, J. F. P. Wojtaszewski, J. Hidalgo, and H. Pilegaard PGC-1{alpha} mediates exercise-induced skeletal muscle VEGF expression in mice Am J Physiol Endocrinol Metab, July 1, 2009; 297(1): E92 - E103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Adhihetty, G. Uguccioni, L. Leick, J. Hidalgo, H. Pilegaard, and D. A. Hood The role of PGC-1{alpha} on mitochondrial function and apoptotic susceptibility in muscle Am J Physiol Cell Physiol, July 1, 2009; 297(1): C217 - C225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. G. Coffey, H. Pilegaard, A. P. Garnham, B. J. O'Brien, and J. A. Hawley Consecutive bouts of diverse contractile activity alter acute responses in human skeletal muscle J Appl Physiol, April 1, 2009; 106(4): 1187 - 1197. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hoene, R. Lehmann, A. M. Hennige, A. K. Pohl, H. U. Haring, E. D. Schleicher, and C. Weigert Acute regulation of metabolic genes and insulin receptor substrates in the liver of mice by one single bout of treadmill exercise J. Physiol., January 1, 2009; 587(1): 241 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Pilegaard and E. A. Richter PGC-1{alpha}: important for exercise performance? J Appl Physiol, May 1, 2008; 104(5): 1264 - 1265. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
Tendency for a difference over time 0.05 < P < 0.1.
