A decline in muscle strength is a common consequence of both aging and prolonged disuse, and muscle atrophy is generally considered as the major cause of this phenomenon. With aging, muscle size declines by approximately 40% from 20 to 80 yr (16), and with disuse, a 30% fall in muscle cross-sectional area (CSA) occurs within 90 d of strict bed rest (Fig. 1) (21). For both conditions, however, the decrease in muscle CSA is smaller than the decline in force, indicating deterioration in force per unit CSA. Although reduced neural drive and a decrease in single fiber-specific tension are known to contribute to this phenomenon, recent evidence, obtained by magnetic resonance and ultrasound imaging, suggests that changes in muscle architecture and in tendon mechanical properties also play a role. In this article, we describe these alterations and the possible underlying mechanisms and discuss how their interaction theoretically affects the mechanical output of the muscle-tendon complex as a whole.
ALTERATIONS OF MUSCLE ARCHITECTURE WITH DISUSE
Since the pioneering studies of Goldspink (7) in the early 1980s, skeletal muscle has been known to display a remarkable plasticity in response to chronic loading or unloading. When mammalian skeletal muscle is immobilized in a stretched position, sarcomeres are added in series at the ends of fibers; when immobilized in a shortened position, sarcomeres in series are lost. Therefore, muscle fibers quickly adapt to the new functional length. In addition to inducing longitudinal growth by the addition of sarcomeres in series at the ends of the fiber, chronic stretch also stimulates an increase in the CSA of muscle fibers by the addition of sarcomeres in parallel. When rabbit tibialis anterior is subjected to stretch combined with overloading, the increase in muscle mass is quite phenomenal, more than 30% over a period of just 7 d (7).
Plasticity of skeletal muscle is also observed in humans under conditions of prolonged disuse. We had the opportunity to study these adaptations in one of the longest simulated microgravity studies organized by the European Space Agency (Long-Term Bed Rest Study 2001-2002) in which 10 healthy men underwent strict bed rest for a period of 90 d (21). The architecture of the gastrocnemius medialis (GM) muscle was investigated by ultrasound imaging, and calf muscle volume was assessed by magnetic resonance imaging (MRI) before and after bed rest. At the end of the bed rest period, calf muscle volume decreased by 29%. Atrophy of the calf muscles not only involved a reduction in muscle volume but also significant alterations in muscle architecture. After bed rest, GM resting fascicle length and pennation angle decreased by 10% and 13%, respectively (Fig. 2). Physiological CSA (PCSA = muscle volume/fascicle length) decreased by 22%. Assuming that the PCSA represents all sarcomeres in parallel and that fascicle length indicates all sarcomeres in series, it may be concluded that, in this chronic disuse paradigm, the loss of sarcomere in parallel is approximately twofold greater than the loss of sarcomeres in series. Hence, on purely theoretical grounds, one would then expect maximal isometric force (proportional to the total number of sarcomeres in parallel) to be decreased almost twice as much as maximal shortening velocity (proportional to the total number of sarcomeres in series).
ALTERATIONS OF MUSCLE ARCHITECTURE WITH AGING
Distinct morphological differences exist between disuse atrophy and senile sarcopenia, where the latter is defined as the loss of muscle mass associated with aging. Although disuse atrophy is characterized by a decrease in fiber CSA but not in fiber number, sarcopenia involves both a decrease in fiber size and number. However, the changes of muscle architecture observed in aging are similar to those found in disuse. We were the first to report age-associated alterations in fiber fascicle length and pennation angle measured in vivo (17) in the GM muscle (Fig. 2). Fascicle length and pennation angle of older men (age, 70+ yr) were found to be 10% and 13% smaller than those of young men (age, 20-30 yr), and PCSA was 15% smaller in the older men. Similar to disuse atrophy, the decrease in PCSA is greater than that of fascicle length. This suggests that more sarcomeres are lost in parallel than in series with muscle atrophy, which predicts a greater loss in force than in shortening velocity. Indeed, when comparing the muscle power characteristics (force × velocity) of young and older men, maximal isometric torque of the older men was lower by 40%, whereas maximal shortening velocity, estimated from Hill plots, was 10% slower (23).
