AMPK - Mediated Regulation of Protein Degradation Systems in Unloaded Mouse Skeletal Muscle

Objective : The aim of the present study was to investigate the involvement of AMPK in regulating skeletal muscle atrophy during hindlimb unloading. Methods : Transgenic (AMPK-DN) mice expressing a dominant negative mutant of AMPK α 1 in the skeletal muscle and their wild-type littermates (WT) mice were randomly divided into two groups: untreated preexperimental control (n=12/group) and unloading (n=12/group) groups. Mice of the unloading group were subjected to continuous hindlimb suspension for 2 weeks. Results : Soleus muscle weight relative to body weight in WT mice was decreased by 30% in response to hindlimb suspension, whereas by 20% in AMPK-DN mice. The expressions of ubiquitinated proteins and MuRF1 mRNA, markers of ubiquitin-proteasome system activation, were upregulated by hindlimb suspension in WT mice, but no changes were observed in AMPK-DN mice. The expression of phosphorylated FoxO3a was decreased by hindlimb suspension in WT mice, but not in AMPK-DN mice. HSP72 expression was higher in AMPK-DN mice compared to WT mice during the experiment, and reduced more in WT mice by hindlimb suspension than AMPK-DN mice. Conclusions : The present study demonstrated that the repression of skeletal muscle AMPK activation suppressed the progress of unloading-induced skeletal muscle atrophy. Our findings suggest that AMPK is involved in adaptation of skeletal muscle mass to atrophic stimuli.


Introduction
Skeletal muscle has a large capacity to adapt to various stimuli. For example, aging, poor nutrition, several diseases, and decreased loading such as inactivity, result in skeletal muscle atrophy 1)-3) . However, the molecular mechanism involved in atrophic process in skeletal muscle is not fully understood.
Several studies have been revealed that 5ʼ AMP-activated protein kinase (AMPK) has a potential role in skeletal muscle mass regulation.
Elevated AMPK activity led to diminished capacity for hypertrophy of fast-twitch skeletal muscle in aged rat 4) 5) . Moreover, impaired hypertrophy of slow-twitch skeletal muscle during overload in diabetic rat was partly attributed to upregulation of the expression of phosphorylated AMPKα Thr 172 6) . A study using a knockout mouse model showed that overload-induced muscle hypertrophy was accelerated in AMPKα1-deficient mice compared to the wild-type mice 7) . Taken together, AMPK is suggested to involve in the regulation of skeletal muscle mass during hypertrophic conditions. However, there are no studies examining a role of AMPK in the regulation of skeletal muscle mass during atrophic conditions. Therefore, the aim of the present study was to investigate the potential function of AMPK in muscle mass adaptation in response to atrophic stimuli. For the purpose, we examined the changes of muscle mass and molecular responses after 2-week hindlimb unloading using transgenic mice that overexpresses musclespecific dominant-negative mutant of AMPKα1 (AMPK-DN).

Animals
Transgenic (AMPK-DN) mice expressing a dominant negative mutant of AMPKα1 in the skeletal muscle 8) were obtained from JCRB (Japanese Collection of Research Bioresources Cell Bank) Laboratory Animal Resource Bank at NIBIO (National Institute of Biomedical Innovation, Osaka, Japan). Twenty-four male AMPK-DN mice (age: 13.2 ± 3.2 weeks, body weight: 24.4 ± 1.5 g, mean ± SD) and twenty-four their WT mice (age: 13.5 ± 3.5 weeks, body weight: 23.2 ± 2.9 g, mean ± SD) were used. All mice were housed in an animal room maintained at 22-24℃ with a 12:12-h lightdark cycle and fed a standard laboratory diet and water ad libitum. All animal protocols were carried out in accordance with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health (Bethesda, MD).

Procedure of hindlimb unloading
Both AMPK-DN and WT mice were randomly divided into two groups: untreated preexperimental control (n = 12 in each group) and unloading (n = 12 in each group) groups. Mice of the unloading group were subjected to continuous hindlimb suspension for 2 weeks. Hindlimb suspension was performed as described previously 9) . Body weight of each mice was recorded at the end of experiment. Two weeks after the procedure, soleus was dissected from each mice and weighed. Left muscles for real-time RT-PCR analyses and right muscles for western blot analyses and AMPK kinase assay were frozen in liquid nitrogen, and stored at -80℃.

Western blot analyses
Sample preparation and western blot analyses were performed with some modification of the previously reported method 10) 11) .

Real-time RT-PCR analyses
Real-time RT-PCR analyses were performed as was described previously 12) .

Isoform-specific AMPK activity assay
The AMPK activity assay was performed with some modification of the previously reported method 13) .

