**3. Negative regulators of skeletal muscle mass**

#### **3.1. Ubiquitin-Proteasome System (UPS)**

The UPS is essential for protein degradation. The degradation of a protein via the UPS includes two steps: (1) tagging of the substrate by covalent attachment of multiple ubiquitin molecules and (2) degradation of the tagged protein by the 26S proteasome. The ubiquitination of proteins is regulated by at least three enzymes: ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2), and ubiquitin ligase (E3). Consistent increases in two important E3 ubiquitin ligases (atrogin-1 and MuRF-1) gene expression have been observed in a wide range of in vivo models of skeletal muscle atrophy including diabetes, cancer, denervation, unweighting, or glucocorticoid treatment [60].

eIF 3 subunit 5 (eIF3-f) appears to be a key effector of atrogin-1 as targeted increases and decreases in eIF3-f levels cause skeletal muscle hypertrophy and atrophy, respectively [61]. Overexpression of atrogin-1 results in the polyubiquitination of MyoD and an inhibition of MyoD-induced myotube formation [62]. In contrast, the knockdown of atrogin-1 reversed endogenous MyoD proteolysis and the overexpression of a mutant MyoD, unable to be ubiquitinated, prevented muscle atrophy *in vivo* [62]. These results confirmed MyoD as a substrate of atrogin-1 during dexamethasone-induced myotube atrophy [63].

In contrast to atrogin-1, it appears that MuRF-1 mainly interacts with structural proteins. MuRF-1 binds to titin and potentially affects titin signaling. It also binds to and degrades MHC proteins following the treatment of skeletal muscle with dexamethasone. Additionally, MuRF-1 degrades myosin light chain 1 and 2 during denervation and fasting conditions [64]. These studies suggest that while numerous stimuli can activate both two atrogenes, the downstream pathways affected may be separate for each protein.

#### *3.1.1. Adapatative changes in atrogin-1 and MuRF-1 in muscle atrophy models*

#### **UPS signaling in cachectic muscle**

A decrease of SRF expression achieved using a transgenic approach was found to accelerate the atrophic process in muscle fibers with age [53]. These SRF-deleted mice showed marked deposition of intermuscle lipid with aging. One morphologic aspect of sarcopenia is the infiltration of muscle tissue components by lipids, because of the increased frequency of adipocyte or lipid deposition [54] within muscle fibers. As with precursor cells in bone marrow, liver, and kidney, muscle satellite cells expressing the adipocytic phenotype increased with age [55], although this process is still relatively poorly understood in terms of its extent and spatial distribution. Lipid deposition may result from a net buildup of lipids due to the reduced oxidative capacity of muscle fibers with aging [54]. These lines of evidence clearly showed a

SRF appears to be linked to the degenerative process during muscular dystrophy. Significant reduction in the amount of SRF has been observed [56], 40-50% and 50-65% at 2 and 12 weeks of age, respectively, in merosin-deficient congenital muscular dystrophy. Our immunohisto‐ chemical analysis demonstrated that mature normal mice exhibited an abundant SRF protein in the cytoplasm of several muscle fibers, while the *dy* mice did not. There is no direct evidence of a link between SRF disorders and the pathogenesis of disease in the skeletal muscle.

However, Lange et al. [36] observed that a mutation in the TK domain of titin, a possible modulator of SRF, disrupted Nbr1 binding and led to hereditary myopathy with early respiratory failure (HMERF). HMERF patient muscle biopsies revealed a Nbr1 diffusible localization, cytoplasmic p62/SQSTM1 aggregates, and the selective accumulation of MuRF-2 in centralized nuclei. Unfortunately, the localization of SRF has not been determined in the muscle of HMERF patients. In contrast, a natural dominant-negative form of SRF was dem‐ onstrated to be elevated in human heart failure [57]. The dominant negative SRF isoform potently inhibited SRF-dependent function, showing the biochemical phenotype seen in SRFnull mice [57]. In addition, a subsequent human heart failure study showed decreases in fulllength SRF and elevated expression of a caspase-3-cleaved product of SRF [58]. A more recent review [59] proposed various disorders to be linked with the SRF mutation as shown by many

The UPS is essential for protein degradation. The degradation of a protein via the UPS includes two steps: (1) tagging of the substrate by covalent attachment of multiple ubiquitin molecules and (2) degradation of the tagged protein by the 26S proteasome. The ubiquitination of proteins is regulated by at least three enzymes: ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzyme (E2), and ubiquitin ligase (E3). Consistent increases in two important E3 ubiquitin ligases (atrogin-1 and MuRF-1) gene expression have been observed in a wide range of in vivo

defect of SRF-signaling in aged mammalian muscle.

reliable studies using cell-specific SRF-knockout phenotypes.

