Molecular and Cellular Determinants of Skeletal Muscle Atrophy and Hypertrophy

See allHide authors and affiliations

Science's STKE  03 Aug 2004:
Vol. 2004, Issue 244, pp. re11
DOI: 10.1126/stke.2442004re11


The maintenance of adult skeletal muscle mass is ensured by physical exercise. Accordingly, physiological and pathological situations characterized by either impaired motor neuron activity, reduced gravity (microgravity during space flights), or reduced physical activity result in loss of muscle mass. Furthermore, a plethora of clinical conditions, including cancer, sepsis, diabetes, and AIDS, are associated with varying degrees of muscle atrophy. The cellular and molecular pathways responsible for maintaining the skeletal muscle mass are not well defined. Nonetheless, studies aimed at the understanding of the mechanisms underlying either muscular atrophy or hypertrophy have begun to identify the physiological determinants and clarify the molecular pathways responsible for the maintenance of muscle mass.


Two opposing phenomena—muscle growth and muscle atrophy—are mechanistically linked, in that either the activity or inactivity of a common set of molecules controlling a few cellular pathways determines whether the skeletal muscle tissue will respond to defined stimuli with increased protein synthesis and stimulation of cell growth (muscle hypertrophy) or with increased protein breakdown and reduced cell proliferation (muscle atrophy). Among these molecules, insulin-like growth factor–1 (IGF-1) occupies a nodal point.

The IGF-1 Pathway

Pertinent to the discussion of the mechanisms regulating both muscle atrophy and muscle hypertrophy is the description of pathways regulated by IGF-1 [Fig. 1; for a detailed review of the IGF-1 pathway in skeletal muscle, see (1)]. IGF-1 is a secreted growth factor that regulates several cellular biochemical pathways after binding to its membrane receptor, the IGF-1 receptor (IGFR). Upon binding of IGF-1, the IGFR, which is a receptor tyrosine kinase, becomes phosphorylated and recruits the insulin receptor substrate 1 (IRS1), leading to the activation of the lipid kinase phosphatidylinositol 3-kinase (PI3K). PI3K catalyzes the transfer of a phosphate group to the membrane-bound phosphatidylinositol 4,5-bisphosphate (PIP2). Phosphorylation of PIP2 generates phosphatidylinositol 3,4,5-trisphosphate (PIP3), which recruits two additional kinases, Akt1 and phosphoinositide-dependent protein kinase-1 (PDK-1) (2). Atk1 is bound, phosphorylated, and activated by PDK-1 (3). Once Akt1 is activated, it initiates a cascade of phosphorylation events targeting mammalian target of rapamycin (mTOR) (4)—which in turn phosphorylates the p70 S6 kinase (p70S6K)—and glycogen synthase kinase 3β (GSK-3β) (5). Phosphorylated IRS1 also stimulates the Ras-Raf-MEK-ERK pathway, a mitogen-activated protein kinase (MAPK) pathway. Activation of this pathway may actually prevent hypertrophy because Ras-Raf-MEK-ERK inactivation, rather than its activation, appears to characterize skeletal muscle hypertrophy (6). Phosphorylation of mTOR represses, through an adaptor protein known as Raptor, eukaryotic initiation factor 4E (eIF-4E)–binding protein 1 [4EBP1, also known as phosphorylated heat- and acid-stable protein (PHAS-1)], an inhibitor of eIF-4E (7). Thus, mTOR-mediated inhibition of 4EBP1 results in activation of eIF-4E and subsequent increased protein synthesis. An additional level of regulation on mTOR is provided by glucose and amino acids, which influence the interaction between Raptor and mTOR (8). Therefore, mTOR phosphorylation activates both eIF-4E and p70S6K, two positive regulators of protein synthesis. Similarly, the eukaryotic translation initiation factor 2B (eIF-2B) is activated through Akt-mediated phosphorylation and inactivation of GSK-3β. In summary, activation of the IGF-1 pathway leads to a series of phosphorylation events culminating in the activation of molecules that regulate protein synthesis and that are likely involved in the genesis of muscle hypertrophy.

Fig. 1.