Statistical analyses
Values were expressed as mean ± SEM. Statistical significance was analyzed by using two-way analysis of variance (ANOVA) with mice and treatments as main factors. Post hoc analyses were conducted with Tukey-Kramerʼs test. The differences between groups were considered statistically significant at p < 0.05.

AMPK activity
The measurement results of AMPK activity and phosphorylation level of ACC Ser 79 , a marker of AMPK activity, in soleus muscle are shown in Figure-1. Basal activity of AMPKα1 and AMPKα2 was lower by 20% and 95% in AMPK-DN mice than WT mice, respectively (Figure-1). The predominant reduction of AMPKα2 activity rather than AMPKα1 activity in the transgenic mice expressing inactive α1 mutant was corresponding with the results previously reported 8) 14) 15) . A recent study reported that skeletal muscle AMPK signaling was up-regulated at the early stage (3 days) of hindlimb unloading and returned to basal state at 7 days in mice 16) . Correspondingly, our findings showed no activation of AMPK signaling after 2-week hindlimb unloading ( Figure-1). Although we did not examine the time-course changes of AMPK activity, it might be that AMPK signaling was temporary activated following hindlimb unloading and returned to basal state at 2 weeks in the present study.

Muscle mass
It is well known that diminished loading results in skeletal muscle atrophy. In the present study, when the muscle weight was normalized to body weight to correct for the loss of weight after hindlimb unloading, the relative weight of soleus to body weight were decreased following hindlimb unloading ( Figure-2). Notably, soleus muscle was atrophied by~30% in WT mice during hindlimb unloading, while the deficiency of skeletal muscle AMPKα2 activity weakened the progress of atrophy almost by half (~17%, Figure-2). These data indicate that AMPK, mainly AMPKα2, may be a crucial molecule regulating unloading-induced skeletal muscle atrophy.

Ubiquitin-proteasome system
Increased protein degradation is contributed to unloading-induced atrophy in skeletal muscle 17) . Ubiquitin-proteasome system is well known as a major protein degradation pathway 18) . The key enzyme in this pathway is E3 ubiquitin ligases, which is responsible for protein ubiquitination. The two muscle-specific ubiquitin ligases, MuRF1 and atrogin-1/MAFbx, have been considered to be master regulators of skeletal muscle atrophy, because these genes are up-regulated in different models of muscle atrophy and have an important role in increasing protein degradation through ubiquitin-proteasome system 19) 20) . Previous studies have reported that agonist-induced activation of AMPK enhances protein degradation accompanied by increased MuRF1 and atrogin-1/MAFbx mRNA expressions in cultured myotubes 21) 22) . In addition, we have recently demonstrated that pharmacological activation of AMPK up-regulates MuRF1 mRNA expression and this up-regulation is abolished in AMPK-knockdown cells 23) . Thus, AMPK appears to be associated with activation of ubiquitin-proteasome system, and it is possible that AMPK regulates protein degradation through ubiquitin-proteasome system during unloading. In the present study, the accumulation of ubiquitinated proteins was observed after hindlimb unloading in WT mice, but not in AMPK-DN mice (Figure-3). MuRF1 mRNA and protein expressions were significantly increased in response to hindlimb unloading by 4.0-fold in WT mice, but not in AMPK-DN mice (Figure-3). These results indicate that the unloading-induced activation of ubiquitin-proteasome system was attenuated in the suppression of AMPK. Therefore, it is suggested that AMPK regulates ubiquitin-proteasome Values are mean ± SEM. n = 8 per group. †: post hoc multiple comparisons tests following two-way ANOVA showed that the overall differences were statistically significant between wild-type littermates (WT) mice and mice overexpressed muscle-specific AMPK dominant-negative (DN), ¶: post hoc multiple comparisons tests following two-way ANOVA showed that the overall differences were statistically significant between untreated control (C) and hindlimb unloading (HU) groups.  system-mediated protein degradation during skeletal muscle atrophy in response to unloading.

