Meanwhile, spermidine coupled with exercise could induce autophagy through activating AMPK-FOXO3a signal pathway with characterization of increased Beclin1 and LC3-II/LC3-I ratio, up-regulated anti-apoptotic Bcl-2, down-regulated pro-apoptotic Bax and caspase-3, as well as activated AMPK and FOXO3a. Therefore, spermidine combined with exercise can execute the prevention or treatment of D-galinduced aging-related skeletal muscle atrophy through enhanced autophagy and reduced apoptosis mediated by AMPK-FOXO3a signal pathway.

INTRODUCTION

Skeletal muscle mass represents roughly 40- 50% of human body weight as the largest tissue and the major amino acid reservoir, and may vary according to physiological and pathological conditions [1]. One of the major tissues affected during aging process is skeletal muscle, which is characteristics of decreased mass and fiber size [2]. The atrophy of skeletal muscle is a highly prevalent and disabling condition as the rapid increase in the number of aged people, but knowledge about cellular signal pathways mediating muscle atrophy is still limited. During the aging process with gradual loss of capability to adapt changing environments, the decreased regeneration capacity and negative protein turnover as well as mitochondrial dysfunction of skeletal muscle are usually considered to be the molecular mechanisms involving in the pathogenesis of aging-related atrophy of skeletal muscle [2].

According to previous studies, the animals subjected to the treatment of D-galactose (D-gal) have shown several aging hallmarks, such as reduced longevity, increased oxidative stress [3], decreased activity of antioxidant enzymes [4], increased mitochondrial DNA mutation [5] and mitochondrial dysfunction [6], which may be also correlated with skeletal muscle atrophy associated with aging. A number of critical signaling pathways correlated with aging-related hallmarks can further promote the aging process. It is worthy of attention that aging-related myopathy usually presents an exacerbation of apoptosis and an abnormal functional status of autophagy including deficient autophagy or dysfunctional autophagic flux due to the accumulation of damaged products and overall decrease of protein degradation process [7]. The signaling pathways of autophagy and apoptosis in mammalian cells are sensitive to intracellular homeostasis and can be either activated or deactivated by exoteric stress, such as toxins, insufficient or excessive intake of nutrients and exercise [8]. With the extension of age, the accumulation of metabolic wastes and damaged cells or organelles may disrupt the effective regulation of autophagy and apoptosis. In turn, a declined or deficient autophagy as well as dysfunctional autophagic flux during aging process could cause more accumulation of dysfunctional organelles and unfolded proteins [9, 10], which stimulates us to focus on exercise and drug intervention strategies such as caloric restriction mimetics and antioxidants for the prevention and/or amelioration of aging-associated skeletal muscle atrophy.

 Regular and appropriate exercise is a common intervention and a well-known stimulus for muscle adaptation and energy metabolism of mitochondria in skeletal muscle, which is often used to attenuate the loss of skeletal muscle in aged population. During the intervention using appropriate exercise, both autophagy and apoptosis could be induced in skeletal muscle to limit tissue damage, restore tissue integrity, terminate inflammatory responses, or induce direct signal pathways for adaptation [11-13]. Currently, the attenuation of basal autophagy with the extension of age is confirmed, and endurance exercise training has been shown to enhance autophagic signaling in aged mice [14]. Moreover, regular exercise has been demonstrated to alleviate apoptosis and DNA fragmentation in skeletal muscle through downregulating apoptotic proteins such as Bax and caspases, and up-regulating apoptotic inhibitors such as Bcl-2. However, the regulatory mechanisms responsible for the activation or the balance of autophagy and apoptosis in aging skeletal muscle during different types of exercise training are still debatable. Spermidine has been identified as a potent and specific inducer of autophagy for expending lifespan due to the hypoacetylation of histone, which can provide protection for aging process of several tissues including heart, brain and skeletal muscle, thereby accomplishing the longevity [15-19]. A recent study has documented that, intracellular concentration of spermidine, with a declining trend during human aging process, is connected with the lifespan of yeast, flies and worms, and human cells due to the deficient autophagy [20].