In aging, as in disuse, the loss of muscle force and power is greater than that of muscle size and volume, even after accounting for differences in muscle architecture. In the plantarflexors (PF) of older men, for instance, we found specific force (force per unit of PCSA) to be reduced by 30% compared with that of younger men (15). Similarly, even after accounting for differences in muscle volume, PF peak power of the older men was approximately 45% lower than that of younger adults (23). Obviously, other factors besides changes in muscle architecture contribute to the loss of muscle force and power. Decreases in whole muscle specific force are influenced by single fiber-specific tension (because of reductions in cross-bridge number and E-C coupling) and by the effects of changes in muscle architecture and in tendon mechanical properties on the length-force relation. These aspects are discussed in the last section of this review. Decreases in muscle-specific power (power/muscle volume) observed in vivo are likely mediated by a decrease in specific force, a decrease in intrinsic speed of shortening of the myosin molecule (23), and by an increase in antagonist muscle coactivation known to occur in old age. Although cross-sectional studies comparing muscle fiber composition of young and older individuals reported a selective loss of Type II fibers in old age (see (1) for review), recent longitudinal data actually show similar fiber composition between young adults and the very old (1). Hence, selective loss of Type II muscle fibers does not seem a likely cause of the loss in specific muscle power in old age.
FACTORS REGULATING SARCOMERE REMODELING
The findings of changes in fascicle length and pennation with prolonged disuse and aging raise the question of what cellular and molecular factors regulate the removal or addition of sarcomeres in series and in parallel. Mechanotransduction - that is, the conversion of mechanical energy into biochemical reactions - involves the stimulation of mechanoreceptors located in the cell membrane and in structures within the extracellular matrix (ECM)-integrin and cytoskeleton complex (for review, see (8)). Essentially, two major pathways are recognized: 1) a mechanotransduction pathway characterized by the transmission of mechanical load to the ECM and cytoskeleton, leading directly, or via deformation of the nucleus, to expression of genes regulating protein synthesis and 2) a mechanochemical transduction pathway characterized by transmission of the mechanical signals from the ECM to the cytoskeleton via transsarcolemmal structures (integrins, dystroglycans, and stretch-activated calcium channels), followed by induction of biochemical signals regulating protein synthesis. Stimulation of both transduction routes leads, directly or indirectly, to activation of the major signaling pathway: the phosphatidyl-inositol-3 kinase (PI3K), protein kinase B (Akt), and the mammalian target of rapamycin (mTOR), p70 S6 kinase pathway, resulting in downstream activation of targets required for protein synthesis.
Main activators of the Akt/mTOR pathway are insulinlike growth factor (IGF) 1 and its splice variant mechanogrowth factor (MGF). The availability of these growth factors for receptor binding is regulated by the signal transduction pathways sensitive to stretch, overload, and muscle contraction. Insulinlike growth factor 1 and MGF, together with an up-regulation of myogenin and myogenic regulatory factor 4, stimulate satellite cells proliferation and differentiation, promoting muscle growth. With disuse, and with aging, satellite cell proliferation and differentiation is reduced by an increased expression of myostatin through up-regulation of cyclin-dependent kinase inhibitor p21 and by inhibition of the expression of myoblast determination factor. This inhibitory effect on satellite cell proliferation and differentiation by myostatin results in a marked reduction in the number of myonuclei and is thought to lead to muscle atrophy. Whereas the mechanisms regulating the expression of MGF in response to loading are largely unknown, the expression of IGF-1 increases with growth hormone and testosterone levels and decreases in response to glucocorticoids and to inflammatory cytokine levels (tumor necrosis factor (TNF) α and interleukin (IL) 1β). Several studies have demonstrated that TNF-α and IL-1β play a major role in the decline in muscle mass in aging; although a direct role of TNF-α in disuse atrophy has yet to be shown, activation of nuclear factor κB has been found to induce atrophy in disuse by increasing the expression of the muscle atrophy F-box (MAF-bx) gene (also known as atrogin) of the ubiquitin-proteasome pathway. Similarly, immobilization and denervation have been shown to up-regulate the expression of the muscle ring finger (MURF) 1 gene encoding for ubiquitin-ligase activity, therefore promoting muscle atrophy. The involvement of both MAF-bx and MURF1 in the proteolytic pathways clearly suggests that atrophy is not simply the reverse of hypertrophy, by which specific pathways are inhibited (e.g., the Akt/mTOR pathway), but that unique genes are up-regulated to trigger muscle atrophy (for review, see (6)).