The IGF-1 pathway is involved in muscle atrophy and hypertrophy. Upon IGF-1 binding, the IGFR is phosphorylated. The phosphorylated receptor recruits and phosphorylates the IRS1, which activates PI3K. PI3K transfers a phosphate group to the membrane-bound PIP2 to generate PIP3. PIP3 serves as a nucleation site for Akt1 and PDK-1. PDK-1 phosphorylates and activates Akt1, which, in turn, catalyzes the transfer of phosphate groups to several substrates. Akt-mediated phosphorylation of the FOXO transcription factors promotes their cytoplasmic retention and functional inactivation through interaction with the 14-3-3 proteins. Because the FOXO proteins stimulate the transcription of the atrophy-promoting factor MAFbx, FOXO inactivation prevents muscle atrophy. Akt phosphorylates and inhibits GSK-3β, which activates the eukaryotic translation factor eIF-2B and increases protein synthesis. The mTOR is also a phosphorylation substrate for Akt. With the assistance of the interacting protein Raptor, phosphorylated mTOR promotes phosphorylation and inhibition of 4EBP1. Also, mTOR promotes protein synthesis by relieving 4EBP1-mediated inhibition of eIF-4B. When phosphorylated by mTOR, the ribosomal p70S6K becomes activated and increases protein synthesis. Green arrows indicate phosphorylation events that activate the substrate, whereas red arrows point at phosphorylations that result in substrate inactivation.

Molecular Determinants of Muscle Atrophy

Skeletal muscle atrophy is characterized by a decrease in the size of preexisting muscle fibers and is observed in several physiological and pathological settings. Atrophy has been interpreted as the result of a passive adaptation of the muscle to a lack of electrical or mechanical stimuli. Recently, this view has been challenged by several findings indicating that the establishment of an active transcriptional program is necessary for the induction of muscle atrophy.

The ubiquitin protein ligases MuRF1 and MAFbx

Using differential display and complementary DNA (cDNA) microarray approaches, the messenger RNAs (mRNAs) obtained from muscles of either rats or mice in which atrophy was induced by immobilization, denervation, hind-limb suspension, or food deprivation were compared with controls, and the expression of two genes was consistently found to be increased. The two genes encode muscle RING finger 1 (MuRF1) (9) and muscle atrophy F-box (MAFbx; also known as atrogin-1) (9, 10), two ubiquitin protein ligases expressed solely in skeletal and cardiac muscles. Ubiquitin ligases are enzymes involved in catalyzing the ubiquitination of proteins fated to be degraded by the proteasome. During atrophy, there is extensive muscle remodeling characterized by increased proteolysis, and MuRF1 and MAFbx might mediate such a process. A role for the ubiquitination-mediated proteolysis of muscle proteins by MuRF1 is suggested by experiments in which MuRF1 interacts with titin, a giant myofibrillar protein located at the M line of the sarcomere and mediates titin's degradation (11).

An active role for both MuRF1 and MAFbx in mediating muscle atrophy can be inferred from experiments conducted with mice in which both alleles for either MAFbx or MuRF1 have been deleted. MAFbx−/− and MuRF1−/− animals are partially refractory to denervation-induced atrophy (9). It should be possible, in principle, to cross the MAFbx−/− and MuRF1−/− animals to obtain double-knockout MAFbx−/−:MuRF1−/− and to determine whether these animals are viable and whether MAFbx and MuRF1 act cooperatively to mediate atrophy. A complementary approach for evaluating the functional role of these two ubiquitin ligases could be pursued by generating mice that overexpress either MAFbx or MuRF1, because they may develop spontaneous muscle atrophy, have an exacerbated response to atrophic stimuli, or fail to respond to hypertrophic stimuli. Finally, because both MAFbx and MuRF1 are expressed in the heart, it will be important to determine their role in pathological conditions characterized by cardiac remodeling (for example, heart failure, acute myocardial infarction, and cardiomyopathies that accompany several muscular dystrophies). It appears that neither MuRF1 nor MAFbx plays a role in physiological muscle homeostasis, because both MuRF1- and MAFbx-knockout animals have normal skeletal muscles (9).

Nonetheless, the PI3K-Akt-mTOR and ubiquitin-protein ligase pathways seem to be linked, in that IGF-1-dependent activation of the PI3K-Akt-mTOR pathway blocks accumulation of MAFbx mRNA and prevents muscle proteolysis in a cellular model of dexamethasone (DEX)-induced atrophy (12). From a mechanistic and therapeutic point of view, several models of atrophy—including denervation, immobilization, and atrophies accompanying chemically induced diabetes, treatment with dexamethasone or interleukin-1 (IL-1; a cachexia-inducing cytokine), and experimental sepsis—all result in increased transcription, mRNA stabilization, or both, of MuRF1 and MAFbx (9, 13, 14). This suggests the possibility that a common set of transcriptional regulators are stimulated by diverse atrophic stimuli and that pharmacological modulators targeting these regulators may be effective in preventing atrophy observed in disparate pathological conditions.