 Spermidine can not affect the phosphorylation of mTOR and its substrate ribosomal protein S6 kinase, but can induce autophagy through AMP-activated protein kinase (AMPK)-dependent signal pathway [21]. Moreover, spermidine has the function to induce autophagy in most eukaryotes and reduce oxidative stress. Hence, spermidine may decrease endoplasmic reticulum (ER) stress and reduce apoptosis via activating AMPK-dependent autophagy pathway [22]. Excessive apoptosis and deficient autophagy may cause the atrophy of skeletal muscle under pathological events or aging process, and correspondingly may promote cell death and disease progression [2]. Therefore, the functional status of autophagy and apoptosis should be the determinant for maintaining long-term health of skeletal muscle [23]. AMPK has been confirmed to regulate protein metabolism and many cellular processes that are disrupted in skeletal muscle of aging subjects [24]. The activation of AMPK also can slow down the process of senescence by regulating different signal pathways [25], and has been investigated as a therapeutic target in aging-related myopathies [26]. FOXO3a, as a downstream effector of AMPK signaling, is known to be associated with longevity [27]. The up-regulation of FOXO3a can attenuate the expression of atrophy-related proteins in skeletal muscle [28]. Therefore, we hypothesized that long-term regular exercise combined with a specific autophagic inducer spermidine could activate autophagy and inhibit apoptosis to establish this “just right” level for health maintenance of skeletal muscle.

Endurance training could not only delay the process of aging-related muscle loss, but can be benefit for cardiopulmonary function [29], insulin sensitivity [30], and mitochondrial quality control [31], while it could offer less muscle mass than strength development [32]. Although the difference of GWM/ BM ratio between individual intervention group (DS or DE group) and D-gal group did not exhibit a statistical significance, the intervention of spermidine coupled with exercise (DES group) could result in an obvious protection for D-gal-induced damage of skeletal muscle (Table 1). As shown in Figure 1, in order to evaluate whether there are any morphological changes associated with D-gal-induced aging process, we performed masson trichrome staining to characterize the internal structure of skeletal muscle fibers and to examine the morphology of gastrocnemius muscle with or without exercise training and spermidine intervention. The cross-sectional areas of gastrocnemius muscle were qualitatively and quantitatively analyzed under a light microscope. Morphological changes were observed in three randomly selected fields with masson staining. Under normal conditions, myofilaments revealed an orderly and neat arrangement. All nuclei were located at the edge of skeletal muscle fibers, whereas skeletal muscle fibers in D-gal-induced aging group had an obvious damage with loose structure and disordered arrangement. After spermidine and exercise training intervention for 6 weeks, the disordered structure of myofilaments was obviously improved. These results suggest that D-gal-induced aging might accelerate the atrophy of skeletal muscle, and could be alleviated by regular exercise training. 

Meanwhile, histochemistry staining also showed a trace of local collagen fibrils (connective tissue) among skeletal muscle in each group. As one of the important characteristic hallmarks for the aging of skeletal muscle, the fibrosis in gastrocnemius samples of the rats receiving D-gal injection was more severe when compared to the young normal control rats, which was alleviated after daily intervention with spermidine and exercise (Figure 1). At the same time, we also evaluated the cross-sectional area of skeletal muscle fibers between groups. As shown in Figure 1, a corresponding reduction in cross-sectional areas was observed in D-gal-induced aging rats when compared to young normal control rats (P < 0.05). However, a significant increase in cross-sectional areas of skeletal muscle fibers in rats from DE (P < 0.001) and DES (P < 0.01) groups was detected. We have speculated that the loss of skeletal muscle function in rats from D-gal administration group is similar with those from sarcopenia. As one of the most widely used biomarkers for aging cells, the increased β-galactosidase during senescence has been reported and used widely as a biomarker of cellular senescence. As shown in Figure 2, SA-β-gal positive cells with blue staining were rarely observed in Con and DES groups. The D-gal-induced aging was evaluated by relative optical density (ROD) for positive cells with SA-β-gal blue staining. The ROD of positive cells did not reveal statistically significant difference between Con group and DES group. Remarkably, D-gal administration induced a remarkable increase in SA-β-gal staining when compared with Con group (Figure 2). 

 Furthermore, the content of p62 exhibited the highest level in the D-gal-induced aging model group (P < 0.05) than in other groups, suggesting lower autophagic flux in D-galinduced aging skeletal muscle. In contrast, swimming intervention group showed a significantly decreased expression of p62 when compared with D-gal model aging group (P < 0.05). Exercise and spermidine could result in the reduction of excessive apoptosis in D-gal-induced aging skeletal muscle Since autophagic protein levels were increased in exercise and spermidine groups, we next analyzed whether apoptotic signal pathways were affected during exercise and spermidine intervention. In order to investigate the protective effect of spermidine and exercise on aging skeletal muscle cells, D-gal-induced aging rats were pretreated with spermidine and swimming training. Usually, apoptosis occurs mainly in capillary endothelial cells in aging skeletal muscle [33], while we found that the skeletal muscle of the rats from D-galinduced aging group exhibited a significantly higher proportion of apoptotic cells when compared with the normal control group, as shown in the TUNEL analysis (Figure 6A). Moreover, the rate of TUNEL-positive cells was significantly reduced after spermidine and exercise interventions. The population of apoptotic cells induced by D-gal significantly decreased after pretreatment with spermidine. As shown in Figure 6B, D-gal administration caused an obvious increase in apoptosis (52.02% versus 6.77%, P < 0.001) when compared to the normal control group. 

