Cell senescence was first described more than 50 years ago (Hayflick and Moorhead, 1961). Hayflick and Moorhead observed that normal human fibroblasts have a finite proliferative capacity in culture. They termed the cell cycle arrest at the exhaustion of this capacity “replicative senescence”; the word “senescence” is derived from the Latin word senex, meaning “old.” Subsequently, it was recognized that cellular senescence arises from telomere shortening, which is associated with chromosomal instability; accordingly, senescence was viewed as a tumor suppressor mechanism (Serrano et al., 1997; Lowe et al., 2004). Later studies revealed the physiological importance of cellular senescence beyond its tumor suppressor functions in processes such as wound healing (Jun and Lau, 2010; Demaria et al., 2014), embryonic development (Storer et al., 2013), and tissue repair and regeneration (Krizhanovsky et al., 2008; Muñoz-Espín and Serrano, 2014). On the other hand, cellular senescence also contributes to organismal aging and related diseases (Song et al., 2020). Senescent cells are metabolically active and secrete a variety of factors: inflammatory cytokines and chemokines collectively termed the senescence-associated secretory phenotype (SASP), which can induce chronic inflammation (Coppé et al., 2008; Gorgoulis et al., 2019, Birch 2020). A recent study showed that senescent cells can induce other non-senescent cells to undergo senescence by juxtacrine or paracrine effects, including the SASP, a phenomenon known as secondary senescence (Admasu 2021). In addition, cellular senescence can occur in post-mitotic cells, such as neurons and muscle cells (Zglinicki 2021). These discoveries suggested that senescent cells can have a widespread impact on various tissue and pathologies. Therefore, senolytic agents, which eliminate accumulated senescent cells, have attracted a great deal of attention as potential treatments for age-related diseases, which are frequently associated with chronic inflammation (Bussian et al., 2018; Xu et al., 2018; Hickson et al., 2019; Justice et al., 2019). Multiple aspects of cellular senescence are involved in skeletal muscle physiology and disease (Kudryashova et al., 2012; Saito et al., 2020; Sugihara et al., 2020), raising the question of whether fundamentally different mechanisms underlie the beneficial and detrimental effects of senescence in skeletal muscle. This review focuses on how senescent cells are involved in skeletal muscle physiology and pathology and how senolytics or pro-senescent therapies (including exercise) could be used to treat diseases of skeletal muscle.

Cellular senescence is an adaptive response induced by multiple physiological and pathological stresses that results in irreversible cell cycle arrest (Muñoz-Espín and Serrano, 2014; van Deursen, 2014; Gorgoulis et al., 2019; Borghesan et al., 2020). Senescence provides a defense mechanism that limits tumorigenesis to maintain tissue homeostasis and allow tissue remodeling via removal of damaged senescent cells (Muñoz-Espín and Serrano, 2014). However, permanent accumulation of senescent cells is a major cause of age-related disease and chronic inflammation. Leonard Hayflick and Paul Moorhead found that human fibroblasts have a finite in vitro proliferative capacity (Hayflick and Moorhead, 1961), and subsequent work showed that replicative senescence is caused by the shortening of the telomeres at the ends of chromosomes, which triggers the DNA damage response (DDR) and causes cell cycle arrest (Sedelnikova et al., 2004; Campisi and d'Adda di Fagagna, 2007). Similarly, cellular senescence can arise due to DNA damage from various stresses, including radiation (Le et al., 2010), oncogene activation (Di Micco et al., 2006), high levels of reactive oxygen species (ROS) (von Zglinicki, 2002), mitochondrial dysfunction (Chapman et al., 2019), mechanical stress (Xing et al., 2010), protein aggregation (Johmura et al., 2021), failure of protein removal due to diminished autophagy (García-Prat et al., 2016), and inflammatory cytokines and growth factors (Beyne-Rauzy et al., 2004; Hubackova et al., 2012). These stresses activate DDR components including ATR, ATM, and p53, which promote activation of cyclin-dependent kinase (CDK) inhibitors such as p16^INK4a (CDKN2A) and p21^WAF1/Cip1(CDKN1A) (Gorgoulis et al., 2019). Defects in ribosome biogenesis and derepression of retrotransposons also contribute to cell cycle arrest in senescent cells (Lessard et al., 2018; De Cecco et al., 2019). To date, however, no specific markers of the cell cycle have been identified in senescent cells. For example, p16^INK4a is also expressed in non-senescent cells (Sharpless and Sherr, 2015) and is not expressed in all senescent cells (Hernandez-Segura et al., 2017). Furthermore, senescence induced by E2F3 activation or c-Myc inhibition is DDR-independent and involves p16^INK4a and p19ARF (Lazzerini Denchi et al., 2005). Another DDR-independent inducer of cellular senescence is BRAF (V600E), which activates senescence through a metabolic mechanism involving upregulation of mitochondrial pyruvate dehydrogenase (Kaplon et al., 2013; van Deursen, 2014). Senescent cells exhibit characteristic morphological and physiological changes associated with this condition. In other words, senescent cells become hypertrophied and flattened in vitro, and the nuclear envelope is incomplete due to reduced expression of lamin B1 (González-Gualda et al., 2021). Accumulation of senescence-associated β-galactosidase (SA-β-gal) due to changes in lysosomal activity is another characteristic of cellular senescence (Gorgoulis et al., 2019). Chromatin rearrangement, especially the formation of senescence-related heterochromatin foci (SAHFs), is a frequently observed biomarker in oncogene-induced senescent (OIS) cells. SAHFs contain histone H3 methylated on lysine 9 (H3K9Me) (Braig et al., 2005), heterochromatin protein 1 (HP1), and histone H2A variant H2AX phosphorylated on Ser139 (γH2AX), and thus can be used as indicators of DNA damage to assess senescence (Dungan et al., 2020). To date, however, specific and sensitive markers of senescent cells have not been identified. Consequently, combinations of biomarkers, such as nuclear (p16^INK4a, p21^WAF1/Cip1, Ki67, γH2AX), cytoplasmic (SA-β-gal), SASP, context, and cell-type-specific markers, are generally used to define the presence of senescence (Coppé et al., 2008; Childs et al., 2015; Gorgoulis et al., 2019).