**3. Negative regulators of skeletal muscle mass**

**3.1. Ubiquitin-Proteasome System (UPS)**

**2.3. Muscular disorder**

150 Muscle Cell and Tissue

In many acute models of cachexia, including cancer, the UPS is thought to be fundamental in the process of muscle atrophy [60, 65]. Weight-losing mice bearing MAC16 adenocarcinoma exhibited the increased expression of mRNA for both α- and β-proteasome subunits in gastrocnemius muscle [66]. Several experimental models of cancer cachexia [e.g., AH-130, C26, Lewis lung carcinoma (LLC)] had increased UPS activity, as well as overexpression of atrogin-1 and MuRF-1 [67]. However, investigations in humans have failed to be conclusive. An earlier study suggested that the UPS plays a prominent role in the degradation of myofibrillar proteins, particularly in cancer patients with weight loss of > 10% [68]. In gastric cancer patients with average weight loss of 5.2%, increased UPS activity (determined by measurements of RNA and cleavage of specific fluorogenic substrates) was seen compared with that in controls, which was exacerbated by increasing tumor stage and weight loss [69]. In contrast, investiga‐ tions of UPS activity in quadriceps muscle biopsies have shown levels similar to those in healthy controls in patients with lung cancer and weight loss < 10% (termed pre-cachexia) [70]. In a transcriptomic study of UGI patients, candidate genes, including FOXO and ubiquitin E3 ligases, were not related to weight loss. Another study in lung cancer patients with low weight loss demonstrated no change in the components of the UPS using Northern blotting, but there was a suggestion that the activity of the lysosomal pathway was increased [71]. Similarly, a more recent study also failed to show an increase of proteasome activity in muscle of esoph‐ ageal cancer patients [72]. Although more descriptive studies are needed to determine whether UPS is activated in muscle in cancer cachexia, the adaptive manner of UPS components may differ between rodents and humans in cancer-cachectic muscles.

In chronic obstructive pulmonary disease (COPD) patients, the reported findings on the adaptive changes in the UPS in skeletal muscle are highly contradictory. Doucet et al. [73] found significant increases in atrogin-1, MuRF-1, and FOXO1 mRNA in the muscle of COPD patients showing irreversible airflow obstruction [post-bronchodilator forced expiratory volume in one second (FEV1) < 80% predicted and FEV1/forced vital capacity (FVC) < 70%] by pulmonary function testing. They also observed an increase, albeit not significant, in the atrogin-1 protein content in the muscle of COPD patients. Such inconsistency in the protein levels of ubiquitin ligases was recognized in COPD patients by Natanek et al. [74]. In contrast, Natanek et al. [74] reported a significant decrease in atrogin-1 protein content in quadriceps in COPD compared with those in controls. Advanced cardiac heart failure (CHF) patients also did not exhibit high mRNA and protein levels of atrogin-1 in muscle [75], although CHF model mice with cardiac-specific overexpression of calsequestrin exhibited marked increases in atrogin-1 mRNA in both soleus and white vastus lateralis muscles [76]. In human studies, several investigators described a discrepancy between the mRNA expression level on atrogene and the extent of muscle atrophy [77]. Intriguingly, some recent studies have shown that atrogin-1 and MuRF-1 are not essential components for proteasome activity [78]. There have been few studies dealing with the protein levels of atrogin-1 and MuRF-1 in muscle wasting, particularly in human subjects.

#### **UPS signaling in sarcopenic muscle**

Atrogin-1 and/or MuRF-1 mRNA levels in aged muscle are reportedly increased or unchanged in humans and rats, or decreased in rats [50, 79, 80]. Even when the mRNA expression of these atrogenes increased in sarcopenic muscles, this was very limited (1.5- to 2.5-fold). Although mRNA levels of both ubiquitin ligases have been determined in several aged mammalian muscles, the analysis of protein levels did not correspond to age-related increases in the mRNA of ubiquitin ligases. For instance, the marked upregulation of phosphorylated Akt and FOXO4 have been observed in the gastrocnemius muscle of aged female rats [80]. These adaptations of protein levels probably contribute to the downregulation of atrogin-1 and MuRF-1 mRNA in aged muscle. In addition, Léger et al. [81], using human subjects aged 70 years old, dem‐ onstrated decreases in nuclear FOXO1 and FOXO3a in spite of no apparent age-related changes in the atrogin-1 and MuRF-1 mRNA. Interestingly, recent findings indicate that atrogin-1 knockout mice are short-lived and experience higher loss of muscle mass during aging than control mice [82], indicating that the activity of this E3 ubiquitin ligase is required to preserve muscle mass during aging in mice. Moreover, the muscle size of aged MuRF-1-null mice is preserved [83]. However, they exhibited a higher decay of muscle strength than controls. As indicated by Sandri et al. [82], chronic inhibition of these atrogenes should not be considered a therapeutic target to counteract sarcopenia because this does not prevent muscle loss but instead exacerbates weakness. Figure 2 summarizes a possible adaptation of atrogin-1 and MuRF-1 in sarcopenic muscle.