The FOXO transcription factors and muscle atrophy

Two studies have advanced our knowledge of the molecular modulators of muscle atrophy (15, 16). In these studies, the FOXO family of transcription factors was reported to regulate MAFbx and MuRF1 transcription and to influence muscle atrophy. Akt negatively regulates FOXO transcription factors by phosphorylating them and promoting their nuclear export into the cytoplasm, where they are retained through association with the 14-3-3 proteins (17). Muscle atrophy is mimicked in cultured myotubes—that is, terminally differentiated, postmitotic syncytial cells forming the muscle fiber—by exposing them to either DEX or starvation (removal of growth factors, glucose, and amino acids for 6 hours). In these situations, transcription of both MAFbx and MuRF1 is increased. Concomitantly, phosphorylation of Akt and of FOXO1 and FOXO3 is reduced. Animals with denervation atrophy treated with intramuscular injection of IGF-1, which promotes Akt phosphorylation, do not increase expression of MAFbx and MuRF1 and, most important, are spared from muscle loss. FOXO3 dephosphorylation allows transcriptional activation of the MAFbx promoter and consequently increases MAFbx mRNA production. Through Akt-mediated phosphorylation of FOXO1, 3, and 4, IGF-1 counteracts the ability of DEX to activate MAFbx and MuRF1 transcription. When cultured myotubes exposed to DEX are infected with an adenovirus expressing a constitutively active form of FOXO1, IGF-1 becomes incapable of blocking the increase in MAFbx and MuRF1. These findings indicate that the effects of activating the IGF-1-PI3K-Akt pathway that prevents atrophy operate through the FOXO proteins. Nonetheless, it is likely that additional targets besides MAFbx are affected by the FOXO proteins. In fact, although overexpression of MAFbx alone does not cause muscle atrophy, muscle fibers expressing a constitutively active form of FOXO3 display a much reduced diameter (16). Furthermore, in the same study, FOXO3 overexpression resulted in increased MAFbx transcription, whereas in another study (15), MAFbx expression was not affected by FOXO1. These findings raise the possibility that FOXO3 and FOXO1 may regulate different targets. Alternatively, the differences intrinsic to the two experimental settings may be responsible for the different results. Indeed, reduced expression of FOXO1 achieved with RNA interference reduced activation of the MAFbx promoter (16). The identification of additional FOXO targets is expected to shed light on the molecules and mechanisms regulating muscle atrophy.

Another transcription factor, nuclear factor κB (NF-κB), mediates the effects of tumor necrosis factor–α (TNF-α) and interferon–γ (IFN-γ) (18) in promoting the muscle wasting that accompanies several pathological conditions (19). However, forced activation of the NF-κB pathway does not increase MAFbx expression (16), suggesting that muscle atrophy can be elicited through processes that do not involve FOXO proteins, for example, NF-κB-mediated degradation of the master regulatory protein MyoD (18). The possibility exists that different mediators of muscle atrophy may act at distinct stages and may indeed cross-talk to ensure initiation and completion of the atrophic program.

Exercise and Muscle Growth

Postnatal skeletal muscle growth in response to chronic physical exercise is characterized by cell hypertrophy. Skeletal muscle hypertrophy is defined as an increase in fiber diameter without an apparent increase in the number of muscle fibers, accompanied by enhanced protein synthesis and augmented contractile force. Microtraumas resulting from chronic exercise also lead to satellite cell activation and proliferation, which contribute to the increased muscle mass that ensues from extensive physical activity. The molecular determinants of muscle growth are beginning to be characterized.

IGF-1 in muscle growth

Increased muscle loading, such as that resulting from chronic exercise, results in augmented expression of the gene encoding IGF-1 in both animal models (20) and humans (21). More specifically, two IGF-1 isoforms, IGF-1E and mechanogrowth factor (MGF), seem to be selectively expressed in skeletal muscle and to be regulated by mechanical load (22). These observations, along with the findings that the circulating IGF-1 produced by the liver is not necessary for postnatal body growth (23), suggest that specific isoforms of IGF-1 act in an autocrine or paracrine manner to regulate skeletal muscle biology. Availability of IGF-1 for muscle IGFR is controlled by the combination of IGF-1, IGFR, and IGF-1 binding proteins (IGFBPs). Binding of IGF-1 to certain IGFBPs results in opposing biological effects. For instance, binding of IGF-1 to IGFBP-4 appears to inhibit cell proliferation and differentiation, whereas binding to IGFBP-5 has a dual effect (that is, either inhibition or stimulation of IGF-1 effects), depending on the culture conditions (24, 25). The analysis of the physiological contribution of the individual IGFBPs in the animal has been hampered by the redundancy in the IGFBP system (26). Nonetheless, muscle overloading is associated with increased abundance of IGFBP-4 and decreased abundance of IGFBP-5, suggesting that these two isoforms may be involved in regulating muscle adaptation to mechanical forces (27). Thus, whether IGF-1 acts as an extracellular cue in muscle biology depends on a finely tuned and regulated series of events leading to either the stimulation or the inactivity of the IGF-1–IGFR axis.

