Cell Stem Cell Review Tissue-Specific Stem Cells: Lessons from the Skeletal Muscle Satellite Cell Andrew S. Brack1,2,3,* and Thomas A. Rando4,5 1Center of Regenerative Medicine, Massachusetts General Hospital, Boston, MA 02114, USA Medical School, Boston, MA 02115, USA 3Harvard Stem Cell Institute, Cambridge, MA 02138, USA 4Paul F. Glenn Laboratories for the Biology of Aging and the Department of Neurology and Neurological Sciences, Stanford University School of Medicine, Stanford, CA 94305, USA 5Neurology Service and Rehabilitation Research and Development Center of Excellence, Veterans Affairs Palo Alto Health Care System, Palo Alto, CA 94304, USA *Correspondence: abrack@partners.org DOI 10.1016/j.stem.2012.04.001 2Harvard In 1961, the satellite cell was first identified when electron microscopic examination of skeletal muscle demonstrated a cell wedged between the plasma membrane of the muscle fiber and the basement membrane. In recent years it has been conclusively demonstrated that the satellite cell is the primary cellular source for muscle regeneration and is equipped with the potential to self renew, thus functioning as a bona fide skeletal muscle stem cell (MuSC). As we move past the 50th anniversary of the satellite cell, we take this opportunity to discuss the current state of the art and dissect the unknowns in the MuSC field. Introduction The observation that skeletal muscle has the capacity to regenerate after injury was well documented by microscopic examination in the mid-19th century, primarily in German literature (Scharner and Zammit, 2011). Still, the cellular basis of this regenerative potential remained elusive for a century until 50 years ago when Alexander Mauro detected a mononucleated cell, which he termed a ‘‘satellite cell,’’ closely apposed to mature myofibers in electron micrographs of skeletal muscle (Mauro, 1961). Without any functional evidence, he hypothesized that this could represent a kind of muscle progenitor cell akin to those present in the developing embryo, capable of forming new muscle in response to injury. This turned out to be a very accurate prediction: five decades of research on satellite cells have demonstrated that they have the characteristics that Mauro surmised. In modern parlance, the satellite cell is considered a muscle stem cell distinguished from the plethora of adult tissue-specific stem cells that have been described by the fact that it was identified anatomically before it was characterized functionally; most adult stem cells have been first demonstrated to exist in functional assays, which are then followed by a hunt for the cells histologically. The history of the satellite cell has been the subject of several recent reviews (Scharner and Zammit, 2011; Yablonka-Reuveni, 2011), and the regulation and contribution of satellite cell progenitors during lineage progression, differentiation, and contribution to muscle repair has also been extensively documented (Chargé and Rudnicki, 2004; Wang and Rudnicki, 2012; Zammit et al., 2006). In this review, we focus on the current status of satellite cell research through the lens of stem cell biology. We highlight recent studies illustrating that satellite cells are essential for maintenance of the stem cell pool and repair of the differentiated muscle tissue in which they reside. In addition, we discuss the properties that satellite cells possess in common with other stem cell populations and the mechanisms that regulate satellite cell functions. 504 Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. Satellite Cell Identification and Stem Cell Properties After anatomical identification of satellite cells in 1961, their behavior in response to growth and regeneration was investigated. It was noted that in regenerating muscle, undifferentiated cells increase in abundance and align with the periphery of damaged fibers. As regeneration progresses, immature myogenic progenitors are replaced with more mature myoblasts (Allbrook, 1962). At later stages of repair, undifferentiated cells begin to appear in association with the regenerated fibers (Shafiq and Gorycki, 1965). A series of studies using tritiated thymidine confirmed that satellite cells were mitotically dormant in mature muscle and the source for regenerating muscle (Reznik, 1969; Schultz et al., 1978; Snow, 1977) and that daughters of satellite cells contributed to both the satellite cell compartment and differentiated nuclei in growing muscle (Lipton and Schultz, 1979; Moss and Leblond, 1970, 1971; Schultz, 1996). Thus with the evidence that satellite cells were capable of asymmetric divisions and endowed with self-renewal properties, a new era was born, in which satellite cells were considered bona fide MuSCs. Cell transplantation was becoming more commonplace in regenerative biology to test cellular contribution to tissue repair and renewal of progenitor populations. The grafting of committed satellite cell progeny (myoblasts) between mice with different isoenzyme subtypes confirmed that donor cells could fuse with host cells or myofibers (Partridge et al., 1978; Watt et al., 1982), providing evidence of a renewable cell source with regenerative capacity. However, identification and isolation of the self-renewing cells, MuSCs, remained elusive for many years. Eventually, immunotypic analysis identified the paired box 7 transcription factor Pax7 as a uniform marker of satellite cells (Seale et al., 2000). In response to injury, Pax7+ satellite cells enter cycle and differentiate and a subset returns back to quiescence to replenish the dormant satellite cell pool (Abou-Khalil and Brack, 2010). A body of work during the late 1980s and early Cell Stem Cell Review Figure 1. Satellite Cell Functions (A) Adult single muscle fiber showing Pax7+ satellite cells (pink) located on individual multinucleated adult muscle fibers, myonuclei stained with DAPI (blue). (B) An example of satellite cell contribution to muscle repair with Pax7CreERTM.R26Rb-gal is shown by X-gal reaction on an adult lower limb muscle prior to and after injury (reproduced from Shea et al., 2010). Note the many X-gal+ muscle fibers running throughout the regenerating muscle. (C) Pax7+ cells located in satellite cell position in uninjured muscle and their contribution to both myofiber repair and repopulation in the satellite cell position after regeneration (reproduced from Shea et al., 2010). (D) A schematic showing the participating roles of satellite cells. Pax7-expressing satellite cells (red) reside between the plasmalemma (brown line) of the muscle fiber and the basal lamina (black line). A mature adult muscle contains thousands of differentiated nuclei (green) and 10 SCs (this number is highly dependent on age, muscle length, and metabolic character). In response to injury, Pax7+ cells proliferate, most differentiate and fuse to regenerate myofibers (blue), whereas a minority self renew and return back to quiescence to replenish the satellite cell pool (left). After genetic ablation of satellite cells, muscle regeneration is prevented. During rapid hypertrophic stimuli, satellite cells fuse to add differentiated myonuclei. After genetic ablation of satellite cells, muscle hypertrophy occurs without addition of myonuclei, suggesting that satellite cells are dispensable for protein accumulation and hypertrophy (center). During normal tissue homeostasis, the contribution of satellite cells to maintenance of tissue or the satellite cell pool remains unknown (right). 1990s discovered the family of myogenic regulatory factors (MRFs), including Myod, Myf5, Mrf44, and Myogenin, genes that coordinate developmental myogenesis (Bentzinger et al., 2012). In their quiescent state, adult satellite cells express, along with Pax7, the Myf5 transcript but lack Myod. During activation and lineage progression, satellite cells turn on Myod and Myf5 protein, followed by the marker of terminal differentiation, Myogenin (Chargé and Rudnicki, 2004). Demonstration of a tractable muscle-resident cell that contributed to muscle fiber repair and repopulation of the niche was achieved through transplantation of a FACS-isolated subset of muscle-resident cells selected via cell surface receptor expression (such as CD31 , CD45 , Sca1 , CXCR4+, CD31+, Integrin b1+) (Montarras et al., 2005; Sherwood et al., 2004). Engrafting a single myofiber with its satellite cells marked by a nuclear Myf5lacZ reporter provided the first definitive demonstration that satellite cells possess stem cell activity (Collins et al., 2005). In recent years, satellite cells have been proven to self-renew at the single-cell level (Sacco et al., 2008) and to retain stem cell function over seven rounds of serial transplantation and therefore function as MuSCs (Rocheteau et al., 2012). With the advent of conditional Cre/lox systems, lineage tracing of satellite cells has demonstrated that adult Pax7+ cells selfrenew and differentiate in their endogenous environment in response to injury (Lepper et al., 2009; Shea et al., 2010), suggesting that adult Pax7+ satellite cells are sufficient for both MuSC pool maintenance and myofiber repair. Moreover, studies using inducible genetic strategies to ablate adult Pax7+ cells have demonstrated that satellite cells are essential for muscle repair and replenishment of the MuSC pool (Lepper et al., 2011; McCarthy et al., 2011; Murphy et al., 2011; Sambasivan et al., 2011). Cre/lox technology has also enabled the assessment of lineage potential of satellite cells. Based on studies with Pax7CreERTM and MyodCre, it appears that most if not all satellite cells are restricted to the myogenic lineage, arguing in favor of unipotency of adult satellite cells (Lepper et al., 2009; Shea et al., 2010; Starkey et al., 2011). However, it appears that at least a subset of satellite cells are driven out of lineage, becoming fibrogenic in pathological conditions or during aging (Amini-Nik et al., 2011; Brack et al., 2007). Although the contribution of satellite cells to tissue repair is indisputable, their role in other aspects of skeletal muscle biology is less certain (Figure 1). During normal daily activity, muscle fiber size and myonuclear content remain relatively constant throughout most of adult life (Hughes and Schiaffino, 1999), and satellite cells may exhibit a modest decline depending upon the specific muscle and species examined (Brack and Rando, 2007). Are satellite cells participating to maintain muscle fibers during homeostasis? Indelible-marked adult satellite cells chased over 6 weeks do not appear to undergo proliferation or contribute to muscle fibers (Lepper et al., 2009; Shea et al., 2010). Moreover, 2 weeks after ablation of the satellite cell pool by a genetic strategy, muscle fiber size appears unaffected (Lepper et al., 2011). Even 6 months after depletion of satellite Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. 505 Cell Stem Cell Review cells from a muscle, gross histological appearance is normal and fiber size changes negligibly (Cheung et al., 2012). Together, the data suggest that satellite cell contribution to myofiber maintenance is minimal during homeostasis. It would be informative to determine whether ablation of the adult pool would exacerbate muscle atrophy in contexts of higher cellular demand such as exercise or aging. Adult skeletal muscle is capable of hypertrophy, and the question as to satellite cell participation in that process has been debated for many years. Recent studies have provided direct evidence of hypertrophy without any satellite cell contribution. Using a genetic strategy to ablate the adult Pax7+ cell pool, mice deficient in satellite cells were able to rapidly increase muscle fiber size during a 2 week hypertrophic stimulus (McCarthy et al., 2011). This occurred in the absence of satellite cell-derived myonuclear accretion, suggesting that satellite cells are dispensable for increase in muscle mass. However, it is possible that over longer periods of time, additional differentiated nuclei, and hence a maintained myonuclear domain size in hypertrophic muscle, would be required to maintain a larger muscle fiber (Hughes and Schiaffino, 1999). This is supported by the fact that loss of differentiated nuclei in aged muscle fibers precedes muscle fiber atrophy (Brack et al., 2005). Stem Cell Characteristics Developmental Origins Based on anatomical position and expression of Pax7, satellite cell precursors first appear during late fetal stage (Relaix et al., 2004, 2005). Lineage tracing approaches demonstrated that the majority of adult Pax7+ satellite cells are formed from a somitic origin of cells that transition through a Pax7+/Myod+ state (Kanisicak et al., 2009; Lepper and Fan, 2010; Schienda et al., 2006). It is clear that the number of Pax7+ cells declines during development and postnatal growth (Schultz, 1989; White et al., 2010). Therefore, not all embryonic Pax7+ cells will form the self-renewing adult pool, but instead some will primarily contribute to the growing myofiber. This suggests that some fetal Pax7+/Myod+ cells differentiate and other Pax7+/Myod+ cells form the adult satellite cell pool. Myod is considered a master regulator of myogenesis (Weintraub et al., 1991). Based on germline mutants, Myod is required for satellite cell differentiation but not for the formation of the satellite cell pool (Chargé et al., 2008; Rudnicki et al., 1992). In vitro experiments demonstrate that subsets of myogenic progenitors lose Myod and return to a quiescent state (Halevy et al., 2004; Olguin and Olwin, 2004; Yoshida et al., 1998; Zammit et al., 2004). Moreover, selfrenewing adult satellite cells transiently express Myod prior to niche repopulation during repair (Shea et al., 2010). Therefore, Pax7+/Myod+ precursors are able to form the satellite cell pool providing that Myod is repressed. Based on studies with a Myf5CreYFP reporter, it was demonstrated that the majority (90%) of the adult satellite cell pool is formed from a Myf5+ precursor (Kuang et al., 2007). However, the formation of the satellite cell pool in germline Myf5 nulls suggests that Myf5 is not required for satellite cell formation; compensation by Myod cannot be excluded (Gensch et al., 2008; Haldar et al., 2008). While there are multiple caveats to consider when interpreting these studies, including the contribution of non-Cre-recombined cell populations, these results suggest that the adult satellite 506 Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. Figure 2. MuSC Heterogeneity Origins of the MuSC pool. SC precursors (Pax7CreER+) located in the myotome begin expressing Pax7 (red) at E10.5, and most express Myf5+ and Myod+ before occupancy into the niche. During early postnatal growth (P0–P7), a subset of alkaline phosphatase (AP+)-derived pericytes as detected with APCre (gray) express Pax7 (red) and occupy the MuSC niche (dark red). A subset of AP+ pericytes is capable of contributing to myofiber differentiation (dark red myonuclei). Whether pericytes that commit to myogenic fusion express Pax7 during their lineage progression is unknown (thick red arrows). cell pool is formed from precursors that are under different transcriptional control. Contribution from the extrinsic environment may also participate in differential regulation of satellite cell precursors. While the adult satellite cell pool is formed from Pax7+ embryonic precursors, this does not preclude a model whereby an alternative source of cells express Pax7 and home to the satellite cell niche. Using immunotypic markers (CD34+/Sca1+) in combination with donor cell marking for transplantation studies, it has been demonstrated that a subset of PICs (PW1+ Pax7 interstitial cells) self-renew and contribute to myofibers and Pax7+ satellite cells in injured adult muscle (Mitchell et al., 2010). Moreover, myogenic contribution of PICs was Pax7 dependent and temporally restricted to postnatal growing muscle. Based on lineage tracking, a subset of alkaline phosphatase (AP)-expressing pericytes indelibly marked in postnatal mice contributed to the formation of the adult Pax7+ satellite cell compartment (Dellavalle et al., 2011). Therefore the origin of the adult satellite cell pool may be heterogeneous (Figure 2). In the future, it will be important to dissect the unique properties and lineage relationship between pericytes and PICs that express Pax7 relative to Pax7-derived satellite cells. Asymmetric Cell Divisions While asymmetric cell divisions are one of the fundamental characteristics of stem cells underlying the process of self-renewal, stem cells may undergo self-renewal by virtue of symmetric divisions (Tajbakhsh et al., 2009). As long as only one daughter of a symmetric cell division is instructed to differentiate, the stem cell pool can be maintained while differentiated progeny are produced. Historically, there were no seminal observations that provided evidence to either support or refute asymmetric cell divisions in the satellite cell lineage, and it was only in the past decade that this has been the focus of studies in the MuSC field. The first evidence of asymmetric cell divisions came from studies of the role of Notch signaling during satellite cell activation and proliferative amplification. It was found that the Notch inhibitory protein Numb was asymmetrically segregated in progenitors undergoing cell division and that Numb was localized Cell Stem Cell Review Figure 3. Nonrandom Segregation of Chromosomes and Asymmetric Fate Determination The satellite cell pool is composed of primitive (Pax7hi, green) and lineage primed (Pax7lo, gray) subsets. Pax7hi cells are in a dormant (less metabolically active) state. Pax7lo satellite cells are lineage primed. After cytokinesis, satellite cell daughters can each inherit some chromosomes bearing newer (purple) or older (dark green) template strands. Pax7hi subsets undergo nonrandom segregation of chromosomes, whereby one daughter contains exclusively chromosomes bearing older template strands, while its sister contains only chromosomes bearing newer template strands. In contrast, Pax7lo cells undergo random segregation of chromosomes ensuring that each daughter cell inherits an older and newer template strand. Subsets of Pax7hi and Pax7lo are capable of multiple rounds of self-renewal (curved arrows) and differentiation (blue cells), but Pax7hi cells are endowed with enhanced selfrenewal capability. Daughter cells containing newer template strands express high levels of Numb, Myod, and Dek1 and are biased to differentiate (blue cell). Pax7hi cells can give rise to Pax7lo, but not vice versa, suggesting a hierarchical relationship in the satellite cell pool based on Pax7 expression. In addition to cell fate determination, nonrandom segregation of chromosomes may act to protect genome integrity during cell proliferation. Because of low metabolic output of primitive satellite cells, they may be protected from oxidative damage. Lineage-primed satellite cells are capable of self-renewal but may be acutely sensitive to proliferative demands and other forms of genotoxic insults. to one pole of dividing cells, perpendicular to the plane of cell division (Conboy and Rando, 2002). This asymmetric Numb localization is reminiscent of what had been previously described in specific lineages during Drosophila development (Petersen et al., 2002; Zhong et al., 1996). The result of such an asymmetric segregation in the satellite cell lineage is the inheritance of Numb by one daughter and not the other, the former being destined to differentiate and the latter remaining an undifferentiated progenitor (Conboy and Rando, 2002). The asymmetric segregation of Numb was subsequently investigated during satellite cell activation, and asymmetric segregation of Numb was also detected in cells prior to the onset of differentiation (Shinin et al., 2009). Other proteins, such as Myod (Zammit et al., 2004) and Dek (Cheung et al., 2012), are asymmetrically segregated in activated satellite cells, and those proteins are more highly expressed in cells that have progressed further along the myogenic lineage. Asymmetric cell divisions have also been suggested in relation to the aforementioned distinction of satellite cell subsets based on developmental history of Myf5 expression (Kuang et al., 2007). In this model, satellite cells undergo asymmetric cell divisions, giving rise to two daughters, one being another stem cell (still with no history of Myf5 gene expression) and the other becoming part of the larger pool of satellite cells (and expressing the Myf5 transcript). In this case, there is no direct evidence as yet of an asymmetric segregation of any specific protein or transcript in the mother cell, and the asymmetry may instead be a divergent cell fate dictated by the different environments of the two daughters. This is particularly likely given the observation that daughters exhibiting divergent fates, one self-renewing and the other committing toward differentiation, were much more frequent when the plane of division was perpendicular to the axis of the myofiber (Kuang et al., 2007). By contrast, divisions that occurred parallel to the myofiber axis were much more likely to give rise to two daughters with identical characteristics as opposed to divergent fates, suggesting that the local environment of the satellite cell niche might be dictating the fate of satellite cell progeny. Yet another asymmetry that has been described is the asymmetric segregation of sister chromatids as satellite cells divide and the population expands (Conboy et al., 2007; Shinin et al., 2006). Recent evidence suggests that a subset of satellite cells, in fact those with characteristics of long-term stem cells, are more likely to exhibit nonrandom sister chromatid segregation (Rocheteau et al., 2012). This further supports the notion that asymmetric segregation of sister chromatids might be a property of stem cells as originally proposed (Figure 3; Cairns, 1975). Although the functional significance of this remains theoretical, the proposed function is to segregate DNA strands that have been damaged during replication to the more differentiated daughter and reserve for the self-renewing stem cells those strands that served as templates during the previous round of cell division, thus acquiring few replication-induced errors (Charville and Rando, 2011). Although this teleological function has not been demonstrated and there may be detrimental consequences of template strand segregation during aging (Charville and Rando, 2011), it is supported by the fact that the oldest template strands segregate with the less differentiated satellite cell daughter (Conboy et al., 2007; Shinin et al., 2006). Satellite Cell Heterogeneity It is becoming more apparent that adult stem cell populations are heterogeneous. Heterogeneity in the adult satellite cell pool has been demonstrated based on multiple criteria, such as expression profile (Beauchamp et al., 2000; Rocheteau et al., 2012), proliferation kinetics in vitro (Day et al., 2009), self-renewal potential (Kuang et al., 2007), and molecular regulation (Kitamoto and Hanaoka, 2010; Kuang et al., 2007; Shea et al., 2010). Functional heterogeneity within a pool of adult stem cells can arise from differentially specified subsets that retain distinct properties during cellular division or evolution of a subset of cells derived from a single homogeneous cell population that will emerge during cellular division. It is widely accepted that Pax7 expression declines in differentiating progenitors (Zammit et al., 2004). Evidence for varied Pax7 levels within the satellite cell pool was recently demonstrated via a transgenic GFP reporter of Pax7 transcript (Rocheteau et al., 2012). Functional analysis of satellite cells at extreme ends of the Pax7 expression distribution demonstrated that Pax7hi cells had reduced metabolic activity and slower cell cycle entry kinetics, whereas Pax7lo cells were primed to differentiate in vitro. Although cells with lower Pax7 expression appeared to be more prone to differentiation, all Pax7+ cells could support more than six serial transplantations, demonstrating that both subsets of satellite cells possess remarkable regenerative capacity and the ability to self-renew. Using an inducible Cre/ lox system, Pax7 was deleted from satellite cells in growing postnatal muscle and adult muscle prior to injury. The results Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. 507 Cell Stem Cell Review demonstrated that Pax7 is essential for satellite cell function during early postnatal growth and is dispensable for adult muscle tissue repair (Lepper et al., 2009). Given that a subset of postnatal Pax7+ cells seeds the adult Pax7+ pool, it supports the hypothesis that there is a small subset of postnatal Pax7+ cells that evolve during the selective pressure of muscle growth to propagate satellite cells that function independently of Pax7. It raises the possibility that functionally distinct satellite cell subsets can interconvert for selective advantage under conditions of high cellular demand as observed in spermatogonial stem cells (Nakagawa et al., 2007, 2010). Adult stem cells with limited proliferative history are endowed with greater stem cell capability than more frequently dividing counterparts (Foudi et al., 2009; Wilson et al., 2008), suggesting that stem cell self-renewal capacity may decline in proportion with the number of divisions a stem cell has undergone in its history. Interestingly, subsets of presumptive satellite cells in growing postnatal muscle divide with slower proliferation kinetics, based on tritiated thymidine uptake (Schultz, 1996). These data suggest that one level of functional heterogeneity in the satellite cell pool may exist based on their proliferative history. In dystrophic muscle, disease progression correlates with proliferative capacity of satellite cells, suggesting that restricting proliferative output of stem cells relative to their downstream progeny may extend self-renewal capacity and ameliorate disease pathogenesis (Sacco et al., 2010). Whether functional heterogeneity across a pool of satellite cells is stable once specified or adaptive based on cellular demand, it is clear that the satellite cell pool is functionally heterogeneous. The challenge in the future is to determine how molecular heterogeneity at the cell-intrinsic and -extrinsic levels instructs functional diversity within the satellite cell pool. Regulation of Satellite Cell Function Stem Cell Niche The stem cell niche refers to the microenvironment that maintains ‘‘stemness’’ (Schofield, 1978). The attributes of the niche as originally conceptualized were (1) a defined anatomical site, (2) a location where stem cells could reproduce, (3) a place where differentiation is inhibited, and (4) a space that also limits the numbers of stem cells. Therefore, the niche is a protector of stem cell number and function, restraining proliferation and differentiation of stem cells, and maintaining a quiescent phenotype. The location of the satellite cell, i.e., residing in a depression in the plasmalemma and beneath the basal lamina of the muscle fiber, provided a defined anatomical site of the putative MuSC (Mauro, 1961). The hypothesis that the muscle fiber is a satellite cell niche is supported by evidence generated from single muscle fiber experiments. Removal of the myofiber plasmalemma drives quiescent satellite cells into cycle, suggesting a role in inhibition of mitogen-induced cell cycle entry (Bischoff, 1986a). Quiescence is a conserved property of stem cells mediated by the niche (Orford and Scadden, 2008). Studies on freshly isolated single muscle fibers have shown that a subset of quiescent satellite cells can proliferate and return back to quiescence when in contact with the single muscle fiber, whereas a larger subset commits to differentiation but their fusion is inhibited (Bischoff, 1986a; Olguin and Olwin, 2004; Zammit et al., 2004). 