#### **The adaptation of UPS in muscular dystrophy**

Gene expression profiling in Limb-girdle muscular dystrophy (LGMD)2A showed overex‐ pression of UPS-related genes [84, 85]. While the expression of atrogin-1 and MuRF-1 was not increased in mouse models of LGMD2A, FOXO1 was strongly upregulated and induced muscle atrophy in calpain-3-deficient mice [86]. More recently, LGMD2A patients have been shown to exhibit significantly higher expression of MuRF-1 protein but not atrogin-1 protein

found significant increases in atrogin-1, MuRF-1, and FOXO1 mRNA in the muscle of COPD patients showing irreversible airflow obstruction [post-bronchodilator forced expiratory volume in one second (FEV1) < 80% predicted and FEV1/forced vital capacity (FVC) < 70%] by pulmonary function testing. They also observed an increase, albeit not significant, in the atrogin-1 protein content in the muscle of COPD patients. Such inconsistency in the protein levels of ubiquitin ligases was recognized in COPD patients by Natanek et al. [74]. In contrast, Natanek et al. [74] reported a significant decrease in atrogin-1 protein content in quadriceps in COPD compared with those in controls. Advanced cardiac heart failure (CHF) patients also did not exhibit high mRNA and protein levels of atrogin-1 in muscle [75], although CHF model mice with cardiac-specific overexpression of calsequestrin exhibited marked increases in atrogin-1 mRNA in both soleus and white vastus lateralis muscles [76]. In human studies, several investigators described a discrepancy between the mRNA expression level on atrogene and the extent of muscle atrophy [77]. Intriguingly, some recent studies have shown that atrogin-1 and MuRF-1 are not essential components for proteasome activity [78]. There have been few studies dealing with the protein levels of atrogin-1 and MuRF-1 in muscle wasting,

Atrogin-1 and/or MuRF-1 mRNA levels in aged muscle are reportedly increased or unchanged in humans and rats, or decreased in rats [50, 79, 80]. Even when the mRNA expression of these atrogenes increased in sarcopenic muscles, this was very limited (1.5- to 2.5-fold). Although mRNA levels of both ubiquitin ligases have been determined in several aged mammalian muscles, the analysis of protein levels did not correspond to age-related increases in the mRNA of ubiquitin ligases. For instance, the marked upregulation of phosphorylated Akt and FOXO4 have been observed in the gastrocnemius muscle of aged female rats [80]. These adaptations of protein levels probably contribute to the downregulation of atrogin-1 and MuRF-1 mRNA in aged muscle. In addition, Léger et al. [81], using human subjects aged 70 years old, dem‐ onstrated decreases in nuclear FOXO1 and FOXO3a in spite of no apparent age-related changes in the atrogin-1 and MuRF-1 mRNA. Interestingly, recent findings indicate that atrogin-1 knockout mice are short-lived and experience higher loss of muscle mass during aging than control mice [82], indicating that the activity of this E3 ubiquitin ligase is required to preserve muscle mass during aging in mice. Moreover, the muscle size of aged MuRF-1-null mice is preserved [83]. However, they exhibited a higher decay of muscle strength than controls. As indicated by Sandri et al. [82], chronic inhibition of these atrogenes should not be considered a therapeutic target to counteract sarcopenia because this does not prevent muscle loss but instead exacerbates weakness. Figure 2 summarizes a possible adaptation of atrogin-1 and

Gene expression profiling in Limb-girdle muscular dystrophy (LGMD)2A showed overex‐ pression of UPS-related genes [84, 85]. While the expression of atrogin-1 and MuRF-1 was not increased in mouse models of LGMD2A, FOXO1 was strongly upregulated and induced muscle atrophy in calpain-3-deficient mice [86]. More recently, LGMD2A patients have been shown to exhibit significantly higher expression of MuRF-1 protein but not atrogin-1 protein

particularly in human subjects.

152 Muscle Cell and Tissue

MuRF-1 in sarcopenic muscle.