Animals overexpressing IGF-1 under the control of muscle-specific regulatory elements display muscle hypertrophy and increased muscle regeneration during senescence (28, 29). These findings, along with the observation that IGF-1 overexpression improves the phenotype of the mdx mouse model for Duchenne muscular dystrophy, suggest that IGF-1 may positively influence satellite cell proliferation, activation, or both (30). Consistent with the existence of a linear pathway proceeding from IGF-1 to Akt, deliberate activation of the Akt-mTOR pathway can stimulate fiber hypertrophy both in control and denervated muscles (31). Together, these results suggest a potential therapeutic value of manipulating the IGF-1–Akt-mTOR pathway in the treatment of muscle atrophy. In contrast, acute physical exercise is associated with increased levels of total and phosphorylated Akt, increased phosphorylation and reduced activity of GSK-3β, and reduced levels of phosphorylated β-catenin (32). Thus, the Akt to GSK-3β pathway may also be involved in muscle hypertrophy and in the initial phases of muscle activity that contribute to hypertrophy.

Although there is little doubt that IGF-1 is involved in mediating experimental muscle hypertrophy, additional molecules and stimuli may be required to assist IGF-1 in this process. Transgenic mice in which IGF-1 expression is driven by the muscle-specific regulatory regions of the gene encoding myosin light chain 1/3 (MLC1/3) display pronounced muscle hypertrophy (33). Expression of the endogenous MLC1/3 gene and activation of the IGF-1 transgene occurred early during development. Nonetheless, muscle hypertrophy in MLC1/3–IGF-1 mice first becomes detectable only 10 days after birth (33). It is possible that moderate physical exercise may be required for the overexpressed IGF-1 to induce hypertrophy or that molecules expressed or activated after birth may be necessary for the hypertrophic effects of IGF-1. Alternatively, the rapid postnatal muscle growth in both the transgenic and the wild-type animals may mask the effects of IGF-1, which become evident in transgenic animals only after muscle growth slows in the control animals. The physiological role of the Akt-mTOR pathway in ensuring muscle trophism also remains to be fully clarified, because pharmacological treatment with the mTOR-specific inhibitor rapamycin fails to induce muscle atrophy in control animals (31).


Myostatin is a transforming growth factor–β (TGF-β) family member expressed in the myotome compartment of developing somites and in the skeletal muscles of adult animals (34). Naturally occurring mutations in the myostatin [Mstn, also known as growth and differentiation factor 8 (GDF-8)] gene are responsible for the "double-muscling" phenotype, which is characterized by a dramatic increase in muscle mass of certain breeds of cattle (35). Mstn-null mice show an increase in muscle mass due to muscle hyperplasia (increased number of muscle fibers) and hypertrophy (increased muscle fiber diameter) (34). Recently, a child with muscle hypertrophy was found to have a loss-of-function mutation in the Mstn gene (36), although whether the child's muscles have an increased number of fibers or a normal number of enlarged myofibers remains to be established. These findings indicate that inactivation of myostatin has similar effects in humans, mice, and cattle.

After an enzymatic cleavage, the active myostatin peptide interacts with the membrane-bound type IIB activin receptor, initiating a signaling cascade (likely mediated by the Smad transcription factors) that ultimately blocks muscle differentiation, possibly by inhibiting the master regulatory genes MyoD (37) and Pax-3 (38), which encode proteins controlling skeletal myogenesis (39, 40). These findings raise the possibility that pharmacological inhibition of myostatin may be of therapeutic value to promote muscle growth and differentiation in human diseases (41). Accordingly, antibody-mediated myostatin blockade effectively ameliorates anatomical and physiological muscle parameters in the mdx mouse (42). Consistent with these findings, knockout of the Mstn gene in the mdx mouse improves its dystrophic muscle phenotype (43). Whether down-regulation of Mstn gene expression is involved in exercise-induced muscle hypertrophy remains to be firmly determined (44). Interestingly, Mstn-null mice have a substantial reduction in fat accumulation despite normal food intake. Furthermore, when Mstn-null mice were bred with two different models of obesity, agouti lethal yellow [A(y)] and obese [Lep(ob/ob)], the double-mutant mice exhibited partly corrected glucose metabolism (45). Therefore, pharmacological inactivation of myostatin may have the double benefit of increasing muscle mass and retarding or preventing the development of obesity and type II diabetes.