Among the various structures sensitive to alterations in mechanical loading (stretch, overloading, and unloading) the transsarcolemmal proteins that seem to play a major role in the regulation of muscle mass are the integrins, the dystroglycans, and the stretch-sensitive calcium channels. Integrins and dystroglycans are the main transsarcolemmal proteins of adhesive complexes, termed costameres, that link the ECM to the Z disk of the sarcomere via the cytoskeleton (Fig. 3). Because of their link to the sarcomere via γ-actin filaments and other cytoskeletal proteins (talin, vinculin, and α-actin), costameres are thought to provide lateral transmission of the force generated by the myofibers to the ECM. In addition to this outward force transmission, recent evidence suggests that they may act also as lateral mechanical sensors. After activation of focal adhesion kinase (FAK) and also of ρ proteins belonging to the Ras superfamily of small GTPases (ρ-guanosine triphosphatase), they interact with various signaling pathways, such as the mitogen-activated protein kinases and the PI3k-mTOR pathway, regulating protein synthesis. Chronic loading of avian skeletal muscle has indeed shown that hypertrophy is accompanied by an up-regulation of FAK activity and an increase in FAK content (5). Several animal studies, performed in vivo (5) and in culture, suggest that costameric FAK meets the criteria of being an upstream regulator of sarcomere number (4) in response to changes in mechanical loading. It has been shown in cultured cardiac myocytes that mechanical overload by cyclic stretch results in activation of the FAK and the Z disk protein kinase Cε which mediate fast sarcomere remodeling to maintain optimal sarcomere length (14). Hence, the integrins and dystroglycan complexes act as mechanoreceptors transmitting information between the outside and inside of the muscle fiber and are also involved in the conversion of mechanical stimuli into biochemical signals, regulating protein synthesis. It is likely that these protein complexes play a key role also in disuse atrophy because their presence is greatly diminished in rat muscle exposed to tenotomy (3).
In addition to integrins and dystroglycans, the cytoskeletal protein desmin has also been suggested to play a role in the remodeling of sarcomeres in response to loading and unloading. This protein forms intermediate filaments concentrated at the Z disks of the sarcomeres. Desmin filaments connect sarcomeres in the lateral direction and form connections to the nucleus and to the costameres at the sarcolemma. Because of its structural location in the sarcomere, desmin may play a role in mediating mechanical signal transduction regulating muscle mass (22) as disruption of muscle architecture has been found in knockout mice lacking desmin. However, comparison of the effects of 28-d hind limb suspension on muscles of wild-type and desmin-null mice showed that, despite a significant loss of sarcomeres in series, no significant differences were found between the responses of the two types of mice (22). Thus, desmin does not seem to be essential for sarcomere remodeling in response to disuse, although it probably plays an important role in mechanical signal transduction.
ALTERATIONS OF TENDON MECHANICAL PROPERTIES WITH DISUSE AND AGING
Tensile tests in isolated material, mainly nonhuman, have established that disuse and aging affect both the collagenous structures and the mechanical characteristics of muscles and tendons (2). It was not until the late 1990s, however, that advances in the application of ultrasonography made it possible to characterize the mechanical properties of human tendons in vivo (10), thus enabling the study of tendon adaptability to conditions of altered mechanical loading in the intact human body.
Studies based on ultrasound scanning have shown that disuse can have a sizable negative effect on the mechanical properties of human tendons. In one study (19), mechanical unloading induced by 90 d of bed rest reduced the gastrocnemius tendon stiffness by 60%; stiffness was measured as the slope of a force-elongation graph constructed using ultrasound-based recordings of tendon elongation during isometric ankle plantar flexion of graded intensity. The tendon length and CSA remained unaltered after bed rest, so the Young modulus of the tendon (stiffness normalized to tendon dimensions), which reflects the material properties of the tendon, decreased also by approximately 60% as a result of bed rest (Fig. 4).