**The adaptation of UPS in muscular dystrophy**

**UPS signaling in sarcopenic muscle**

**Figure 2.** The comparison of ubiquitin-proteasome system between young and sarcopenic muscle. In contrast to young muscle, sarcopenic muscle exhibits no activation of atrogin-1 and MuRF-1-dependent signaling to destroy the degener‐ ative proteins.

in skeletal muscle. LGMD2B is due to deficiency of the protein dysferlin. The loss of dysferlin causes failure in resealing of the membrane lesions generated during eccentric muscle contractions [88]. Similar to LGMD2A, dysferlinopathy patients exhibited more abundant mRNA and protein of MuRF-1 but not atrogin-1 [89]. Activation of UPS in dysferlinopathy has also been reported in cellular models (patient-derived muscle cells) [90]. UCMD is a common form of muscular dystrophy associated with defects in collagen VI, characterized by loss of muscle fibers and proliferation of connective and adipose tissues. More recently, Paco et al. [91] studied muscle biopsies of Ullrich congenital muscular dystrophy (UCMD) (n = 6), other myopathy [Duchenne muscular dystrophy (DMD)], calpain-3-deficient, Kearns-Sayre, and nemaline myopathy (n = 12), and control patients (n = 10) and found reduced expression of atrogin-1 and MuRF-1 mRNAs in UCMD cases.

Pharmacological inhibition of UPS appears to exert some beneficial effect on muscular dystrophy. Velcade, once injected locally into the gastrocnemius muscles of *mdx* mice, appears to increase the expression and membrane localization of dystrophin and members of the dystrophin-associated protein complex (DAPC) [92]. Treatment with Velcade (0.8 mg/Kg) over a 2-week period has been shown to reduce muscle degeneration and necrotic features, and to increase muscle size (gastrocnemius and diaphragm), in mdx muscle fibers [93]. In addition, Gazzerro et al. [93] observed that Velcade administration generates many myotubes and/or immature myofibers expressing embryonic myosin heavy chain in mdx muscle, probably due to upregulation of myogenic differentiating modulators.

#### **3.2. Autophagy-dependent signaling**

Macroautophagy (herein autophagy) is a catabolic process that involves the bulk degradation of cytoplasmic components by interacting a lysosome [94, 95]. This process is characterized by the engulfment of part of the cytoplasm inside double-membrane vesicles (autophagosomes). Autophagosomes subsequently fuse with lysosomes to form autophagolysosomes in which the cytoplasmic cargo is degraded. The turnover of most proteins, biological membranes, and whole organelles such as mitochondria and ribosomes is mediated by autophagy [96].

Autophagy represents an extremely refined collector of altered organelles, abnormal protein aggregates, and pathogens, similar to a selective recycling center [97]. The selectivity of the autophagy process is conferred by a growing number of specific cargo receptors such as p62/ SQSTM1 and Nix (Bnip3L) [98]. These adaptor proteins are equipped with both a cargobinding domain, with the capability to recognize and attach directly to molecular tags on organelles. At the same time, these adaptor proteins bind a microtubule-associated protein light chain LC3)-interacting region domain to recruit and bind essential autophagosome membrane proteins. Three molecular complexes mainly regulate the formation of autopha‐ gosomes: the LC3 conjugation system and the regulatory complexes governed by unc51-like kinase-1 and beclin-1. The conjugation complex is composed of different proteins encoded by autophagy-related genes (Atg). The Atg12-Atg5-Atg16L1 complex, along with Atg7, plays an essential role in the conjugation of LC3 to phosphatidylethanolamine, which is required for the elongation and closure of the isolation membrane.

The UPS and the lysosomal-autophagy system in skeletal muscle are interconnected [15, 16]. Both these studies identified FOXO3 as a regulator of these two pathways in muscle wasting [Fig. 3]. FOXO3 is a transcriptional regulator of the atrogin-1 and MuRF-1. FOXO3 modulates the expression of autophagy-related genes in mammalian skeletal muscle and C2C12 myo‐ tubes [16]. Masiero et al. [99] found an intriguing characteristic using muscle-specific autoph‐ agy knockout mice, which exhibit fiber atrophy, weakness, and mitochondrial abnormalities. Autophagy-dependent protein degradation seems to be also modulated by tumor necrosis factor (TNF) receptor associated factor 6 and peroxisome proliferator-activated receptor γ coactivator-1α (PGC1α) [100, 101]. Wenz et al. [101] recognized no significant age-related increase in the ratio of LC3-II to LC3-I in MCK-PGC1α mice. Therefore, PGC1α would attenuate the autophagic process probably through increased anti-oxidant defense and mitochondrial biogenesis.