Signaling molecules associated with protein degradation systems
Our findings suggest a role of AMPK that regulates unloading-induced skeletal muscle atrophy through modulating protein degradation systems. In this context, there are some possible mechanisms by which AMPK activates protein degradation systems during unloading. FoxOs are transcriptional factors that regulate transcription of genes associated with skeletal muscle homeostasis including skeletal muscle atrophy 24) 25) . Previous reports have suggested that AMPK-mediated modulation of FoxO3a expression and/or nuclear translocation contributes to activation of ubiqui-tin-proteasome and autophagy systems in skeletal muscle cells 22) 26) 27) . Thus, it is possible that AMPK regulates protein degradation systems in unloadedassociated skeletal muscle atrophy through a FoxO3a-dependent mechanism. In the present study, the expression of phosphorylated FoxO3a Ser 253 was decreased by hindlimb unloading in WT mice, whereas that in AMPK-DN mice was not affected by hindlimb unloading (Figure-4), suggesting that AMPK participates in the activation of FoxO3a during skeletal muscle unloading. Therefore, FoxO3a is a possible molecule related to AMPK-mediated up-regulation of protein degradation systems in response to unloading.
On the other hand, a recent study have suggested that nuclear factor-κB (NF-κB) signaling is more important than FoxO signaling in disuse muscle  atrophy 28) , since NF-κB sites, but not FoxO sites, are required for the transcription of MuRF1 during hindlimb unloading. NF-κB is a transcriptional factor that is sequestered in the cytoplasm by a family of inhibitory proteins called IκBα 29) . The IκB kinase complex phosphorylates IκBα, resulting in its degradation, thereby leading to nuclear translocation of NFκB and activation. It has been reported that disruption of NFκB prevents skeletal muscle atrophy induced by hindlimb unloading 30) . In the present study, the expression of IκBα tended to decrease during muscle atrophy in WT mice, and the expression was high in AMPK-DN mice compared to WT mice after hindlimb unloading (Figure-4). These results suggest that AMPK regulates NFκB signaling via the expression of IκBα during unloading-associated muscle atrophy and this might affect the different activation of ubiquitinproteasome system including MuRF1 expressions.
HSP72 might be another candidate molecule involved in the regulation of AMPK-mediated protein degradation systems during unloading. HSP72 is one of the most prominent member of HSPs family and considered to have an important role in preventing skeletal muscle atrophy 31) . In the present study, it was observed that HSP72 expression in AMPK-DN mice was high and decreased less by unloading compared to WT mice (Figure-4). It has been reported that overexpression of HSP72 in skeletal muscle prevents immobilization-induced atrophy in rat 32) . Furthermore, a molecular mechanism of the resistance to skeletal muscle atrophy by HSP72 seems to be that HSP72 directly prevents FoxO3a activation during unloading 25) 32) . We have also previously demonstrated that AMPK negatively regulates HSP72 expression in skeletal muscle cells and that HSP72 controls AMPK-mediated activation of ubiquitinproteasome system 23) . Considering these findings, it is suggested that a high expression of HSP72 due to the suppression of AMPK activity is a possible mechanism that attenuates the unloading-induced activation of protein degradation system, partly through FoxO3a deactivation.

Autophagy system
Autophagy is another important cell proteolytic system that controls protein turnover in skeletal muscle 33) . During autophagosome formation, LC3I is converted to LC3II through lipidation that allows for LC3 to become associated with autophagic vesicles. The presence of LC3 in autophagosomes and the conversion of LC3 to the lower migrating form LC3II have been used as indicators of autophagy activity 34) . Recently, it has been reported that AMPK activation stimulates autophagosome formation in skeletal muscle cells 26) , thus a modulation of autophagy process is possible to involve in AMPK-mediated regulation of protein degradation during unloading. In the present study, the relative expression of LC3II to LC3I was increased in response to hindlimb unloading by 8.0-fold in WT mice but by 2.0-fold in AMPK-DN mice (Figure-5). These findings indicate that AMPK mediates autophagosome formation during unloadinginduced skeletal muscle atrophy.
The ubiquitin-binding protein p62 which binds to LC3 is preferentially degraded by autophagy 35) , and thus breakdown of p62 is generally used as a marker of autophagy flux 36) . In the present study, accumulation of p62 after hindlimb unloading was also observed in WT mice but not in AMPK-DN mice (Figure-5). This is consistent with the previous findings that p62 mRNA is up-regulated in mouse soleus 37) and gastrocnemius 16) muscle following 3-day hindlimb unloading and that p62 protein is increased by 4-week hindlimb unloading in mouse tibialis anterior and gastrocnemius muscle 38) . Accumulation of p62 generally indicates an impairment of autophagy flux 36) , but p62 hyperexpression was also observed in cancer cachexiainduced skeletal muscle atrophy despite the autophagy induction 39) . Although our findings indicate that AMPK modulates the expression of autophagy-related proteins during unloading-induced muscle atrophy, we cannot ascertain whether AMPK-mediated autophagy regulation is associated with the progress of muscle atrophy in response to hindlimb unloading.

Conclusions
We showed that the suppression of musclespecific AMPK activity (mainly AMPKα2) partially attenuated unloading-induced atrophy of slowtwitch soleus muscle. The protective effect of muscle atrophy might be attributed to attenuation of the activity of ubiquitin-proteasome-mediated protein degradation. This is supported by the alterations of signaling molecules including FoxO3a, IκBα, and HSP72. Overall, we suggest that AMPK is required for proper adaptation of muscle mass and its related molecules during skeletal muscle unloading.
This article was adapted from Egawa et al. 40) with permission by the publisher.