Influence of exercise and aging on extracellular matrix composition in the skeletal muscle stem cell niche

Abstract

Skeletal muscle is endowed with a remarkable capacity for regeneration, primarily due to the reserve pool of muscle resident satellite cells. The satellite cell is the physiologically quiescent muscle stem cell that resides beneath the basal lamina and adjacent to the sarcolemma. The anatomic location of satellite cells is in close proximity to vasculature where they interact with other muscle resident stem/stromal cells (e.g., mesenchymal stem cells and pericytes) through paracrine mechanisms. This mini-review describes the components of the muscle stem cell niche, as well as the influence of exercise and aging on the muscle stem cell niche. Although exercise promotes ECM reorganization and stem cell accumulation, aging is associated with dense ECM deposition and loss of stem cell function resulting in reduced regenerative capacity and strength. An improved understanding of the niche elements will be valuable to inform the development of therapeutic interventions aimed at improving skeletal muscle regeneration and adaptation over the life span.

SKELETAL MUSCLE EXTRACELLULAR MATRIX

The extracellular matrix (ECM) in skeletal muscle plays an integral role in tissue development, structural support, and force transmission (78). The ECM accounts for up to 10% of muscle weight and is organized into three layers: the endomysium that surrounds individual muscle fibers, the perimysium that divides the muscle into fascicles, and the epimysium that provides external support to the entire muscle (47). Intramuscular connective tissue is dominated by type I collagen, a fibrous ECM protein that can vary widely in content and alignment between layers and different muscle types to accommodate function. Whereas collagen fibrils are more loosely oriented near the individual fiber to allow for contraction, collagen is densely layered and highly oriented in the perimysium to optimally transmit force to the tendon.

The basement membrane is a specialized layer of ECM that exists between the sarcolemma of each muscle fiber and the surrounding endomysium. It is subdivided into 1) the outer reticular lamina, composed mainly of collagenous fibrils (type I, III, and VI collagens) and fibronectin and 2) the inner basal lamina, composed of nonfibrillar collagen (collagen type IV) and laminin (47). Collagen IV forms a flexible network throughout the basal lamina and connects with laminins near the sarcolemma. As a cross-shaped, heterotrimeric glycoprotein consisting of α-, β-, and γ-subunits, laminin serves as a primary ligand for two sarcolemmal-associated receptors located at the costamere of Z bands, the dystrophin-associated glycoprotein complex (DGC) and the α7β1 integrin (35). Thus the collagen IV/laminin/transmembrane receptor complex provides a critical scaffold necessary for lateral transfer of mechanical force from the myofiber to the surrounding fibrous connective tissue (inside-outside force transmission) during contraction. Diminished expression of any of the structures in the basal lamina (type IV collagen, laminin, transmembrane receptors) can severely impair sarcolemmal and myofiber cytoskeletal integrity, providing the basis for different forms of muscular dystrophy (6, 58). The ECM appears to regulate transmembrane linkage protein expression in a complex manner at the level of gene transcription, because α7β1 integrin expression is severely limited in patients with laminin α2 chain congenital dystrophy, as well as mice that do not synthesize the laminin α2 chain (dy/dy) (45).

SKELETAL MUSCLE STEM CELL NICHE

A wide variety of mononuclear cell types reside in skeletal muscle that are essential for tissue repair and maintenance, including stem cells, fibroblasts, immune cells (Fig. 1A). In response to cues provided by damage or mechanical strain, quiescent and unipotent satellite cells (Pax7⁺) localized outside the sarcolemma and beneath the basal lamina become activated [express myogenic regulatory factors (MRF) Myf5 and MyoD and proliferate] and form myoblasts that can fuse together to regenerate lost tissue or fuse with existing fibers to allow for myofiber repair (14). Pax7⁻ stem/stromal cells specifically residing in the perivascular niche within the interstitium, including pericytes (NG2⁺CD146⁺PDGFRβ⁺), fibroadipogenic progenitor cells (FAPs, PDGFRα⁺), and muscle-derived mesenchymal stem cells (mMSCs, Sca-1⁺), may also directly or indirectly assist in fiber repair, yet redundant cell surface marker expression among apparent subpopulations limits our ability to assess their relative contribution at this time (8). Regardless of this gap in knowledge, it is evident that satellite cell and perivascular stem/stromal cell migration, activation, proliferation, and/or function are dependent on cues provided by the niche, including ECM composition, stiffness, topography, and porosity.

Fig. 1.

Schematic representation of the skeletal muscle stem cell niche (A) and its alteration postexercise (B) and aging (C). Exercise results in increased mesenchymal stem cell (MSC) accumulation and ECM reorganization facilitated by matrix metalloproteinases (MMPs). In contrast, aging is associated with increased ECM deposition and reduced growth factor concentration resulting in stem cell dysfunction. [Modified with permission from Taylor & Francis Ltd. (http://www.tandfonline.com) (78).]