508 Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. Therefore the muscle fiber fulfills three of Schoffield’s criteria of a stem cell niche. Whether the muscle fiber constrains stem cell numbers is to be determined. In response to injury, the muscle fiber degenerates, which probably leads to niche destruction and a loss of inhibitory signals. Besides losing inhibitory factors that restrict proliferation and differentiation from the degenerated niche, muscle injury also promotes the release of stimulatory factors present at the basal lamina of the muscle fiber that drive proliferation and differentiation (Bischoff, 1986b; Sheehan and Allen, 1999). The observation that microenvironment stiffness is a regulator of stem cell potential and fate decisions suggests that maintenance of stemness from the niche is regulated at many levels (Engler et al., 2006; Gilbert et al., 2010). Resolving the signals that drive self-renewal versus differentiation will require a careful characterization of the basal and apical aspects of the niche (Figure 4A). The satellite cell niche may be composed of different cell types, as observed for other stem cell compartments (Morrison and Spradling, 2008). Besides the muscle fiber, there are other muscle-resident cells in close proximity to the satellite cell, such as Ang1-expresssing endothelial cells (Abou-Khalil et al., 2009; Christov et al., 2007). It is becoming clear that other cell types, such as fibro-adipogenic progenitors (FAPs) and TCF4+ fibroblasts, influence satellite cell behavior in contexts of regeneration and growth (Joe et al., 2010; Murphy et al., 2011). It remains to be determined whether such cell types constitute the niche and therefore regulate stemness or function as paracrine agents involved in proliferation and differentiation of satellite cell progeny (Figure 4A). As cell-specific Cre drivers become more readily available, it will be possible and essential to determine the cell types and mode of regulation that contribute to a functional satellite cell niche. Signaling Pathway Regulation of Satellite Cell Function A balance between extrinsic cues and intracellular signals converge to preserve stem cell function. Over the past 50 years, many excellent studies have demonstrated that multiple extrinsic signaling pathways, such as IGF, FGF, Wnt, Notch, BMP, and TGF-b, function in the activation of satellite cells, their downstream progeny, and their lineage progression (Kuang et al., 2008). More recently, the signaling pathways that modulate functions specific to stem cells, such as maintenance of quiescence during homeostasis, reversible quiescence and self-renewal after proliferation, asymmetric fate decisions, and symmetric expansion of stem cells, are becoming more defined (Figure 4B). It was recently shown that active Notch signaling is required to maintain satellite cells in the quiescent state, implying a role of niche-derived Notch ligand that binds to a Notch receptor on the quiescent satellite cell (Bjornson et al., 2011; Mourikis et al., 2011). In these studies, Rbp-j, the downstream transcriptional coactivator in the Notch pathways, was deleted from adult satellite cells in uninjured muscle. This led to activation and ectopic differentiation of satellite cells, possibly in the absence of cell cycle entry and therefore bypassing the transient amplifying (TA) progenitor stage (Bjornson et al., 2011; Mourikis et al., 2011). Therefore, not only is RBP-J important for restricting cell cycle entry, it also argues that in nonquiescent satellite cells, RBP-J, possibly as a mediator of Notch1 signaling, may be required for normal step-wise lineage progression (Conboy Cell Stem Cell Review Figure 4. Regulation of MuSC Functions in Satellite Cells (A) The MuSC niche. The satellite cell (red) resides in a quiescent state in contact with the plasmalemma (brown) of the myofiber and basal lamina (black) in a milieu of inhibitory (white arrows) and stimulatory (red arrows) factors that retain satellite cells in a stem cell state and inhibit their differentiation. Other cells in close proximity to satellite cells in vivo such as TCF4+ fibroblasts, FAPs, PICs, and pericytes may constitute components of the MuSC niche and signal to the satellite cell and vice versa. (B) Schematic of MuSC regulation. Satellite cells transition from quiescence into a transient amplification stage; most progress along a myogenic lineage to differentiate while a subset self-renew and return back to quiescence (thick blue arrows). Markers denoting each stage of lineage progression are shown in black, and molecules required for distinct satellite cell functions are shown in red. Quiescent satellite cells express Spry1, Notch-3, and miR-489. RBP-J and Myf5 are expressed in quiescent and cycling satellite cells. RBP-J and miR-489 (red) are required to retain satellite cells in quiescence. Spry1 and Notch-3 are required in cycling satellite cells for their return to quiescence and homeostasis of satellite cell pool after injury. Myf5 is required for normal transient amplification of progenitors. Myod, Myogenin, and Myosin Heavy Chain (MyHC) are required for differentiation. Normal step-wise lineage progression of satellite cells depends on RBP-J (thin blue arrow), possibly through modulation of Notch-1 signaling. and Rando, 2002). That the satellite cell pool is lost when the TA stage is bypassed has implications for understanding the relationship between self-renewal and differentiation potential of downstream progenitors within the stem cell hierarchy. To replenish the satellite cell pool after injury, stem cells return back to quiescence after proliferating. Our understanding of reversion back to quiescence in vivo is limited. Based on in vitro experiments it is becoming apparent that multiple signaling pathways are actively involved in promoting satellite cell quiescence. Activation of the Ang1/Tie2 signaling complex promotes the return to quiescence of myoblasts in vitro (AbouKhalil et al., 2009). P38/MAPK pathway has also been implicated in regulating quiescence of satellite cells (Jones et al., 2005). Finally, Myostatin, a secreted growth factor belonging to the TFG-b family, was shown to negatively regulate quiescence and the return to quiescence of satellite cells in vitro (McCroskery et al., 2003; McFarlane et al., 2008). Together these data illustrate a critical role of growth factor signaling in the return of cycling satellite cell progenitors back to a quiescent state. Further in vivo analysis and genetic approaches that directly target satellite cells will clarify the signaling cascades that regulate the maintenance of quiescence and return to quiescence after proliferation. Recently, signaling through Notch-3 was found to negatively regulate satellite cell pool size in regenerating muscle (Kitamoto and Hanaoka, 2010). Whether this is regulated at the cellautonomous level of the satellite cell or by a niche-based mechanism will require cell-specific genetic mutants. Using a satellite cell-specific mutant for Spry1, it was demonstrated that Spry1, an inhibitor of growth factor signaling, is required to restore satellite cell pool size after injury (Shea et al., 2010). This suggests that a balance between inhibitory and stimulatory cues from the microenvironment is required for quiescence to be achieved. One can theorize that Spry1 and other intracellular growth factor inhibitors help to dampen the stimulatory factors at the basal lamina to direct fate toward quiescence and the repression of differentiation. The challenge in the future is to identify niche-derived inhibitory signals that promote and retain stemness after injury. Stem cells possess a unique ability to divide symmetrically or asymmetrically depending on tissue requirements (Tajbakhsh et al., 2009). Noncanonical Wnt signaling through the Wnt7AFrz7-Vangl2 cascade was recently implicated in driving symmetric expansion of satellite cells located on single muscle fibers in vitro (Le Grand et al., 2009). It will be interesting to unravel the relationship between asymmetric divisions via Numb and symmetric divisions via Wnt7A-Vangl2 that occur to control stem cell expansion and differentiation. In general, the cellular output within a pool of cells in response to manipulation of signaling cascades is not equivalent. Through loss-of-function approaches, it has been demonstrated that satellite cells differentially respond to manipulation of signaling molecules such as Spry1 and Notch-3 (Kitamoto and Hanaoka, 2010; Shea et al., 2010). It will be important to decipher whether heterogeneous cell signaling requirements are due to extrinsic differences, such as localization of satellite cells in discrete niches, or to cell-intrinsic differences, based on lineage relationships. While knowledge of the signaling cascades that regulate stem cell properties of satellite cells is progressing, it is likely that more Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. 509 Cell Stem Cell Review pathways participate than have been identified. During cell division and lineage progression, signaling molecules are used reiteratively. For example, Numb is partitioned to initially allow asymmetric fate of neural precursors and subsequently to maintain progenitors (Petersen et al., 2004). Likewise, Numb may be important for satellite cell self-renewal and also, at a later stage of myogenic lineage progression, for regulating the divergent fates of proliferating progenitors (Conboy and Rando, 2002; Jory et al., 2009; Shinin et al., 2006). Therefore, while canonical Wnt, FGF, and BMP signaling cascades participate in myogenic lineage progression and differentiation (Kuang et al., 2008), it may be that they are also deployed in some fashion to direct MuSC properties. Epigenetic Regulation of Satellite Cell Function Based on what we know about extrinsic cues controlling stem cell function, the role of the stem cell niche cannot be underestimated. However, more stable modes of regulation may also maintain key functions of satellite cells. Control of the function of stem cells, particularly embryonic stem cells (ESCs), has focused on regulatory mechanisms at the level of epigenetics, namely the ensemble of heritable changes in gene function that occur without modifications of the primary DNA sequence (Bird, 2007). Such changes include DNA methylation and chromatin structural modifications mediated by histone modifications and nucleosome positioning. Not too long after the identification of the satellite cell, electron micrographs demonstrated that there are changes in chromatin organization in satellite cells from adult muscle and growing muscle consistent with their transition from a quiescent to a proliferative state (Church, 1969). As our knowledge in the field of epigenetics becomes more sophisticated, it is becoming appreciated that the regulators of epigenetic states in stem cells include mediators of DNA methylation and demethylation, microRNAs and other noncoding RNAs, and modifiers of histones, including acetylases, deacetylases, methylases, and demethylases. This field is redefining cellular states at a molecular level that are much more varied and complex and probably reflect the dynamic interaction of stem cells with their environment resulting in greater functional heterogeneity than was previously envisioned. It is likely that epigenetic states will better define such key satellite cell features as prolonged quiescence and lineage fidelity. It is also likely that DNA and histone modifications will underlie many of the changes in aged satellite cells that account for age-related declines in functionality and rejuvenation through exposure to the systemic environment (Brack et al., 2007; Conboy et al., 2005). Although there has been extensive research on the epigenetic control of myogenesis, primarily using the C2C12 myoblast cell line (Juan et al., 2011), studies in satellite cells have been limited because of the need for large numbers of cells for such analysis. Advances in both satellite cell purification and methodologies for genome- and epigenome-wide analyses of limited cell numbers will allow for rapid advances in these areas in the coming years. There have, however, been studies of regulators of epigenetic states in satellite cells. It was shown that Pax7 functions with the Wdr5-Ash2L-MLL2 histone methyltransferase (HMT) complex to direct methylation of histone H3 lysine 4 (H3K4). Binding of the Pax7-HMT complex resulted in H3K4 trimethylation of 510 Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. chromatin at the Myf5 locus (McKinnell et al., 2008), suggesting that Pax7 may act through chromatin modifications to stably and hereditably maintain the myogenic program. Conversely, the regulation of Pax7 expression may be determined by the regulation of the repressive Polycomb repressive complex 2 (PRC2) and its enzymatic component, EZH2 (Palacios et al., 2010), suggesting a network providing stability of the myogenic program related to Pax7 expression. The regulation of myogenesis by microRNAs has likewise focused mainly on developmental myogenesis and the regulation of myogenic differentiation (Braun and Gautel, 2011). Recent studies have also examined the role of miRNAs in the regulation of quiescent satellite cells. miR-206 was recently shown to target Pax3 in quiescent satellite cells, accounting for the differential expression in Pax3 in cellular subsets (Boutet et al., 2012). This study also revealed that alternate polyadenylation of the Pax3 transcript is an important determinant not only of satellite cell heterogeneity but also of the susceptibility of the Pax3 transcript to regulation by any miRNA (Boutet et al., 2012). Another microRNA, miR-489, was recently shown to be highly expressed in quiescent satellite cells and to regulate specifically the quiescent state by targeting the transcript of the Dek gene, a gene known to be important for cell cycle regulation and alternative splicing (Cheung et al., 2012). Knockdown of miR-489 in vivo led to spontaneous activation of quiescent satellite cells, thus demonstrating the importance of a stable miRNA network for maintaining the quiescent state. Conclusions and Perspectives The last fifty years have led to a remarkable understanding of the satellite cell. It is widely accepted that the satellite cell is essential for regenerative myogenesis and can maintain itself after injury. It has been demonstrated that the muscle fiber constitutes a major functional component of the satellite cell niche. More recently, experiments have illustrated the functional heterogeneity within the satellite cell pool. As we move forward, what will the next fifty years bring? Remarkably, we do not know whether satellite cells maintain themselves or differentiate to maintain muscle tissue during normal daily wear and tear. This information is critical as biologists devise strategies to harness skeletal muscle strength of the ever-expanding human aged population. To date, direct experimental evidence is lacking to determine whether stem cell number is limiting. If MuSCs have a finite capacity, then a quorum will be required to functionally repair muscle tissue and maintain homeostasis. In the context of regenerating a tissue, such as skeletal muscle, it may be too simplistic to consider MuSC number and function as independent entities. Rather, there is probably a cooperative relationship between MuSC number and function in response to a regenerative insult. For example, in the presence of few MuSCs, a greater functional demand will be imposed on those rare cells, leading to their exhaustion. In contrast, increasing the number of MuSCs in a muscle will lessen the burden on each cell imposed by the regenerative insult. In scenarios of limiting satellite cell number, other MuSCs may participate, contributing as facultative stem cells. Other cell types such as PICs, mesangioblasts, and pericytes are able to contribute to muscle tissue repair (Péault et al., 2007). Further analysis is Cell Stem Cell Review needed to determine whether such cell types function as facultative stem cells that are dependent on cellular context. Understanding whether their contribution is dependent on the presence of a functional pool of satellite cells may reveal further complexities of the regulation of skeletal muscle repair. Development of genetic tools to ablate fractions of satellite cells will illustrate the intimate cooperative relationship between satellite cell number and function and their interdependency with other contributing cell types. Technological advances in recent years have provided greater resolution of stem cell heterogeneity. With the advent of fluorescent reporters of proliferative history, it is now possible to isolate and compare such heterogeneous cells to dissect their function and molecular regulation (Foudi et al., 2009; Tumbar et al., 2004). Marking satellite cells based on Cre drivers has proven invaluable for tracking them and their contribution to growth and repair (Lepper and Fan, 2010; Lepper et al., 2011; Shea et al., 2010). More recent advances through the use of multicolor Cre reporters or genetic barcoding utilized in other stem cell compartments have provided improved cellular resolution that facilitates the tracking of clonally diverse cells in a population to be studied (Lu et al., 2011; Snippert et al., 2010). Such techniques reveal that the level of stem cell heterogeneity is more dynamic and context dependent than previously appreciated (Lu et al., 2011). Information gleaned from such approaches will allow the relationship between molecular and functional heterogeneity to be resolved. The anatomical location of the satellite cell provides a landmark for the satellite cell niche. Based on its conceptual framework, although the niche impacts stem cell properties of the satellite cell, the mechanism by which this is achieved remains a mystery. In the future, functional components of the satellite cell niche will be identified through technological advances in live imaging, microdissection, and biochemical analysis. As our understanding of satellite cell heterogeneity evolves, it will be interesting to identify whether there are distinct niches that, in some way, establish satellite cell heterogeneity. Advances in our knowledge of cell-intrinsic mechanisms that regulate stem cell function, such as epigenetic regulation of satellite cells, will illustrate the communication between the extrinsic environment and intrinsic effectors to specify and maintain stem cell states. In conclusion, the seminal observations of the satellite cell in its niche made by Alexander Mauro almost fifty years ago have spawned many great advances in our understanding of this tissue-specific stem cell. It is apparent that there is greater heterogeneity in the satellite cell pool in terms of cellular subsets that specify the pool, their function, and the signaling cascades that regulate them. We are just beginning to unravel the microenvironmental influences that mediate stem cell properties and how epigenetic regulation governs stable properties of satellite cells during homeostasis and repair. REFERENCES Abou-Khalil, R., and Brack, A.S. (2010). Muscle stem cells and reversible quiescence: the role of sprouty. Cell Cycle 9, 2575–2580. Abou-Khalil, R., Le Grand, F., Pallafacchina, G., Valable, S., Authier, F.J., Rudnicki, M.A., Gherardi, R.K., Germain, S., Chretien, F., Sotiropoulos, A., et al. (2009). Autocrine and paracrine angiopoietin 1/Tie-2 signaling promotes muscle satellite cell self-renewal. Cell Stem Cell 5, 298–309. Allbrook, D. (1962). An electron microscopic study of regenerating skeletal muscle. J. Anat. 96, 137–152. Amini-Nik, S., Glancy, D., Boimer, C., Whetstone, H., Keller, C., and Alman, B.A. (2011). Pax7 expressing cells contribute to dermal wound repair, regulating scar size through a b-catenin mediated process. Stem Cells 29, 1371– 1379. Beauchamp, J.R., Heslop, L., Yu, D.S., Tajbakhsh, S., Kelly, R.G., Wernig, A., Buckingham, M.E., Partridge, T.A., and Zammit, P.S. (2000). Expression of CD34 and Myf5 defines the majority of quiescent adult skeletal muscle satellite cells. J. Cell Biol. 151, 1221–1234. Bentzinger, C.F., Wang, Y.X., and Rudnicki, M.A. (2012). Building muscle: molecular regulation of myogenesis. Cold Spring Harb. Perspect. Biol. 4, a008342. Bird, A. (2007). Perceptions of epigenetics. Nature 447, 396–398. Bischoff, R. (1986a). Proliferation of muscle satellite cells on intact myofibers in culture. Dev. Biol. 115, 129–139. Bischoff, R. (1986b). A satellite cell mitogen from crushed adult muscle. Dev. Biol. 115, 140–147. Bjornson, C.R., Cheung, T.H., Liu, L., Tripathi, P.V., Steeper, K.M., and Rando, T.A. (2011). Notch signaling is necessary to maintain quiescence in adult muscle stem cells. Stem Cells 30, 232–242. Boutet, S.C., Cheung, T.H., Quach, N.L., Liu, L., Prescott, S., Edalati, A., Iori, K., and Rando, T.A. (2012). Alternative polyadenylation mediates microRNA regulation of muscle stem cell function. Cell Stem Cell 10, 327–336. Brack, A.S., and Rando, T.A. (2007). Intrinsic changes and extrinsic influences of myogenic stem cell function during aging. Stem Cell Rev. 3, 226–237. Brack, A.S., Bildsoe, H., and Hughes, S.M. (2005). Evidence that satellite cell decrement contributes to preferential decline in nuclear number from large fibres during murine age-related muscle atrophy. J. Cell Sci. 118, 4813– 4821. Brack, A.S., Conboy, M.J., Roy, S., Lee, M., Kuo, C.J., Keller, C., and Rando, T.A. (2007). Increased Wnt signaling during aging alters muscle stem cell fate and increases fibrosis. Science 317, 807–810. Braun, T., and Gautel, M. (2011). Transcriptional mechanisms regulating skeletal muscle differentiation, growth and homeostasis. Nat. Rev. Mol. Cell Biol. 12, 349–361. Cairns, J. (1975). Mutation selection and the natural history of cancer. Nature 255, 197–200. Chargé, S.B., and Rudnicki, M.A. (2004). Cellular and molecular regulation of muscle regeneration. Physiol. Rev. 84, 209–238. Chargé, S.B., Brack, A.S., Bayol, S.A., and Hughes, S.M. (2008). MyoD- and nerve-dependent maintenance of MyoD expression in mature muscle fibres acts through the DRR/PRR element. BMC Dev. Biol. 8, 5. Charville, G.W., and Rando, T.A. (2011). Stem cell ageing and non-random chromosome segregation. Philos. Trans. R. Soc. Lond. B Biol. Sci. 366, 85–93. ACKNOWLEDGMENTS Cheung, T.H., Quach, N.L., Charville, G.W., Liu, L., Park, L., Edalati, A., Yoo, B., Hoang, P., and Rando, T.A. (2012). Maintenance of muscle stem-cell quiescence by microRNA-489. Nature 482, 524–528. This work was supported by grants from the NIH (R01 AR060868 to A.S.B. and P01 AG036695, R37 AG23806, [MERIT Award], and R01 AR056849 to T.A.R.). We thank Jenna Norton for artwork and Josef Christensen for critical reading of the manuscript. Christov, C., Chrétien, F., Abou-Khalil, R., Bassez, G., Vallet, G., Authier, F.J., Bassaglia, Y., Shinin, V., Tajbakhsh, S., Chazaud, B., and Gherardi, R.K. (2007). Muscle satellite cells and endothelial cells: close neighbors and privileged partners. Mol. Biol. Cell 18, 1397–1409. Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. 511 Cell Stem Cell Review Church, J.C. (1969). Satellite cells and myogenesis; a study in the fruit-bat web. J. Anat. 105, 419–438. Collins, C.A., Olsen, I., Zammit, P.S., Heslop, L., Petrie, A., Partridge, T.A., and Morgan, J.E. (2005). Stem cell function, self-renewal, and behavioral heterogeneity of cells from the adult muscle satellite cell niche. Cell 122, 289–301. Conboy, I.M., and Rando, T.A. (2002). The regulation of Notch signaling controls satellite cell activation and cell fate determination in postnatal myogenesis. Dev. Cell 3, 397–409. Conboy, I.M., Conboy, M.J., Wagers, A.J., Girma, E.R., Weissman, I.L., and Rando, T.A. (2005). Rejuvenation of aged progenitor cells by exposure to a young systemic environment. Nature 433, 760–764. Conboy, M.J., Karasov, A.O., and Rando, T.A. (2007). High incidence of nonrandom template strand segregation and asymmetric fate determination in dividing stem cells and their progeny. PLoS Biol. 5, e102. Day, K., Paterson, B., and Yablonka-Reuveni, Z. (2009). A distinct profile of myogenic regulatory factor detection within Pax7+ cells at S phase supports a unique role of Myf5 during posthatch chicken myogenesis. Dev. Dyn. 238, 1001–1009. Dellavalle, A., Maroli, G., Covarello, D., Azzoni, E., Innocenzi, A., Perani, L., Antonini, S., Sambasivan, R., Brunelli, S., Tajbakhsh, S., et al. (2011). Pericytes resident in postnatal skeletal muscle differentiate into muscle fibres and generate satellite cells. Nat. Comm. 2, 499. Engler, A.J., Sen, S., Sweeney, H.L., and Discher, D.E. (2006). Matrix elasticity directs stem cell lineage specification. Cell 126, 677–689. Kitamoto, T., and Hanaoka, K. (2010). Notch3 null mutation in mice causes muscle hyperplasia by repetitive muscle regeneration. Stem Cells 28, 2205– 2216. Kuang, S., Kuroda, K., Le Grand, F., and Rudnicki, M.A. (2007). Asymmetric self-renewal and commitment of satellite stem cells in muscle. Cell 129, 999–1010. Kuang, S., Gillespie, M.A., and Rudnicki, M.A. (2008). Niche regulation of muscle satellite cell self-renewal and differentiation. Cell Stem Cell 2, 22–31. Le Grand, F., Jones, A.E., Seale, V., Scimè, A., and Rudnicki, M.A. (2009). Wnt7a activates the planar cell polarity pathway to drive the symmetric expansion of satellite stem cells. Cell Stem Cell 4, 535–547. Lepper, C., and Fan, C.M. (2010). Inducible lineage tracing of Pax7-descendant cells reveals embryonic origin of adult satellite cells. Genesis 48, 424–436. Lepper, C., Conway, S.J., and Fan, C.M. (2009). Adult satellite cells and embryonic muscle progenitors have distinct genetic requirements. Nature 460, 627–631. Lepper, C., Partridge, T.A., and Fan, C.M. (2011). An absolute requirement for Pax7-positive satellite cells in acute injury-induced skeletal muscle regeneration. Development 138, 3639–3646. Lipton, B.H., and Schultz, E. (1979). Developmental fate of skeletal muscle satellite cells. Science 205, 1292–1294. Lu, R., Neff, N.F., Quake, S.R., and Weissman, I.L. (2011). Tracking single hematopoietic stem cells in vivo using high-throughput sequencing in conjunction with viral genetic barcoding. Nat. Biotechnol. 29, 928–933. Foudi, A., Hochedlinger, K., Van Buren, D., Schindler, J.W., Jaenisch, R., Carey, V., and Hock, H. (2009). Analysis of histone 2B-GFP retention reveals slowly cycling hematopoietic stem cells. Nat. Biotechnol. 27, 84–90. Mauro, A. (1961). Satellite cell of skeletal muscle fibers. J. Biophys. Biochem. Cytol. 9, 493–495. Gensch, N., Borchardt, T., Schneider, A., Riethmacher, D., and Braun, T. (2008). Different autonomous myogenic cell populations revealed by ablation of Myf5-expressing cells during mouse embryogenesis. Development 135, 1597–1604. McCarthy, J.J., Mula, J., Miyazaki, M., Erfani, R., Garrison, K., Farooqui, A.B., Srikuea, R., Lawson, B.A., Grimes, B., Keller, C., et al. (2011). Effective fiber hypertrophy in satellite cell-depleted skeletal muscle. Development 138, 3657–3666. Gilbert, P.M., Havenstrite, K.L., Magnusson, K.E., Sacco, A., Leonardi, N.A., Kraft, P., Nguyen, N.K., Thrun, S., Lutolf, M.P., and Blau, H.M. (2010). Substrate elasticity regulates skeletal muscle stem cell self-renewal in culture. Science 329, 1078–1081. McCroskery, S., Thomas, M., Maxwell, L., Sharma, M., and Kambadur, R. (2003). Myostatin negatively regulates satellite cell activation and self-renewal. J. Cell Biol. 162, 1135–1147. Haldar, M., Karan, G., Tvrdik, P., and Capecchi, M.R. (2008). Two cell lineages, myf5 and myf5-independent, participate in mouse skeletal myogenesis. Dev. Cell 14, 437–445. Halevy, O., Piestun, Y., Allouh, M.Z., Rosser, B.W., Rinkevich, Y., Reshef, R., Rozenboim, I., Wleklinski-Lee, M., and Yablonka-Reuveni, Z. (2004). Pattern of Pax7 expression during myogenesis in the posthatch chicken establishes a model for satellite cell differentiation and renewal. Dev. Dyn. 231, 489–502. Hughes, S.M., and Schiaffino, S. (1999). Control of muscle fibre size: a crucial factor in ageing. Acta Physiol. Scand. 167, 307–312. Joe, A.W., Yi, L., Natarajan, A., Le Grand, F., So, L., Wang, J., Rudnicki, M.A., and Rossi, F.M. (2010). Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis. Nat. Cell Biol. 12, 153–163. Jones, N.C., Tyner, K.J., Nibarger, L., Stanley, H.M., Cornelison, D.D., Fedorov, Y.V., and Olwin, B.B. (2005). The p38alpha/beta MAPK functions as a molecular switch to activate the quiescent satellite cell. J. Cell Biol. 169, 105–116. Jory, A., Le Roux, I., Gayraud-Morel, B., Rocheteau, P., Cohen-Tannoudji, M., Cumano, A., and Tajbakhsh, S. (2009). Numb promotes an increase in skeletal muscle progenitor cells in the embryonic somite. Stem Cells 27, 2769–2780. McFarlane, C., Hennebry, A., Thomas, M., Plummer, E., Ling, N., Sharma, M., and Kambadur, R. (2008). Myostatin signals through Pax7 to regulate satellite cell self-renewal. Exp. Cell Res. 314, 317–329. McKinnell, I.W., Ishibashi, J., Le Grand, F., Punch, V.G., Addicks, G.C., Greenblatt, J.F., Dilworth, F.J., and Rudnicki, M.A. (2008). Pax7 activates myogenic genes by recruitment of a histone methyltransferase complex. Nat. Cell Biol. 10, 77–84. Mitchell, K.J., Pannérec, A., Cadot, B., Parlakian, A., Besson, V., Gomes, E.R., Marazzi, G., and Sassoon, D.A. (2010). Identification and characterization of a non-satellite cell muscle resident progenitor during postnatal development. Nat. Cell Biol. 12, 257–266. Montarras, D., Morgan, J., Collins, C., Relaix, F., Zaffran, S., Cumano, A., Partridge, T., and Buckingham, M. (2005). Direct isolation of satellite cells for skeletal muscle regeneration. Science 309, 2064–2067. Morrison, S.J., and Spradling, A.C. (2008). Stem cells and niches: mechanisms that promote stem cell maintenance throughout life. Cell 132, 598–611. Moss, F.P., and Leblond, C.P. (1970). Nature of dividing nuclei in skeletal muscle of growing rats. J. Cell Biol. 44, 459–462. Moss, F.P., and Leblond, C.P. (1971). Satellite cells as the source of nuclei in muscles of growing rats. Anat. Rec. 170, 421–435. Juan, A.H., Derfoul, A., Feng, X., Ryall, J.G., Dell’Orso, S., Pasut, A., Zare, H., Simone, J.M., Rudnicki, M.A., and Sartorelli, V. (2011). Polycomb EZH2 controls self-renewal and safeguards the transcriptional identity of skeletal muscle stem cells. Genes Dev. 25, 789–794. Mourikis, P., Sambasivan, R., Castel, D., Rocheteau, P., Bizzarro, V., and Tajbakhsh, S. (2011). A critical requirement for Notch signaling in maintenance of the quiescent skeletal muscle stem cell state. Stem Cells 30, 243–252. Kanisicak, O., Mendez, J.J., Yamamoto, S., Yamamoto, M., and Goldhamer, D.J. (2009). Progenitors of skeletal muscle satellite cells express the muscle determination gene, MyoD. Dev. Biol. 332, 131–141. Murphy, M.M., Lawson, J.A., Mathew, S.J., Hutcheson, D.A., and Kardon, G. (2011). Satellite cells, connective tissue fibroblasts and their interactions are crucial for muscle regeneration. Development 138, 3625–3637. 512 Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. Cell Stem Cell Review Nakagawa, T., Nabeshima, Y., and Yoshida, S. (2007). Functional identification of the actual and potential stem cell compartments in mouse spermatogenesis. Dev. Cell 12, 195–206. Nakagawa, T., Sharma, M., Nabeshima, Y., Braun, R.E., and Yoshida, S. (2010). Functional hierarchy and reversibility within the murine spermatogenic stem cell compartment. Science 328, 62–67. Olguin, H.C., and Olwin, B.B. (2004). Pax-7 up-regulation inhibits myogenesis and cell cycle progression in satellite cells: a potential mechanism for selfrenewal. Dev. Biol. 275, 375–388. Schultz, E. (1996). Satellite cell proliferative compartments in growing skeletal muscles. Dev. Biol. 175, 84–94. Schultz, E., Gibson, M.C., and Champion, T. (1978). Satellite cells are mitotically quiescent in mature mouse muscle: an EM and radioautographic study. J. Exp. Zool. 206, 451–456. Seale, P., Sabourin, L.A., Girgis-Gabardo, A., Mansouri, A., Gruss, P., and Rudnicki, M.A. (2000). Pax7 is required for the specification of myogenic satellite cells. Cell 102, 777–786. Orford, K.W., and Scadden, D.T. (2008). Deconstructing stem cell self-renewal: genetic insights into cell-cycle regulation. Nat. Rev. Genet. 