Therefore, in the present study, we applied D-gal-induced aging rat model to simulate aging-related damage of skeletal muscle. D-gal-treated rats showed typical aging symptoms as evidenced by increased senescence-associated β-galactosidase and cell damage, which is consistent with previous demonstration that D-gal can cause the accumulation of reactive oxygen species (ROS) with a final oxidative stress [39]. Although the optimal dose and duration of D-gal administration are required to produce an aging rat model, the effect and toxicity of D-gal (150, 200 and 300 mg/kg) and SPD (1, 5 and 10 mg/kg) at various time durations have been systematically evaluated in our pilot experiments (data not shown). Therefore, the daily concentrations of D-gal and SPD with 6 weeks intervention is optimized to be 200 mg/kg and 5 mg/kg based on our preliminary experiments, which not only can induce the sub-acute aging model, but also can result in less drug toxicity. In the present study, the SA-β-gal staining and TEM assay showed that D-gal treatment could result in the aging of skeletal muscle in rats (Figure 3). Similarly, increased MDA content and reduced SOD activity was observed in serum of D-gal-induced aging rats, indicating that the animal model of mimetic aging by D-gal administration was successfully established. According to the mitochondrial-lysosomal theory, autophagy, or even mitophagy, reveals an obvious decline during aging process in long-lived post-mitotic tissues such as skeletal muscle, thus leading to an accumulation of dysfunctional and enlarged “giant” mitochondria [41].

 The significant enlargement with the increased branching and length of mitochondria reported in aged skeletal muscle [42] is also consistent with our finding (Figure 3), which further supports the mitochondrial-lysosomal theory for the homeostasis and mitochondrial quality control in skeletal muscle. It is well known that mitochondrial fusion is important for the maintenance of mitochondrial function and integrity; therefore, the change in mitochondrial morphology observed here might be an increase in fusion to compensate for impaired mitophagy. The purpose of our study is to evaluate the protective effects of spermidine and exercise on skeletal muscle mass in sub-acute aging model induced by D-gal and to explore corresponding mechanisms. We also investigated the signal pathways for the regulation of autophagic and apoptotic signals in the presence of spermidine and exercise. As a ubiquitin-like protein conjugated to PE, LC3 is involved in cargo recruitment and autophagosome biogenesis [43]. During the formation of autophagosomes, LC3-I is processed to LC3-II by lipidation. Lipidation of LC3 results in an increased LC3- II/LC3-I ratio as a golden marker of autophagy induction in different tissues including skeletal muscle. Thus, the level of LC3-II reveals an increase during the generation of autophagosomes [44, 45]. A decline in autophagic levels has been described in skeletal muscle of aging animals. The qualitative or quantitative monitoring of LC3 is also applied for evaluating autophagic flux via LC3-II/LC3-I ratio coupled with the degradation status of p62. In aged animals, chronic exercise training can significantly reduce DNA fragmentation, cleaved caspase-3 content, Bax level, and Bax/Bcl-2 ratio [46].

 Correspondingly, these data support the notion that regular exercise may be beneficial to skeletal muscle by decreasing DNA fragmentation and promoting an anti-apoptotic environment. Our results have also confirmed that regular exercise combined with spermidine could ameliorate skeletal muscle atrophy and excessive apoptosis in D-gal-induced aging skeletal muscle with deficient autophagy through comprehensively evaluating above indicators associated with autophagy and apoptosis, and achieving the consistent results. There are enough publications as the evidence to support the fact that the activation of autophagy can reduce or inhibit apoptosis, and vice versa [47, 48]. The genetic inhibition of autophagy by knockout or knockdown of ATG genes often leads to apoptotic or necrotic cell death [49- 51], which clearly reveals that the inhibition of autophagy contributes to the activation or prevalence of apoptosis. Bcl-2 is an anti-apoptotic protein acting as an indirect negative regulator of autophagy by binding and preventing Beclin1 from inducing autophagy [52]. Conversely, the Bcl-2 family member, Bax, is pro-apoptotic, and also can suppress autophagy through caspase-mediated cleavage of Beclin1, which prevents Beclin1 from initiating autophagy [53]. The interaction between the anti-apoptotic protein Bcl-2 and the autophagy protein Beclin1 is essential to regulate the switch between autophagy and apoptosis. Bcl2 binds to Beclin1 and segregates Beclin1 away from class III PI3K, thereby leading to an inhibition of autophagic response. This suggests that apoptosis activation results in autophagy suppression in a context-dependent manner. In contrast, the activation of autophagy could inhibit apoptosis by exercise or autophagic activator [54, 55].