Satellite cells and myoblasts express the α7β1 integrin and are highly reliant on the presence of laminin in the basal lamina for multiple activities, including proliferation, adhesion, migration, and differentiation within skeletal muscle (15, 22, 34, 75, 89). Laminins exist as multiple isoforms in a variety of tissues, and at least four are expressed in muscle during development, including laminin-111 (α1β1γ1; LM-111, formerly laminin-1), LM-211 (formerly laminin-2, also known as merosin), LM-121 (formerly laminin-3), and LM-221 (formerly laminin-4) (35). LM-111 is only present in skeletal muscle during early development, and LM-211/LM-221 are the preferential integrin binding isoforms present during adulthood (18). Loss of regenerative capacity in laminin-deficient (dy/dy) mice, as well as improved satellite cell expansion and activity with exogenous LM-111 injection in vivo, suggest a vital role for this ECM protein in the regulation of stem cell function postinjury (45, 71, 93). The influence of other ECM proteins on stem cell activity is also evident from recent studies. For example, fibronectin binding to syndecan-4 can stimulate Wnt7a-dependent expansion of satellite cells (5) and lack of collagen VI in Col6a^−/− mice can impair satellite cell self-renewal and repair after injury (7, 81). A recent study also suggests a role for tenascin C in the regulation of satellite cell expansion (79). In vitro studies, although limited in translational capacity, also can delineate the stem cell response to ECM proteins. By direct comparison, laminin can promote muscle stem cell proliferation and differentiation to a greater extent than collagen type I, fibronectin, and gelatin in culture (86). ECM influence on stem cells is clearly not restricted to satellite cells, because our recent work suggests collagen type I can suppress mMSC function (19) and other studies implicate collagen in the initiation of fibroblast-mediated fibrosis (positive feedback loop) (67).

ECM components not only influence stem cell behavior, but interactions are reciprocal such that stem cells also dictate niche composition. This reciprocal relationship is highlighted by studies that demonstrate reconstitution of laminin in dy/dy mice upon myoblast and CD90⁺ (mMSC) transplantation (25, 84, 85) and accumulation of collagen in wild-type mouse skeletal muscle upon conditional ablation of satellite cells (23, 64). Therefore, although fibroblasts are considered the main contributor to the ECM composition in skeletal muscle, satellite cells and perivascular stem/stromal cells also synthesize a wide variety of ECM components that are necessary for tissue remodeling, including collagens, laminins, fibronectin, and matrix metalloproteinases (MMPs) (5, 19, 36, 63, 74). Interestingly, the ECM gene signature of satellite cells suggests that several ECM components are preferentially expressed in the quiescent state (laminin α2 and γ1), and fibronectin gene and protein expression are uniquely upregulated during proliferation and differentiation (5). Thus it is clear the ECM does not simply serve as a supportive scaffold for skeletal muscle, but dynamically regulates mononuclear cell behavior in a manner that can dictate tissue development, repair, remodeling, and overall function.

EXERCISE-MEDIATED REGULATION OF THE STEM CELL NICHE

Physical activity and mechanical loading provide a strong stimulus for ECM production and degradation in skeletal muscle (Fig. 1B) (42, 43, 46, 47, 56). Collagen type I, III, and IV gene expression, their posttranslational modifying enzymes, and the concentration of hydroxyproline are increased in response to an acute bout of downhill running (39, 48, 49). Collagen type I, III, and IV gene expression is slowly elevated and is sustained in human muscle several weeks after an acute bout of eccentric exercise (46). Damage does not appear to be necessary for ECM synthesis, as an acute bout of nondamaging kicking exercise can increase collagen protein synthesis 3.5 times its resting value by 6 h (61). Collagen synthesis after acute exercise is transient as levels peak at 24 h and gradually return to baseline by 72 h. Similar increases in collagen protein synthesis are observed after heavy resistance exercise (4-fold at 8 h postexercise), regardless of contraction mode (62). MMP-2 and MMP-9, zinc- and calcium-dependent proteolytic enzymes that target collagen type IV and laminin in the basal lamina are also increased in response to both endurance and resistance exercise training (11, 66, 72, 73). Conversely, there is evidence to suggest that lack of loading in the form of immobilization can suppress both protein synthesis and gene expression of collagen type III and IV in skeletal muscle (40, 82).