9, 115–128. Shafiq, S.A., and Gorycki, M.A. (1965). Regeneration in skeletal muscle of mouse: some electron-microscope observations. J. Pathol. Bacteriol. 90, 123–127. Palacios, D., Mozzetta, C., Consalvi, S., Caretti, G., Saccone, V., Proserpio, V., Marquez, V.E., Valente, S., Mai, A., Forcales, S.V., et al. (2010). TNF/p38a/ polycomb signaling to Pax7 locus in satellite cells links inflammation to the epigenetic control of muscle regeneration. Cell Stem Cell 7, 455–469. Shea, K.L., Xiang, W., LaPorta, V.S., Licht, J.D., Keller, C., Basson, M.A., and Brack, A.S. (2010). Sprouty1 regulates reversible quiescence of a selfrenewing adult muscle stem cell pool during regeneration. Cell Stem Cell 6, 117–129. Partridge, T.A., Grounds, M., and Sloper, J.C. (1978). Evidence of fusion between host and donor myoblasts in skeletal muscle grafts. Nature 273, 306–308. Sheehan, S.M., and Allen, R.E. (1999). Skeletal muscle satellite cell proliferation in response to members of the fibroblast growth factor family and hepatocyte growth factor. J. Cell. Physiol. 181, 499–506. Péault, B., Rudnicki, M., Torrente, Y., Cossu, G., Tremblay, J.P., Partridge, T., Gussoni, E., Kunkel, L.M., and Huard, J. (2007). Stem and progenitor cells in skeletal muscle development, maintenance, and therapy. Mol. Ther. 15, 867–877. Sherwood, R.I., Christensen, J.L., Conboy, I.M., Conboy, M.J., Rando, T.A., Weissman, I.L., and Wagers, A.J. (2004). Isolation of adult mouse myogenic progenitors: functional heterogeneity of cells within and engrafting skeletal muscle. Cell 119, 543–554. Petersen, P.H., Zou, K., Hwang, J.K., Jan, Y.N., and Zhong, W. (2002). Progenitor cell maintenance requires numb and numblike during mouse neurogenesis. Nature 419, 929–934. Shinin, V., Gayraud-Morel, B., Gomès, D., and Tajbakhsh, S. (2006). Asymmetric division and cosegregation of template DNA strands in adult muscle satellite cells. Nat. Cell Biol. 8, 677–687. Petersen, P.H., Zou, K., Krauss, S., and Zhong, W. (2004). Continuing role for mouse Numb and Numbl in maintaining progenitor cells during cortical neurogenesis. Nat. Neurosci. 7, 803–811. Shinin, V., Gayraud-Morel, B., and Tajbakhsh, S. (2009). Template DNA-strand co-segregation and asymmetric cell division in skeletal muscle stem cells. Methods Mol. Biol. 482, 295–317. Relaix, F., Rocancourt, D., Mansouri, A., and Buckingham, M. (2004). Divergent functions of murine Pax3 and Pax7 in limb muscle development. Genes Dev. 18, 1088–1105. Relaix, F., Rocancourt, D., Mansouri, A., and Buckingham, M. (2005). A Pax3/ Pax7-dependent population of skeletal muscle progenitor cells. Nature 435, 948–953. Reznik, M. (1969). Thymidine-3H uptake by satellite cells of regenerating skeletal muscle. J. Cell Biol. 40, 568–571. Snippert, H.J., van der Flier, L.G., Sato, T., van Es, J.H., van den Born, M., Kroon-Veenboer, C., Barker, N., Klein, A.M., van Rheenen, J., Simons, B.D., and Clevers, H. (2010). Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143, 134–144. Snow, M.H. (1977). Myogenic cell formation in regenerating rat skeletal muscle injured by mincing. II. An autoradiographic study. Anat. Rec. 188, 201–217. Rocheteau, P., Gayraud-Morel, B., Siegl-Cachedenier, I., Blasco, M.A., and Tajbakhsh, S. (2012). A subpopulation of adult skeletal muscle stem cells retains all template DNA strands after cell division. Cell 148, 112–125. Starkey, J.D., Yamamoto, M., Yamamoto, S., and Goldhamer, D.J. (2011). Skeletal muscle satellite cells are committed to myogenesis and do not spontaneously adopt nonmyogenic fates. J. Histochem. Cytochem. 59, 33–46. Rudnicki, M.A., Braun, T., Hinuma, S., and Jaenisch, R. (1992). Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf-5 and results in apparently normal muscle development. Cell 71, 383–390. Tajbakhsh, S., Rocheteau, P., and Le Roux, I. (2009). Asymmetric cell divisions and asymmetric cell fates. Annu. Rev. Cell Dev. Biol. 25, 671–699. Sacco, A., Doyonnas, R., Kraft, P., Vitorovic, S., and Blau, H.M. (2008). Selfrenewal and expansion of single transplanted muscle stem cells. Nature 456, 502–506. Sacco, A., Mourkioti, F., Tran, R., Choi, J., Llewellyn, M., Kraft, P., Shkreli, M., Delp, S., Pomerantz, J.H., Artandi, S.E., and Blau, H.M. (2010). Short telomeres and stem cell exhaustion model Duchenne muscular dystrophy in mdx/mTR mice. Cell 143, 1059–1071. Sambasivan, R., Yao, R., Kissenpfennig, A., Van Wittenberghe, L., Paldi, A., Gayraud-Morel, B., Guenou, H., Malissen, B., Tajbakhsh, S., and Galy, A. (2011). Pax7-expressing satellite cells are indispensable for adult skeletal muscle regeneration. Development 138, 3647–3656. Scharner, J., and Zammit, P.S. (2011). The muscle satellite cell at 50: the formative years. Skeletal Muscle 1, 28. Schienda, J., Engleka, K.A., Jun, S., Hansen, M.S., Epstein, J.A., Tabin, C.J., Kunkel, L.M., and Kardon, G. (2006). Somitic origin of limb muscle satellite and side population cells. Proc. Natl. Acad. Sci. USA 103, 945–950. Tumbar, T., Guasch, G., Greco, V., Blanpain, C., Lowry, W.E., Rendl, M., and Fuchs, E. (2004). Defining the epithelial stem cell niche in skin. Science 303, 359–363. Wang, Y.X., and Rudnicki, M.A. (2012). Satellite cells, the engines of muscle repair. Nat. Rev. Mol. Cell Biol. 13, 127–133. Watt, D.J., Lambert, K., Morgan, J.E., Partridge, T.A., and Sloper, J.C. (1982). Incorporation of donor muscle precursor cells into an area of muscle regeneration in the host mouse. J. Neurol. Sci. 57, 319–331. Weintraub, H., Davis, R., Tapscott, S., Thayer, M., Krause, M., Benezra, R., Blackwell, T.K., Turner, D., Rupp, R., Hollenberg, S., et al. (1991). The myoD gene family: nodal point during specification of the muscle cell lineage. Science 251, 761–766. White, R.B., Biérinx, A.S., Gnocchi, V.F., and Zammit, P.S. (2010). Dynamics of muscle fibre growth during postnatal mouse development. BMC Dev. Biol. 10, 21. Schofield, R. (1978). The relationship between the spleen colony-forming cell and the haemopoietic stem cell. Blood Cells 4, 7–25. Wilson, A., Laurenti, E., Oser, G., van der Wath, R.C., Blanco-Bose, W., Jaworski, M., Offner, S., Dunant, C.F., Eshkind, L., Bockamp, E., et al. (2008). Hematopoietic stem cells reversibly switch from dormancy to selfrenewal during homeostasis and repair. Cell 135, 1118–1129. Schultz, E. (1989). Satellite cell behavior during skeletal muscle growth and regeneration. Med. Sci. Sports Exerc. 21 (5, Suppl), S181–S186. Yablonka-Reuveni, Z. (2011). The skeletal muscle satellite cell: still young and fascinating at 50. J. Histochem. Cytochem. 59, 1041–1059. Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. 513 Cell Stem Cell Review Yoshida, N., Yoshida, S., Koishi, K., Masuda, K., and Nabeshima, Y. (1998). Cell heterogeneity upon myogenic differentiation: down-regulation of MyoD and Myf-5 generates ‘reserve cells’. J. Cell Sci. 111, 769–779. Zammit, P.S., Partridge, T.A., and Yablonka-Reuveni, Z. (2006). The skeletal muscle satellite cell: the stem cell that came in from the cold. J. Histochem. Cytochem. 54, 1177–1191. Zammit, P.S., Golding, J.P., Nagata, Y., Hudon, V., Partridge, T.A., and Beauchamp, J.R. (2004). Muscle satellite cells adopt divergent fates: a mechanism for self-renewal? J. Cell Biol. 166, 347–357. Zhong, W., Feder, J.N., Jiang, M.M., Jan, L.Y., and Jan, Y.N. (1996). Asymmetric localization of a mammalian numb homolog during mouse cortical neurogenesis. Neuron 17, 43–53. 514 Cell Stem Cell 10, May 4, 2012 ª2012 Elsevier Inc. Review series Repairing skeletal muscle: regenerative potential of skeletal muscle stem cells Francesco Saverio Tedesco,1,2 Arianna Dellavalle,1 Jordi Diaz-Manera,1,3,4 Graziella Messina,1,5 and Giulio Cossu1,5 1Division of Regenerative Medicine, San Raffaele Scientific Institute, Milan, Italy. 2Vita-Salute San Raffaele and Open University, Milan, Italy. Santa Creu i Sant Pau, Neuromuscular Diseases Unit, Barcelona, Spain. 4Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Madrid, Spain. 5Department of Biology, University of Milan, Milan, Italy. 3Hospital Skeletal muscle damaged by injury or by degenerative diseases such as muscular dystrophy is able to regenerate new muscle fibers. Regeneration mainly depends upon satellite cells, myogenic progenitors localized between the basal lamina and the muscle fiber membrane. However, other cell types outside the basal lamina, such as pericytes, also have myogenic potency. Here, we discuss the main properties of satellite cells and other myogenic progenitors as well as recent efforts to obtain myogenic cells from pluripotent stem cells for patient-tailored cell therapy. Clinical trials utilizing these cells to treat muscular dystrophies, heart failure, and stress urinary incontinence are also briefly outlined. Introduction It has been known for more than a century that skeletal muscle, the most abundant tissue of the body, has the ability to regenerate new muscle fibers after it has been damaged by injury or as a consequence of diseases such as muscular dystrophy (1). Muscle fibers are syncytial cells that contain several hundred nuclei within a continuous cytoplasm. Therefore, whether the process of regeneration depends upon the fusion of mononucleated precursor cells or upon the fragmentation of dying muscle fibers, which release new cells, remained controversial for a long time, even after the demonstration by Beatrice Mintz and Wilber Baker (2) that multinucleated fibers are formed by the fusion of single cells. In 1961, Alexander Mauro (3) observed mononuclear cells between the basal lamina that surrounds each muscle fiber and the plasma membrane of the muscle fiber and named them satellite cells (SCs) (Figure 1). SCs were later accepted to be, and are still considered today, the main players in skeletal muscle regeneration. SCs also contribute to the postnatal growth of muscle fibers, which in adults contain approximately 6–8 times more nuclei than in neonates, all of them being irreversibly postmitotic. In addition to SCs, other progenitors located outside the basal lamina, including pericytes, endothelial cells, and interstitial cells, have been shown to have some myogenic potential in vitro or after transplantation. The developmental origin of these progenitors is unclear, as is their lineage relationship with SCs, even though they may feed, to some extent, into the SC compartment (4). There is much interest in understanding the cellular and molecular mechanisms underlying skeletal muscle regeneration in different contexts because such knowledge might help in the development of cell therapies for diseases characterized by skeletal muscle degeneration. These diseases include muscular dystrophy, the term for a group of inherited disorders characterized by progressive muscle wasting and weakness leading to a variable degree of mobility limitation, including confinement to a wheelchair and, in the most severe forms, heart and/or respiratory failure (5). Many muscular dystrophies arise from loss-of-function mutations in genes encoding cytoskeletal and membrane proteins, the most common Conflict of interest: The authors have declared that no conflict of interest exists. Citation for this article: J. Clin. Invest. 120:11–19 (2010). doi:10.1172/JCI40373. and severe being Duchenne muscular dystrophy (DMD), which is caused by mutations in the gene encoding dystrophin, an integral part of a complex that links the intracellular cytoskeleton with the extracellular matrix in muscle. Muscular dystrophies are some of the most difficult diseases to treat, as skeletal muscle is composed of large multinucleated fibers whose nuclei cannot divide. Consequently, cell therapy has to restore proper gene expression in hundreds of millions of postmitotic nuclei (6). In this Review, we discuss recent work indicating the possible existence of a stem/progenitor cell compartment in adult muscle (see also ref. 7) as well as studies related to the derivation of myogenic cells from embryonic and induced pluripotent stem cells (PSCs) for the development of new cell therapy strategies for diseases of skeletal muscle. An overview of clinical trials based upon transplantation of skeletal muscle stem cells is also provided. Neither the role of SCs in aging skeletal muscle nor the SC niche are discussed here due to space constraints, and readers are directed to excellent recent reviews on these topics by Suchitra Gopinath and Thomas Rando (8) and Michael Rudnicki and colleagues (9), respectively. SCs Identification and characterization. The most stringent way to classify cells as SCs remains by determining their anatomical location: SCs are found underneath the basal lamina of muscle fibers, closely juxtaposed to the plasma membrane (3). SCs originate from somites (10, 11), spheres of paraxial mesoderm that generate skeletal muscle, dermis, and axial skeleton, but the exact progenitor that gives rise to SCs remains to be identified. SCs are present in healthy adult mammalian muscle as quiescent cells and represent 2.5%–6% of all nuclei of a given muscle fiber. However, when activated by muscle injury, they can generate large numbers of new myofibers within just a few days (12). Quiescent SCs (13) express characteristic (although not unique) markers. In the mouse, the most widely used of these markers is the transcription factor paired box 7 (Pax7) (14), which is essential for SC specification and survival (15). In contrast, Pax3 is expressed only in quiescent SCs in a few specific muscle groups such as the diaphragm (16). The basic helix-loop-helix (bHLH) gene myogenic regulatory factor 5 (Myf5) is expressed in the large majority of quiescent SCs, and for this reason, mice expressing nuclear LacZ under the control of the Myf5 promoter (Myf5nlacZ/+ mice) have The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 11 review series Figure 1 Asymmetric cell division during activation of SCs. This is a drawing representing the anatomy of a muscle fiber with adjacent small vessels. SCs and other myogenic cells (pericytes and hematopoietic, endothelial, and interstitial cells) are depicted. SC activation in vivo is followed by an asymmetric division, with Pax7, MyoD, and Myf5 being expressed in differentiating cells and Pax7 in cells returning to quiescence in order to maintain a pool of progenitors (25, 41). been useful for identifying and characterizing SCs (17). Many other markers (18–29) have been identified and are listed in Table 1. Some of these surface markers are useful for isolating “purified” SC populations by cell sorting, but since each marker is not exclusively expressed on SCs, a combination of different markers must be used. Alternatively, transgenic mice such as those expressing GFP under the control of promoters that drive the expression of genes encoding SC markers — for example, the Pax3 promoter — can be used to isolate SCs (29–31). In humans, markers of both quiescent and activated SCs do not fully correspond to those in the mouse, and relatively little is known about them due to the difficulty of obtaining human tissue. For example, although CD34 is a marker of SCs in mice, it does not mark SCs in human muscle (32); and M-cadherin is not as consistent a marker of SCs in humans as it is in mice. Among the more reliable markers of SCs in human muscle is CD56, although it also marks natural killer lymphocytes (33). Activation. In response to a muscle injury, SCs are activated and start to proliferate; at this stage, they are often referred to as either myogenic precursor cells (mpc) or myoblasts (34, 35). Several signals, deriving both from damaged fibers and infiltrating cells, are involved in SC activation, including HGF (36), FGF (37), IGF (38), and NO (39). The progression of activated SCs toward myogenic differentiation is mainly controlled by Myf5 and myogenic differentiation 1 (MyoD) (17) and is followed by fusion into regenerating fibers. The whole process takes approximately 7 days in the mouse (40), during which time SCs undergo different fates, giving rise to a few Pax7+MyoD– cells, which return to quiescence (to maintain the progenitor pool), and many Pax7+MyoD+ cells, which are commit12 ted to differentiation (41) (Figure 1). Notch signaling is thought to regulate this process through promotion of asymmetric divisions, although there is not agreement on the role of Numb (a Notch inhibitor and cell-fate determinant) in inducing differentiation (42) and sustaining self renewal (43). The occurrence of asymmetric cell division is also supported by the identification of a subpopulation of SCs able to retain BrdU after pulse-chase labeling, with some cells displaying selective template DNA strand segregation during mitosis (43, 44). In addition, Rudnicki and colleagues validated the label-retention model of SCs and demonstrated that approximately 10% of Pax7+ mouse SCs had never expressed Myf5 and that these cells remain adherent to the basal lamina during asymmetric mitosis, generating one Pax7+Myf5– satellite “stem cell” and one Pax7+Myf5+ SC “progenitor,” eventually destined to differentiate (25) (Figure 1). The same group also elegantly described Wnt7a as regulating the symmetric expansion of Pax7+Myf5– SCs (45). Transplantation. Because of their features, SCs were considered obvious candidates for a cell-therapy approach to treating muscular dystrophy. Pioneer studies demonstrated that intramuscular injection of normal myoblasts (46) into mdx mice, which lack dystrophin and are a model for DMD, resulted in fusion with host fibers and extensive dystrophin production. These studies led to several clinical trials in the early 1980s (see Muscular dystrophies section for details) that failed for a number of reasons, including poor survival and migration of donor cells after transplantation and rejection of the donor cells due to an immune response by the patients (47). Many subsequent preclinical studies aimed to improve the survival, proliferation, and differentiation of the SCs after engraftment. For example, transplantation in dystrophic mouse mus- The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 review series Table 1 SC markers Marker SC expression Localization Function Prospective isolationA Pax7 100% of quiescent and Nucleus Transcription Pax7-GFP activated SCs factor Pax3 Quiescent SCs Nucleus Transcription Pax3-GFP (only in a subset of muscles) factor Myf5 Most quiescent SCs and all Nucleus Transcription Myf5-nLacZ proliferating SCs and myoblasts factor Syndecan-3 and -4 98% of quiescent Membrane Transmembrane Cell and activated SCs heparan sulfate sorting proteoglycan VCAM-1 Quiescent and activated SCs Membrane Adhesion Cell molecule sorting c-met Quiescent and activated SCs Membrane HGF receptor Not used Foxk1 Quiescent and activated SCs Nucleus Nuclear factor Not used Cd34 Quiescent and activated SCs Membrane Membrane Cell protein sorting M-cadherin Quiescent and activated SCs; Membrane Adhesion Not used myoblasts protein Caveolin-1 Quiescent and activated SCs; Membrane Membrane Not used myoblasts protein α7 Integrin Quiescent and activated SCs; Membrane Adhesion Cell myoblasts protein sorting β1 Integrin Quiescent and activated SCs Membrane Adhesion Cell protein sorting Cd56 Quiescent and activated SCs; Membrane Homophilic binding Cell myoblasts glycoprotein sorting SM/C2.6B Quiescent and activated SCs; Unknown Unknown Cell myoblasts sorting Cxcr4 Subset of quiescent SCs Membrane SDF1 Cell receptor sorting Nestin Around 98% of quiescent SCs Intermediate Intermediate Nestin and myoblasts filament filament protein GFP Expression in other tissues/cells Ref Absent 14 Melanocyte stem cells Absent 16 Brain, dermis, BM, bone, smooth muscle, tumors Activated endothelial cells 18 Many tissues and tumors Neurons Hematopoietic, endothelial, mast, and dendritic cells Absent Endothelial fibrous and adipose tissue Vessel-associated cells Many tissues Glia, neurons, and natural killer cells Unknown HSCs, vascular endothelial cells, and neuronal cells Neuronal precursor cells 17 19 20 21 13 22 23 24 25 26 27 28 29 AProspective BNovel isolation: direct isolation of cells from tissue, usually based upon cytofluorimetric sorting with antibodies directed against cell surface markers. monoclonal antibody directed against an unknown antigen present on SCs. Foxk1, forkhead box k1. cles of a single muscle fiber that contained as few as seven SCs led to an increasing number of new SCs that in turn generated more than 100 new muscle fibers and could also be activated after injury (48). This is a much more efficient way to generate new muscle fibers than transplantation of cultured SCs, in which normally the number of donor-derived new fibers that are generated is several orders of magnitude less than the number of injected cells. Unfortunately, this method would be difficult to translate into clinical protocols. In the past few years, several groups have succeeded in prospectively isolating “pure” populations of SCs by using a combination of different markers (Table 1), such as Pax3-GFP (30), CXCR4 and β1 integrin (49), α7 integrin and CD34 (50), or syndecan-3 and -4 (51). It is still unknown whether the different protocols allow isolation of the same cell population, enriched to different extents for a more primitive “stem-like” fraction. However, all these studies revealed that freshly isolated cells have a much greater capacity to generate dystrophin-expressing fibers in mdx mice than the same cells after in vitro expansion (30); the simplest explanation for this is that the “stem” fraction either dies in culture or generates differentiation-committed SCs. Importantly, a short culture period of 3–4 days, without subculture, allowed lentiviral-mediated transduction and thus genetic correction of SCs freshly isolated from mdx mice, without compromising myogenic potency in vivo (52). Despite these encouraging results, previous unsolved problems still prevent the use of SCs to systemically treat patients with muscular dystrophy: in particular, the inability of these cells to cross the endothelial wall makes systemic delivery impossible and prevents possible healing of the diaphragm and cardiac muscles, both critical for patient survival (34). Other myogenic progenitors The availability of cell-autonomous, tissue-specific transgenic markers allowed the unequivocal demonstration of the existence of myogenic progenitors originating from tissues other than skeletal muscle (53). Upon transplantation (either BM transplantation [BMT] or direct injection into skeletal muscle), these cells, identified by transgene expression, participate in muscle regeneration in wild-type and/or dystrophic mice (Figures 1 and 2; Table 2) and eventually enter the SC pool. The possibility that myogenic differentiation may depend upon fusion (and hence exposure to the dominant activity of MyoD) remains, but for skeletal muscle, this would be part of the physiological mechanism that creates the tissue. Below, we describe briefly some examples of these unorthodox myogenic cells. The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 13 review series cal β-gal–positive nuclei were detected in regenerated fibers, demonstrating that murine BM contains transplantable progenitors that can be recruited to an injured muscle through the circulation, where they participate in muscle repair (56). This opened the possibility of treating muscular dystrophy by BMT, but work in mice indicated that, unfortunately, the frequency of this event was too low, even in a chronically regenerating dystrophic muscle and even if the side population (SP) progenitor–enriched fraction was transplanted (57, 58). To address this issue, subsequent experiments were directed to identifying a rare, potentially highly myogenic progenitor, but those studies have so far had modest success. The hematopoietic, CD45+ fraction of the BM has been identified as the cell population with myogenic potential (59), and retrospective analysis in a DMD patient that had undergone BMT confirmed the persistence of donorderived skeletal muscle cells over a period of many years, again at very low frequency (60). Together, these data suggested that HSCs or a yet-to-be-identified cell that expresses several markers in common with true HSCs has myogenic potential. More recent approaches confirmed that hematopoietic cells have myogenic potential but disagreed on the stage at which myogenic differentiation would occur. One study reported that the progeny of a single mouse hematopoietic progenitor cell can both reconstitute the hematopoietic system and contribute, at low frequency, to muscle regeneration (61). However, a similar study showed that in response to injury, CD45+ hematopoietic progenitors contribute to regenerating mouse skeletal muscle through fusion of mature myeloid cells rather than fusion of the HSCs (62). A subpopulation of circulating cells expressing CD133 (also known as Ac133), a well-characterized marker of HSCs, also expresses early myogenic markers (63). When injected into the circulation of dystrophic scid/mdx mice, CD133+ cells have been found to contribute to muscle repair, recovery of force, and replenishment of Figure 2 the SC pool. The group that discovered this also isolatDerivation of different stem cells for possible use in skeletal muscle regeneration. The figure shows a flow chart with the sources of different myogenic stem cells. ed a population of muscle-derived stem cells (MDSCs) + The possibility of deriving the same cells from a reprogrammed cell of the dermis expressing CD133 (64). Furthermore, when CD133 is also outlined (blue boxes and arrows). Once obtained, the stem cells can be cells from DMD patients were genetically corrected by lentivirus-mediated exon skipping for dystrophin exon characterized, expanded, genetically corrected, and transplanted. 51, these cells were able to mediate morphological and functional recovery in scid/mdx mice (64). Thus, differCells from ectoderm: neural stem cells. To date, neural stem cells ent subpopulations of hematopoietic cells, whose characterization (both murine and human) are the only ectoderm-derived stem cells is still incomplete, seem to possess myogenic potency, but none of that have been shown to differentiate into skeletal muscle when these exhibit this property at high frequency. cocultured with skeletal myoblasts or transplanted into regeneratCells derived from mesoderm (other than hematopoietic cells). Many difing skeletal muscle (54). Interestingly, cells expressing Myf5 exist ferent types of mesoderm stem/progenitor cells have been shown in the brain and spinal cord, suggesting a cryptic potency that to exhibit myogenic potential, usually after drug treatment, genetbecomes apparent in vitro (55). ic modification, or coculture with SCs or myoblasts. In some cases, Hematopoietic cells. The first evidence of in vivo generation of evidence of in vivo myogenesis has been documented. The list of skeletal muscle from BM cells was reported in 1998 (56) in a such cells includes mesenchymal stem cells (MSCs), multipotent study that used transgenic mice expressing a nuclear LacZ under adult progenitor cells (MAPCs), MDSCs, CD133+ cells, mesoanthe control of the striated muscle promoter myosin light chain gioblasts (MABs), endothelial progenitor cells (EPCs), and adi1/3 fast (MLC3f). After transplantation of BM from the trans- pose-derived stem cells, all of which are briefly described below or genic mice and subsequent injury to the host muscle, unequivo- in Table 2. More details can be found in previous reviews (6, 32). 