Follistatin and deacetylase inhibitors

Follistatin is a secreted protein that interacts with and inhibits the activity of several TGF-β family members, including GDF-11 [also known as bone morphogenetic protein 11 (BMP-11)] (46), a protein with high sequence similarity to myostatin (47). Furthermore, myostatin circulates as a part of a latent complex containing myostatin propeptide and either follistatin-related gene (FLRG) or GDF-associated serum protein-1 (GASP-1), two molecules closely related to follistatin (48, 49). Finally, follistatin interacts with myostatin and reverses the inhibition exerted by the latter on muscle differentiation of primary chick cells (38). Biochemical experiments indicate that follistatin can block the binding of myostatin to the activin IIB receptor by interacting with the C-terminal region of a myostatin dimer (50). Mice overexpressing follistatin under the control of muscle-specific regulatory elements display a muscle phenotype that recapitulates that observed in the absence of myostatin (50). Thus, follistatin counteracts the biological activity of myostatin and possibly of other TGF-β family members.

Follistatin (Fst) gene expression is activated by exposure of skeletal muscle cells to deacetylase inhibitors (51), a class of small, membrane-permeable molecules that block the activity of class I-II histone deacetylases (HDACs). Some of these molecules, including valproic acid, are currently employed to treat various clinical conditions such as epilepsy (52) and mood disorders (53). Exposure of cultured skeletal muscle cells to HDAC inhibitors results in the formation of myotubes with increased diameter, increased number of myonuclei (Fig. 2), and enhanced abundance of muscle structural proteins such as myosins (54). Similarly, mouse embryos exposed to HDAC inhibitors display a premature formation of caudal somites and augmented muscle gene expression when compared to control animals (54, 55). At the doses employed in those studies, valproic acid (a known teratogen) did not cause apparent toxicity or malformations. In cultured cells, the effects of HDAC inhibitors on follistatin are restricted to skeletal muscle cells, because treatment with these inhibitors of mouse primary keratinocytes and of other cell lines fated to differentiate toward the adipose, bone, or pituitary phenotypes fails to activate Fst gene expression (51). These findings are relevant because they suggest that systemic delivery of HDAC inhibitors may activate follistatin expression selectively in skeletal muscles. Iezzi et al. (51) found that the effects exerted by HDAC inhibitors on skeletal muscle are due to an increased ability of undifferentiated myoblasts to be recruited and to fuse with already formed myotubes, and appear not to be mediated by either IGF-1 or IL-4 (56), two molecules involved in myoblast fusion and growth. Follistatin overexpression in skeletal muscle cells can fully recapitulate the morphological and biochemical modifications induced by the HDAC inhibitors, indicating the pivotal role of follistatin in promoting myoblast accretion. Consistent with these observations, blockade of follistatin, by either functional inactivation with recombinant myostatin or RNA interference-meditated "knock-down" of follistatin, render HDAC inhibitors incapable of promoting myoblast accretion. A role for follistatin during skeletal muscle regeneration is indicated by its expression in activated satellite cells during both initial (51, 57) and later (51) stages of muscle regeneration. Systemic delivery of HDAC inhibitors in a mouse model of muscle degeneration and regeneration results in the activation of markers, suggesting the existence of an active and sustained process of muscle regeneration. Increased muscle regeneration promoted by HDAC inhibitors requires the presence of activated satellite cells, indicating that the activity of HDAC inhibitors may be not only cell-type specific, but also limited to damaged muscles (51).

Fig. 2.

The deacetylase inhibitor trichostatin A induces an increase in both the diameter of differentiated myotubes and the number of myonuclei. Undifferentiated C2C12 skeletal myoblasts were either mock-treated (control) or exposed to 50 nM trichostatin A and allowed to differentiate in low growth factor–containing medium for 2 days before immunostaining with an antibody to MHC.

In conclusion, an alternative method to blocking myostatin activity to promote muscle growth may be by inducing Fst gene expression with HDAC inhibitors. Although it is not known whether exercise-induced muscle hypertrophy is associated with an increase of Fst expression, it is tempting to speculate that this might be the case, because Fst expression is observed in activated satellite cells (57).