Senescent cells limit their own proliferation but remain metabolically active, secreting a variety of factors: inflammatory cytokines such as IL-6, IL-8, and TNF-⍺; chemokines; growth factors such as TGFβ; matrix metalloproteinases (MMPs); and micro-RNAs. Collectively, these secreted factors are referred to as the SASP (Coppé et al., 2008). The SASP is considered a hallmark of cellular senescence, and some of the secreted factors exert various autocrine/paracrine effects on the microenvironment of surrounding tissues. The SASP has both beneficial and detrimental consequences, depending on the context. For example, the SASP recruits immune cells to initiate tissue repair through removal of damaged cells (Krizhanovsky et al., 2008; Chikenji et al., 2019) but is also associated with angiogenesis and ECM remodeling, which may promote tumor cell progression (Gonzalez-Meljem et al., 2018; Levi et al., 2020). Although the SASP regulates beneficial effects such as developmental senescence (Muñoz-Espín et al., 2013; Storer et al., 2013) and wound healing (Demaria et al., 2014), it also contributes to the pathology of chronic inflammation (Franceschi and Campisi, 2014). The composition of the SASP varies depending on senescence trigger and cell type. For example, mitochondria dysfunction causes a distinct senescence response, termed mitochondrial dysfunction–associated senescence (MiDAS). MiDAS is associated with lower NAD+/NADH ratios, which both cause growth arrest and prevent the IL-1–associated SASP through AMPK-mediated p53 activation (Wiley et al., 2016). A recent analysis demonstrated that soluble SASP factors, including exosomes, differ markedly among different senescence triggers and distinct cell types, and also identified common SASP factors representing the “core SASP” (Basisty et al., 2020). The authors also found that several SASP factors, including growth/differentiation factor 15 (GDF15), stanniocalcin 1 (STC1), and serine protease inhibitors (SERPINs), correlated with age in plasma from a human cohort (Basisty et al., 2020). The composition and strength of the SASP are dynamic, changing at intervals after senescence induction (Hernandez-Segura et al., 2017). The dynamic and complex nature of the SASP is likely critical to the diverse biological functions associated with senescence.