ECM turnover during and postexercise likely alters the satellite cell niche in a manner that can accommodate participation, all events leading to repair, including proliferation, migration, and differentiation. Pax7⁺ satellite cells accumulate in skeletal muscle after acute and repeated bouts of strength training (4, 69), with preferential localization observed in the niche surrounding type II fibers after an acute bout of eccentric exercise (13). The fact that high-intensity eccentric exercise selectively recruits type II fibers that incur more damage than type I suggests preferential changes in the type II fiber niche that allow for satellite cell expansion. Accordingly, laminin-211 is expressed to a greater extent in predominantly type II rectus femoris muscle compared with predominantly type I soleus muscle in rats (50). Prominent localization of intracellular MMP-2 expression and gelatinolytic activity in type II fibers compared with type I fibers suggest that remodeling can occur in a muscle fiber type-specific or spatially discrete manner (11, 38). Degradation of the ECM by MMP-2 can facilitate satellite cell migration through the interstitium, as well as release hepatocyte growth factor (HGF) tethered to the matrix (88), an event that can facilitate satellite cell migration and proliferation. Interestingly, MMP-2 is not only necessary for regeneration postinjury (52) and activity-induced angiogenesis (37), but also overload-induced muscle growth (91). Although adult myofiber growth may not be dependent on satellite cell activation and/or fusion (59), it is clear that satellite cells prevent fibroblast-mediated collagen deposition and are necessary for maintenance of growth with long-term chronic loading in mice (23). In addition, our recent data suggest that perivascular stem/stromal cells positively contribute to eccentric exercise-induced gains in muscle fiber size and function in mice (94). Thus ECM remodeling of the basal lamina postexercise may not only facilitate myofiber repair when necessary, but may promote healthy expansion of the interstitial connective tissue environment to accommodate myofiber growth.

AGING AND ECM

Structural changes in the muscle ECM are associated with a decline in mechanical properties and muscle strength with age (51). Collagen deposition has been shown to increase with age using several quantitative and semiquantitative techniques such as hydroxyproline content measurement, picosirius red, or Masson's trichrome staining (54, 70). In-depth evaluation of aged muscle demonstrates that the proportion of collagen subtypes is altered such that levels of collagen type I increase and collagen type III decrease (44). Collagen type IV concentration is enhanced in the basal lamina of slow twitch muscles, whereas laminin concentration decreases with age (50). Similar to immobilization, age can dramatically decrease extracellular gene expression in rat skeletal muscle, including procollagen type I (68), suggesting that accumulation reflects decreased degradation rather than increased synthesis. In support of this hypothesis, aged skeletal muscle exhibits lower resting levels of MMP-2 (20) and baseline MMP-2 gene expression strongly correlates with gains in strength and muscle size posttraining in older adults (21). Interestingly, transplantation of fibroblasts modified to secrete MMP-9 and placenta growth factor diminished collagen and fat deposition in aged dystrophic skeletal muscle (28), suggesting the capacity to reverse ECM dysfunction with disease.

The lack of ECM turnover in aged skeletal muscle may increase susceptibility to posttranslational modifications. In fact, enzymatically mediated crosslinks and advanced glycation end products (AGE) are increased in muscle with age (41, 87). The Young's modulus of skeletal muscle is increased with age, and increased stiffness is associated with excessive accumulation of crosslinked collagen (54, 87).

Alterations in ECM composition, density, and biophysical properties (stiffness) with age can negatively influence satellite cell function. With regard to ECM composition and density, increased thickening of the basal lamina due to increased collagen deposition (Fig. 1C) can influence the physical association between satellite cells and the myofiber (76). Increased accumulation of proteolytic fragments derived from the impaired degradation of ECM proteins such fibronectin, elastin, and laminin may further limit the interaction of satellite cells with neighboring cells and muscle fibers (53). A reduction in the activity of the primary pathways responsible for the removal of damaged or modified proteins (e.g., proteasome) in aged skeletal muscle exacerbates the problem (16). In addition, ECM stiffness alone can impact the overall proliferation and differentiation capability of the satellite cell (31). By utilizing stiffness-tunable hydrogels, it was shown that satellite cells were more likely to proliferate on a hydrogel with a Young's modulus of 2 kPa (80). However, when stiffness was increased to 18 kPa (similarly observed in damaged aged myofibers), satellite cells were more likely to undergo differentiation (54). Therefore, changes in both ECM composition and mechanical properties may play an important role in reducing the regenerative capacity of the skeletal muscle with age.