Cell Determinants of Muscle Growth

Satellite cells

Satellite cells are lineage-committed adult stem cells (58), located between the basal lamina and the sarcolemma of myofibers, that contribute to postnatal muscle growth (59) and muscle regeneration after injury (60). Upon myotrauma, quiescent satellite cells become activated, proliferate, and ultimately fuse to existing damaged muscle fibers or among themselves to form new myofibers (61). Satellite cells are activated in response to hypertrophic stimuli, such as those occurring during muscle mechanical overload (59, 62, 63). In several animal models of compensatory hypertrophy (6466) or after resistance training in humans (6769), the total number of activated satellite cells is substantially increased. In most of these early works, activation, proliferation, and incorporation of satellite cells into myofibers was documented by electron microscopy. Quiescent satellite cells are recognizable for their distinct location and morphological features, such as a very high nuclear-to-cytoplasm ratio with few organelles and a smaller nuclear size with increased amounts of heterochromatin compared to fiber myonuclei (70). When activated, satellite cells are easily identified, because they appear as swellings on the myofiber, have an increased cytoplasmic-to-nuclear ratio, and have increased numbers of intracellular organelles (61). Satellite cells have been the object of extensive studies aimed at the identification of specific molecular markers. Among the specific satellite cell markers so far identified are Pax7, a paired-box-containing transcription factor essential for the specification of satellite cells (71); myocyte nuclear factor (MNF, also called Foxk1), a winged-helix transcription factor; c-Met, the receptor for hepatocyte growth factor (HGF); M-cadherin, a calcium-dependent cell adhesion molecule that is expressed only in a small subset of quiescent satellite cells; neural cell adhesion molecule (NCAM) and vascular adhesion molecule-1 (VCAM-1); and syndecan-3 and syndecan-4 [reviewed in (72)]. All these markers are present in either quiescent or proliferating satellite cells. Upon exposure to signals emanating from the damaged area, quiescent satellite cells become activated and start proliferating (73). Once satellite cells are activated, they express the genes encoding the two myogenic regulatory factors Myf 5 and MyoD, which are not detectable in quiescent cells (74). The appearance of MyoD temporally precedes any signs of cellular division, such as the expression of proliferative cell nuclear antigen (PCNA) (75). After proliferation, satellite cells initiate expression of the genes encoding myogenin, muscle regulatory factor 4 (MRF4), cyclin-dependent kinase inhibitor p21, and, after a permanent exit from the cell cycle, they fuse with damaged myofibers or give rise to new myofibers (76, 77). Newly formed myofibers can be identified by the central location of myonuclei and expression of embryonic or developmental forms of myosin heavy chain (MHC) (78). In a process called self-renewal, a small proportion of satellite cells return to the quiescent state to replenish the satellite cell pool.

Because satellite cells in their mitotic phase are sensitive to DNA damage, γ-irradiation has been used to impair their activation and proliferation and to test for their functional relevance in the development of skeletal muscle hypertrophy. Compensatory hypertrophy can be largely prevented by muscle irradiation, providing compelling evidence that the presence of activated and proliferating satellite cells is indeed necessary for this process (7981). However, compensatory hypertrophy is not completely prevented by muscle irradiation, suggesting the possibility that some satellite cells could enter terminal differentiation and fuse with damaged muscle fibers after one cell division at most, thus evading the antiproliferative effect of radiation damage (82). Alternatively, some satellite cells may have survived, given the existence of cell subpopulations resistant to radiation (83). It should also be possible to address the satellite cell requirement for the genesis of muscle hypertrophy in Pax7-deficent mice subjected to compensatory hypertrophy. In fact, because Pax7-deficient mice lack satellite cells, any increase in muscle mass would have to be ascribed either to the expansion of other nonsatellite adult stem cells (see Nonsatellite Adult Stem Cells, below) or to increased protein synthesis in existing myofibers.

Satellite cell activation

The mechanisms leading to satellite cell activation during muscle hypertrophy are not entirely understood. It is postulated that extensive physical activity, such as resistance training or muscle overloading (chronic stretch, agonist muscle ablation, tenotomy), inflicts muscle injury (84, 85) that, similar to more severe muscle damage, may initiate a process of regeneration. An indirect proof of muscle damage after mechanical stress is given by an increase of serum markers such as muscle creatine kinase, an enzyme that is usually restricted to the myofiber cytosol (86). Muscle injury initiates an inflammatory response with the attraction of nonmuscle mononucleated cells, such as neutrophils and macrophages, into the damaged zone (87, 88), with consequent release of several growth factors (by either the infiltrating cells or the damaged myofibers themselves). These growth factors may ultimately regulate satellite cell proliferation and differentiation (89). Indeed, several cytokines have been described that modulate proliferation and differentiation of satellite cells in vitro or during regeneration after muscle injury (72, 90). HGF, which was initially detected in hepatic tissue, is considered to be a key regulator of satellite cell activity during muscle regeneration (91, 92). HGF is secreted by damaged tissue during the early phase of muscle regeneration in amounts proportional to the extent of muscle injury (93, 94). It seems that HGF directly regulates satellite cell activation, as suggested by the presence of its receptor c-Met in quiescent and activated satellite cells. Upon HGF binding, the c-Met receptor tyrosine kinase transmits intracellular signals through a unique multidocking site at its C terminus that is generated by autophosphorylation of two tyrosine residues. The multisubstrate docking site mediates the binding of several adaptor proteins, such as Grb2, SHC, Crk (also known as CRKL), PI3K, and the large adaptor protein Gab1. These adaptor proteins then participate in transducing extracellular signals elicited by HGF to downstream targets (95). The PI3K-Akt pathway, for instance, mediates HGF-induced survival and protection from apoptosis (96).