#### *3.2.1. Adaptation of autophagy-linked signaling during muscle atrophy*

#### **A possible contribution of autophagic signaling to cachexia**

As for cancer cachexia, earlier results obtained on muscles isolated from cachectic animals led us to rule out a substantial role for lysosomes in overall protein degradation [102]. In contrast, an elevation of total lysosome protease activity has been observed in the skeletal muscle and liver of tumor-bearing rats [103]. In addition, increased levels of cathepsin L mRNA have been reported in the skeletal muscle of septic or tumor-bearing rats, whereas cathepsin B gene expression has been shown to be enhanced in muscle biopsy samples obtained from patients

**3.2. Autophagy-dependent signaling**

154 Muscle Cell and Tissue

the elongation and closure of the isolation membrane.

*3.2.1. Adaptation of autophagy-linked signaling during muscle atrophy*

**A possible contribution of autophagic signaling to cachexia**

mitochondrial biogenesis.

Macroautophagy (herein autophagy) is a catabolic process that involves the bulk degradation of cytoplasmic components by interacting a lysosome [94, 95]. This process is characterized by the engulfment of part of the cytoplasm inside double-membrane vesicles (autophagosomes). Autophagosomes subsequently fuse with lysosomes to form autophagolysosomes in which the cytoplasmic cargo is degraded. The turnover of most proteins, biological membranes, and whole organelles such as mitochondria and ribosomes is mediated by autophagy [96].

Autophagy represents an extremely refined collector of altered organelles, abnormal protein aggregates, and pathogens, similar to a selective recycling center [97]. The selectivity of the autophagy process is conferred by a growing number of specific cargo receptors such as p62/ SQSTM1 and Nix (Bnip3L) [98]. These adaptor proteins are equipped with both a cargobinding domain, with the capability to recognize and attach directly to molecular tags on organelles. At the same time, these adaptor proteins bind a microtubule-associated protein light chain LC3)-interacting region domain to recruit and bind essential autophagosome membrane proteins. Three molecular complexes mainly regulate the formation of autopha‐ gosomes: the LC3 conjugation system and the regulatory complexes governed by unc51-like kinase-1 and beclin-1. The conjugation complex is composed of different proteins encoded by autophagy-related genes (Atg). The Atg12-Atg5-Atg16L1 complex, along with Atg7, plays an essential role in the conjugation of LC3 to phosphatidylethanolamine, which is required for

The UPS and the lysosomal-autophagy system in skeletal muscle are interconnected [15, 16]. Both these studies identified FOXO3 as a regulator of these two pathways in muscle wasting [Fig. 3]. FOXO3 is a transcriptional regulator of the atrogin-1 and MuRF-1. FOXO3 modulates the expression of autophagy-related genes in mammalian skeletal muscle and C2C12 myo‐ tubes [16]. Masiero et al. [99] found an intriguing characteristic using muscle-specific autoph‐ agy knockout mice, which exhibit fiber atrophy, weakness, and mitochondrial abnormalities. Autophagy-dependent protein degradation seems to be also modulated by tumor necrosis factor (TNF) receptor associated factor 6 and peroxisome proliferator-activated receptor γ coactivator-1α (PGC1α) [100, 101]. Wenz et al. [101] recognized no significant age-related increase in the ratio of LC3-II to LC3-I in MCK-PGC1α mice. Therefore, PGC1α would attenuate the autophagic process probably through increased anti-oxidant defense and

As for cancer cachexia, earlier results obtained on muscles isolated from cachectic animals led us to rule out a substantial role for lysosomes in overall protein degradation [102]. In contrast, an elevation of total lysosome protease activity has been observed in the skeletal muscle and liver of tumor-bearing rats [103]. In addition, increased levels of cathepsin L mRNA have been reported in the skeletal muscle of septic or tumor-bearing rats, whereas cathepsin B gene expression has been shown to be enhanced in muscle biopsy samples obtained from patients