Brack et al. (9) recently provided a comprehensive review regarding the impact of age on satellite cell senescence and factors that limit the ability to retain quiescence. Aging is a complex and insidious process, and whether deficiencies in satellite cell function occur as a result of structural aberrations in the niche or whether intrinsic satellite cell dysfunction provides the basis for structural modifications is not clear. Lifelong reduction of muscle satellite cells does not appear to accelerate or exacerbate atrophy or further restrict mechanical load-induced growth in aged mice. However, long-term satellite cell depletion does increase collagen content under sedentary conditions compared with age-matched control mice (24, 55), and in vitro studies support the concept that satellite cells secrete factor that prevents fibrosis. Thus deficiencies in the ability for the satellite cell to directly or indirectly modify the ECM may provide the underlying basis for age-related deficits in growth and strength.

THERAPEUTIC APPROACHES TOWARD IMPROVING THE AGED SKELETAL MUSCLE STEM CELL NICHE

Exercise training can reverse the effects of aging on skeletal muscle in both rodents (77) and humans (60). Exercise training can positively impact the passive viscoelastic properties of the skeletal muscle by reducing collagen cross-linking levels in both young (12) and aged (32, 92) rats. Lower levels of ECM crosslinking reduce the stiffness of skeletal muscle, resulting in improved mechanical properties and mechanotransduction to the resident stem cells. Resistance training is known to reduce fat infiltration in aging skeletal muscle and is an effective strategy to maintain skeletal muscle mass and cross-sectional area with age (57, 90). Unfortunately, disabilities due to injury or disease can reduce physical activity and full engagement may not be possible. Thus alternative clinical approaches must be explored.

TGF-β1 signaling promotes fibroblast-mediated collagen deposition and inhibits satellite cell activation and differentiation (27). A recent study demonstrated that administration of losartan, an angiotensin II type 1 receptor blocker, inhibits TGF-β1 signaling and improves muscle regeneration and function after cardiotoxin injury in aged mice. Losartan also enhanced muscle mass and myofiber cross-sectional area in aged, immobilized mice (10). In another study, inhibition of myostatin (a member of the TGF-β1 super-family) using an injectable propeptide resulted in increased muscle mass and muscle fiber size in aged mice (1). Thus manipulation of TGF-β1 signaling may provide a therapeutic approach to aging.

Satellite cell transplantation fails to produce any beneficial effects in mouse or human skeletal muscle due to early cell death, limited proliferation, and migration (65). Several studies have shown that decellularized ECM scaffolds (in which the ECM is preserved) can guide stem cell migration and differentiation and positively influence regenerative outcomes when implanted into injured tissues (2, 3). ECM-based biomaterials engineered to mimic the mechanical properties of healthy young muscle have been effective in rejuvenating aged muscle stem cells. It has been shown that culturing aged satellite cells on soft hydrogels (mimicking physiological stiffness of young skeletal muscle) can restore their regenerative capacity when transplanted into injured muscles of aged mice (17). In a similar study, the myogenic and angiogenic capacity of aged pericytes was improved when they were cultured on fibrinogen-based hydrogels that mimicked the stiffness and mechanical cues of young muscle ECM (26).

LM-111 supplementation has demonstrated remarkable regenerative capacity in several models of disease (33, 71, 83) and injury (93), primarily by upregulating α7 integrin protein expression in the myofiber and influencing satellite cell activity. Laminin α1 chain has also been reported to substitute for the missing endogenous laminin α2 chains and improve the overall health and longevity of mice with congenital muscular dystrophy (29, 30). We speculate that age-related collagen accumulation inhibits the physical interaction of laminin with the α7 integrin within the sarcolemma, as well as integrin expressed by satellite cells, and that exogenous LM-111 or LM-211/221 possess the potential to restore binding and overall muscle function in the context of aging.

CONCLUSION

Exercise training can potently stimulate stem cell activation and positively influence skeletal muscle ECM remodeling in a manner that suggests both factors are important and perhaps codependent in their ability to improve and/or maintain muscle structure and function following a physiological stimulus. Although most investigators acknowledge and appreciate the importance of the ECM in the maintenance of muscle health across the life span, detailed information regarding the impact of age on ECM composition, density, and posttranslational modifications and the subsequent influence on stem cell activation and muscle health is limited. Incorporation of large scale technologies in combination with in vivo and in vitro analyses are necessary to elucidate the complex interaction between the ECM and stem cells, as well as the interaction between the niche and myofiber. The work will undoubtedly be challenging, but will be necessary to develop new paradigms that can effectively treat age-related disabilities.

Present address of K. Garg: Saint Louis University, Department of Biomedical Engineering, 3507 Lindell Blvd., Suite 2011, St. Louis, MO 63103.

GRANTS

This work was partially supported by a grant from the National Institutes of Health (NIH R21 AR065578 to M.D.B.).

AUTHOR CONTRIBUTIONS

K.G. and M.D.B. prepared figures; K.G. and M.D.B. drafted manuscript; K.G. and M.D.B. edited and revised manuscript; K.G. and M.D.B. approved final version of manuscript.