Several studies have reported a relationship between senescence and muscle diseases, including aging. In skeletal muscle tissue of aged mice, the mRNA expression of Cdkn2a, Cdkn1a, Trp53, Gadd45a, Il6, Serpine1, Mmp1, and Mmp3 is elevated, and muscle mass and muscle functions are reduced (Edwards et al., 2007; Baker et al., 2016; Solovyeva et al., 2021). In humans, skeletal muscle tissue from older women (65–71 years old) expresses higher levels of CDKN1A than skeletal tissue from young women (20–29 years old) (Welle et al., 2004). Several potential mechanisms of muscle cell senescence–mediated, aging-associated muscle loss (sarcopenia) were proposed based on an in vitro study using C2C12 cells, a murine myoblast cell line (Alcalde-Estévez et al., 2020; Mijares et al., 2021; Moustogiannis et al., 2021). These studies revealed that elevated intracellular Ca²⁺ concentration or endothelin-1 receptor in C2C12 cells induces cellular senescence, and senescent C2C12 cells increase the expression of apoptotic, atrophic, and inflammatory factors, which might in turn induce sarcopenia mediated by muscle cell senescence (Alcalde-Estévez et al., 2020; Mijares et al., 2021; Moustogiannis et al., 2021). On the other hand, recent in vivo analysis of human skeletal muscle revealed that there is no difference in the abundance of γH2AX-positive myonuclei between young individuals (21–30 years old) and old individuals (70–86 years old) (Dungan et al., 2020). Interestingly, young obese individuals (21–24 years old, BMI: 34–46) have higher γH2AX expression in myonuclei than young lean individuals (21–24 years old, BMI: 20–25) (Dungan et al., 2020). These studies suggested that aging and obesity, which are related to the impairment of muscle function, regenerative capacity, and muscle volume, tend to increase the abundance of senescent muscle cells; however, this idea remains still controversial.

Immunosenescence is an age-related process of immune dysfunction that contributes to morbidity and mortality (Kennedy et al., 2014; Ventura et al., 2017; Duggal et al., 2019). Among several biomarkers for immunosenescence that have been reported, p16^INK4a is a reliable marker for senescence in T cells, B cells, and macrophages (Liu et al., 2009; Cudejko et al., 2011; Liu et al., 2011; Liu et al., 2019). In T cells, expression of the proliferation marker Ki-67 and activated T cell marker are observed with aging, and depletion of p16^INK4a attenuates these features of senescence (Liu et al., 2009). The authors of the same study showed that depletion of p16^INK4a in B cells attenuates senescence hallmarks that arise with aging (Liu et al., 2009). Bone marrow–derived macrophages isolated from p16^INK4a-deficient mice downregulate genes associated with inflammatory M1 macrophages and increase expression of genes associated with M2 macrophages (Cudejko et al., 2011). The expression level of the p16^INK4a in macrophages does not affect M1/M2 polarization, although p16^INK4a-high macrophages have higher phagocytic activity than p16^INK4a-low macrophages (Liu et al., 2019). Skeletal muscle regeneration is regulated not only by MuSCs and FAPs but also by immune cells; hence, immunosenescence must affect muscle regeneration, although evidence of a connection between the two phenomena remains limited. Recently, a study using Vav-iCre(+/−); Ercc1 (−/fl) mice showed that immune cell–specific induction of senescence decreased muscle strength and impaired regeneration after muscle injury (Yousefzadeh et al., 2021). Furthermore, immune cell–specific senescence induction increased the number of infiltrating macrophages and decreased the ratio of M2 to M1 macrophages after muscle injury (Yousefzadeh et al., 2021). On the other hand, the Newcastle 85 + Study, a prospective, population-based study of very old adults living in the Newcastle and Tyneside regions, United Kingdom, showed that immunosenescence profiles were not associated with muscle function and sarcopenia risk (Granic et al., 2020). Although it remains controversial whether immunosenescence affects muscle inflammation and regeneration, a deeper understanding of the mechanistic regulation of immunosenescence and muscle regeneration could help promote the progress of senotherapy for muscle aging and disease.