**Figure 3.** Contribution of the proteolytic pathways to muscle atrophy during catabolic conditions. In catabolic condi‐ tions such as denervation, cancer, and fasting, an atrophy program is induced to degrade muscle proteins and organ‐ elles. Proteins can have a double fate, being recognized and removed by the proteasome or docked to the autophagosome. In the latter case the chains of polyubiquitins are interacting with the p62. These proteins have also a domain for the interaction with LC3 therefore bringing the ubiquitinated proteins to the growing autophagosome. Less anabolic stimulation (IGF-I, mechanical loading, amino acids, etc.) reduces the amount of activated Akt, not promoting protein synthesis by activating the mTOR/p70S6K pathway. Lower Akt activity also does not block the nuclear translo‐ cation of FOXO3 to enhance the expression of autophagy-related genes (Bnip3, LC3, Atg12) and Atrogin-1 and the con‐ sequent protein degradation. FOXO: forkhead box O; IGF-I: insulin-like growth factor-I; LC3: microtubule-associated protein light chain; mTOR: mammalian target of rapamycin; p70S6K: 70 kDa ribosomal protein S6 kinase; Ub: ubiqui‐ tin. Data from Sakuma et al. [130]

with lung cancer [71, 104]. Furthermore, a few general observations suggested that autophagy can be activated in the muscle of animals bearing LLC or colon 26 (C26) tumor [105, 106]. More recently, Penna et al. [107] investigated whether autophagy signaling was elevated in muscle using three different models of cancer cachexia. They observed marked increases in the levels of beclin-1, p62/SQSTM1, and LC3B-II (the lipidated form; a reliable marker of autophagosome formation) in muscle in C26-bearing mice. In addition, Penna et al. [107] evaluated autophagic markers in the gastrocnemius muscle of rats bearing Yoshida AH-130 hepatoma or of mice transplanted with LLC. Several autophagic markers were upregulated in the muscle of these two cancer cachexia rodent models, although there was some difference in the adaptive manner. Furthermore, OP den Kamp et al. [70] indicated that the levels of both LC3B-I and - II proteins but not LC3B mRNA were significantly increased in the vastus lateralis muscle of patients with lung cancer cachexia. Esophageal cancer patients also appear to exhibit higher LC3B-II/I ratios and levels of cathepsin B and L expression in muscle [72]. Since they did not detect a significant change of proteasome, calpain, or caspase 3 activity in the muscle of these patients, they considered that the autophagic-lysosomal pathway is the main proteolytic system in the muscle in esophageal cancer cachexia

The functional importance of autophagy in the pathogenesis of lung disease in COPD patients has recently been demonstrated by Chen et al. [108] who described significant increases of autophagy in clinical lung samples taken from COPD patients. LC3B, beclin-1, Atg7, and Atg5 were all upregulated, and autophagosome formation was visualized using electron microsco‐ py. In addition, Ryter et al. [109] have also described increased autophagy in clinical specimens of the lung from patients with COPD. They showed the increased expression and activation of autophagic regulator proteins (i.e., LC3B, beclin-1, Atg5, Atg7) in lung. Similar evidence of increased autophagy was observed in mice subjected to chronic inhalation of cigarette smoke [108] and in lung epithelial cells exposed to aqueous cigarette smoke extracts [110]. Taking these findings together, autophagy seems to be activated in the lungs as a stress response. To date, little research has been completed on the contribution of the autophagy system to protein degradation and loss of skeletal muscle mass in COPD patients. Using muscle biopsy samples obtained from severe COPD patients with marked atrophy [forced expiratory volume in 1 s (FEV1) value of 35±2% of predicted], Plant et al. [111] demonstrated that there was no difference in the levels of beclin-1 and LC3 transcripts in the quadriceps muscle of patients with COPD compared with those in control individuals. On the basis of these results, Plant et al. [111] concluded that autophagy is not activated in muscles of COPD patients. However, they assessed the degree of autophagy by measuring mRNA levels only. More recently, Guo et al. [112] performed a pilot experiment using Western blot and real-time PCR mRNA measure‐ ments to evaluate autophagy-related gene expression of muscle biopsies obtained from cases of severe COPD. These experiments revealed significant increases in the intensity of LC3B-II protein in muscle of COPD patients compared with that in control subjects. In addition, they also observed significant increases in beclin-1 and p62/SQSTM1 protein levels in muscle biopsies of COPD patients indicating the activation of autophagy. More complete elucidation of the functional role of autophagy in muscle of COPD patients remains to be determined, but some research in this field has been undertaken. It is probable that the activation of autophagy in the muscle of COPD patients is modulated by several factors, such as oxidative stress, inflammation, malnutrition, and therapeutic medication, as proposed in an excellent system‐ atic review by Hussain and Sandri [113].

One original study investigated the relationship between CHF and autophagy signaling in skeletal muscle [114]. It was suggested that there is a difference in the manner of autophagic adaptation between soleus (slow-type) and plantaris (fast-type) muscles by using rats with myocardial infarction. In fact, the transcription levels of GABARAPL-1 and Atg7 were increased in the plantaris but not the soleus muscle. However, the expression levels of other autophagic markers (beclin-1 and Atg12) did not change significantly. In addition, an autoph‐ agy-activating marker (LC3B-II/I) also did not increase in both muscles. However, there have been no studies examining the autophagy in muscle in cases of CHF. It remains to be elucidated whether CHF includes autophagic activation in skeletal muscle similar to muscle in cancer cachexia and COPD.