14 The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 review series Table 2 Properties of myogenic progenitors other than SCs Cell type CD133+ EPC HSC MAB MADS MAPC MDSC MEC MSC Origin Proliferation In vitro myogenic differentiation Dystrophin expression in vivo Systemic delivery Ref Blood/skeletal muscle Vessel wall BM Vessel wall Adipose tissue Vessel wall Skeletal muscle Vessel wall Vessel wall Low/high Low Low High High High High High High Induced by muscle cells /spontaneous Induced by muscle cells Induced by muscle cells Induced by muscle cells/spontaneous Spontaneous Induced by Aza-cytidine Induced by muscle cells Spontaneous Induced by Aza-cytidine Yes Not tested Yes Yes Yes Not tested Yes Yes Yes Yes Yes Yes Yes Not done Not done Not done Not done Not done 64 71 56 68 74 32 32 73 65 MEC, myoendothelial cell. MSCs have been shown to be capable of skeletal myogenesis (65). However, recently, Perlingeiro and colleagues demonstrated that although Pax3 activation enabled the in vitro differentiation of murine and human MSCs into MyoD+ myogenic cells, these cells failed to cause functional muscle recovery in mdx mice, despite good engraftment (66). The reason for this failure remains unclear. MABs are vessel-associated progenitors (67) that express early endothelial markers when isolated from the embryo and pericyte markers when isolated from postnatal tissues. Since MABs are able to cross the vessel wall and are easily transduced with lentiviral vectors, they have been used in preclinical models of cell therapy for muscular dystrophy. Intraarterial delivery of either wild-type or genetically corrected MABs morphologically and functionally ameliorated the dystrophic phenotype of mice lacking α-sarcoglycan (Sgca), which model limb-girdle muscular dystrophy 2D, a muscular dystrophy caused by mutations in the SGCA gene (68). In addition, intraarterial delivery of wild-type postnatal canine MABs resulted in extensive recovery of dystrophin expression and ameliorated pathologic muscle morphology and function in golden retriever dogs that model DMD (69). Similar cells have been isolated from human postnatal skeletal muscle and shown to represent a subset of pericytes and to give rise to dystrophin-positive muscle fibers when transplanted into scid/mdx mice (70). Based on these studies, a phase I clinical trial with MAB allotransplantation in DMD patients is planned for the near future. Initially identified as circulating cells expressing CD34 and fetal liver kinase-1 (Flk-1; also known as VEGFR2), EPCs (71) were shown to be transplantable and to participate actively in angiogenesis in various physiologic and pathologic conditions (71). It was then shown that freshly isolated human cord blood CD34+ cells injected into ischemic adductor muscles gave rise not only to endothelial but also to skeletal muscle cells in mice (72). Consistent with this, Péault and colleagues have identified cells with high myogenic potential within the vascular endothelium of human adult skeletal muscle (73). These human myoendothelial cells, which represent less than 0.5% of the cells in dissociated adult skeletal muscles, expressed both myogenic and endothelial cell markers (CD56 +CD34 +CD144 +CD45 –), exhibited long term proliferation, had a normal karyotype, and when transplanted into scid mice, were able to regenerate fibers in injured muscle (73). Human multipotent adipose-derived stem (hMADS) cells, isolated from adipose tissue, differentiate into adipocytes, osteo blasts, and myoblasts (74). Recently, the myogenic and muscle repair capacities of hMADS cells have been enhanced by transient expression of MyoD (75). The easy availability of their tissue source, their strong capacity for expansion ex vivo, their multipotent differentiation, and their immune-privileged behavior suggest that hMADS cells could be an important tool for cell-mediated therapy for skeletal muscle disorders. PSCs for muscle regeneration PSCs can give rise to all cell types. Among the various PSCs, we limit our discussion to ES cells (76, 77) and induced pluripotent stem (iPS) cells (78), as they are, in practice, the two types of PSC most commonly used to direct differentiation toward a given cell type, skeletal muscle, for the purpose of this Review. PSCs hold tremendous hopes for the cell therapy of degenerative diseases; and iPS cells further offer the possibility of deriving patient-specific PSCs (79) to study diseases in vitro (80) and the potential for genetic correction for autologous cell therapy. Turning PSCs into skeletal muscle. A critical step in establishing the potential of PSCs as a therapeutic for skeletal muscle diseases is the development of techniques for the differentiation of these cells into tissue-specific progenitors suitable for transplantation. The most elegant way to obtain specific transplantable cell types is by exposing them in vitro to the same molecules that control their in vivo commitment during embryogenesis (reviewed in ref. 81), although the empirical testing of molecules and substrates could generate equally useful cells. Seminal studies from the mid-1990s described how ES cell– derived embryoid bodies (EBs), tridimensional structures formed when ES cells are grown in the absence of an embryonic fibroblast feeder layer, contained multinucleated muscle fibers that express skeletal muscle myosin heavy-chain genes (82, 83). Ten years later, it was documented for the first time in vivo that intramuscular injection of mouse EBs cocultured with mdx muscle–derived progenitors in mdx mice led to the production of a few clusters of donor-derived dystrophin-positive fibers (84). Recently, Studer and coworkers have derived transplantable myoblasts from human ES cells (85), while Chang and colleagues have generated transplantable satellite-like cells from mouse ES cells (86). Upon transplantation into mdx mice, the latter cells have been found to regenerate acutely and chronically injured muscle and could also be secondarily transplanted (86). MyoD-mediated myogenic conversion of ES cell–derived cells is another intriguing approach, The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 15 review series Table 3 Clinical trials of myoblast transplantation in DMD patients Author Year Blind Number Pt age of pts Drugs administered Number of myoblasts (×106) Expression of dystrophin Law 1991 No 3 9–10 Cyclosporine 8 3/3 by IHC and WB at 3 months Huard 1992 No 4 13–20 None 25 3/4 by IHC and WB at 4 months (up to 80% fibers positive) Law 1992 No 21 6–14 Cyclosporine 5000 Not evaluated Gussoni 1992 Yes 8 6–10 None 100 3/8 by PCR at 1 month Karpati 1993 Yes 8 6–10 Cyclophosphamide 55 0/8 by IHC at 1-year follow-up Tremblay 1993 Yes 5 4–9 None 100–240 3/4 by IHC at 1 month (up to 36% fibers positive); 1/4 by IHC at 6 months Morandi 1995 Yes 3 6–9 Cyclosporine 55 0/3 by IHC or PCR Mendell 1995 Yes 12 5–9 Cyclosporine 110 1/12 by IHCA Miller 1997 Yes 10 5–10 Cyclosporine 100 3/10 by PCR at 1 month; 1/10 at 6 months Neumeyer 1998 Yes 6B >21 Cyclosporine 73-100 0/3, by IHCC Skuk 2006 Yes 9 8–17 Tacrolimus 30 8/9 patients by IHCC 3.5%–26% fibers expressed dystrophin Improved strength Ref 3/3 patients at 3 months 3/4 patients 95 43% muscles examined improved Not evaluated 3/8 98 99 97 0/5 100 Not evaluated 0/12 0/10 101 102 103 0/3 Not evaluated 94 104 96 Pt, patient; IHC, immunohistochemistry; WB, Western blot. AMultiple injection site protocol. BStudy performed with Becker muscular dystrophy patients. CIHC was performed using a peptide-specific antibody recognizing only donor dystrophin. whose proof of principle dates to the early 1990s (87). On this front, Perlingeiro and colleagues recently achieved in vivo skeletal muscle differentiation from purified PDGFRα+Flk1– progenitors isolated from EBs generated from mouse ES cells containing an inducible Pax3 gene (88). At the time of writing, there are no reports on the generation of myogenic cells from iPS cells, but assuming that the present protocols for ES cells can be adapted for iPS cells, we believe that in the next months, papers on this topic are likely to appear. Muscle regeneration from ES/iPS cells via mesodermal progenitors. There is a large body of evidence indicating the existence of nonconventional muscle progenitors (see Other myogenic progenitors). Thus, the possibility of deriving mesodermal myogenic progenitors (89) from ES/iPS cells offers an alternative route for cell therapy for skeletal muscle regeneration. This approach has the advantage of producing myogenic progenitors that could be delivered systemically through the circulation. In 2005, Studer and colleagues described the derivation from ES cells of mesenchymal precursors (90) that differentiated in vitro into different mesodermal lineages, including skeletal muscle. A recent article described the derivation from mouse ES cells of PDGFRα+ mesodermal progenitors that, after in vivo transplantation, expressed markers of SCs and contributed to muscle regeneration (91). Issues to be solved. Despite the excitement for these novel strategies for treating degenerative muscular conditions, a number of safety and efficacy issues, some common to other cells (immunogenicity, survival, and differentiation), some prominent for ES/iPS cells, such as tumor formation, still need to be solved. For example, cytofluorimetric purification of committed progenitors (88) may dramatically decrease the possibility of transplanting undifferentiated tumorigenic cells. The use of standardized protocols for generat16 ing iPS cells (92), together with stringent tumorigenic assays for the derived cell types, will certainly be a fundamental step toward their clinical application. Skeletal muscle stem cells: past and ongoing clinical trials Until now only SCs, and, to a very minor extent, CD133+ cells have been used in human clinical trials. The pathologies treated include forms of muscular dystrophy, heart failure associated with myocardial infarction (HFMI), and stress urinary incontinence (SUI). Muscular dystrophies. In 1990, Peter Law and collaborators reported the first SC transplant in a 9-year-old boy affected by DMD, showing safety and dystrophin production (93). Soon after, 11 clinical trials in DMD patients were conducted using intramuscular injection of SCs (Table 3) (94–104). Although there were no adverse effects, new dystrophin production was demonstrated in many but not all cases and clinical benefit in none (6, 105). This is not surprising considering that intramuscular injection in several locations of a single muscle (or at most a few muscles) cannot elicit a general effect, although improved strength of the injected muscles was detected in 15% of the patients treated. Treating muscular dystrophies by intramuscular injection of myoblasts presents several problems that have not been solved yet. First, intramuscular injected cells distribute locally, implying that a huge number of injections will have to be performed in order to treat a complete muscle (96). Second, immune responses toward the injected SCs have been described, even in the case of major histocompatibility locus coincidence. Finally, the rapid death of most of the SCs in the first 72 hours after injection has been extensively reported (106, 107). Subsequent experimentation has been devoted to solving these problems, and a phase I clinical trial has been com- The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 review series pleted (104). Although encouraging results have been obtained, this method is still limited by the impossibility of delivering myoblasts systemically through the circulation. Recently, Torrente and colleagues reported the first CD133+ cell transplant (108). They designed a phase I double-blind trial with an autologous transplant of unmodified, and thus still dystrophic, muscle-derived CD133+ cells in 8 boys affected by DMD exclusively to test safety; and indeed, no adverse events were reported. Heart failure. Among different options to treat heart infarction, skeletal muscle–derived myoblasts were considered an optimal cell therapy, as they can be easily obtained from the same patient (avoiding the need for long-term immune suppression), rapidly expanded in vitro, and transplanted back in the patient heart. Preclinical experiments performed in animal models demonstrated their ability to engraft correctly, survive in postinfarction scars, differentiate into contractile skeletal muscle cells, and improve heart function (109–111), possibly also because they release angiogenic factors. Unfortunately, in these models, myoblasts were not able to differentiate into cardiomyocytes and did not integrate electrically with the host cardiomyocytes (112, 113). Despite this significant problem, several nonrandomized clinical trials using myoblasts to treat the infarcted heart demonstrated thickening of the LV, an increase in LV ejection fraction, and prevention of LV dilatation, with clinical improvement in some patients (114–120). Histological analysis showed the presence of new myofibers in the scar zone expressing skeletal muscle–specific myosin heavy chain (121). In general, the transplantation procedure was clinically well tolerated, but a high incidence of arrhythmias, some of which were fatal, was reported. In 2007, the results of the MAGIC study, an international phase II double-blind trial were published (122). Ninety-seven patients affected by HFMI were randomized to receive placebo, a low dose of myoblasts (400 × 106 cells), or a high dose of myoblasts (800 × 106 cells). LV end-diastolic volumes decreased substantially in patients receiving myoblasts, supporting a role for myoblasts in remodeling of the heart muscle. The incidence of secondary events, including arrhythmias, was not different between the groups, although all patients received the antiarrhythmic agent amiodarone and were implanted with a cardioverter defibrillator. The results of the CAuSMIC (123) and SEISMIC trials (124) have demonstrated safety and some clinical improvement. Ongoing trials include the MARVEL study, a phase II/III randomized, double-blind, placebo-controlled trial for which results are expected during the fall of 2009. SUI. SC-derived myoblasts have also been used as cell therapy for individuals with SUI, which is characterized by the loss of small amounts of urine upon coughing, laughing, sneezing, exercising, or other movements that increase intraabdominal pressure. Myoblasts may represent an interesting approach for the treatment of this disease (125), as the main cause of SUI is impaired tone of the urethral smooth and striated muscle, which is associated with atrophy of the supporting structures of the urethra, the mucosa, and 1.Carlson BM. The regeneration of skeletal muscle. A review. Am J Anat. 1973;137(2):119–149. 2.Mintz B, Baker WW. Normal mammalian muscle differentiation and gene control of isocitrate dehydrogenase synthesis. Proc Natl Acad Sci U S A. 1967;58(2):592–598. 3.Mauro A. Satellite cells of skeletal muscle fibers. J Biophys Biochem Cytol. 1961;9:493–495. 4.Cossu G, Biressi S. Satellite cells, myoblasts and other occasional myogenic progenitors: possible vascular submucosa (126, 127). Treatment of SUI with SC-derived myoblasts has been performed using two different strategies: injection of autologous myoblasts to improve the sphincter tone; and injection of myoblasts together with fibroblasts (isolated by differential adhesion from the same muscle biopsy) to both improve sphincter tone and treat mucosa atrophy. Until now, all published studies have been nonrandomized, open studies, demonstrating a remarkable clinical improvement in most of the patients treated (127, 128). Moreover, structural and functional techniques have demonstrated thickening of the urinary sphincter and an increase in maximum urethral closure pressure. The onset of improvement is not immediate and may be delayed up to six months after cell injection; however, the benefit, once obtained, lasts for a long period of time, at least up to 12 months (126). Cystoscopic studies have not demonstrated overgrowth of myoblasts nor obstruction of the lower urinary tract (129). Unfortunately, these successful results have not been confirmed yet in a randomized study. Conclusions This is an exciting period for those studying the biology of skeletal muscle stem cells and seeking to harness the information for clinical applications. By learning more about SCs and other mesodermal skeletal muscle progenitors, we can learn how to better use them to repair muscle. The main limitations of SCs are loss of “stemness” upon culture and an inability to cross the vessel wall for systemic delivery. Limitations for other cell types are incomplete characterization and their overall minor myogenic potency. Nevertheless, a phase I clinical trial with donor-derived MABs is planned for the end of 2010, given the fact that, because of extensive preclinical work, these cells appear at the moment as the best candidates for the cell therapy of muscular dystrophy. Moreover, the terrific prospective of deriving countless autologous, genetically corrected iPS cells from patients certainly will set the stage for future cell therapies. Finally, we should remember that SCs have already been used for clinical trials for DMD, myocardial infarction, and SUI. The last seems to be a case of success for this approach and a situation from which we may also learn in order to redirect efforts toward therapies for myocardial infarction and muscular dystrophies, which have thus far been less successful. Acknowledgments Work in the authors’ laboratory is supported by grants from the European Community (OptiStem, Angioscaff, and MyoAmp), European Research Council, Téléthon, Association Française contre les Myopathies, Duchenne Parent Project, CureDuchenne, and the Italian Ministries of Research (FIRB) and Health. Address correspondence to: Giulio Cossu, Division of Regenerative Medicine, San Raffaele Scientific Institute, 58 via Olgettina, 20132 Milan, Italy. Phone: 39-02-2643-4954; Fax: 39-02-2643-4621; E-mail: cossu.giulio@hsr.it. origin, phenotypic features and role in muscle regeneration. Semin Cell Dev Biol. 2005;16(45):623–631. 5.Emery AE. The muscular dystrophies. Lancet. 2002;359(9307):687–695. 6.Cossu G, Sampaolesi M. New therapies for Duchenne muscular dystrophy: challenges, prospects and clinical trials. Trends Mol Med. 2007;13(12):520–526. 7.Buckingham M, Montarras D. Skeletal muscle stem cells. Curr Opin Genet Dev. 2008;18(4):330–336. 8.Gopinath SD, Rando TA. Stem cell review series: aging of the skeletal muscle stem cell niche. Aging Cell. 2008;7(4):590–598. 9.Kuang S, Gillespie MA, Rudnicki MA. Niche regulation of muscle satellite cell self-renewal and differentiation. Cell Stem Cell. 2008;2(1):22–31. 10.Shi X, Garry DJ. Muscle stem cells in development, regeneration, and disease. Genes Dev. 2006;20(13):1692–1708. 11.Sambasivan R, Tajbakhsh S. Skeletal muscle The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 17 review series stem cell birth and properties. Semin Cell Dev Biol. 2007;18(6):870–882. 12.Whalen RG, Harris JB, Butler-Browne GS, Sesodia S. Expression of myosin isoforms during notexininduced regeneration of rat soleus muscles. Dev Biol. 1990;141(1):24–40. 13.Beauchamp JR, et al. Expression of CD34 and Myf5 defines the majority of quiescent adult skeletal muscle satellite cells. J Cell Biol. 2000;151(6):1221–1234. 14.Zammit PS, et al. Pax7 and myogenic progression in skeletal muscle satellite cells. J Cell Sci. 2006;119(pt 9):1824–1832. 15.Kuang S, Charge SB, Seale P, Huh M, Rudnicki MA. Distinct roles for Pax7 and Pax3 in adult regenerative myogenesis. J Cell Biol. 2006;172(1):103–113. 16.Relaix F, et al. Pax3 and Pax7 have distinct and overlapping functions in adult muscle progenitor cells. J Cell Biol. 2006;172(1):91–102. 17.Tajbakhsh S, et al. Gene targeting the myf-5 locus with nlacZ reveals expression of this myogenic factor in mature skeletal muscle fibres as well as early embryonic muscle. Dev Dyn. 1996;206(3):291–300. 18.Cornelison DD, Filla MS, Stanley HM, Rapraeger AC, Olwin BB. Syndecan-3 and syndecan-4 specifically mark skeletal muscle satellite cells and are implicated in satellite cell maintenance and muscle regeneration. Dev Biol. 2001;239(1):79–94. 19.Rosen GD, et al. Roles for the integrin VLA-4 and its counter receptor VCAM-1 in myogenesis. Cell. 1992;69(7):1107–1119. 20.Cornelison DD, Wold BJ. Single-cell analysis of regulatory gene expression in quiescent and activated mouse skeletal muscle satellite cells. Dev Biol. 1997;191(2):270–283. 21.Garry DJ, et al. Myogenic stem cell function is impaired in mice lacking the forkhead/winged helix protein MNF. Proc Natl Acad Sci U S A. 2000;97(10):5416–5421. 22.Irintchev A, Zeschnigk M, Starzinski-Powitz A, Wernig A. Expression pattern of M-cadherin in normal, denervated, and regenerating mouse muscles. Dev Dyn. 1994;199(4):326–337. 23.Gnocchi VF, White RB, Ono Y, Ellis JA, Zammit PS. Further characterisation of the molecular signature of quiescent and activated mouse muscle satellite cells. PLoS One. 2009;4(4):e5205. 24.Blanco-Bose WE, Yao CC, Kramer RH, Blau HM. Purification of mouse primary myoblasts based on alpha 7 integrin expression. Exp Cell Res. 2001;265(2):212–220. 25.Kuang S, Kuroda K, Le Grand F, Rudnicki MA. Asymmetric self-renewal and commitment of satellite stem cells in muscle. Cell. 2007;129(5):999–1010. 26.Betsholtz C. Insight into the physiological functions of PDGF through genetic studies in mice. Cytokine Growth Factor Rev. 2004;15(4):215–228. 27.Fukada S, Higuchi S, Segawa M, Koda K, Yamamoto Y, et al. Purification and cell-surface marker characterization of quiescent satellite cells from murine skeletal muscle by a novel monoclonal antibody. Exp Cell Res. 2004;296(2):245–255. 28.Sherwood RI, et al. Isolation of adult mouse myogenic progenitors: functional heterogeneity of cells within and engrafting skeletal muscle. Cell. 2004;119(4):543–554. 29.Day K, Shefer G, Richardson JB, Enikolopov G, Yablonka-Reuveni Z. Nestin-GFP reporter expression defines the quiescent state of skeletal muscle satellite cells. Dev Biol. 2007;304(1):246–259. 30.Montarras D, et al. Direct isolation of satellite cells for skeletal muscle regeneration. Science. 2005;309(5743):2064–2067. 31.Bosnakovski D, et al. Prospective isolation of skeletal muscle stem cells with a Pax7 reporter. Stem Cells. 2008;26(12):3194–3204. 32.Peault B, et al. Stem and progenitor cells in skeletal muscle development, maintenance, and therapy. Mol Ther. 2007;15(5):867–877. 18 33.Illa I, Leon-Monzon M, Dalakas MC. Regenerating and denervated human muscle fibers and satellite cells express neural cell adhesion molecule recognized by monoclonal antibodies to natural killer cells. Ann Neurol. 1992;31(1):46–52. 34.Price FD, Kuroda K, Rudnicki MA. Stem cell based therapies to treat muscular dystrophy. Biochim Biophys Acta. 2007;1772(2):272–283. 35.Dhawan J, Rando TA. Stem cells in postnatal myogenesis: molecular mechanisms of satellite cell quiescence, activation and replenishment. Trends Cell Biol. 2005;15(12):666–673. 36.Tatsumi R, Anderson JE, Nevoret CJ, Halevy O, Allen RE. HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells. Dev Biol. 1998;194(1):114–128. 37.Floss T, Arnold HH, Braun T. A role for FGF-6 in skeletal muscle regeneration. Genes Dev. 1997;11(16):2040–2051. 38.Musaro A. Growth factor enhancement of muscle regeneration: a central role of IGF-1. Arch Ital Biol. 2005;143(3-4):243–248. 39.Wozniak AC, Anderson JE. Nitric oxide-dependence of satellite stem cell activation and quiescence on normal skeletal muscle fibers. Dev Dyn. 2007;236(1):240–250. 40.Zammit PS, et al. Kinetics of myoblast proliferation show that resident satellite cells are competent to fully regenerate skeletal muscle fibers. Exp Cell Res. 2002;281(1):39–49. 41.Zammit PS, et al. Muscle satellite cells adopt divergent fates: a mechanism for self-renewal? J Cell Biol. 2004;166(3):347–357. 42.Conboy IM, Rando TA. The regulation of Notch signaling controls satellite cell activation and cell fate determination in postnatal myogenesis. Dev Cell. 2002;3(3):397–409. 43.Shinin V, Gayraud-Morel B, Gomes D, Tajbakhsh S. Asymmetric division and cosegregation of template DNA strands in adult muscle satellite cells. Nat Cell Biol. 2006;8(7):677–687. 44.Conboy MJ, Karasov AO, Rando TA. High incidence of non-random template strand segregation and asymmetric fate determination in dividing stem cells and their progeny. PLoS Biol. 2007;5(5):e102. 45.Le Grand F, Jones AE, Seale V, Scime A, Rudnicki MA. Wnt7a activates the planar cell polarity pathway to drive the symmetric expansion of satellite stem cells. Cell Stem Cell. 2009;4(6):535–547. 46.Partridge TA, Morgan JE, Coulton GR, Hoffman EP, Kunkel LM. Conversion of mdx myofibres from dystrophin-negative to -positive by injection of normal myoblasts. Nature. 1989;337(6203):176–179. 47.Cossu G, Sampaolesi M. New therapies for muscular dystrophy: cautious optimism. Trends Mol Med. 2004;10(10):516–520. 48.Collins CA, et al. Stem cell function, self-renewal, and behavioral heterogeneity of cells from the adult muscle satellite cell niche. Cell. 2005;122(2):289–301. 49.Cerletti M, et al. Highly efficient, functional engraftment of skeletal muscle stem cells in dystrophic muscles. Cell. 2008;134(1):37–47. 50.Sacco A, Doyonnas R, Kraft P, Vitorovic S, Blau HM. Self-renewal and expansion of single transplanted muscle stem cells. Nature. 2008;456(7221):502–506. 51.Tanaka KK, et al. Syndecan-4-expressing muscle progenitor cells in the SP engraft as satellite cells during muscle regeneration. Cell Stem Cell. 2009;4(3):217–225. 52.Ikemoto M, et al. Autologous transplantation of SM/C-2.6(+) satellite cells transduced with microdystrophin CS1 cDNA by lentiviral vector into mdx mice. Mol Ther. 2007;15(12):2178–2185. 53.Cossu G. Unorthodox myogenesis: possible developmental significance and implications for tissue histogenesis and regeneration. Histol Histopathol. 1997;12(3):755–760. 54.Galli R, et al. Skeletal myogenic potential of human and mouse neural stem cells. Nat Neurosci. 2000;3(10):986–991. 55.Tajbakhsh S, et al. A population of myogenic cells derived from the mouse neural tube. Neuron. 1994;13(4):813–821. 56.Ferrari G, et al. Muscle regeneration by bone marrow-derived myogenic progenitors. Science. 1998;279(5356):1528–1530. 57.Gussoni E, et al. Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature. 1999;401(6751):390–394. 58.Ferrari G, Stornaiuolo A, Mavilio F. Failure to correct murine muscular dystrophy. Nature. 2001;411(6841):1014–1015. 59.McKinney-Freeman SL, et al. Muscle-derived hematopoietic stem cells are hematopoietic in origin. Proc Natl Acad Sci U S A. 2002;99(3):1341–1346. 60.Gussoni E, et al. Long-term persistence of donor nuclei in a Duchenne muscular dystrophy patient receiving bone marrow transplantation. J Clin Invest. 2002;110(6):807–814. 61.Corbel SY, et al. Contribution of hematopoietic stem cells to skeletal muscle. Nat Med. 2003;9(12):1528–1532. 62.Camargo FD, Green R, Capetanaki Y, Jackson KA, Goodell MA. Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat Med. 2003;9(12):1520–1527. 63.Torrente Y, et al. Human circulating AC133(+) stem cells restore dystrophin expression and ameliorate function in dystrophic skeletal muscle. J Clin Invest. 2004;114(2):182–195. 64.Benchaouir R, et al. Restoration of human dystrophin following transplantation of exon-skippingengineered DMD patient stem cells into dystrophic mice. Cell Stem Cell. 2007;1(6):646–657. 65.Dezawa M, et al. Bone marrow stromal cells generate muscle cells and repair muscle degeneration. Science. 2005;309(5732):314–317. 66.Gang EJ, et al. Engraftment of mesenchymal stem cells into dystrophin-deficient mice is not accompanied by functional recovery. Exp Cell Res. 2009;315(15):2624–2636. 67.Minasi MG, et al. The meso-angioblast: a multipotent, self-renewing cell that originates from the dorsal aorta and differentiates into most mesodermal tissues. Development. 2002;129(11):2773–2783. 68.Sampaolesi M, et al. Cell therapy of alpha-sarcoglycan null dystrophic mice through intraarterial delivery of mesoangioblasts. Science. 2003;301(5632):487–492. 69.Sampaolesi M, et al. Mesoangioblast stem cells ameliorate muscle function in dystrophic dogs. Nature. 2006;444(7119):574–579. 70.Dellavalle A, et al. Pericytes of human skeletal muscle are myogenic precursors distinct from satellite cells. Nat Cell Biol. 2007;9(3):255–267. 71.Asahara T, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science. 1997;275(5302):964–967. 72.Pesce M, et al. Myoendothelial differentiation of human umbilical cord blood-derived stem cells in ischemic limb tissues. Circ Res. 2003;93(5):e51–e62. 73.Zheng B, et al. Prospective identification of myogenic endothelial cells in human skeletal muscle. Nat Biotechnol. 2007;25(9):1025–1034. 74.Meliga E, Strem BM, Duckers HJ, Serruys PW. Adipose-derived cells. Cell Transplant. 2007;16(9):963–970. 75.Goudenege S, et al. Enhancement of myogenic and muscle repair capacities of human adipose-derived stem cells with forced expression of MyoD. Mol Ther. 2009;17(6):1064–1072. 76.Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature. 1981;292(5819):154–156. 77.Thomson JA, et al. Embryonic stem cell lines derived from human blastocysts. Science. The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 review series 1998;282(5391):1145–1147. 78.Yamanaka S. A fresh look at iPS cells. Cell. 2009;137(1):13–17. 79.Nishikawa S, Goldstein RA, Nierras CR. The promise of human induced pluripotent stem cells for research and therapy. Nat Rev Mol Cell Biol. 2008;9(9):725–729. 80.Park IH, et al. Disease-specific induced pluripotent stem cells. Cell. 2008;134(5):877–886. 81.Murry CE, Keller G. Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell. 2008;132(4):661–680. 82.Robbins J, Gulick J, Sanchez A, Howles P, Doetschman T. Mouse embryonic stem cells express the cardiac myosin heavy chain genes during development in vitro. J Biol Chem. 1990;265(20):11905–11909. 83.Sanchez A, Jones WK, Gulick J, Doetschman T, Robbins J. Myosin heavy chain gene expression in mouse embryoid bodies. An in vitro developmental study. J Biol Chem. 1991;266(33):22419–22426. 84.Bhagavati S, Xu W. Generation of skeletal muscle from transplanted embryonic stem cells in dystrophic mice. Biochem Biophys Res Commun. 2005;333(2):644–649. 85.Barberi T, et al. Derivation of engraftable skeletal myoblasts from human embryonic stem cells. Nat Med. 2007;13(5):642–648. 86.Chang H, et al. Generation of transplantable, functional satellite-like cells from mouse embryonic stem cells. Faseb J. 2009;23(6):1907–1919. 87.Dekel I, Magal Y, Pearson-White S, Emerson CP, Shani M. Conditional conversion of ES cells to skeletal muscle by an exogenous MyoD1 gene. New Biol. 1992;4(3):217–224. 88.Darabi R, et al. Functional skeletal muscle regeneration from differentiating embryonic stem cells. Nat Med. 2008;14(2):134–143. 89.Bianco P, Cossu G. Uno, nessuno e centomila: searching for the identity of mesodermal progenitors. Exp Cell Res. 1999;251(2):257–263. 90.Barberi T, Willis LM, Socci ND, Studer L. Derivation of multipotent mesenchymal precursors from human embryonic stem cells. PLoS Med. 2005;2(6):e161. 91.Sakurai H, Okawa Y, Inami Y, Nishio N, Isobe K. Paraxial mesodermal progenitors derived from mouse embryonic stem cells contribute to muscle regeneration via differentiation into muscle satellite cells. Stem Cells. 2008;26(7):1865–1873. 92.Maherali N, Hochedlinger K. Guidelines and techniques for the generation of induced pluripotent stem cells. Cell Stem Cell. 2008;3(6):595–605. 93.Law PK, et al. Dystrophin production induced by myoblast transfer therapy in Duchenne muscular dystrophy. Lancet. 1990;336(8707):114–115. 