Proper regulation of two types of senescent cells, detrimental-senescent cells and beneficial-senescent cells, could exert beneficial effects on muscle regeneration. Several previous studies showed that senolytics restore muscle loss and inflammation in aged mice. Upon genetic elimination of senescent cells using INK-ATTAC mice, mRNA expression of p16, p21, p19, and SASP genes including Pai1, Il6, Mmp6, and Mmp13 were reduced in aged C57BL/6 mice and BubR1H/H mice, a model of early aging (Baker et al., 2011, 2016). Pharmacological elimination of senescent cells by dasatinib/quercetin treatment also decreased muscle strength and function in aged C57BL/6 mice, which develop age-related muscle loss and inflammation in skeletal muscle (Xu et al., 2018). Another senolytic drug, ABT263, also inhibits production of the SASP factors IL-6, TGF-β, and IL-1β in a rat model of Duchenne muscular dystrophy (Sugihara et al., 2020). Treatment with the NAD + precursor nicotinamide riboside (NR) prevents MuSC senescence and decreases production of SASP factors in aged mice as well as in mdx mice, a model of muscular dystrophy (Zhang et al., 2016). On the other hand, other studies suggested that pro-senescence therapy can promote muscle regeneration. The combination of exercise and AICAR, a cell-permeable AMPK activator, promotes muscle regeneration by inducing FAP senescence in chronic inflammatory myopathy model mice (Wang et al., 2003; Saito et al., 2020). Another study found that transplantation of functional senescent mesenchymal stromal cells (MSC) treated with placenta extract promoted muscle regeneration in chronic inflammatory myopathy model mice (Figure 2) (Chikenji et al., 2019). In mechanistic terms, transplantation of functional senescent MSC, which have phenotypes distinct from those of cells induced to senesce by continuous cultivation, promotes FAP senescence followed by phagocytic cell recruitment and MuSC proliferation (Figure 2) (Chikenji et al., 2019). In cardiac muscle, fibroblast senescence plays important roles in heart regeneration (Feng et al., 2019; Sarig et al., 2019). One study found that treatment with CCN1, a matricellular protein, induced PDGFRα+ fibroblast senescence, decreased cardiac fibrosis, and triggered the expression of SASP factors including IL-1a and IL-6, thereby promoting cardiomyocyte proliferation and heart regeneration (Feng et al., 2019). Interestingly, the senolytic drug ABT263 decreases cardiomyocyte proliferation and increases proliferation of PDGFRα+ fibroblasts, thus promoting heart fibrosis (Feng et al., 2019). Another study found that treatment with the extracellular matrix molecule agrin induced transient senescence in vimentin + fibroblasts and promoted heart regeneration after myocardial infarction (Sarig et al., 2019). Together, the results of these studies suggest that pro-senescent therapy, especially targeting fibroblasts or mesenchymal stromal cells, represents a novel strategy for regulating muscle inflammation and regeneration (Figure 2).

Understanding the mechanism of cellular senescence–mediated tissue regeneration and degeneration is essential for maintaining healthy skeletal muscle. Mounting evidence supports the hypothesis that cellular senescence in skeletal muscle plays diverse roles in muscle regeneration and degeneration. In this review, we described the basic mechanisms of cellular senescence and the beneficial and detrimental effects of senescence on muscle regeneration and degeneration (Figure 4). Unfortunately, there are no distinguishing features of beneficial and detrimental senescent cells; however, several pieces of evidence suggested that a transient increase in the proportion of senescent cells exerts beneficial effects, whereas prolonged accumulation of senescent cells exerts detrimental effects. In regard to PDGFRα+ mesenchymal cells in skeletal muscle and cardiac muscle, induction of senescence is likely to have a positive effect on muscle regeneration by promoting the proliferation of parenchymal cells and inhibiting fibrosis by activating phagocytic cells.

A deeper understanding of the relationship between cellular senescence and muscle physiology and pathology will lead to advances in research on skeletal muscle. However, there are still some outstanding questions. First, what are the molecular and cellular differences between beneficial senescence and detrimental senescence? Second, which cell types, such as MuSCs and FAPs, frequently enter senescent states during acute injury, chronic muscle disease, and aging? Third, what kind of SASP factors contribute to muscle regeneration or chronic inflammation. Fourth, can senotherapy attenuate chronic muscle inflammation and promote muscle regeneration? In addition, which type of senotherapy (e.g., senolytics, senomorphics, or pro-senescence therapies) can exert positive effects on muscle health? Fifth, can exercise-mediated senescence regulation provide a therapeutic effect against chronic muscle diseases or aging? If so, how can we optimize the intensity and frequency of exercise to regulate senescence in each cell type? Addressing these questions will lead to the development of therapies that target cellular senescence.

Finally, many senotherapy drugs have been developed; some have proceeded to clinical trials, and even newer senotherapy drugs continue to be reported today. Undoubtedly, senotherapies, including not only pharmacological therapies, but also cell-based therapies, exercise therapies, and a combination of these therapeutic approaches could provide a beneficial outcome in patients with chronic inflammatory muscular disease and aging-related muscle dysfunction. Importantly, an in-depth understanding of the complex roles of senescent cells, which remain largely unknown, is essential for the development of truly effective senescence-targeting therapies.