#### **A possible contribution of autophagic signaling to sarcopenia**

patients with lung cancer cachexia. Esophageal cancer patients also appear to exhibit higher LC3B-II/I ratios and levels of cathepsin B and L expression in muscle [72]. Since they did not detect a significant change of proteasome, calpain, or caspase 3 activity in the muscle of these patients, they considered that the autophagic-lysosomal pathway is the main proteolytic

The functional importance of autophagy in the pathogenesis of lung disease in COPD patients has recently been demonstrated by Chen et al. [108] who described significant increases of autophagy in clinical lung samples taken from COPD patients. LC3B, beclin-1, Atg7, and Atg5 were all upregulated, and autophagosome formation was visualized using electron microsco‐ py. In addition, Ryter et al. [109] have also described increased autophagy in clinical specimens of the lung from patients with COPD. They showed the increased expression and activation of autophagic regulator proteins (i.e., LC3B, beclin-1, Atg5, Atg7) in lung. Similar evidence of increased autophagy was observed in mice subjected to chronic inhalation of cigarette smoke [108] and in lung epithelial cells exposed to aqueous cigarette smoke extracts [110]. Taking these findings together, autophagy seems to be activated in the lungs as a stress response. To date, little research has been completed on the contribution of the autophagy system to protein degradation and loss of skeletal muscle mass in COPD patients. Using muscle biopsy samples obtained from severe COPD patients with marked atrophy [forced expiratory volume in 1 s (FEV1) value of 35±2% of predicted], Plant et al. [111] demonstrated that there was no difference in the levels of beclin-1 and LC3 transcripts in the quadriceps muscle of patients with COPD compared with those in control individuals. On the basis of these results, Plant et al. [111] concluded that autophagy is not activated in muscles of COPD patients. However, they assessed the degree of autophagy by measuring mRNA levels only. More recently, Guo et al. [112] performed a pilot experiment using Western blot and real-time PCR mRNA measure‐ ments to evaluate autophagy-related gene expression of muscle biopsies obtained from cases of severe COPD. These experiments revealed significant increases in the intensity of LC3B-II protein in muscle of COPD patients compared with that in control subjects. In addition, they also observed significant increases in beclin-1 and p62/SQSTM1 protein levels in muscle biopsies of COPD patients indicating the activation of autophagy. More complete elucidation of the functional role of autophagy in muscle of COPD patients remains to be determined, but some research in this field has been undertaken. It is probable that the activation of autophagy in the muscle of COPD patients is modulated by several factors, such as oxidative stress, inflammation, malnutrition, and therapeutic medication, as proposed in an excellent system‐

One original study investigated the relationship between CHF and autophagy signaling in skeletal muscle [114]. It was suggested that there is a difference in the manner of autophagic adaptation between soleus (slow-type) and plantaris (fast-type) muscles by using rats with myocardial infarction. In fact, the transcription levels of GABARAPL-1 and Atg7 were increased in the plantaris but not the soleus muscle. However, the expression levels of other autophagic markers (beclin-1 and Atg12) did not change significantly. In addition, an autoph‐ agy-activating marker (LC3B-II/I) also did not increase in both muscles. However, there have been no studies examining the autophagy in muscle in cases of CHF. It remains to be elucidated

system in the muscle in esophageal cancer cachexia

156 Muscle Cell and Tissue

atic review by Hussain and Sandri [113].