94.Neumeyer AM, et al. Pilot study of myoblast transfer in the treatment of Becker muscular dystrophy. Neurology. 1998;51(2):589–592. 95.Law PK, et al. Myoblast transfer therapy for Duchenne muscular dystrophy. Acta Paediatr Jpn. 1991;33(2):206–215. 96.Huard J, et al. Human myoblast transplantation between immunohistocompatible donors and recipients produces immune reactions. Transplant Proc. 1992;24(6):3049–3051. 97.Karpati G, et al. Myoblast transfer in Duchenne muscular dystrophy. Ann Neurol. 1993;34(1):8–17. 98.Law PK, et al. Feasibility, safety, and efficacy of myoblast transfer therapy on Duchenne muscular dystrophy boys. Cell Transplant. 1992;1(2-3):235–244. 99.Gussoni E, et al. Normal dystrophin transcripts detected in Duchenne muscular dystrophy patients after myoblast transplantation. Nature. 1992;356(6368):435–438. 100.Tremblay JP, et al. Results of a triple blind clinical study of myoblast transplantations without immunosuppressive treatment in young boys with Duchenne muscular dystrophy. Cell Transplant. 1993;2(2):99–112. 101.M orandi L, et al. Lack of mRNA and dystrophin expression in DMD patients three months after myoblast transfer. Neuromuscul Disord. 1995;5(4):291–295. 102.Mendell JR, et al. Myoblast transfer in the treatment of Duchenne’s muscular dystrophy. N Engl J Med. 1995;333(13):832–838. 103.Miller RG, et al. Myoblast implantation in Duchenne muscular dystrophy: the San Francisco study. Muscle Nerve. 1997;20(4):469–478. 104.Skuk D, Goulet M, Tremblay JP. Use of repeating dispensers to increase the efficiency of the intramuscular myogenic cell injection procedure. Cell Transplant. 2006;15(7):659–663. 105.Partridge T. The current status of myoblast transfer. Neurol Sci. 2000;21(suppl 5):S939–S942. 106.Fan Y, Maley M, Beilharz M, Grounds M. Rapid death of injected myoblasts in myoblast transfer therapy. Muscle Nerve. 1996;19(7):853–860. 107.Guerette B, et al. Prevention by anti-LFA-1 of acute myoblast death following transplantation. J Immunol. 1997;159(5):2522–2531. 108.Torrente Y, et al. Autologous transplantation of muscle-derived CD133+ stem cells in Duchenne muscle patients. Cell Transplant. 2007;16(6):563–577. 109.Atkins BZ, et al. Intracardiac transplantation of skeletal myoblasts yields two populations of striated cells in situ. Ann Thorac Surg. 1999;67(1):124–129. 110.Taylor DA, et al. Regenerating functional myocardium: improved performance after skeletal myoblast transplantation. Nat Med. 1998;4(8):929–933. 111.Pouzet B, et al. Factors affecting functional outcome after autologous skeletal myoblast transplantation. Ann Thorac Surg. 2001;71(3):844–850; discussion 850–851. 112.Reinecke H, Poppa V, Murry CE. Skeletal muscle stem cells do not transdifferentiate into cardiomyocytes after cardiac grafting. J Mol Cell Cardiol. 2002;34(2):241–249. 113.L eobon B, et al. Myoblasts transplanted into rat infarcted myocardium are functionally isolated from their host. Proc Natl Acad Sci U S A. 2003;100(13):7808–7811. 114.Smits PC, et al. Catheter-based intramyocardial injection of autologous skeletal myoblasts as a primary treatment of ischemic heart failure: clinical experience with six-month follow-up. J Am Coll Cardiol. 2003;42(12):2063–2069. 115.Menasche P, et al. Myoblast transplantation for heart failure. Lancet. 2001;357(9252):279–280. 116.M enasche P, et al. Autologous skeletal myoblast transplantation for severe postinfarction left ventricular dysfunction. J Am Coll Cardiol. 2003;41(7):1078–1083. 117.Dib N, et al. Safety and feasibility of autologous myoblast transplantation in patients with ischemic cardiomyopathy: four-year follow-up. Circulation. 2005;112(12):1748–1755. 118.Hagege AA, et al. Skeletal myoblast transplantation in ischemic heart failure: long-term follow-up of the first phase I cohort of patients. Circulation. 2006;114(suppl 1):I108–113. 119.Herreros J, et al. Autologous intramyocardial injection of cultured skeletal muscle-derived stem cells in patients with non-acute myocardial infarction. Eur Heart J. 2003;24(22):2012–2020. 120.Steendijk P, et al. Intramyocardial injection of skeletal myoblasts: long-term follow-up with pressure-volume loops. Nat Clin Pract Cardiovasc Med. 2006;3(suppl 1):S94–S100. 121.Pagani FD, et al. Autologous skeletal myoblasts transplanted to ischemia-damaged myocardium in humans. Histological analysis of cell survival and differentiation. J Am Coll Cardiol. 2003;41(5):879–888. 122.M enasche P, et al. The Myoblast Autologous Grafting in Ischemic Cardiomyopathy (MAGIC) trial: first randomized placebo-controlled study of myoblast transplantation. Circulation. 2008;117(9):1189–1200. 123.Dib N, et al. One-year follow-up of feasibility and safety of the first U.S., randomized, controlled study using 3-dimensional guided catheter-based delivery of autologous skeletal myoblasts for ischemic cardiomyopathy (CAuSMIC study). JACC Cardiovasc Interv. 2009;2(1):9–16. 124.Serruys PW. Final results of a phase I/II randomized, open-label trial to evaluate intramyocardial autologous skeletal myoblast transplantation in congestive heart failure patients: The SEISMIC Trial. Paper presented at: 57th Annual Scientific Session of the American College of Cardiology; March 29-April 1 2008; Chicago, IL, USA. 125.Smaldone MC, Chancellor MB. Muscle derived stem cell therapy for stress urinary incontinence. World J Urol. 2008;26(4):327–332. 126.Mitterberger M, et al. Adult stem cell therapy of female stress urinary incontinence. Eur Urol. 2008;53(1):169–175. 127.Carr LK, et al. 1-year follow-up of autologous muscle-derived stem cell injection pilot study to treat stress urinary incontinence. Int Urogynecol J Pelvic Floor Dysfunct. 2008;19(6):881–883. 128.Strasser H, et al. Transurethral ultrasonographyguided injection of adult autologous stem cells versus transurethral endoscopic injection of collagen in treatment of urinary incontinence. World J Urol. 2007;25(4):385–392. 129.M itterberger M, et al. Autologous myoblasts and fibroblasts for female stress incontinence: a 1-year follow-up in 123 patients. BJU Int. 2007;100(5):1081–1085. The Journal of Clinical Investigation http://www.jci.org Volume 120 Number 1 January 2010 19 PERSPECTIVES OPINION Satellite cells, the engines of muscle repair Yu Xin Wang and Michael A. Rudnicki Abstract | Satellite cells are a heterogeneous population of stem and progenitor cells that are required for the growth, maintenance and regeneration of skeletal muscle. The transcription factors paired-box 3 (PAX3) and PAX7 have essential and overlapping roles in myogenesis. PAX3 acts to specify embryonic muscle precursors, whereas PAX7 enforces the satellite cell myogenic programme while maintaining the undifferentiated state. Recent experiments have suggested that PAX7 is dispensable in adult satellite cells. However, these findings are controversial, and the issue remains unresolved. It has been 50 years since Mauro1 first postulated that satellite cells could be resident progenitor cells involved in skeletal muscle regeneration. Satellite cells were initially characterized according to their anatomica­l position as sublaminar mononuclear cells ‘wedged’ between the basal lamina and the plasma membrane (also known as the sarcolemma in this context) of myofibres. Within this niche, satellite cells are commonly adjacent to a myonucleus of their host myofibre and an endothelial cell of a nearby capillary. The close proximity of satellite cells to myofibres suggested that they may have a role in muscle regeneration. Since the discovery of satellite cells, evidence has accumulated showing that they are the primary contributors to the postnatal growth, maintenance and repair of skeletal muscle. In adult muscle, satellite cells express the transcription factor pairedbox 7 (PAX7) (FIG. 1a) and remain quiescent under normal physiological conditions2,3. Readily responsive to molecular triggers from exercise, injuries or disease, satellite cells have a remarkable ability to self-renew, expand, proliferate as myoblasts or undergo myogenic differentiation to fuse and restore damaged muscle4 (FIG. 1b). Most import­ antly, satellite cells are maintained through repeated cycles of growth and regeneration, which supports the notion that they are a heterogeneous population containing stem cells that sustain their self-renewal. Recent studies using transgenic mice have further examined the role of satellite cells during postnatal regeneration and in hypertrophy. In this Opinion article, we focus on the evidence that satellite cells are essential during these processes, describe the mechanisms maintaining their homeostasis through many rounds of regeneration and discuss findings showing that PAX7 is required to specify this lineage. Myogenesis Many similarities between the activation of satellite cells and myogenesis in the somite (BOX 1) have reinforced the idea that adult muscle regeneration to a certain extent re­capitulates embryonic development through analogous, but not necessarily identical, mechanisms. Below, we briefly introduce the genetic networks involved in myo­genesis — for recent, comprehensive reviews on myogenesi­s, see REFS 4–7. Myogenic regulatory factors. Similarly to the establishment of embryonic progenitors in the myotome (BOX 1), the activation of satellite cells into myoblasts involves the upregulation of the basic helix–loop–helix (bHLH) transcription factors myogenic facto­r 5 (MYF5) and myoblast determination protein NATURE REVIEWS | MOLECULAR CELL BIOLOGY (MYOD) (FIG. 1b). Together with musclespecific regulatory factor 4 (MRF4; also known as MYF6) and myogenin, which are upregulated during myoblast differentiation, these myogenic regulatory factors (MRFs) transcriptionally and epigenetically determine the myogenic capacity of muscle progenitors7. Mice lacking MyoD have seemingly normal muscle but express about fourfold higher levels of MYF5 (REF. 8). MYF5‑deficient animals also have normal muscle9,10. However, mice lacking both Myf5 and MyoD are totally devoid of myoblasts and myofibres11,12, indicating that these genes are required for the determination of myogenic precursors. Although MRF4 has some capacity to function as a determination factor 12, it is has not been established that it does so in the presenc­e of MYF5 and MYOD. As a downstream target of MYOD, myogenin regulates the transition from myoblasts into myocytes and myotubes (FIG. 1b) and, although myogenin-knockout mice show proper compartmentalization of muscle groups, they almost completely lack myofibres and accumulate undifferentiated myoblasts13,14. Mice lacking both MyoD and Mrf4 display a phenotype that is similar to the myogenin-null phenotype, suggesting that, for myoblasts only expressing MYF5, the differentiation into myocytes requires MRF4 (REF. 15). Thus, MYF5 and MYOD are required for the determination of myogenic precursors and act upstream of myogenin and MRF4, which are required for terminal differentiation11. Together, these findings suggest that various overlapping regulatory networks control myogenic differentiation. PAX3 and PAX7 specify myogenic progeni­ tors. The expression of MRFs is regulated by PAX3 and PAX7, both of which have been shown to directly bind proximal promoters of MyoD and distal enhancer elements of Myf5, thereby regulating their expression16–18. PAX3 and PAX7 play key parts in maintaining the proliferation of progen­ itors and preventing early differentiation. Moreover, ectopic expression of Pax3 or Pax7 in mouse embryonic stem cells showed that either one is sufficient to promote a myogenic fate19. VOLUME 13 | FEBRUARY 2012 | 127 © 2012 Macmillan Publishers Limited. All rights reserved PERSPECTIVES which requires the activation of hepatocyte growth factor receptor (HGFR; also known as MET) by PAX3 (REFS 21,25) (BOX 1). This suggests that PAX3 is solely responsible for the migration of progenitors to establish myogenic compartments within the limb. By contrast, PAX7 seems to be uniquely required in satellite cells (discussed in detail below). Indeed, PAX3 is downregulated postnatally in most muscle progenitors, except for the satellite cells of a subset of muscle groups such as the diaphragm25, whereas PAX7 expression is maintained in all satellite cells2. Furthermore, PAX7‑null mice have normal embryonic myogenesis but lack functional satellite cells2. Together, these findings argue that PAX3 and PAX7 have overlapping but non-redundant roles in the myogenic programme. a Merge PAX7 YFP DAPI b Quiescence Basal lamina Self-renewal PAX7+MYF5– Satellite stem cell Sarcolemma Proliferation Differentiation MYOD– Commitment ↑ MYF5 PAX7+MYF5+ ↓ PAX7 ↑ Myogenin Activation ↑ MYOD Commited satellite cell Myoblasts PAX7+MYF5+MYOD+ ↑ MRF4 ↑ MYHC Myocyte MYOD+myogenin+ Figure 1 | Satellite cell fate in regeneration. a | Immunofluorescence micrograph of quiescent satellite Nature Reviews | Molecular Cell Biology cells on a resting extensor digitorum longus muscle fibre isolated from a MYF5–Cre R26R–YFP trans‑ genic mouse, in which MYF5 expression is marked by yellow fluorescent protein (YFP; green): these mice express MYF5–Cre (knock-in of the Cre coding region into the Myf5 coding region) and R26R–YFP (loxP-based lineage reporter; loxP–tPA–loxP–EYFP knocked in at the Rosa locus coding region). Satellite cells represent <5% of all nuclei (labelled with DAPI (4′,6‑diamidino-2‑phenylindole); blue) in skeletal muscle and express the transcription f­actor paired-box 7 (PAX7; red). Although most satellite cells (~90%) are committed to the muscle lineage and have expressed Myf5 at some point in their ancestry (arrowhead), a subpopulation (~10%) has never expressed Myf5 and remains YFP negative (arrow). b | In quiescence, satellite cells are a heterogeneous population residing between the basal lamina and the sarcolemma of their host fibres; ~90% are committed satellite cells (PAX7+MYF5+), whereas ~10% are satellite stem cells (PAX7+MYF5–). Following activation, committed satellite cells upregulate myoblast determination protein (MYOD; PAX7+MYF5+MYOD+ cells), re-enter the cell cycle to proliferate as myo‑ blasts and then progress into differentiation by downregulating PAX7 and upregulating myogenin as myocytes (MYOD+myogenin+ cells). Regeneration of myofibres requires the fusion of myocytes into new myotubes or with host fibres (not shown). During proliferation, heterogeneity regarding MYOD expres‑ sion suggests that some MYOD– myoblasts are able to reverse into quiescence (dashed arrow). MRF4, muscle-specific regulatory facto­r 4; MYHC, myosin heavy chain. Belonging to the PAX family of transcription factors, PAX3 and PAX7 are paralogues with conserved amino acid sequences and have almost identical sequence-specific DNA-binding motifs20. Despite this hom­ ology, studies using knockout mice suggest that PAX3 and PAX7 have overlapping roles only in myogenic specification; the distinct phenotypes of these mice indicate that PAX3 has unique functions during embryonic development 21 and that PAX7 is involved in the specification of satellite cells2. PAX3‑null mice (also known as splotch mice) have many developmental defects, including a lack of limb muscles and reduced MYOD expression in the myotome22 (BOX 1). Interestingly, mice lacking both PAX3 and Specification of satellite cells It has long been postulated that satellite cells are the remnants of embryonic muscle development1. Somitic progenitors that eventually give rise to satellite cells express PAX3 and/or PAX7 and do not express MRFs. These progenitor PAX3­­+PAX7+ cells upregulate MYF5 and MYOD when they enter the myogenic differentiation programme, or remain as satellite cells during late fetal myogenesis without upregulating MRFs. PAX3­­+PAX7+ cells that do not express MRFs are first found to align with nascent myotubes at embryonic day 15.5 and then become satellite cells by taking a sublaminar position24. Notably, once they arrive at the nascent myotubes, most satellite cells rapidly upregulate MYF5 and downregulate PAX3 (REF. 26). Lineage-tracing studies suggest that PAX3+ cells contribute to embryonic myoblasts and the endothelial lineage, whereas PAX7+ cells contribute to fetal myoblasts, supporting the notion that these cells rep­resent distinct myogenic lineages27. Therefore, satellite stem cells (PAX7+MYF5–; see below) in adult muscle may represent a lineage continuum of the embryonic PAX3+PAX7+MRF– progenitors28. However, these studies have not conclusively ruled out the possibility that satellite cells are a distinct myogenic lineage, independent from those that give rise to fetal myoblasts, or that they arise from atypical myogenic stem cells in postnatal muscle (BOX 2). MYF5 cannot upregulate MYOD to compensate for the lack of MYF5 and thus lack muscle23, which further indicates that PAX3 activates MYOD16. Even though PAX3 is expressed at an earlier embryonic stage than PAX7, PAX7 is upregulated in PAX3‑null mice, presumably to compensate for the loss of PAX3 (REF. 22). In fact, muscle formation requires PAX3 or PAX7, as progenitor cell populations from mice lacking both Pax3 and Pax7 undergo apoptosis, and these mice do not form muscle24. Although Pax7 knock-in at the Pax3 locus can rescue the PAX3‑null phenotype during somitogenesis, it does not fully rescue the lack of delamination and long-range migration of muscle progenitors to the limb bud, 128 | FEBRUARY 2012 | VOLUME 13 Relative roles of PAX3 and PAX7 in satellit­e cells. PAX7 expression is maintained in all satellite cells and proliferating myoblasts but is sharply downregulated before differentiation29. PAX3 and PAX7 co-expression in adult satellite cells seems to be limited to www.nature.com/reviews/molcellbio © 2012 Macmillan Publishers Limited. All rights reserved PERSPECTIVES only a few muscle groups — mostly the diaphragm and body wall muscles25. Although PAX3 expression is robust in the embryo, its downregulation after birth suggests that it has little function in most satellite cells. Prolonged expression of PAX3, by preventing its proteosomal degradation, results in the inhibition of terminal differentiation30, consistent with the ability of ectopically expressed PAX7 to inhibit differentiation29. This result has been used to argue that PAX3 regulates satellite cell activation. Although studies have indicated a role for PAX3 in satellite cell development and mainten­ance30,31,32,33, this currently remains controversial. By contrast, PAX7 is required for the development and maintenance of satellite cells. Indeed, cells found in the satellite cell positions of PAX7‑null mice are not myogenic and do not express satellite cell markers (such as syndecan 4 and CD34; see below)31. Importantly, although rare PAX3+ myogenic cells are observed in PAX7‑null mutants, PAX3 cannot compensate for the loss of PAX7 in the establishment of the satellite cell lineage. Moreover, although satellite cells in the diaphragm normally express abundant PAX3, the phenotype of PAX7‑null satellite cells that express PAX3 in the diaphragm is identical to other muscle groups: they cannot maintain the undifferentiated state, they fuse into myofibres early or they have reduced survival, ultimately compromising their capacity for muscle growth and regeneration2,25,31,34,35. This suggests that PAX3 alone is insufficient to rescue these cells2,32. Two possible functions have been proposed for PAX7 in the specification of satel­ lite cells: the specification of the myogenic identity and the maintenance of the undifferentiated state. Ectopic overexpression of PAX7 in C2C12 myoblasts and primary myoblasts suggests that PAX7 drives the expression of Myf5 but not MyoD18,29. Furthermore, PAX7 has been shown to directly activate Myf5 by promoting the recruitment of the ASH2‑like (ASH2L)–MLL2 histone methyltransferase complex at a regulatory element 57 kb upstream of the Myf5 coding region18, which indicates that PAX7 can epigenetically specify the myogenic identity of satellite cells. However, PAX7 expression correlates with an undifferentiated but committed myogenic state36, which suggests that PAX7 in the early postnatal period probably functions to prevent the transition into terminal differentiation while maintaining the myogenic identity by promoting MYF5 expression. Both functions together allow PAX7 to specify the satellite cell lineage. Box 1 | Embryonic myogenesis Early embryonic myogenic precursors giving rise to body wall muscles are marked by paired-box 3 (PAX3) expression in the paraxial mesoderm67. At this early stage, PAX3+ progenitors are multipotent and give rise to the dorsal dermis and vascular progenitors of the aorta68. Morphogens expressed by surrounding embryonic structures eventually determine the cellular commitment within the somites (mesodermal structures found on either side of the neural tube in vertebrate embryos that eventually give rise to muscles, skin and vertebrae)69. Sonic hedgehog signalling from the notochord acts on the ventral medial segment of the somite and leads to the formation of the sclerotome, which contains precursors to bone and cartilage70. WNT signalling in the dorsal portion of the somite leads to the retention of an epithelial morphology and the formation of the dermomyotome71. Combined expression of PAX3 and PAX7 in the dermomyotome further specifies progenitor cells towards the muscle lineage, but transient Notch activation from neural crest cells is required before the expression of myogenic regulatory factors (MRFs; which define muscle progenitors)22,24,72. Notch signalling is essential for retaining myogenic progenitors in a proliferative state; however, its downregulation is required to allow progression into terminal differentiation. MRF+ progenitors migrate and proliferate to give rise to the myotome, in which cells downregulate PAX3 and PAX7 to undergo differentiation and form primitive nascent myofibres21,24,26. These primitive myofibres act as templates for the formation of additional myofibres during postnatal growth. The development of limb muscles requires the establishment of myogenic compartments in the limb bud. This process involves the delamination and migration of progenitors from the dermomyotome73. Induction of hepatocyte growth factor (HGF) in the limb bud activates myogenic progenitors expressing HGF receptor (HGFR; also known as MET), which delaminate from the ventrolateral lip of the dermomyotome. PAX3, and not PAX7, is able to induce the expression of HGFR in embryonic progenitors25. Thus, PAX3 is required to activate the long-range migration of progenitors to the limb bud. This is consistent with the lack of limb muscles in PAX3‑null mice21. Requirement for PAX7 in satellite cells. Recently, a study by Lepper et al.35 using an inducible knockout mouse strain lacking PAX7 specifically in satellite cells when injected with tamoxifen challenged the proposed requirement for PAX7 expression in adult satellite cells. Interestingly, when postnatal time points were examined, the phenotype of mice in which Pax7 had been deleted between postnatal day 7 (P7) and P21 was the same as that of the previously characterized PAX7‑null mice2,25,31,34,35. This result confirms a crucial function for PAX7 during postnatal growth, and this is further supported by the observed early fusion of progenitors in PAX7‑null mice and the conversion of fetal cells from conditional PAX7‑knockout mice into fibroblasts35. Surprisingly, induced ablation of both Pax3 and Pax7 after P21 did not lead to any deficiency in muscle regeneration, satellite cell number or primary myoblast growth or differentiation35. This suggests a rather dramatic change in the genetic requirement of muscle regeneration around P21, the time point when myonuclear accretion is completed (see below) and when postnatal growth is thought to end and satellite cell numbers to reach a steady state37. The distinct regeneration phenotypes obtained by PAX7 deletion between P7–P21 and adulthood challenge the notion that PAX7 is required throughout adulthood to activate downstream MRFs. It remains unclear NATURE REVIEWS | MOLECULAR CELL BIOLOGY whether satellite cells are poised for activation and extrinsic signalling is sufficient to initiate MRF expression or whether temporal specification by PAX7 is sufficient to maintain satellite cell identity after P21. The temporal specification hypothesis is supported by the finding that tamoxifeninduced knockout of Pax7 leads to loss of myogenic identify in fetal cells but not adult cells35. One possibility is that the epigenetic function of PAX7 in recruiting ASH2L– MLL2 to target genes has been completed by P21 and that target genes that define myogenic identity are capable of maintaining their epigenetic memory in the absence of PAX7. However, the ability of satellite cells to function after P21 without PAX7 is difficult to reconcile with the published literature. For example, in vitro knockdown of Pax7 in any aged myoblast or satellite cell has been shown to result in growth arrest and marked reduction in MYF5 expression18,34. Why would a knockdown of PAX7 but not a deletion affect satellite cell maintenance? The notion that proliferating myogenic cells have become ‘addicted’ to PAX7 might be considered; that is, that continued PAX7 expression might be required to keep the cells in a proliferative state. Another interesting possibility is that the floxed allele analysed by Lepper et al. is a hypermorph that is expressing a truncated but functional PAX7 lacking a paired domain. VOLUME 13 | FEBRUARY 2012 | 129 © 2012 Macmillan Publishers Limited. All rights reserved PERSPECTIVES Box 2 | Atypical muscle stem cells Several laboratories have observed that stem cell populations other than satellite cells also have myogenic capacity. These include pericytes, mesangioblasts, side population cells, CD45+SCA1+ cells and even bone marrow-derived haematopoietic stem cells (HSCs)74–80. Their myogenic specification requires activation by injury or co-culturing with myoblasts; however, even under such conditions, only small fractions of these cells adopt myogenic fates. Paired-box 7 (PAX7) is required for the myogenic specification of CD45+SCA1+ cells during regeneration79, whereas the myogenic identity of pericytes, side population cells and HSCs can be activated by alternative pathways. Limited engraftment of these cells to muscle is observed in transplantation experiments, but the extent of their contribution during in vivo muscle regeneration will require further lineage-tracing studies. PW1+ interstitial cells (PICs) were recently described as non-satellite cell myogenic progenitors during postnatal muscle growth that can also adopt either vascular or myogenic fates81. The myogenic specification of PICs depends on PAX7 and is enhanced by the presence of satellite cell-derived myoblasts, suggesting that the recruitment of PICs depends on community effects. As most newly specified fetal satellite cells also express PW1, and because lineage-tracing experiments suggest that PICs are not derived from satellite cells, it was speculated that PICs could be a postnatal source of satellite cells. However, in recent analyses of satellite cell-depleted muscle53,54, atypical muscle stem cells could not promote muscle regeneration or replenish the satellite cell pool. This challenges the previous evidence and suggests either that these atypical muscle stem cells have very limited myogenic potential or that they are not recruited during regeneration in this model owing to a lack of paracrine signals from satellite cells (FIG. 3). The functional participation of atypical myogenic stem cells will need to be studied further. Moreover, the paracrine factors secreted by satellite cells to induce the recruitment of atypical muscle stem cells could be an interesting therapeutic target to modulate regeneration kinetics. If PAX7 is required only in the specification of new satellite cells, such a defect would become apparent only after several rounds of regeneration, when satellite cell pools have become exhausted. Moreover, this defect can be masked by alternative mechanisms of selfrenewal in satellite cells, such as reversible quiescence (see below). Nevertheless, Lepper et al.35 challenged the paradigm that PAX7 is crucial for satellite cells in adults. Thus, further analysis is needed to elucidate the requirement and function of PAX7 in adult satellite cells. Satellite cells in postnatal life After postnatal growth (~P21), maturation of skeletal muscle slowly reaches a homeostasis of rest and regeneration. Satellite cells also enter quiescence as their numbers decrease dramatically from ~30% of sublaminar nuclei to <5% by 8 weeks, which reflects fusion into myofibres, a process known as myonuclear accretion37,38. This entry into quiescence was recently shown to depend on Notch signalling, as mice lacking two Notch targets, Hairy and Enhancer of Split-related 1 (HESR1) and HESR3, fail to generate quiescent satellite cells39. Inhibition or disruption of Notch signalling in quiescent satellite cells leads to their early differentiation and fusion40,41, similarly to what is observed in PAX7‑null mice. Originally defined by their anatomica­l position to the host myofibre1, quiescent satellite cells are reliably identifiable by the expression of molecular markers, such as PAX7, α7β1 integrin, CD34, syndecan 3, syndecan 4, myotubule cadherin (M‑cadherin), caveolin 1 and CXC chemokine receptor 4 (CXCR4; also known as fusin and CD184)42. The identification of these markers, as well as the generation of transgenic reporter mouse lines, has only recently allowed the prospective isolation and characterization of satellite cells in their quiescent state28,43. Satellite cells are heterogeneous. All adult satellite cells express PAX7, and this expression is conserved through evolution42. However, subpopulations of satellite cells express a mixture of different surface markers as well as altered expression levels of MYF5 and MYOD44. Because satellite cells are self-renewing, within this heterogeneity there must be a stem cell-like population that is resistant to differentiation and maintains the satellite cell pool through many rounds of regeneration. Experiments have shown that small fractions (~20%) of satellite cells have slower proliferation kinetics and can resist differentiation45. Furthermore, satellite cells from different muscle groups also display differential engraftment potential after transplantation46. In these transplantation experiments, only a small fraction of satellite cells shows stem cell-like properties and engrafts into the satellite cell compartment. 