Other secreted factors may also participate in satellite cell activation. For example, fibroblast growth factor 6 (FGF-6) is another potential regulator of satellite cell activity (97, 98). However, FGF-6-deficient mice generated by two different laboratories display different phenotypes during muscle regeneration. In one case, FGF-6-null animals have impaired satellite cell proliferation and defects in muscle regeneration in response to a crush injury (99), whereas another study found no evidence of regenerative defects in response to several kinds of muscle damage (100). IL-4 is secreted by both myotubes and macrophages present in the regenerating muscles. By promoting fusion of myoblasts with preformed myotubes (56), IL-4 may actively participate in the regenerative process. Leukemia inhibitory factor (LIF), IL-6, IL-15, TNF-α, and several other factors have been described as potentially involved in the muscle remodeling response after injury (101). However, a definitive role for all these factors in regulating regeneration, hypertrophy, or both remains to be further elucidated.

As already described, a large body of evidence supports the importance of IGF-1 in the genesis of skeletal muscle hypertrophy. The expression of genes encoding two IGF-1 muscle-specific isoforms, IGF-1E and MGF, is stimulated by mechanical overload (102) with apparently different kinetics. MGF seems to be expressed early after mechanical damage and to temporally precede the activation of satellite cells, whereas IGF-1E expression peaks later and is likely to be involved in maintaining protein synthesis required to complete the muscle repair (103). IGF-1 can promote both proliferation (104, 105) and differentiation of cultured satellite cells (106, 107), and these findings have been confirmed in animal models. For instance, both IGF-1–mediated muscle hypertrophy and compensatory hypertrophy are halved by γ-irradiation (108). Furthermore, IGF-1–induced muscle hypertrophy is accompanied by increased DNA content of myofibers (109, 110), appearance of centrally located nuclei, and appearance of developmental forms of myosin (29, 111). All these findings support the involvement of satellite cells in IGF-1–mediated muscle hypertrophy. Further evidence for a role of satellite cells is suggested by experiments in which muscle-localized expression of IGF-1 prevented, through an increase of the regenerative potential of satellite cells, the aging-related loss of muscle mass (29). Finally, satellite cells derived from mice overexpressing IGF-1 display an increased proliferative potential (112), which seems to be mediated by activation of the PI3K-Akt pathway and down-regulation of the activity of cyclin-dependent inhibitor p27kip1 (113), which results from an inhibitory effect on FOXO1 (114). Therefore, the molecular pathways activated by IGF-1 in the muscle fibers to promote increased protein translation appear to also be activated in satellite cells.

Satellite cells and aging

Aging is accompanied by a decline in muscle mass and strength, a phenomenon referred as to sarcopenia (115). It is clear that fitness is greater at any age in individuals who exercise regularly versus those who do not and that sarcopenia is reduced—albeit not abolished—in physically active elderly people. A decline in the number of satellite cells, their proliferative capacity, or both may contribute to sarcopenia. One of the mechanisms responsible for the reduced regenerative potential of old muscle seems to be the decline in Notch signaling (116), a pathway involved in the formation of several tissues during embryogenesis. Notch signaling also contributes to satellite cell activation, proliferation, and cell lineage determination (117). The regenerative potential of old muscle can be experimentally restored either by forced activation of Notch signaling or, as mentioned, by IGF-1. Whether physical exercise activates the Notch signaling pathway is not known. However, manipulations involving the Notch and IGF-1 pathways may result in the activation and proliferation of satellite cells and may be useful in retarding age-related loss of muscle mass and strength.