Autophagic defect has been described for invertebrates and higher organisms during normal aging. Inefficient autophagy has been attributed a major role in the age-dependent accumu‐ lation of damaged cellular components, such as undegradable lysosome-bound lipofuscin, protein aggregates, and damaged mitochondria [115]. The function of the autophagy/lysosome system of protein degradation seems to decline during aging in the *Drosophila* skeletal muscle [116]. Senescent *Drosophila* muscle exhibits the progressive accumulation of the aggregates of poly-ubiquitin protein. Intriguingly, overexpression of the FOXO upregulates the many autophagy-related genes, preserves the function of the autophagy pathway, and prevents the accumulation of poly-ubiquitin protein aggregates in sarcopenic *Drosophila* muscle [116]. Several investigators reported autophagic changes in aged mammalian skeletal muscle [101, 117, 118]. Compared with those in young male Fischer 344 rats, amounts of beclin-1 were significantly increased in the plantaris muscles of senescent rats [117]. Using Western blot of fractionated homogenates and immunofluorescence microscopy, we recently demonstrated the selective induction of p62/SQSTM1 and beclin-1 but not LC3 in the cytosol of sarcopenic muscle fibers in mice [119]. In addition, we also observed a significant smaller p62/SQSTM1 positive muscle fibers in aged muscle compared to the surrounding p62/SQSTM1-negative fibers [119]. In contrast, aging did not influence the amounts of Atg7 and Atg9 proteins in rat plantaris muscle [117]. More recently, Wohlgemuth et al. [117] clearly showed a marked increase in the amount of LC3 in muscle during aging using analysis of Western blot. However, they failed to detect an aging-related increase of the ratio of LC3-II to LC3-I, a better biochem‐ ical marker of ongoing autophagy. In addition, we failed to detect marked increases in LC3-I and LC3-II (active form) proteins in aged quadriceps muscle [119]. In contrast, a significant increase in the ratio of LC3-II to LC3-I during aging has been demonstrated in the biceps femoris muscle of wild-type mice [101]. None of the studies determining the mRNA expression level of autophagy-linked molecules found a significant increase with age [117, 118]. Not all contributors to autophagy signaling seem to change similarly at both mRNA and protein levels in senescent skeletal muscle. Therefore, sarcopenia may include a partial defect of autophagy signaling, although more exhaustive investigation is needed in this field. Intriguingly, more recent study [120] using biopsy samples of young and aged human volunteers clearly showed the age-dependent autophagic defect such as the decrease in the amount of Atg7 protein and in the ratio of LC3-II/LC3-I protein. Figure 4 summarizes a possible adaptation of autophagylinked molecules (LC3 and p62/SQSTM1) in sarcopenic muscle.

#### **Autophagic signaling in muscular dystrophy**

Inhibition/alteration of autophagy contributes to myofiber degeneration leading to accumu‐ lation of abnormal (dysfunctional) organelles and of unfolded and aggregation-prone proteins [94, 99], which are typical features of several myopathies [121, 122]. Generation of Atg5 and Atg7 muscle-specific knockout mice confirmed the physiological importance of the autophagy system in muscle mass maintenance [99, 123]. The muscle-specific Atg7 knockout mice are characterized by the presence of mitochondrial abnormality, accumulation of polyubiquiti‐

**Figure 4.** The comparison of an autophagy-dependent system between young and sarcopenic muscle. In contrast to young muscle, sarcopenic muscle exhibits abundant p62/SQSTM1 proteins with no activation of LC3, showing appa‐ rent autophagy defects, which cannot destroy the degenerative proteins.

nated proteins, and sarcomere disorganization [99]. In addition, the central role of the autoph‐ agy-lysosome system in muscle homeostasis is highlighted by lysosomal storage diseases (Pompe disease, Danon disease, and X-linked myopathy). These diseases are a group of debilitating muscle disorders characterized by alterations in lysosomal proteins and autopha‐ gosome buildup [124]. Intriguingly, the accumulation of autophagic vacuoles inside myofibers is recognized in all of these myopathies because of defects in their clearance.

Various muscular dystrophies also exhibit the apparent defect of autophagy-dependent signaling. The first evidence of impaired autophagy in these models was provided by studies in mice and patients with mutations in collagen VI [125]. Mutations that inactivate Jumpy, a phosphatase that counteracts the activation of VPS34 for autophagosome formation and reduces autophagy, are associated with centronuclear myopathy [126]. De Palma et al. [127] have described a decreased expression of autophagic regulator proteins (i.e., LC3 II, Atg12, GABARAPL-1, Bnip3) in dystrophin-deficient mdx mice and DMD patients. In addition, starvation and treatment with chloroquine, potent inducers of autophagy, did not activate autophagy-dependent signaling in both tibialis anterior and diaphragm muscles of mdx mice [127]. Furthermore, mdx mice and DMD patients exhibited an unnecessary accumulation of p62/SQSTM1 protein, which was lost after prolonged autophagy induction by a low-protein diet [127]. A similar block in autophagy progression was described in lamin A/C null mice [128]. LGMD2A muscles showed up-regulation of p62/SQSTM1 (2.1-fold) and Bnip3 (3-fold) mRNA and slightly increased LC3-II/LC3-I protein ratio and p62/SQSTM1 [87]. Conversely, laminin-mutated (*dy/dy*) animals displayed excessive levels of autophagy, which is equally detrimental [129]. These findings suggest that the defect of autophagy signaling has a central role in the degenerative symptoms in various types of muscular dystrophy.