130 | FEBRUARY 2012 | VOLUME 13 PAX7+MYF5– satellite stem cells. Using lin­ eage tracing in a mouse strain in which cells that have expressed MYF5 are permanently labelled with yellow fluorescent protein (YFP), we identified that ~10% of adult satellite cells express PAX7 but not YFP, indicating that they have never expressed MYF5 (PAX7+MYF5– cells)28 (FIG. 1a). Compared with their committed YFP+ counterparts, PAX7+MYF5– satellite cells were able to engraft as satellite cells, self-renew more efficiently, give rise to committed YFP+ (PAX7+MYF5+) satellite cells and better resist differentiation. Because of these stem cell characteristics and their uncommitted status, we concluded that these PAX7+MYF5– satellit­e cells were in fact satellite stem cells. Similarly to other adult stem cells (reviewed in REF. 47), satellite stem cells follow traditional stochastic (symmetric) and asymmetric paradigms of self-renewal28 (FIG. 2a). The asymmetric nature of the satellite cell niche, with the basal lamina acting as the basement membrane and the sarcolemma of the host myofibre as the apical surface, determines the cell fate of stem cell divisions according to the orientation of the mitotic axis in relation to the host fibre. During in vivo regeneration, satellite stem cell divisions in the apicobasal orientation give rise to a committed MYF5+ satellite cell, which is pushed into the apical surface, and a daughter cell, which remains attached to basal lamina and retains the satellite stem cell identity. By contrast, planar divisions of satellite stem cells occur along the host myofibre, after which both daughter cells remain in contact with the basal lamina, symmetrically expand, giving rise to two PAX7+MYF5– cells. It remains unclear what intrinsic mech­ anism is responsible for the orientation of divisions and how subsequent events lead to commitment of the apical daughter cell. However, insights into extrinsic regulators revealed that satellite stem cells are similar to embryonic PAX7+MYF5– cells in the dermomyotome (BOX 1), in that inhibition of Notch signalling drives satellite stem cells to commit­ment and differentiation28,33. Furthermore, WNT signalling is thought to promote symmetric cell divisions by acting through the planar cell polarity pathway 48. Specifically, WNT7A signalling through Frizzled 7 markedly stimulates the symmetric expansion of satellite stem cells but does not affect the growth of myoblasts or the kinetics of their differentiation (FIG. 2b). This induces the polarized distribution of the planar cell polarity effector VANG-like 2 (VANGL2), promoting a planar division plane and www.nature.com/reviews/molcellbio © 2012 Macmillan Publishers Limited. All rights reserved PERSPECTIVES therefore symmetric division of the daughter cells. This pathway is thought to regulate the regenerative potential of muscle. In support of this, WNT7A is upregulated during regen­­ eration and increases both the number of satel­lite cells and the muscle mass, and muscle lacking WNT7A exhibits a decrease in satellit­e cell number following regeneration. Reversible quiescence. Replenishment of the satellite cell pool requires stem cells to expand but then return to quiescence, a common feature of adult stem cells. This process is very elusive and only applies to a subpopulation of satellite cells, so our current understanding is minimal. In vitro observations of satellite cells on cultured myofibres suggest that a population of MYOD– reserve cells appear in late regeneration and can return to quiescence by sustaining PAX7 expression and resisting differentiation49,50. This could arise from the expansion of a MYOD– population of uncommitted satellite cells or the downregulation of MYOD in proliferating myoblasts. Most PAX7+MYF5– satellite stem cells correlate with this population and do not express MYOD or myogenin in fibre culture28. One known mechanism regulating the reversible quiescence of satellite stem cells is mediated by the receptor Tyr kinase TIE2 (also known as TEK)51. TIE2 activation by binding of the hormone angiopoietin 1, which is secreted from neighbouring fibroblasts and vascular cells, activates extracellular signal-regulated kinase (ERK) signalling downstream of TIE2. This enhances PAX7 expression and thereby sustains the undifferentiated state, ultimately increasing the number of stem cells reverting to quiescence. Reversible quiescence of committed myoblast­s challenges the paradigm that MYOD, as a determination factor, is an irreversible epigenetic checkpoint in the myogenic programme. For a subpopulation of myoblasts, this reversion was reported to require the activation of Sprouty hom­ ologue 1 (SPRY1), a negative regulator of ERK signalling 52. However, myoblasts un­affected in SPRY1 deletion can maintain a reduced, but homeostatic, number of satellite cells through subsequent injuries. This indicates that alternative mechanisms of reversion into quiescence exist. Although contradictory, TIE2–ERK and SPRY1–ERK signalling apply to two distinct cell types at different time points during regeneration. This suggests that temporal regulation of the ERK signalling pathway controls the number of satellite cells returning to quiescence and reveals the possibility of modulating this homeostasis with extrinsic factors. a Self-renewal b PAX7+MYF5– WNT7A Basal lamina Sarcolemma Satellite stem cell Reverse into quiescence WNT7A Asymmetric division Symmetric expansion PAX7+MYF5– PAX7+MYF5– FZD7 VANGL2 Planar cell polarity pathway PAX7+MYF5+ Figure 2 | Schematic of satellite stem cell fate in self-renewal. a | During regeneration, satellite stem Nature Reviewsdepending | Molecular Biology cells expand and undergo stochastic (symmetric) or asymmetric self-renewal, onCell the orienta‑ tion of divisions with respect to the basal lamina (basal surface) and the sarcolemma of the host fibre (apical surface). In the stochastic paradigm, divisions parallel to the muscle fibre (planar) result in the symmetric expansion of satellite stem cells. Divisions that are oriented perpendicular to the host fibre (apicobasal) follow the asymmetric paradigm and give rise to two distinct fates: one daughter cell is pushed into the apical surface, upregulates myogenic factor 5 (MYF5) and becomes PAX7+MYF5+; and the other daughter cell remains adhered to the basal lamina and retains its stem cell identity (PAX7+MYF5–). WNT7A, which is upregulated late in regeneration, promotes the symmetric expansion of satellite stem cells to replenish the stem cell pool and maintain a homeostatic level of stem cells through rounds of regeneration. b | WNT7A signals through its receptor Frizzled 7 (FZD7) to mediate the polarized distribution of its effector VANG-like 2 (VANGL2). Acting through the planar cell polarity pathway to induce cytoskeletal changes and orientation of spindles during division, WNT7A promotes satellite stem cells to divide in a planar orientation, ultimately enhancing symmetric expansion. PAX7, paired-box 7. Satellite cells in regeneration The main function of satellite cells is to proliferate and differentiate into muscle cells to regenerate muscle during exercise or injury. Satellite cells are integral to regeneration. The ability for satellite cells to contribute to myogenesis has been demonstrated in vitro and in vivo. In a series of diphtheria toxininduced ablation experiments, it was revealed that muscle failed to regenerate when PAX7+ satellite cells were depleted53–55. Two different transgenic alleles were used to eliminate PAX7+ satellite cells from muscle, and both led to the same conclusion: no other cell types can rescue the myogenic function of PAX7+ cells during exercise or injury 53,54 (BOX 2). Instead of the formation of myofibres, adipocyte infiltrates expanded in place of muscle54. This genetic ablation result resembles the phenotypes of PAX7‑null mice and supports the notion that satellite cells are the primary, if not the sole, contributors to muscle regeneration. Healthy adult muscle has a slow turnover rate, so satellite cells remain mostly quiescent (see above) unless activated by exercise or injury 3. Consistent with this, ablation of satellite cells in resting conditions does not result in progressive muscle mass loss54. However, in degenerative muscle diseases, such as Duchenne muscular dystrophy, chronic regeneration of muscle triggers higher proliferative demand on satellite cells, leading to NATURE REVIEWS | MOLECULAR CELL BIOLOGY their eventual depletion56. Interfering with the function of satellite cells in a mouse model for Duchenne muscular dystrophy (mdx mice), by either inducing reduced myogenic differ­ entiation (for example, in MYOD-null mice) or reducing the telomere lengths (for example, in mice lacking telomerase mRNA), results in an exaggerated dystrophic phenotype57,58. Together, these experiments suggest that satellite cells are absolutely necessary for regeneration and that satellite cells lost in disease are not replaced by other stem cell sources. Interactions with other cell types during regen­­ eration. Although satellite cells are the main players in repairing muscle, various other cells types are also recruited during regeneration and can modulate the behaviour of satellite cells by secreting cytokines (FIG. 3). During the initial phase of regeneration, secretion of pro-inflammatory cytokines by macrophages can induce myoblast entry into terminal differentiation through the p38 kinase-mediated repression of the Pax7 locus59. Expansion of resident fibroadipogenic progenitors (FAPs) during regeneration also promotes the termin­al differentiation of myoblasts60. However, not all signals during regeneration are pro-myogenic. The presence of muscle connective tissue (MCT) fibroblasts expressing T cell factor 4 (TCF4) seems to prevent early differentiation of myoblasts by creating a transitional niche. Reciprocally, myoblasts promote MCT fibroblast VOLUME 13 | FEBRUARY 2012 | 131 © 2012 Macmillan Publishers Limited. All rights reserved PERSPECTIVES Recruitment Atypical myogenic stem cell Proliferation Fibroblast Macrophage Factors? Angiopoietin 1 TNF IL-6 TGFβ p38 activation Proliferation Quiescent satellite cell Reversal Myoblast Differentiation Activation FAPs Regeneration Damaged myofibre Figure 3 | Schematic of the cell interactions during regeneration. During regeneration, paracrine signals (bold arrows) between necrotic fibres, infiltrating macrophages, fibroblasts and proliferating Nature Reviews | Molecular Cell Biology fibroadipogenic progenitors (FAPs) modulate satellite cell activation, proliferation and differentiation (grey arrows). Secretion of angiopoietin 1 by fibroblasts and vascular cells has also been found to promote satellite cell reversal into quiescence by activating extracellular signal-regulated kinase (ERK) signalling. Signals coming from myoblasts also affect fibroblast proliferation. The lack in regenerative capacity of atypical myogenic stem cells in satellite cell-depleted muscles suggests a possible require‑ ment for satellite cell- or myoblast-secreted factors in their recruitment. IL‑6, interleukin‑6; TGFβ, transforming growth factor‑β; TNF, tumour necrosis factor. proliferation55. Late in regeneration, newly formed myotubes prevent the adipogenic differentiation of mesenchymal progenitors expressing platelet-derived growth factor receptor-α (PDGFRα)61. These interactions give rise to a complex interaction network between satellite cells and other cell types, which allows the extrinsic regulation of satellite cells in regeneration. Satellite cell requirement in hypertrophy. Hypertrophy of muscle (that is, an increase in myofibre size resulting from the production of additional contractile proteins) caused by overloading or exercise requires additional myonuclei for protein synthesis. Because myo­nuclei are mitotically arrested and cannot replicate within the myofibre to meet such a demand, hypertrophy is commonly associated with the activation of satellite cells62, which are thought to proliferate and fuse into existing myofibres. Indeed, classic studies using γ‑radiation to inhibit the proliferation of sate­llite cells found a dependence on satellite cell, activation in overloading-induced hyper­ trophy 63. However, controversy over the specificity of γ‑radiation to satellite cells, as well as a better understanding of intrinsic pathways leading to the hypertrophy of myotubes, such as calcineurin and AKT signalling, indicated that further examination is needed64. A recent study 65 provided evidence that satellite cells are not required for overloadinginduced hypertrophy and that intrinsic mechanisms within the myofibres can compensate for protein synthesis without additional myonuclei. Specifically, significant hyper­trophy was observed in satellite cell-depleted muscles that were exposed to overloading conditions. However, in controls, activation of satellite cells and fusion into myofibres were observed. This indicates that, in a physiological context, satellite cells indeed participate in hypertrophy. It is not surprising that myofibres can undergo short-term hypertrophy without additional myonuclei, but the addition of myonuclei should reduce the ‘workload’ placed on the existing myonuclei. Importantly, productive hyper­trophy resulting from the participation of sate­llite cells would be expected to couple tissue mass with appropriate numbers of stem and progenitor cells. Thus, satellite cells are essential to achieve tissue homeostasis and allow growth and repair over the long term. Concluding remarks 50 years after their discovery, we have solidified the absolute requirement for muscle satellite cells during muscle regeneration and in hypertrophy. Composed primarily of a nucleus and very little cytoplasm, the rege­n­ erative capacity of satellite cells can only be described as remarkable. Indeed, whereas the requirement for satellite cells has been questioned for short-term hypertrophy, it is clear that maintenance of muscle and productive hypertrophy absolutely require satellite cells. 132 | FEBRUARY 2012 | VOLUME 13 We have only scratched the surface of a complex genetic network that requires intrinsic and extrinsic activation to regulate the myogenic fate of satellite cells. We know that PAX3 and PAX7 regulate the expression of MRFs and are crucial for muscle development, and that early postnatal specification of the satellite cell lineage requires PAX7 to establish an undifferentiated but myogenic state. However, further work is needed to definitively confirm whether PAX7 is required in adult satellite cells. Within the past two decades, our understanding of satellite cell biology has formed new paradigms for the maintenance and genetic regulation of adult stem cells in general. Without question, our advancing know­l­ edge of satellite cell biology has opened new doors for the development of therapeutics targeting stem cell growth48, in vitro expansion of satellite cells66 and even induction of embryonic stem cells for transplantation19 as promising avenues for the treatment of patients suffering from degenerative muscle diseases. Yu Xin Wang and Michael A. Rudnicki are at the Sprott Centre for Stem Cell Research, Ottawa Hospital Research Institute Ottawa, 501 Smyth Road Ottawa, Ontario K1H 8L6, Canada. Yu Xin Wang and Michael A. Rudnicki are also at the Faculty of Medicine, Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario K1H 8M5, Canada. Correspondence to M.A.R. e‑mail: mrudnicki@ohri.ca doi:10.1038/nrm3265 Published online 21 December 2011 Mauro, A. Satellite cell of skeletal muscle fibers. J. Biophys. Biochem. Cytol. 9, 493–495 (1961). Seale, P. et al. Pax7 is required for the specification of myogenic satellite cells. Cell 102, 777–786 (2000). 3. Schultz, E., Gibson, M. C. & Champion, T. Satellite cells are mitotically quiescent in mature mouse muscle: an EM and radioautographic study. J. Exp. Zool. 206, 451–456 (1978). 4. Bentzinger, C. F., von Maltzahn, J. & Rudnicki, M. A. Extrinsic regulation of satellite cell specification. Stem Cell Res. Ther. 1, 27 (2010). 5. Braun, T. & Gautel, M. Transcriptional mechanisms regulating skeletal muscle differentiation, growth and homeostasis. Nature Rev. Mol. Cell Biol. 12, 349–361 (2011). 6. Buckingham, M. & Vincent, S. D. Distinct and dynamic myogenic populations in the vertebrate embryo. Curr. Opin. Genet. Dev. 19, 444–453 (2009). 7. Punch, V. G., Jones, A. E. & Rudnicki, M. A. Transcriptional networks that regulate muscle stem cell function. Wiley Interdiscip. Rev. Syst. Biol. Med. 1, 128–140 (2009). 8. Rudnicki, M. A., Braun, T., Hinuma, S. & Jaenisch, R. Inactivation of MyoD in mice leads to up-regulation of the myogenic HLH gene Myf‑5 and results in apparently normal muscle development. Cell 71, 383–390 (1992). 9. Braun, T., Rudnicki, M. A., Arnold, H. H. & Jaenisch, R. Targeted inactivation of the muscle regulatory gene Myf‑5 results in abnormal rib development and perinatal death. Cell 71, 369–382 (1992). 10. Kaul, A., Köster, M., Neuhaus, H. & Braun, T. Myf‑5 revisited: loss of early myotome formation does not lead to a rib phenotype in homozygous Myf‑5 mutant mice. Cell 102, 17–19 (2000). 11. Rudnicki, M. A. et al. MyoD or Myf‑5 is required for the formation of skeletal muscle. Cell 75, 1351–1359 (1993). 1. 2. www.nature.com/reviews/molcellbio © 2012 Macmillan Publishers Limited. All rights reserved PERSPECTIVES 12. Kassar-Duchossoy, L. et al. Mrf4 determines skeletal muscle identity in Myf5:Myod double-mutant mice. Nature 431, 466–471 (2004). 13. Hasty, P. et al. Muscle deficiency and neonatal death in mice with a targeted mutation in the myogenin gene. Nature 364, 501–506 (1993). 14. Nabeshima, Y. et al. Myogenin gene disruption results in perinatal lethality because of severe muscle defect. Nature 364, 532–535 (1993). 15. Rawls, A. et al. Overlapping functions of the myogenic bHLH genes MRF4 and MyoD revealed in double mutant mice. Development 125, 2349–2358 (1998). 16. Hu, P., Geles, K. G., Paik, J.‑H., DePinho, R. A. & Tjian, R. Codependent activators direct myoblast specific MyoD transcription. Dev. Cell 15, 534–546 (2008). 17. Bajard, L. et al. A novel genetic hierarchy functions during hypaxial myogenesis: Pax3 directly activates Myf5 in muscle progenitor cells in the limb. Genes Dev. 20, 2450–2464 (2006). 18. McKinnell, I. W. et al. Pax7 activates myogenic genes by recruitment of a histone methyltransferase complex. Nature Cell Biol. 10, 77–84 (2008). 19. Darabi, R. et al. Assessment of the myogenic stem cell compartment following transplantation of Pax3/Pax7‑ induced embryonic stem cell-derived progenitors. Stem Cells 29, 777–790 (2011). 20. Schäfer, B. W., Czerny, T., Bernasconi, M., Genini, M. & Busslinger, M. Molecular cloning and characterization of a human PAX‑7 cDNA expressed in normal and neoplastic myocytes. Nucleic Acids Res. 22, 4574–4582 (1994). 21. Bober, E., Franz, T., Arnold, H. H., Gruss, P. & Tremblay, P. Pax‑3 is required for the development of limb muscles: a possible role for the migration of dermomyotomal muscle progenitor cells. Development 120, 603–612 (1994). 22. Borycki, A. G., Li, J., Jin, F., Emerson, C. P. & Epstein, J. A. Pax3 functions in cell survival and in pax7 regulation. Development 126, 1665–1674 (1999). 23. Tajbakhsh, S., Rocancourt, D., Cossu, G. & Buckingham, M. Redefining the genetic hierarchies controlling skeletal myogenesis: Pax‑3 and Myf‑5 act upstream of MyoD. Cell 89, 127–138 (1997). 24. Relaix, F., Rocancourt, D., Mansouri, A. & Buckingham, M. A Pax3/Pax7‑dependent population of skeletal muscle progenitor cells. Nature 435, 948–953 (2005). 25. Relaix, F., Rocancourt, D., Mansouri, A. & Buckingham, M. Divergent functions of murine Pax3 and Pax7 in limb muscle development. Genes Dev. 18, 1088–1105 (2004). 26. Kassar-Duchossoy, L. et al. Pax3/Pax7 mark a novel population of primitive myogenic cells during development. Genes Dev. 19, 1426–1431 (2005). 27. Hutcheson, D. A., Zhao, J., Merrell, A., Haldar, M. & Kardon, G. Embryonic and fetal limb myogenic cells are derived from developmentally distinct progenitors and have different requirements for β-catenin. Genes Dev. 23, 997–1013 (2009). 28. Kuang, S., Kuroda, K., Le Grand, F. & Rudnicki, M. A. Asymmetric self-renewal and commitment of satellite stem cells in muscle. Cell 129, 999–1010 (2007). 29. Olguin, H. C. & Olwin, B. B. Pax‑7 up-regulation inhibits myogenesis and cell cycle progression in satellite cells: a potential mechanism for self-renewal. Dev. Biol. 275, 375–388 (2004). 30. Boutet, S. C., Disatnik, M.‑H., Chan, L. S., Iori, K. & Rando, T. A. Regulation of Pax3 by proteasomal degradation of monoubiquitinated protein in skeletal muscle progenitors. Cell 130, 349–362 (2007). 31. Kuang, S., Chargé, S. B., Seale, P., Huh, M. & Rudnicki, M. A. Distinct roles for Pax7 and Pax3 in adult regenerative myogenesis. J. Cell Biol. 172, 103–113 (2006). 32. Relaix, F. et al. Pax3 and Pax7 have distinct and overlapping functions in adult muscle progenitor cells. J. Cell Biol. 172, 91–102 (2006). 33. Conboy, I. M. & Rando, T. A. The regulation of Notch signaling controls satellite cell activation and cell fate determination in postnatal myogenesis. Dev. Cell 3, 397–409 (2002). 34. Oustanina, S., Hause, G. & Braun, T. Pax7 directs postnatal renewal and propagation of myogenic satellite cells but not their specification. EMBO J. 23, 3430–3439 (2004). 35. Lepper, C., Conway, S. J. & Fan, C.‑M. Adult satellite cells and embryonic muscle progenitors have distinct genetic requirements. Nature 460, 627–631 (2009). 36. Olguin, H. C., Yang, Z., Tapscott, S. J. & Olwin, B. B. Reciprocal inhibition between Pax7 and muscle 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. regulatory factors modulates myogenic cell fate determination. J. Cell Biol. 177, 769–779 (2007). White, R., Bierinx, A.‑S., Gnocchi, V. & Zammit, P. Dynamics of muscle fibre growth during postnatal mouse development. BMC Dev. Biol. 10, 21 (2010). Bischoff, R. in Myogenesis (eds Engel, A. G. & Franszini-Armstrong, C.) 97–118 (McGraw-Hill, New York,1994). Fukada, S. et al. Hesr1 and Hesr3 are essential to generate undifferentiated quiescent satellite cells and to maintain satellite cell numbers. Development 138, 4609–4619 (2011). Bjornson, C. R. R. et al. Notch signaling is necessary to maintain quiescence in adult muscle stem cells. Stem Cells 1 Nov 2011 (doi:10.1002/stem.773). Mourikis, P. et al. A critical requirement for Notch signaling in maintenance of the quiescent skeletal muscle stem cell state. Stem Cells 8 Nov 2011 (doi:10.1002/stem.775). Kuang, S. & Rudnicki, M. A. The emerging biology of satellite cells and their therapeutic potential. Trends Mol. Med. 14, 82–91 (2008). Bosnakovski, D. et al. Prospective isolation of skeletal muscle stem cells with a Pax7 reporter. Stem Cells 26, 3194–3204 (2008). Sherwood, R. I. et al. Isolation of adult mouse myogenic progenitors: functional heterogeneity of cells within and engrafting skeletal muscle. Cell 119, 543–554 (2004). Schultz, E. Satellite cell proliferative compartments in growing skeletal muscles. Dev. Biol. 175, 84–94 (1996). Collins, C. A. et al. Stem cell function, self-renewal, and behavioral heterogeneity of cells from the adult muscle satellite cell niche. Cell 122, 289–301 (2005). Yamashita, Y. M. Cell adhesion in regulation of asymmetric stem cell division. Curr. Opin. Cell Biol. 22, 605–610 (2010). Le Grand, F., Jones, A. E., Seale, V., Scimè, A. & Rudnicki, M. A. Wnt7a activates the planar cell polarity pathway to drive the symmetric expansion of satellite stem cells. Cell Stem Cell 4, 535–547 (2009). Yoshida, N., Yoshida, S., Koishi, K., Masuda, K. & Nabeshima, Y. Cell heterogeneity upon myogenic differentiation: down-regulation of MyoD and Myf‑5 generates ‘reserve cells’. J. Cell Sci. 111, 769–779 (1998). Zammit, P. S. et al. Muscle satellite cells adopt divergent fates: a mechanism for self-renewal? J. Cell Biol. 166, 347–357 (2004). Abou-Khalil, R. et al. Autocrine and paracrine angiopoietin 1/Tie‑2 signaling promotes muscle satellite cell self-renewal. Cell Stem Cell 5, 298–309 (2009). Shea, K. L. et al. Sprouty1 regulates reversible quiescence of a self-renewing adult muscle stem cell pool during regeneration. Cell Stem Cell 6, 117–129 (2010). Lepper, C., Partridge, T. A. & Fan, C.‑M. An absolute requirement for Pax7‑positive satellite cells in acute injury-induced skeletal muscle regeneration. Development 138, 3639–3646 (2011). Sambasivan, R. et al. Pax7‑expressing satellite cells are indispensable for adult skeletal muscle regeneration. Development 138, 3647–3656 (2011). Murphy, M. M., Lawson, J. A., Mathew, S. J., Hutcheson, D. A. & Kardon, G. Satellite cells, connective tissue fibroblasts and their interactions are crucial for muscle regeneration. Development 138, 3625–3637 (2011). Jejurikar, S. S. & Kuzon, W. M. Jr. Satellite cell depletion in degenerative skeletal muscle. Apoptosis 8, 573–578 (2003). Megeney, L. A., Kablar, B., Garrett, K., Anderson, J. E. & Rudnicki, M. A. MyoD is required for myogenic stem cell function in adult skeletal muscle. Genes Dev. 10, 1173–1183 (1996). Sacco, A. et al. Short telomeres and stem cell exhaustion model Duchenne muscular dystrophy in mdx/mTR mice. Cell 143, 1059–1071 (2010). Palacios, D. et al. TNF/p38α/polycomb signaling to Pax7 locus in satellite cells links inflammation to the epigenetic control of muscle regeneration. Cell Stem Cell 7, 455–469 (2010). Joe, A. W. B. et al. Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis. Nature Cell Biol. 12, 153–163 (2010). Uezumi, A., Fukada, S., Yamamoto, N., Takeda, S. & Tsuchida, K. Mesenchymal progenitors distinct from satellite cells contribute to ectopic fat cell formation in skeletal muscle. Nature Cell Biol. 12, 143–152 (2010). Salleo, A., La Spada, G., Falzea, G., Denaro, M. G. & Cicciarello, R. Response of satellite cells and muscle NATURE REVIEWS | MOLECULAR CELL BIOLOGY 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. fibers to long-term compensatory hypertrophy. J. Submicrosc. Cytol. 15, 929–940 (1983). Rosenblatt, J. D. & Parry, D. J. Gamma irradiation prevents compensatory hypertrophy of overloaded mouse extensor digitorum longus muscle. J. Appl. Physiol. 73, 2538–2543 (1992). McCarthy, J. J. & Esser, K. A. Counterpoint: satellite cell addition is not obligatory for skeletal muscle hypertrophy. J. Appl. Physiol. 103, 1100–1102 (2007). McCarthy, J. J. et al. Effective fiber hypertrophy in satellite cell-depleted skeletal muscle. Development 138, 3657–3666 (2011). Gilbert, P. M. et al. Substrate elasticity regulates skeletal muscle stem cell self-renewal in culture. Science 329, 1078–1081 (2010). Williams, B. A. & Ordahl, C. P. Pax‑3 expression in segmental mesoderm marks early stages in myogenic cell specification. Development 120, 785–796 (1994). Lagha, M. et al. Pax3:Foxc2 reciprocal repression in the somite modulates muscular versus vascular cell fate choice in multipotent progenitors. Dev. Cell 17, 892–899 (2009). Aoyama, H. & Asamoto, K. Determination of somite cells: independence of cell differentiation and morphogenesis. Development 104, 15–28 (1988). Borycki, A. G., Mendham, L. & Emerson, C. P. Jr. Control of somite patterning by Sonic hedgehog and its downstream signal response genes. Development 125, 777–790 (1998). Tajbakhsh, S. et al. Differential activation of Myf5 and MyoD by different Wnts in explants of mouse paraxial mesoderm and the later activation of myogenesis in the absence of Myf5. Development 125, 4155–4162 (1998). Rios, A. C., Serralbo, O., Salgado, D. & Marcelle, C. Neural crest regulates myogenesis through the transient activation of NOTCH. Nature 473, 532–535 (2011). Heymann, S., Koudrova, M., Arnold, H., Köster, M. & Braun, T. Regulation and function of SF/HGF during migration of limb muscle precursor cells in chicken. Dev. Biol. 180, 566–578 (1996). Dellavalle, A. et al. Pericytes of human skeletal muscle are myogenic precursors distinct from satellite cells. Nature Cell Biol. 9, 255–267 (2007). Dellavalle, A. et al. Pericytes resident in postnatal skeletal muscle differentiate into muscle fibres and generate satellite cells. Nature Commun. 