Nonsatellite adult stem cells

In recent years, there have been several reports indicating that cells deriving from bone marrow can contribute to the regeneration of injured muscle (118, 119). The ontology and effectiveness of these cells in promoting muscle regeneration are currently being intensively pursued. In one study, multipotent progenitors of mesodermal tissues [termed "mesoangioblasts" (120)] that express Flk1 (vascular endothelial growth factor–receptor 2), the key marker of angiopoietic progenitors, were intraarterially delivered into α-sarcoglycan–null dystrophic mice. This resulted in restoration of the α-sarcoglycan–dystrophin complex and, most importantly, in complete functional muscle recovery of the dystrophic animals (121). Although of extreme biological and clinical relevance, the discussion of these nonmuscle resident stem cells is beyond the scope of this review. Novel populations of adult stem cells have been isolated from uninjured skeletal muscle that express the CD45 and Sca1 (122, 123), two surface markers expressed on hematopoietic stem cells. When cultured in vitro, these stem cells form hematopoietic colonies but do not differentiate into muscle, with the exception of a small percentage that can adopt the myogenic phenotype when cocultured with satellite cell-derived myoblasts (124). This population of cells is clearly distinct from satellite cells, because they contain the CD45 and Sca1 markers, which are not present on satellite cells (124). Upon muscle injury, the number of cells in the CD45+:Sca1+ population increases approximately 10-fold and gives rise to myogenic cells (125). This suggests that a factor, or factors, released by the injured or regenerating muscle may switch the fate of the CD45+:Sca1+ population and redirect it from the hematopoietic to the myogenic phenotype. Indeed, Polesskaya et al. (125) demonstrated that the expression of genes encoding selected members of the secreted Wnt protein family are induced in regenerating muscles. Consistent with this, ectopic expression of Wnts causes muscle commitment of CD45+:Sca1+ cells. The results of this study indicate that resident, nonsatellite-derived progenitors can, in response to Wnt signaling, participate in muscle repair. Because mesoangioblasts contain both Sca1 and Wnts (120), it should be informative to compare the expression profiles of mesoangioblasts and of muscle resident CD45+:Sca1+ cells to establish whether these cells have a common origin. Although these findings are important in suggesting a potential clinical use of Wnt modulators to influence muscle regeneration, whether muscle-resident adult stem cells may participate in muscle hypertrophy remains to be experimentally tested. Finally, it is interesting to note that IGF-1 has been reported to enhance the proliferation potential of a subpopulation of bone marrow stem cells, and their recruitment to sites of muscle damage, in lethally irradiated mice that received bone marrow transplants (126). Thus, it is possible that also adult stem cells may respond to IGF-1 signaling.

Conclusions and Future Directions

It has been known for some time that muscle hypertrophy is the result of the activity both of molecules that stimulate growth of existing myofibers and of fiber accretion obtained through recruitment of activated satellite cells. Nonetheless, only recently has a detailed description of the molecular and cellular mechanisms underlying postnatal muscle growth become available. It is now clear that the IGF-1 pathway is central to both increased protein synthesis and satellite cell activation. Similarly, the IGF-1 pathway controls the activity of the FOXO transcription factors. Under physiological conditions, IGF-1 prevents FOXOs from activating the gene expression program leading to muscle atrophy. It does not seem likely that the IGF-1 pathway is dedicated to controlling cell growth exclusively in skeletal muscle, because overexpression of the Drosophila IRS, Akt, or p70S6K can cause hypertrophy in disparate cell lineages (127), including those giving rise to the eye and wing (128). Rather, the autocrine or paracrine activity of locally produced IGF-1 may ensure skeletal muscle specificity.

In several pathological settings, it would be desirable to either prevent muscle atrophy or activate pathways that lead to increased muscle growth and that promote muscle hypertrophy. In this respect, it is of interest to discuss the experimental use of small molecules—whose therapeutic efficacy is currently being evaluated in clinical trials (129, 130)—that modulate the activity of nuclear deacetylases. The transcriptional activity of FOXO proteins is regulated by Sir2, an NAD+-dependent deacetylase, with the expression of several FOXO targets being reduced by Sir2 overexpression (131, 132). Because inhibition of FOXO-transcriptional activity is associated with reduced expression of MAFbx, it will be important to determine whether the Sir2 agonist resveratrol (133) is of therapeutic value in reducing FOXO-dependent activation of the muscle atrophy program. Similarly, the use of class I-II deacetylase inhibitors has given encouraging results in promoting muscle regeneration in an animal model of muscle injury through activation of follistatin gene expression (51). Future experiments will determine whether these agents are effective in modulating the progression of muscle wasting in animal models of cancer, sepsis, and degenerative diseases.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68.
  69. 69.
  70. 70.
  71. 71.
  72. 72.
  73. 73.
  74. 74.
  75. 75.
  76. 76.
  77. 77.
  78. 78.
  79. 79.
  80. 80.
  81. 81.
  82. 82.
  83. 83.
  84. 84.
  85. 85.
  86. 86.
  87. 87.
  88. 88.
  89. 89.
  90. 90.
  91. 91.
  92. 92.
  93. 93.
  94. 94.
  95. 95.
  96. 96.
  97. 97.
  98. 98.
  99. 99.
  100. 100.
  101. 101.
  102. 102.
  103. 103.
  104. 104.
  105. 105.
  106. 106.
  107. 107.
  108. 108.
  109. 109.
  110. 110.
  111. 111.
  112. 112.
  113. 113.
  114. 114.
  115. 115.
  116. 116.
  117. 117.
  118. 118.
  119. 119.
  120. 120.
  121. 121.
  122. 122.
  123. 123.
  124. 124.
  125. 125.
  126. 126.
  127. 127.
  128. 128.
  129. 129.
  130. 130.
  131. 131.
  132. 132.
  133. 133.
View Abstract

Stay Connected to Science Signaling

Navigate This Article