2, 499 (2011). Sampaolesi, M. et al. Mesoangioblast stem cells ameliorate muscle function in dystrophic dogs. Nature 444, 574–579 (2006). Asakura, A., Seale, P., Girgis-Gabardo, A. & Rudnicki, M. A. Myogenic specification of side population cells in skeletal muscle. J. Cell Biol. 159, 123–134 (2002). Tanaka, K. K. et al. Syndecan-4‑expressing muscle progenitor cells in the SP engraft as satellite cells during muscle regeneration. Cell Stem Cell 4, 217–225 (2009). Seale, P., Ishibashi, J., Scimè, A. & Rudnicki, M. A. Pax7 is necessary and sufficient for the myogenic specification of CD45+:Sca1+ stem cells from injured muscle. PLoS Biol. 2, e130 (2004). Xynos, A. et al. Bone marrow-derived hematopoietic cells undergo myogenic differentiation following a Pax‑7 independent pathway. Stem Cells 28, 965–973 (2010). Mitchell, K. J. et al. Identification and characterization of a non-satellite cell muscle resident progenitor during postnatal development. Nature Cell Biol. 12, 257–266 (2010). Acknowledgements We thank F. Bentzinger and H. Yin for critical reading of the manuscript. Y.X.W. is supported by a fellowship from the Ontario Research Fund and by the Canadian Institutes of Health Research (CIHR) Training Program in Regenerative Medicine. M.A.R. holds the Canada Research Chair in Molecular Genetics and is an International Research Scholar of the Howard Hughes Medical Institute. The work from the laboratory of M.A.R. is supported by grants from the US National Institutes of Health, the Howard Hughes Medical Institute, the CIHR, the Muscular Dystrophy Association and the Canada Research Chair Program. Competing interests statement The authors declare no competing financial interests. FURTHER INFORMATION Michael A. Rudnicki’s homepage: http://www.rudnickilab.ca ALL LINKS ARE ACTIVE IN THE ONLINE PDF VOLUME 13 | FEBRUARY 2012 | 133 © 2012 Macmillan Publishers Limited. All rights reserved Cell, Vol. 122, 659–667, September 9, 2005, Copyright ©2005 by Elsevier Inc. DOI 10.1016/j.cell.2005.08.021 Cellular and Molecular Signatures Review of Muscle Regeneration: Current Concepts and Controversies in Adult Myogenesis Amy J. Wagers1,2,* and Irina M. Conboy3,* 1 Section on Developmental and Stem Cell Biology Joslin Diabetes Center and Department of Pathology Harvard Medical School Boston, Massachusetts 02215 2 Harvard Stem Cell Institute Cambridge, Massachusetts 02138 3 Department of Bioengineering University of California, Berkeley Berkeley, California 94702 Adult skeletal muscle generates force in a controlled and directed manner through the contraction of highly specialized, postmitotic, multinucleated myofibers. Lifelong muscle function relies on maintenance and regeneration of myofibers through a highly regulated process beginning with activation of normally quiescent muscle precursor cells and proceeding with formation of proliferating progenitors that fuse to generate differentiated myofibers. In this review, we describe the historical basis and current evidence for the identification of satellite cells as adult muscle stem cells, critically evaluate contributions of other cells to adult myogenesis, and summarize existing data regarding the origins, genetic markers, and molecular regulation of satellite cells in normal, diseased, and aged muscle. Satellite Cells as Adult Muscle Precursor Cells and Candidate Muscle Stem Cells Adult skeletal muscle possesses remarkable regenerative capacity, and large numbers of new myotubes normally are formed in only a few days after acute muscle damage (Bintliff and Walker, 1960; LeGros Clark, 1946; Walton and Adams, 1956; Weber, 1863). Early hypotheses proposed that new myofibers were generated via budding of myotubes from existing, injured fibers (LeGros Clark, 1946; Volkmann, 1893); however, further study has demonstrated instead that this rapid repair occurs through the differentiation and subsequent cell fusion of myogenically specified mononuclear precursor cells contained within a population of “satellite cells” positioned between the plasma membrane and the surrounding basal lamina of mature, differentiated muscle fibers (Mauro, 1961; Snow, 1978). Alexander Mauro first proposed in 1961 that satellite cells might represent “dormant myoblasts” left over from embryonic muscle development and capable of recapitulating the developmental program of skeletal myogenesis in response to muscle damage (Mauro, 1961). Subsequent studies, in both the chick (Konigsberg, 1963) and mouse (Yaffe, 1969), demonstrated that multinucleated myotubes could indeed be generated in vitro from single myo*Correspondence: amy.wagers@joslin.harvard.edu (A.J.W.); iconboy@ berkeley.edu (I.M.C.) genic precursor cells, and that these precursors ultimately derived from muscle satellite cells (Bischoff, 1975). Furthermore, in vivo-labeled (Snow, 1978), and clonally cultured in vitro-labeled (Lipton and Schultz, 1979), satellite cells were shown to participate in the regeneration of damaged muscle when transplanted in vivo, contributing almost exclusively via fusion with pre-existing myofibers. Pulse-chase experiments using a single dose of tritiated thymidine to label dividing cells indicated that DNA synthesis among sublaminar nuclei was limited to satellite cell nuclei, and that true muscle nuclei do not undergo mitosis (Moss and Leblond, 1970). Although these methods labeled satellite cells relatively infrequently, indicating the relative quiescence of satellite cells (Schultz et al., 1978), in some cases DNA label from satellite cells marked during the pulse phase eventually appeared in myonuclei during the chase phase, demonstrating the capacity of at least some satellite cell progeny to incorporate into existing myofibers, even in the absence of acute tissue injury (Moss and Leblond, 1970). Such studies formed the basis for our current view of muscle satellite cells as the primary mediators of postnatal muscle growth and repair. These cells respond to regenerative cues, such as injury or exercise, by proliferating to form myoblasts, which divide a limited number of times before terminally differentiating and fusing to form multinucleated myotubes (reviewed in Morgan and Partridge, 2003; Sloper and Partridge, 1980) (see Figure 1A). The life-long regenerative capacity of satellite cells implies that they can be robustly and perpetually renewed while maintaining the ability to generate differentiated progeny and suggests that they may represent an adult stem cell population for skeletal muscle. Stem cells are thought to exist in many adult tissues capable of regeneration and are defined by their unique capacity to both self-renew and differentiate. Adult stem cells have been best characterized in the mammalian blood-forming system, where clonogenic, multipotent hematopoietic stem cells (HSC) have been prospectively isolated from bone marrow and demonstrate at the single-cell level the capacity to regenerate the entire hematopoietic system (reviewed in Kondo et al., 2003). While it is clear that satellite cells likewise contain unspecified precursor cells capable of extensive proliferation and differentiation to generate mature myofibers, to formally establish that they indeed function as adult stem cells, it will be necessary to demonstrate that individual cells within the satellite cell pool maintain both self-renewal and differentiation potential in vivo. Although this clonal analysis is still lacking, as outlined below, multiple lines of evidence from in vitro and in vivo studies do suggest that muscle stem cells are contained within at least a subset of satellite cells. First, as noted above, satellite cell number and regenerative capacity normally remain nearly constant through multiple cycles of injury and repair, suggesting satellite cell self-renewal. In addition, studies of isolated single myofibers and primary satellite cells maintained in culture have revealed that some cultured sat- Cell 660 Taken together, these data strongly suggest that satellite cells represent self-renewing adult muscle stem cells and alone are sufficient for muscle repair. This conclusion likely will be strengthened by ongoing studies of the capacity for and regulation of satellite cell self-renewal, particularly those testing the continued regenerative potential of single or clonally marked satellite cells in serial transplantations. Figure 1. Regeneration of Skeletal Muscle via Cell-Cell Fusion (A) Repair by endogenous myogenic satellite cells occurs in two phases. First, a subset of differentiated myoblasts (Mb) fuse together to form a nascent myotube (Mt). Next, the Mt attracts and fuses with additional Mb and, with further accretion of nuclei, eventually forms a mature myofiber (Mf). (B) Contributions from hematopoietic cells. The stage at which heterotypic fusion of myelomonocytic blood-lineage cells with muscle-lineage cells may occur in regenerating muscle has not been defined but could involve interactions with damaged Mf, proliferating Mb, nascent Mt, or newly formed Mf. ellite cells form clusters that contain both differentiated progeny and cells that regain a phenotype characteristic of quiescent satellite cells (Pax-7+ and MyoD− or myogenin−) (Olguin and Olwin, 2004; Zammit et al., 2004). Furthermore, populations of neonatal, satellite cellderived myoblasts have been shown to generate both differentiated myofibers and functional satellite cells in vivo following intramuscular transplantation (Blaveri et al., 1999; Heslop et al., 2001). In ex vivo explant cultures, satellite cells activated by muscle injury give rise to intermediate progenitor cells expressing the myogenic transcription factor Pax-3, which divide asymmetrically and differentiate into Pax-3−, Myf-5hi, desminhi myoblasts (Conboy and Rando, 2002). This progression along a myogenic lineage in adult muscle resembles embryonic myogenesis, where Pax-3 induces the expression of Myf-5 and MyoD (Maroto et al., 1997), and Pax-3+ pre-myoblast progenitor cells delaminate from somites and migrate into the limb buds to generate fusion-competent myoblasts (Goulding et al., 1994). Finally, the most compelling evidence for the stem cell nature of satellite cells comes from recent studies in which rigorously purified, single, intact myofibers, containing an average of only 7–22 associated satellite cells, were transplanted into the radiationablated muscle of immunocompromised, dystrophic (mdx-nude) hosts (Collins et al., 2005). The grafted fibers could give rise to over 100 new myofibers, contributing an estimated 25,000–30,000 differentiated myonuclei and, moreover, could regenerate substantially expanded (up to 10-fold) numbers of functional Pax-7+ satellite cells in vivo. These donor-derived satellite cells persisted in the skeletal muscle for at least several weeks and could be reactivated and expanded in response to additional muscle injury (Collins et al., 2005). Developmental Origins of Skeletal Muscle and Satellite Cells Myogenic precursors are specified during development by signals emanating from neighboring cells of the notochord, neural tube, and dorsal ectoderm. This specification depends critically on the function of myogenic transcription factors, such as Pax-3 and Pax-7 (Borycki et al., 1999; Cossu et al., 1996a, 1996b; Goulding et al., 1994). Once committed, somite-derived cells migrate to multiple sites of embryonic myogenesis, begin to express the myogenic basic helix-loop-helix transcription factors Myf-5 and MyoD (Birchmeier and Brohmann, 2000), and differentiate into muscle fibers. Somitederived myogenic progenitors that do not differentiate into myofibers at this time have been suggested instead to be retained into adulthood as muscle satellite cells (Armand et al., 1983; Mauro, 1961; Yablonka-Reuveni et al., 1987). This concept recently has been confirmed in two papers in which myogenic reporter strains and cell-lineage tracing experiments demonstrated that avian (Gros et al., 2005) and mouse (Relaix et al., 2005) embryonic and fetal myogenic progenitor cells arise from the central dermomyotome, following generation of the primary myotome. Importantly, neonatal Pax-7+ progenitor cells found in the satellite cell position were shown to originate also from the avian embryonic dermomyotome, indicating that some of the precursor cells that contribute to embryonic and fetal muscle are retained after birth as satellite cells (Gros et al., 2005). In rapidly growing neonatal muscle, nuclei of satellite cells and myoblasts comprise w30% of myofiber-associated nuclei, but after cessation of muscle growth, quiescent satellite cells represent only w5% of adult myofiber nuclei (Cardasis and Cooper, 1975). It remains to be determined, however, whether the quiescent satellite cells in adult muscle have the same developmental origin as embryonic, fetal, and neonatal cells, or, alternatively, if proliferating dermomyotome-derived myofibers-associated progenitors are exhausted during the muscle growth, and subsequent regeneration of the adult muscle invokes a distinct lineage of precursor cells. Additional analysis of myogenic reporter markers (Relaix et al., 2005) in older animals, and after repeated rounds of muscle injury and regeneration, will be particularly informative to address this issue and also may give further insight into the self-renewal potential of embryonically specified dermomyotome-derived satellite cells. Phenotypic and Functional Heterogeneity of Satellite Cells Satellite cells are classically defined by their position beneath the basal lamina and by their ability to undergo myogenic differentiation (Beauchamp et al., 2000; Mauro, Review 661 1961). However, accumulating evidence suggests that the satellite cell compartment contains cells of distinct ontogeny and function. First, although several markers have been associated with satellite cells, no single marker defines all satellite cells. For example, while most satellite cells express the surface protein CD34 (Beauchamp et al., 2000; Conboy and Rando, 2002; Sherwood et al., 2004a), they can variably express other surface markers (Sherwood et al., 2004a), as well as myogenic transcription factors, such as Pax-7, MyoD, and Myf-5 (Beauchamp et al., 2000; Cornelison and Wold, 1997; Zammit et al., 2004). CD34 and Pax-7 identify quiescent satellite cells, and Pax-7, M-Cadherin, MyoD, and Myf-5 are actually upregulated with differentiation of satellite cells into myoblasts (Morgan and Partridge, 2003; Seale et al., 2000, 2004). This heterogeneity of marker expression may reflect functional differences among satellite cells or distinct stages of myogenic lineage specification or may distinguish myogenic from nonmyogenic cell types within myofiber compartment. In this regard, it is interesting that in vitro studies have shown that some cells emerging from explanted single myofibers can express markers of osteocytes or adipocytes, rather than myogenic markers (Asakura et al., 2001; Csete et al., 2001). Recent studies from our group (Sherwood et al., 2004a) and others (Shefer et al., 2004) have indicated that single cells from within the satellite-cell compartment exhibit mutually exclusive abilities to generate either myogenic or fibroblastic and adipogenic colonies in clonal in vitro assays. Importantly, activated immune or inflammatory cells may also populate the satellite-cell compartment, infiltrating beneath the basal lamina of damaged muscle fibers (Stauber et al., 1988), although such infiltrating hematopoietic cells display no myogenic activity (Sherwood et al., 2004a). Finally, intrinsic differences in proliferation (Beauchamp et al., 1999; Rantanen et al., 1995; Rouger et al., 2004; YablonkaReuveni et al., 1987), differentiation (Rantanen et al., 1995; Zammit et al., 2004), and fusogenic capacity (Rouger et al., 2004) among individual satellite cells have been reported. More detailed studies employing prospective isolation and/or clonal marking to further delineate the precise lineage relationships, cellular interaction, and myogenic capacities of these distinct populations of myofiber-associated cells will greatly enhance our understanding of their normal roles and relative importance during muscle regeneration. Adult Myogenic Cells Distinct from Satellite Cells Recent reports have suggested that adult skeletal muscle progenitors distinct from satellite cells may function in some models of muscle injury and repair. For example, muscle-resident side population (muSP) cells, defined by their ability to exclude Hoechst 33342 (Asakura et al., 2002), have been shown to contribute to myofibers when injected intramuscularly (McKinney-Freeman et al., 2002) or when cocultured with myoblasts (Asakura et al., 2002), although these cells lack myogenic activity when cultured alone (Asakura et al., 2002). Similarly, although they do not generate myogenic cells when cultured alone, interstitial muscle-resident CD45+Sca-1+ cells reportedly acquire myogenic activity when cocultured with primary myoblasts or in response to muscle injury or Wnt signaling (Polesskaya et al., 2003). Finally, cells with high proliferative potential and surprisingly broad differentiation capacity, and thus termed “muscle-derived stem cells” (MDSC) or “multipotent adult progenitor cells” (MAPC), have been isolated following prolonged culture of cells from muscle (Cao et al., 2003; Jiang et al., 2002; Qu-Petersen et al., 2002). Descriptions of these distinct subsets of myogenic cells have raised the possibility that multiple mechanisms may support adult skeletal muscle regeneration. Yet, as compared to conventional satellite cells, many of these populations, with the notable exception of cultured MDSC (Qu-Petersen et al., 2002), display vanishingly small myogenic potential. In addition, in all cases, the contribution of these cells to the maintenance or repair of skeletal muscle under physiologic conditions is uncertain, and their therapeutic potential has not been clearly established. Moreover, recent data suggest that satellite cells are alone sufficient to mediate extensive regeneration of damaged adult skeletal muscle in vivo (Collins et al., 2005). Thus, in further assessing the intrinsic myogenic function of these populations, it will be essential to exclude the possibility that their activity derives from “contamination” with a small subset of highly myogenic satellite cells or their progeny or from a change in developmental potential induced by cell culture. The use of lineage-tracing methods (Gros et al., 2005; Relaix et al., 2005) and single myofiber transplantation (Collins et al., 2005) will be a key first step in determining the myogenicity of these populations in relationship to conventional satellite cells. Bone marrow cells (Ferrari et al., 1998; Fukada et al., 2002; Gussoni et al., 1999; LaBarge and Blau, 2002), and even single hematopoietic stem cells (Camargo et al., 2003; Corbel et al., 2003; Sherwood et al., 2004b), also have been reported to contribute to myofibers when injected directly into injured muscle or intravenously into irradiated injured or dystrophic animals. The frequency with which these unexpected contributions to skeletal muscle have been detected has varied widely and, while generally quite low (<1% of total myofibers; Camargo et al., 2003; Corbel et al., 2003; Ferrari et al., 1998; Gussoni et al., 1999; Sherwood et al., 2004b), has been reported in some cases to reach levels of 5%–12% of differentiated cells (Abedi et al., 2004; LaBarge and Blau, 2002). In this regard, it is worth noting that the method of detection of donor-derived myofibers can significantly influence the estimation of donor contributions, and assay systems employing highly diffusible markers, such as GFP (Jockusch and Voigt, 2003), cannot accurately quantify the numbers of donor myonuclei incorporated due to spread of the marker throughout the myofiber. In any case, such observations initially generated a great deal of excitement, suggesting that bone marrow could represent a reasonably accessible, novel source of regenerative cells for muscle repair; however, as with other nonsatellite cell populations, the physiological significance of bone marrow contributions to skeletal muscle remains uncertain (Ferrari et al., 2001; Gussoni et al., 2002), and accumulating evidence suggests that these events may in fact represent an accidental or even pathological re- Cell 662 sponse elicited by severe muscle damage and inflammation (see below). Interestingly, while early efforts aimed at eliciting myogenic activity from bone marrow cells focused largely on the hematopoietic lineages, recent studies have suggested that nonhematopoietic elements isolated from adult bone marrow or from embryonic sites of hematopoietic development might eventually be harnessed for therapeutic skeletal muscle regeneration. For instance, cultured “mesangioblasts” (Minasi et al., 2002), a subset of blood vessel-associated cells originating from the embryonic dorsal aorta region, which have been shown to generate multiple mesodermal cell types following in vivo transplantation, can, when injected intra-arterially into dystrophic α-sarcoglycan knockout mice, contribute to myofiber formation and significantly improve muscle function (Sampaolesi et al., 2003). In addition, clonal, cultured marrow-derived stromal cells, lacking expression of CD34, c-Kit, and CD45, have been shown to participate in myogenesis in vitro and in vivo, regenerating myofibers and sublaminar Pax-7+ satellite cells in immunocompromised dystrophic mice (Dezawa et al., 2005). Significantly, the myogenic contributions from marrow stromal cells observed in this study were much more robust than those typical of BM-derived HSC or hematopoietic progenitor cell subsets; however, it is important to note that the activation of a myogenic program by these stromal cells absolutely required in vitro induction, first by culturing the cells with certain growth factors and then by ectopic expression of constitutively active Notch-1 (Dezawa et al., 2005). Thus, while this work suggests a promising therapeutic potential of marrow stromal cells, these data do not necessarily suggest that analogous populations of marrow cells normally contribute to physiologic muscle regeneration. In summary, while under particular conditions some cells apparently distinct from satellite cells can contribute to muscle tissue, the preponderance of evidence indicates that the myogenic activity normally responsible for robust adult muscle regeneration is largely restricted to sublaminar CD34+ satellite cells. Mechanisms of Bone Marrow Contribution to Skeletal Muscle: Cell Fusion or Transdifferentiation? In most studies in which hematopoietic or bone marrow-derived contributions to skeletal muscle have been detected, significant muscle injury has been necessary, except in particular muscles (e.g., panniculus carnosus) (Corbel et al., 2003; Sherwood et al., 2004b). The mechanisms underlying these contributions have been an area of intense research. While initial reports appeared to favor the direct “transdifferentiation” of transplanted BM cells to generate satellite cells (Fukada et al., 2002; LaBarge and Blau, 2002), more recent studies have indicated that donor-marker-expressing myofibers arise via fusion of donor hematopoietic cells with host muscle cells (Camargo et al., 2003; Doyonnas et al., 2004; Sherwood et al., 2004b) (Figure 1B). While the precise cell types involved in these fusion events have not been fully defined, transplantation of BM cells from transgenic mice expressing Cre in a hematopoietic lineagerestricted manner suggests that at least one of the fu- sion partners is likely to be a committed blood myeloid cell (Camargo et al., 2003; Doyonnas et al., 2004). The stage of differentiation at which muscle-lineage cells participate in heterotypic cell fusions with infiltrating hematopoietic cells remains unknown (Figure 1B). However, when considering the mechanism by which hematopoietic cells fuse with muscle cells, it is important to remember that normally muscle is repaired by fusion of myoblasts with each other, with nascent myotubes, and/or with damaged multinucleated fibers. Thus, macrophages and other inflammatory cells known to infiltrate injured muscle fibers (Stauber et al., 1988) almost certainly are exposed to physiologic fusogenic signals. Significantly, macrophages themselves are known to undergo cell-cell fusion both physiologically, to generate osteoclasts, and pathologically, to generate multinucleated giant cells (Vignery, 2000). Therefore, heterotypic cell fusion between endogenous muscle cells and infiltrating blood cells may occur via mechanisms that normally allow the homotypic fusion of these cells in the context of tissue maintenance and repair. At present the molecular mediators, as well as the overall significance, of these rare events remain unclear. Future studies using coculture or transplantation models will be required to identify the secreted factors, cell-surface receptors, and signaling pathways involved in bloodcell/muscle-cell fusion. Knowledge of these mechanisms ultimately should allow experiments to block its occurrence and thereby test its physiologic importance. Prospective Isolation of Muscle Precursor Cells We recently reported a methodology that permits the prospective isolation by fluorescence-activated cell sorting (FACS) of highly myogenic muscle precursor cells from other cell populations contained within the satellite-cell compartment (Sherwood et al., 2004a). Combinatorial analysis of multiple cell-surface markers indicated that autonomously myogenic colony-forming cells (CFC) were highly enriched among the CD45−Sca1−Mac-1−CXCR4+β1-integrin+ (CSM4B) subset of myofiber-associated satellite cells, and that individual CSM4B cells efficiently form myogenic colonies, which express myosin heavy chain (MyHC) upon induction of myogenic differentiation in vitro. CSM4B cells are contained within a population of cells (CD45−Sca-1−CD34+) that also generates myofibers in vivo upon intramuscular injection and expresses mRNA encoding the myogenic transcription factors MyoD, Myf-5, and Pax-7 (Sherwood et al., 2004a). The precise relationship of the CSM4B subset of myofiber-associated cells to previously described satellite cells is clearly of interest in synthesizing recent literature and developing a further understanding of skeletal muscle biology and regeneration. These cells coexpress CD34 and c-met, previously reported satellite-cell surface markers (Beauchamp et al., 2000; Cornelison et al., 2001), but are not contained in c-kit+, CD13+, CD71+, Flk-1+, CD105+, CD44+, α1-integrin+, or α6-integrin+ populations of myofiber-associated cells. They also fail to express the pan-hematopoietic marker CD45 and the surface protein Sca-1, in either resting or regenerating (2 days following cardiotoxin injection) muscle; both CD45 and Sca-1 have been reported by others (Polesskaya Review 663 et al., 2003) to identify a subset of Wnt-responsive muscle-regenerative cells distinct from satellite cells and likely residing in the muscle interstitium. Finally, CSM4B cells are not derived from transplanted bone marrow cells and are not repopulated from the circulation at detectable levels, indicating that they likely arise from and are maintained by a lineage of cells distinct from bone marrow-derived, hematopoietic, and fibroblastic cells also present in muscle. These data are consistent with previous studies in muscle transplantation systems indicating that muscle repair does not generally involve the recruitment of regenerative cells from a distance (Jockusch and Voigt, 2003; Washabaugh et al., 2004). Other groups similarly have undertaken phenotypic analysis and prospective isolation of myogenic satellite-cell populations. Using Pax-3/GFP “knockin” mice, in which a GFP marker reports transcriptional activity of the Pax-3 locus, Montarras and colleagues also isolated a highly regenerative population of myogenic precursor cells. Like CSM4B cells, these Pax-3+ cells are largely CD34+, CD45−, and Sca-1− (Montarras et al., 2005). In these studies, intramuscular delivery of only a few thousand freshly sorted CD34+CD45−Sca-1− cells yielded substantial regeneration of normal myofibers in irradiated mdx-nude recipients; however, strikingly, cell culture prior to transplant dramatically reduced the efficiency of muscle engraftment (by as much as 10-fold). Biochemical Pathways Regulating Muscle Regeneration Muscle remodeling involves myogenesis, reinnervation, and revascularization and is regulated by multiple biochemical pathways, including those initiated by inflammatory cytokines, growth factors, and the evolutionarily conserved Notch, Wnt, and Sonic Hedgehog (Shh) signaling pathways (Conboy and Rando, 2002; Husmann et al., 1996; Pola et al., 2003; Polesskaya et al., 2003; Tidball, 2005). Muscle repair coincides with injury-induced inflammation, and some inflammatory cytokines, such as IL-4, LIF, TGF-β, IL-6, and TNF-α regulate myogenic potential (Tidball, 2005). Damaged muscle produces monocyte and macrophage chemoattractants, and blockade of inflammatory cell infiltration impairs muscle regeneration (Chazaud et al., 2003; Jejurikar and Kuzon, 2003; Lescaudron et al., 1999), possibly due to a reduction in macrophage-secreted factors inducing myoblast proliferation (Bondesen et al., 2004; Robertson et al., 1993). In addition to initiating the inflammatory response, injury promotes the release of growth factors that bind to extracellular matrix (ECM) proteins, such as proteoheparan sulfates (Husmann et al., 1996). This process involves the activity of matrix metalloproteinases, recently shown to play a role in muscle repair (Carmeli et al., 2004). The most studied growth factors participating in muscle maintenance and regeneration are FGFs, HGF, IGF-1, and GDF8/myostatin (Heszele and Price, 2004; Husmann et al., 1996; Lee, 2004; Miller et al., 2000). FGF-2 and HGF promote proliferation of myogenic progenitors and delay their differentiation, in part by inhibiting the expression of myogenic regulatory factors, such as MyoD (Maley et al., 1994; Miller et al., 2000). Both growth factors require heparan sulfate proteoglycans for signaling via their receptors, and syndecan-3 and -4 have been identified as the relevant cell-surface proteoglycans expressed by satellite cells (Cornelison et al., 2001). IGF-1 promotes myogenic differentiation and enhances protein synthesis in differentiated myofibers by activating the translation factor 4E-BP and the ribosomal protein S6 kinase (p70S5K) and by inhibiting muscle-specific E3 ligases that promote protein degradation (reviewed in Heszele and Price, 2004). These mechanisms and the antiapoptotic effects of IGF-1 via the suppression of caspases and activation of Akt (Downward, 2004; Lawlor and Rotwein, 2000) likely explain why IGF-1 attenuates experimentally induced muscle wasting (Shavlakadze et al., 2005). On the other end of the proliferative spectrum, the muscle-specific TGF-β family member, GDF8, inhibits cell-cycle progression in myogenic progenitors during embryonic development and in adult muscle via induction of p21 and suppression of the cyclin-dependent kinase CDK25 (McCroskery et al., 2003; McPherron et al., 1997; Thomas et al., 2000; Zimmers et al., 2002). The appropriate expansion followed by the timely differentiation of myogenic progenitor cells during muscle repair appear to be regulated by the same conserved mechanisms that orchestrate embryonic organogenesis. Namely, proliferation of satellite cells in response to muscle injury is positively regulated by the Notch pathway, while terminal myogenic differentiation of these cells requires the inhibition of Notch by its intracellular antagonist Numb (Conboy and Rando, 2002). Notch receptor is expressed in quiescent satellite cells, and Notch signal transduction responsible for satellitecell activation is initiated by the upregulation of Notch ligand, Delta, on the fibers adjacent to the damaged muscle and on the satellite cells themselves (Conboy et al., 2003; Conboy and Rando, 2002). The recent demonstration that BM-derived stromal cells become myogenic after stable expression of constitutively active Notch-1 (Dezawa et al., 2005) may suggest that activation of the Notch pathway generally regulates specification of organ precursor cells toward a myogenic lineage when other myogenic factors, such as FGF-2, are present. In addition to Notch, Shh mRNA and protein, as well as the Patched receptor, become upregulated in regenerating skeletal muscle (Pola et al., 2003). Moreover, ectopic Shh appears to ameliorate experimentally induced muscle atrophy (Alzghoul et al., 2004), thus demonstrating that yet another classic regulator of embryonic development likely also participates in adult myogenesis. Additionally, Wnt signaling has been reported to be important for the presence of CD45+ cells in regenerating adult muscle, although the physiologic myogenic potential of these cells remains unclear (Polesskaya et al., 2003; Sherwood et al., 2004a). Future experiments that decipher how muscle regulates its own inflammatory response and clarify the temporal and spatial crosstalk between Notch, Shh, and Wnt pathways will be instrumental for providing a better understanding of how cell fate is determined during muscle repair. Cell 664 Satellite-Cell Activity in Diseased Muscle In certain pathological states, including congenital myopathies, denervation, and muscle atrophy, satellitecell numbers and proliferative potential may decrease (Jejurikar and Kuzon, 2003). In muscular dystrophy (MD), repeated cycles of muscle regeneration, brought on by repeated loss of differentiated tissue, may lead to an early loss of the proliferative potential of satellite cells in these patients, and a subsequent failure to maintain muscle homeostasis (Luz et al., 2002). Although the underlying mechanism for this loss of satellite-cell responsiveness in diseased muscle has not been fully elucidated, these findings indicate that under particular circumstances satellite cells may be functionally exhausted. Satellite-cell exhaustion may relate, at least in part, to shortening of telomere ends after repeated rounds of DNA replication (Collins et al., 2003; Di Donna et al., 2003), to recurrent exposure to inflammatory conditions and/or oxidative stress (Renault et al., 2002), to an accumulation of mutations in key satellite-cell regulatory genes, introduced during repeated rounds of proliferation, or to a combination of these and other factors. Nonmyogenic cells in the muscle may also contribute to failed muscle regeneration, as fibroblasts in dystrophic patients have been shown to secrete increased levels of IGF-1 binding proteins, which could sequester this cytokine away from myogenic cells (Melone et al., 2000). A better understanding of the dynamic interplay among distinct populations of cells resident in the muscle and recruited by muscle damage will aid in developing a full picture of the complex cellular networks responsible for myogenesis in healthy and diseased muscle and for designing therapeutic strategies to promote muscle repair. Age-Related Changes in the Molecular Regulation of Satellite Cell Activity One characteristic of aging is a decline in the ability of organ stem cells to repair damaged tissues. Adult skeletal muscle is a perfect example of a tissue that robustly regenerates throughout adult life but fails to do so in old age (Grounds, 1998). The reason for this decline in regenerative potential is not completely understood and may involve both intrinsic molecular changes in the stem cells themselves and/or alterations in their aged environment. As mentioned above, muscle repair relies on Notch activity, which is necessary and sufficient for the activation of satellite cells (Artavanis-Tsakonas et al., 1999; Conboy et al., 2003; Conboy and Rando, 2002). Importantly, Notch receptor continues to be expressed in aged satellite cells, while injury-specific induction of the Notch ligand Delta, and therefore subsequent signal transduction, fail with age, resulting in grossly inefficient regeneration of old muscle tissue (Conboy et al., 2003). Strikingly, productive tissue repair can be restored to old muscle by enforced activation of Notch, while the repair of young muscle is severely perturbed when Notch signaling is inhibited (Conboy et al., 2003). Therefore, it seems that Notch is the key age-related determinant of muscle regenerative potential. Is the age-related decline in satellite cell regenerative potential intrinsic or dependent on the cell environ- ment? In early muscle transplantation studies, small minced or intact muscle from young or old rodents was transplanted into either young or old muscle beds, and the ability of donor muscle pieces to regenerate in the middle of either young or old host limbs was examined. Amazingly, the efficiency of muscle regeneration was clearly determined in these experiments by the age of the host environment, rather than by the age of the muscle donor (Carlson and Faulkner, 1989; Zacks and Sheff, 1982). In these important studies, the small transplanted muscle was physically isolated from the host satellite cells, revealing the effects of prevalent local and organismal environments on the regenerative potential of the donor satellite cells. In more recent studies, the age of the systemic environment likewise dominated over the intrinsic agerelated regenerative properties of satellite cells when the efficiency of muscle repair was examined in young and old mice with a shared blood circulation (Conboy et al., 2005). Significantly, regeneration-specific DeltaNotch signaling, appropriate activation of satellite cells, and general success in muscle repair were all rejuvenated by the exposure of aged tissue to a young systemic milieu (Conboy et al., 2005). In concert with the above-mentioned pivotal role of satellite cells as muscle stem cells, the aged satellite cells endogenous to the old muscle successfully engaged in tissue repair without any recruitment of young cells from the shared circulation (Conboy et al., 2005). These data strongly suggest that the molecular pathways responsible for muscle repair are regulated by systemic factors and that these factors change with age in ways precluding the activation of satellite cells. Multiple lines of evidence suggest that many cellintrinsic changes occur with tissue aging, including the accumulation of oxidative damage, a decline in genome maintenance, and diminished mitochondrial function (Ames, 2004; Golden et al., 2002; Hasty et al., 2003). The rejuvenation of aged stem and progenitor cells by the young extrinsic milieu, even in the presence of these age-related changes, suggests the intriguing possibility that the aging of organ stem cells might be regulated extrinsically and that the molecular changes underlying the loss of the tissue-regenerative potential with age can be reversed or overcome if the stem cell niche is young. Future identification of the relevant extrinsic age-related components will be instrumental for therapies aimed at enhancing the regenerative potential of organ stem cells in aged individuals. Future Avenues and Perspectives It is quite clear that endogenous skeletal muscle satellite cells associated with myofibers account for most, if not all, physiologic muscle-regenerative potential and likely represent muscle stem cells. Recent advances have allowed a more refined determination of their origin, position in the myogenic cell lineage, and molecular pathways regulating their function. Other avenues of muscle repair, e.g., by bone marrow-derived cells, may exist; however, unambiguous determination of the precise cell types and specific fusion and reprogramming mechanisms involved in this process will be needed in order to establish whether such cells form muscle tis- Review 665 sue under physiologic conditions or can be used therapeutically. Current advances in our understanding of the cellular and molecular mechanisms regulating cellfate determination and tissue specification during adult muscle repair have indicated a remarkable conservation of developmentally regulated signal transduction pathways, and age-related analysis of these pathways indicates that at least some of them deteriorate in old muscle, causing ineffective tissue repair. The identification of age-related systemic factors that regulate the regenerative capacity of organ stem cells will improve our understanding of aging as a conserved biological process and will help to develop therapies for the enhancement of the regenerative potential often lost in old age or disease. Acknowledgments We thank T. Partridge and D. Montarras for communication of unpublished data and for helpful discussions. This work was supported in part by a Burroughs Wellcome Fund Career Award and a Harvard Stem Cell Institute Seed Funding Grant to A.J.W. and NIA KO-1 AG25652 to I.M.C. References Abedi, M., Greer, D.A., Colvin, G.A., Demers, D.A., Dooner, M.S., Harpel, J.A., Weier, H.U., Lambert, J.F., and Quesenberry, P.J. (2004). Robust conversion of marrow cells to skeletal muscle with formation of marrow-derived muscle cell colonies: a multifactorial process. Exp. Hematol. 32, 426–434. Alzghoul, M.B., Gerrard, D., Watkins, B.A., and Hannon, K. (2004). Ectopic expression of IGF-I and Shh by skeletal muscle inhibits disuse-mediated skeletal muscle atrophy and bone osteopenia in vivo. FASEB J. 18, 221–223. phic mouse muscle: studies on isolated fibers. Dev. Dyn. 216, 244–256. Bondesen, B.A., Mills, S.T., Kegley, K.M., and Pavlath, G.K. (2004). The COX-2 pathway is essential during early stages of skeletal muscle regeneration. Am. J. Physiol. Cell Physiol. 287, C475–C483. Borycki, A.G., Li, J., Jin, F., Emerson, C.P., and Epstein, J.A. (1999). Pax3 functions in cell survival and in pax7 regulation. Development 126, 1665–1674. Camargo, F.D., Green, R., Capetenaki, Y., Jackson, K.A., and Goodell, M.A. (2003). Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat. Med. 9, 1520–1527. Cao, B., Zheng, B., Jankowski, R.J., Kimura, S., Ikezawa, M., Deasy, B., Cummins, J., Epperly, M., Qu-Petersen, Z., and Huard, J. (2003). Muscle stem cells differentiate into haematopoietic lineages but retain myogenic potential. Nat. Cell Biol. 5, 640–646. Cardasis, C.A., and Cooper, G.W. (1975). An analysis of nuclear numbers in individual muscle fibers during differentiation and growth: a satellite cell-muscle fiber growth unit. J. Exp. Zool. 191, 347–358. Carlson, B.M., and Faulkner, J.A. (1989). Muscle transplantation between young and old rats: age of host determines recovery. Am. J. Physiol. 256, C1262–C1266. Carmeli, E., Moas, M., Reznick, A.Z., and Coleman, R. (2004). Matrix metalloproteinases and skeletal muscle: a brief review. Muscle Nerve 29, 191–197. Chazaud, B., Sonnet, C., Lafuste, P., Bassez, G., Rimaniol, A.C., Poron, F., Authier, F.J., Dreyfus, P.A., and Gherardi, R.K. (2003). Satellite cells attract monocytes and use macrophages as a support to escape apoptosis and enhance muscle growth. J. Cell Biol. 163, 1133–1143. Collins, C.A., Olsen, I., Zammit, P.S., Heslop, L., Petrie, A., Partridge, T.A., and Morgan, J.E. (2005). Stem cell function, selfrenewal, and behavioral heterogeneity of cells from the adult muscle satellite cell niche. Cell 122, 1–13. Ames, B.N. (2004). Delaying the mitochondrial decay of aging. Ann. N Y Acad. Sci. 1019, 406–411. Collins, M., Renault, V., Grobler, L.A., St Clair Gibson, A., Lambert, M.I., Wayne Derman, E., Butler-Browne, G.S., Noakes, T.D., and Mouly, V. (2003). Athletes with exercise-associated fatigue have abnormally short muscle DNA telomeres. Med. Sci. Sports Exerc. 35, 1524–1528. Armand, O., Boutineau, A.M., Mauger, A., Pautou, M.P., and Kieny, M. (1983). Origin of satellite cells in avian skeletal muscles. Arch. Anat. Microsc. Morphol. Exp. 72, 163–181. Conboy, I.M., and Rando, T.A. (2002). The regulation of Notch signaling controls satellite cell activation and cell fate determination in postnatal myogenesis. Dev. Cell 3, 397–409. Artavanis-Tsakonas, S., Rand, M.D., and Lake, R.J. (1999). Notch signaling: cell fate control and signal integration in development. Science 284, 770–776. Conboy, I.M., Conboy, M.J., Smythe, G.M., and Rando, T.A. (2003). Notch-mediated restoration of regenerative potential to aged muscle. Science 302, 1575–1577. Asakura, A., Komaki, M., and Rudnicki, M. (2001). Muscle satellite cells are multipotential stem cells that exhibit myogenic, osteogenic, and adipogenic differentiation. Differentiation 68, 245–253. Conboy, I.M., Conboy, M.J., Wagers, A.J., Girma, E.R., Weissman, I.L., and Rando, T.A. (2005). Rejuvenation of aged progenitor cells by exposure to a young systemic environment. Nature 433, 760– 764. Asakura, A., Seale, P., Girgis-Gabardo, A., and Rudnicki, M.A. (2002). Myogenic specification of side population cells in skeletal muscle. J. Cell Biol. 159, 123–134. Beauchamp, J.R., Morgan, J.E., Pagel, C.N., and Partridge, T.A. (1999). Dynamics of myoblast transplantation reveal a discrete minority of precursors with stem cell-like properties as the myogenic source. J. Cell Biol. 144, 1113–1122. Beauchamp, J.R., Heslop, L., Yu, D.S., Tajbakhsh, S., Kelly, R.G., Wernig, A., Buckingham, M.E., Partridge, T.A., and Zammit, P.S. (2000). Expression of CD34 and Myf5 defines the majority of quiescent adult skeletal muscle satellite cells. J. Cell Biol. 151, 1221–1234. Bintliff, S., and Walker, B.E. (1960). Radioautographic study of skeletal muscle regeneration. Am. J. Anat. 106, 233. Corbel, S.Y., Lee, A., Yi, L., Duenas, J., Brazelton, T.R., Blau, H.M., and Rossi, F.M. (2003). Contribution of hematopoietic stem cells to skeletal muscle. Nat. Med. 9, 1528–1532. Cornelison, D.D., Filla, M.S., Stanley, H.M., Rapraeger, A.C., and Olwin, B.B. (2001). Syndecan-3 and syndecan-4 specifically mark skeletal muscle satellite cells and are implicated in satellite cell maintenance and muscle regeneration. Dev. Biol. 239, 79–94. Cornelison, D.D., and Wold, B.J. (1997). Single-cell analysis of regulatory gene expression in quiescent and activated mouse skeletal muscle satellite cells. Dev. Biol. 191, 270–283. Birchmeier, C., and Brohmann, H. (2000). Genes that control the development of migrating muscle precursor cells. Curr. Opin. Cell Biol. 12, 725–730. Cossu, G., Kelly, R., Tajbakhsh, S., Di Donna, S., Vivarelli, E., and Buckingham, M. (1996a). Activation of different myogenic pathways: myf-5 is induced by the neural tube and MyoD by the dorsal ectoderm in mouse paraxial mesoderm. Development 122, 429– 437. Bischoff, R. (1975). Regeneration of single skeletal muscle fibers in vitro. Anat. Rec. 182, 215–235. Cossu, G., Tajbakhsh, S., and Buckingham, M. (1996b). How is myogenesis initiated in the embryo? Trends Genet. 12, 218–223. Blaveri, K., Heslop, L., Yu, D.S., Rosenblatt, J.D., Gross, J.G., Partridge, T.A., and Morgan, J.E. (1999). Patterns of repair of dystro- Csete, M., Walikonis, J., Slawny, N., Wei, Y., Korsnes, S., Doyle, J.C., and Wold, B. (2001). Oxygen-mediated regulation of skeletal Cell 666 muscle satellite cell proliferation and adipogenesis in culture. J. Cell. Physiol. 189, 189–196. Dezawa, M., Ishikawa, H., Itokazu, Y., Yoshihara, T., Hoshino, M., Takeda, S., Ide, C., and Nabeshima, Y. (2005). Bone marrow stromal cells generate muscle cells and repair muscle degeneration. Science 309, 314–317. Di Donna, S., Mamchaoui, K., Cooper, R.N., Seigneurin-Venin, S., Tremblay, J., Butler-Browne, G.S., and Mouly, V. (2003). Telomerase can extend the proliferative capacity of human myoblasts, but does not lead to their immortalization. Mol. Cancer Res. 1, 643–653. Downward, J. (2004). PI 3-kinase, Akt and cell survival. Semin. Cell Dev. Biol. 15, 177–182. Doyonnas, R., LaBarge, M.A., Sacco, A., Charlton, C., and Blau, H.M. (2004). Hematopoietic contribution to skeletal muscle regeneration by myelomonocytic precursors. Proc. Natl. Acad. Sci. USA 101, 13507–13512. Ferrari, G., Cusella-De Angelis, G., Coletta, M., Paolucci, E., Stornaiuolo, A., Cossu, G., and Mavilio, F. (1998). Muscle regeneration by bone marrow-derived myogenic progenitors. Science 279, 1528–1530. Ferrari, G., Stornaiuolo, A., and Mavilio, F. (2001). Failure to correct murine muscular dystrophy. Nature 411, 1014–1015. Fukada, S., Miyagoe-Suzuki, Y., Tsukihara, H., Yuasa, K., Higuchi, S., Ono, S., Tsujikawa, K., Takeda, S., and Yamamoto, H. (2002). Muscle regeneration by reconstitution with bone marrow or fetal liver cells from green fluorescent protein-gene transgenic mice. J. Cell Sci. 115, 1285–1293. Golden, T.R., Hinerfeld, D.A., and Melov, S. (2002). Oxidative stress and aging: beyond correlation. Aging Cell 1, 117–123. Goulding, M., Lumsden, A., and Paquette, A.J. (1994). Regulation of Pax-3 expression in the dermomyotome and its role in muscle development. Development 120, 957–971. Gros, J., Manceau, M., Thome, V., and Marcelle, C. (2005). A common somitic origin for embryonic muscle progenitors and satellite cells. Nature 435, 954–958. Beilhack, G.F., Shizuru, J.A., and Weissman, I.L. (2003). Biology of hematopoietic stem cells and progenitors: implications for clinical application. Annu. Rev. Immunol. 21, 759–806. Konigsberg, I.R. (1963). Clonal analysis of myogenesis. Science 140, 1273–1284. LaBarge, M.A., and Blau, H.M. (2002). Biological progression from adult bone marrow to mononucleate muscle stem cell to multinucleate muscle fiber in response to injury. Cell 111, 589–601. Lawlor, M.A., and Rotwein, P. (2000). Coordinate control of muscle cell survival by distinct insulin-like growth factor activated signaling pathways. J. Cell Biol. 151, 1131–1140. Lee, S.J. (2004). Regulation of muscle mass by myostatin. Annu. Rev. Cell Dev. Biol. 20, 61–86. LeGros Clark, W.E. (1946). An experimental study of regeneration of mammalian striped muscle. J. Anat. 80, 24–36. Lescaudron, L., Peltekian, E., Fontaine-Perus, J., Paulin, D., Zampieri, M., Garcia, L., and Parrish, E. (1999). Blood borne macrophages are essential for the triggering of muscle regeneration following muscle transplant. Neuromuscul. Disord. 9, 72–80. Lipton, B.H., and Schultz, E. (1979). Developmental fate of skeletal muscle satellite cells. Science 205, 1292–1294. Luz, M.A., Marques, M.J., and Santo Neto, H. (2002). Impaired regeneration of dystrophin-deficient muscle fibers is caused by exhaustion of myogenic cells. Braz. J. Med. Biol. Res. 35, 691–695. Maley, M.A., Fan, Y., Beilharz, M.W., and Grounds, M.D. (1994). Intrinsic differences in MyoD and myogenin expression between primary cultures of SJL/J and BALB/C skeletal muscle. Exp. Cell Res. 211, 99–107. Maroto, M., Reshef, R., Munsterberg, A.E., Koester, S., Goulding, M., and Lassar, A.B. (1997). Ectopic Pax-3 activates MyoD and Myf-5 expression in embryonic mesoderm and neural tissue. Cell 89, 139–148. Mauro, A. (1961). Satellite cells of muscle skeletal fibers. J. Biophys. Biochem. 9, 493–495. Grounds, M.D. (1998). Age-associated changes in the response of skeletal muscle cells to exercise and regeneration. Ann. N Y Acad. Sci. 854, 78–91. McCroskery, S., Thomas, M., Maxwell, L., Sharma, M., and Kambadur, R. (2003). Myostatin negatively regulates satellite cell activation and self-renewal. J. Cell Biol. 162, 1135–1147. Gussoni, E., Soneoka, Y., Strickland, C.D., Buzney, E.A., Khan, M.K., Flint, A.F., Kunkel, L.M., and Mulligan, R.C. (1999). Dystrophin expression in the mdx mouse restored by stem cell transplantation. Nature 401, 390–394. McKinney-Freeman, S.L., Jackson, K.A., Camargo, F.D., Ferrari, G., Mavilio, F., and Goodell, M.A. (2002). Muscle-derived hematopoietic stem cells are hematopoietic in origin. Proc. Natl. Acad. Sci. USA 99, 1341–1346. Gussoni, E., Bennett, R.R., Muskiewicz, K.R., Meyerrose, T., Nolta, J.A., Gilgoff, I., Stein, J., Chan, Y.M., Lidov, H.G., Bonnemann, C.G., et al. (2002). Long-term persistence of donor nuclei in a Duchenne muscular dystrophy patient receiving bone marrow transplantation. J. Clin. Invest. 110, 807–814. McPherron, A.C., Lawler, A.M., and Lee, S.J. (1997). Regulation of skeletal muscle mass in mice by a new TGF-beta superfamily member. Nature 387, 83–90. Hasty, P., Campisi, J., Hoeijmakers, J., van Steeg, H., and Vijg, J. (2003). Aging and genome maintenance: lessons from the mouse? Science 299, 1355–1359. Melone, M.A., Peluso, G., Galderisi, U., Petillo, O., and Cotrufo, R. (2000). Increased expression of IGF-binding protein-5 in Duchenne muscular dystrophy (DMD) fibroblasts correlates with the fibroblast-induced downregulation of DMD myoblast growth: An in vitro analysis. J. Cell. Physiol. 185, 143–153. Heslop, L., Beauchamp, J.R., Tajbakhsh, S., Buckingham, M.E., Partridge, T.A., and Zammit, P.S. (2001). Transplanted primary neonatal myoblasts can give rise to functional satellite cells as identified using the Myf5nlacZl+ mouse. Gene Ther. 8, 778–783. Miller, K.J., Thaloor, D., Matteson, S., and Pavlath, G.K. (2000). Hepatocyte growth factor affects satellite cell activation and differentiation in regenerating skeletal muscle. Am. J. Physiol. Cell Physiol. 278, C174–C181. Heszele, M.F., and Price, S.R. (2004). Insulin-like growth factor I: the yin and yang of muscle atrophy. Endocrinology 145, 4803–4805. Minasi, M.G., Riminucci, M., De Angelis, L., Borello, U., Berarducci, B., Innocenzi, A., Caprioli, A., Sirabella, D., Baiocchi, M., De Maria, R., et al. (2002). The meso-angioblast: a multipotent, self-renewing cell that originates from the dorsal aorta and differentiates into most mesodermal tissues. Development 129, 2773–2783. Husmann, I., Soulet, L., Gautron, J., Martelly, I., and Barritault, D. (1996). Growth factors in skeletal muscle regeneration. Cytokine Growth Factor Rev. 7, 249–258. Jejurikar, S.S., and Kuzon, W.M., Jr. (2003). Satellite cell depletion in degenerative skeletal muscle. Apoptosis 8, 573–578. Jiang, Y., Jahagirdar, B.N., Reinhardt, R.L., Schwartz, R.E., Keene, C.D., Ortiz-Gonzalez, X.R., Reyes, M., Lenvik, T., Lund, T., Blackstad, M., et al. (2002). Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418, 41–49. Jockusch, H., and Voigt, S. (2003). Migration of adult myogenic precursor cells as revealed by GFP/nLacZ labelling of mouse transplantation chimeras. J. Cell Sci. 116, 1611–1616. Kondo, M., Wagers, A.J., Manz, M.G., Prohaska, S.S., Scherer, D.C., Montarras, D., Morgan, J., Collins, C., Relaix, F., Zaffran, S., Cumano, A., Partridge, T., and Buckingham, M. (2005). Direct isolation of muscle satellite cells demonstrates their major role in skeletal muscle self renewal. Science. Published online September 1, 2005. 10.1126/science.1114758. Morgan, J.E., and Partridge, T.A. (2003). Muscle satellite cells. Int. J. Biochem. Cell Biol. 35, 1151–1156. Moss, F.P., and Leblond, C.P. (1970). Nature of dividing nuclei in skeletal muscle of growing rats. J. Cell Biol. 44, 459–462. Olguin, H.C., and Olwin, B.B. (2004). Pax-7 up-regulation inhibits Review 667 myogenesis and cell cycle progression in satellite cells: a potential mechanism for self-renewal. Dev. Biol. 275, 375–388. Pola, R., Ling, L.E., Aprahamian, T.R., Barban, E., Bosch-Marce, M., Curry, C., Corbley, M., Kearney, M., Isner, J.M., and Losordo, D.W. (2003). Postnatal recapitulation of embryonic hedgehog pathway in response to skeletal muscle ischemia. Circulation 108, 479–485. Polesskaya, A., Seale, P., and Rudnicki, M.A. (2003). Wnt signaling induces the myogenic specification of resident CD45+ adult stem cells during muscle regeneration. Cell 113, 841–852. Qu-Petersen, Z., Deasy, B., Jankowski, R., Ikezawa, M., Cummins, J., Pruchnic, R., Mytinger, J., Cao, B., Gates, C., Wernig, A., and Huard, J. (2002). Identification of a novel population of muscle stem cells in mice: potential for muscle regeneration. J. Cell Biol. 157, 851–864. Rantanen, J., Hurme, T., Lukka, R., Heino, J., and Kalimo, H. (1995). Satellite cell proliferation and the expression of myogenin and desmin in regenerating skeletal muscle: evidence for two different populations of satellite cells. Lab. Invest. 72, 341–347. Relaix, F., Rocancourt, D., Mansouri, A., and Buckingham, M. (2005). A Pax3/Pax7-dependent population of skeletal muscle progenitor cells. Nature 435, 948–953. Renault, V., Thornell, L.E., Butler-Browne, G., and Mouly, V. (2002). Human skeletal muscle satellite cells: aging, oxidative stress and the mitotic clock. Exp. Gerontol. 37, 1229–1236. Robertson, T.A., Maley, M.A., Grounds, M.D., and Papadimitriou, J.M. (1993). The role of macrophages in skeletal muscle regeneration with particular reference to chemotaxis. Exp. Cell Res. 207, 321–331. Rouger, K., Brault, M., Daval, N., Leroux, I., Guigand, L., Lesoeur, J., Fernandez, B., and Cherel, Y. (2004). Muscle satellite cell heterogeneity: in vitro and in vivo evidences for populations that fuse differently. Cell Tissue Res. 317, 319–326. Sampaolesi, M., Torrente, Y., Innocenzi, A., Tonlorenzi, R., D’Antona, G., Pellegrino, M.A., Barresi, R., Bresolin, N., De Angelis, M.G., Campbell, K.P., et al. (2003). Cell therapy of alpha-sarcoglycan null dystrophic mice through intra-arterial delivery of mesoangioblasts. Science 301, 487–492. Shavlakadze, T., White, J.D., Davies, M., Hoh, J.F., and Grounds, M.D. (2005). Insulin-like growth factor I slows the rate of denervation induced skeletal muscle atrophy. Neuromuscul. Disord. 15, 139–146. Schultz, E., Gibson, M.C., and Champion, T. (1978). Satellite cells are mitotically quiescent in mature mouse muscle: an EM and radioautographic study. J. Exp. Zool. 206, 451–456. Seale, P., Sabourin, L.A., Girgis-Gabardo, A., Mansouri, A., Gruss, P., and Rudnicki, M.A. (2000). Pax7 is required for the specification of myogenic satellite cells. Cell 102, 777–786. Seale, P., Ishibashi, J., Holterman, C., and Rudnicki, M.A. (2004). Muscle satellite cell-specific genes identified by genetic profiling of MyoD-deficient myogenic cell. Dev. Biol. 275, 287–300. Shefer, G., Wleklinski-Lee, M., and Yablonka-Reuveni, Z. (2004). Skeletal muscle satellite cells can spontaneously enter an alternative mesenchymal pathway. J. Cell Sci. 117, 5393–5404. Sherwood, R.I., Christensen, J.L., Conboy, I.M., Conboy, M.J., Rando, T.A., Weissman, I.L., and Wagers, A.J. (2004a). Isolation of adult mouse myogenic progenitors; functional heterogeneity of cells within and engrafting skeletal muscle. Cell 119, 543–554. Sherwood, R.I., Christensen, J.L., Weissman, I.L., and Wagers, A.J. (2004b). Determinants of skeletal muscle contribution from circulating cells, bone marrow cells, and hematopoietic stem cells. Stem Cells 22, 1292–1304. Sloper, J.C., and Partridge, T.A. (1980). Skeletal muscle: regeneration and transplantation studies. Br. Med. Bull. 36, 153–158. Snow, M.H. (1978). An autoradiographic study of satellite cell differentiation into regenerating myotubes following transplantation of muscles in young rats. Cell Tissue Res. 186, 535–540. Stauber, W.T., Fritz, V.K., Vogelbach, D.W., and Dahlmann, B. (1988). Characterization of muscles injured by forced lengthening. I. Cellular infiltrates. Med. Sci. Sports Exerc. 20, 345–353. Thomas, M., Langley, B., Berry, C., Sharma, M., Kirk, S., Bass, J., and Kambadur, R. (2000). Myostatin, a negative regulator of muscle growth, functions by inhibiting myoblast proliferation. J. Biol. Chem. 275, 40235–40243. Tidball, J.G. (2005). Inflammatory processes in muscle injury and repair. Am. J. Physiol. Regul. Integr. Comp. Physiol. 288, R345– R353. Vignery, A. (2000). Osteoclasts and giant cells: macrophage-macrophage fusion mechanism. Int. J. Exp. Pathol. 81, 291–304. Volkmann, R. (1893). Uber die regeneration des quergestrieften muskelgewebes beim menschen und saugetier. Bieitr. Path. Anat. 12, 233–332. Walton, J.N., and Adams, R.D. (1956). The response of the normal, the denervated and the dystrophic muscle-cell to injury. J. Pathol. Bacterol. 72, 273–298. Washabaugh, C.H., Ontell, M.P., and Ontell, M. (2004). Nonmuscle stem cells fail to significantly contribute to regeneration of normal muscle. Gene Ther. 11, 1724–1728. Weber, C.O. (1863). Ueber die regeneration quergestreifter muskelfasern. Zentrabl. Med. Wis. 1, 529–531. Yablonka-Reuveni, Z., Quinn, L.S., and Nameroff, M. (1987). Isolation and clonal analysis of satellite cells from chicken pectoralis muscle. Dev. Biol. 119, 252–259. Yaffe, D. (1969). Cellular aspects of muscle differentiation in vitro. Curr. Top. Dev. Biol. 4, 37–77. Zacks, S.I., and Sheff, M.F. (1982). Age-related impeded regeneration of mouse minced anterior tibial muscle. Muscle Nerve 5, 152– 161. Zammit, P.S., Golding, J.P., Nagata, Y., Hudon, V., Partridge, T.A., and Beauchamp, J.R. (2004). Muscle satellite cells adopt divergent fates: a mechanism for self-renewal? J. Cell Biol. 166, 347–357. Zimmers, T.A., Davies, M.V., Koniaris, L.G., Haynes, P., Esquela, A.F., Tomkinson, K.N., McPherron, A.C., Wolfman, N.M., and Lee, S.J. (2002). Induction of cachexia in mice by systemically administered myostatin. Science 296, 1486–1488.