If the deterioration of tendon properties is proportional to the duration of disuse, as indicated by comparing the above two studies, then severe disuse - caused, for example, by several years of paralysis - might endanger the mechanical integrity of tendon. To address this important issue, we used ultrasonography and compared the mechanical properties of the patellar tendon in able-bodied (AB) men and in spinal cord-injured (SCI) men, with duration of lesion ranging from 1.5 to 24 yr (11). Contrary to the above hypothesis of a proportional change in properties with disuse, we found that the SCI tendons only had a reduced stiffness by 77% and a reduced Young modulus by 59% compared with the AB tendons (Fig. 5).
The difference between these values and those obtained in the 90-d bed rest study is relatively small. Furthermore, there was not any apparent relation between tendon properties and duration of lesion (Fig. 6). These findings may indicate that most of the deterioration of the tendon occurs rapidly within the first few months of immobilization. However, the potential effect of differences in joint positioning between the bed rest and SCI models should also be considered because joint positions corresponding to longer tendon lengths may attenuate some of the detrimental effects of disuse (13). The patellar tendon of the SCI patients who are chronically confined to a wheelchair with the knees flexed at approximately 90 degrees would be more stretched, and thus less vulnerable to deterioration, than the gastrocnemius tendon in the bed rest model because of the slackness of this tendon caused by the natural plantarflexion of the foot while bedridden.
In contrast to the bed rest study, the disused tendons in the SCI study had smaller CSAs (by 17%; (Fig. 7)), indicating that they may have undergone atrophy. The possibility that some of the differences in CSA between the SCI and AB groups relates to their anthropometric characteristics rather than atrophy cannot be excluded, although it should be noted that the length of the tendon was similar in the two groups. A longitudinal study with SCI patients tested at different time points after the lesion is required to confirm the degree of atrophy and shed more light on the dose-response relation of adaptations in tendon mechanical properties with disuse.
As is also the case for human muscle architecture, the changes of in vivo human tendon mechanical properties caused by aging seem to be in the same direction with those caused by disuse. Onambele et al. (18) examined the mechanical properties of the gastrocnemius tendon in younger (age, 24 ± 1 yr), middle-aged (age, 46 ± 1 yr), and older (age, 68 ± 1 yr) individuals. They found that both the stiffness and Young modulus of tendon decreased gradually with age, with the difference between the younger and the older groups reaching 36% and 48%, respectively. The CSA of the tendon decreased with age (by ~19%), but this result may indicate an anthropometric difference rather than tendon atrophy because a similar intergroup difference was observed for the tendon length (reduced by ~16%). In contrast to this finding, Magnusson et al. (12) reported that the CSA of the Achilles tendon in older women was larger (by 22%) than that in younger women. In combination with a reduced plantarflexion torque, it was proposed that the older women would be less likely than younger women to break their tendons in tension. Again, the possibility that these tendon CSA differences reflect anthropometric differences rather than an adaptation cannot be excluded.
The studies indicate that deterioration in the material of the tendon, as evidenced by the Young modulus reduction, is a common adaptive response of the tendon to both disuse and aging. But what exactly changes in the tendon? The collagen in the tendon becomes stiffer with disuse and aging because there is an increase in the intermolecular cross-linking through accumulation of advanced glycation end products (9). This indicates that there is some mechanism, activated similarly by disuse and aging, that prevails over the stiffening effect of increased collagen cross-linking. Potential mechanisms include a reduction in ground substances, a reduction in the number of longitudinally aligned collagen fibers, and a reduction in fibril diameter (9). One mechanism associated with the latter adaptation is the secretion of cytokines, such IL-1β and TNF-α, which increase the activity of matrix metalloproteinases (MMP), leading to collagen degradation (9). Interestingly, overexpression of cytokines with a positive effect on collagen degradation may occur simultaneously with overexpression of other cytokines, such as transforming growth factor α, which have the opposite effect of promoting collagen synthesis through activation of tissue inhibitors of matrix metalloproteinases (TIMPs) (9). It is interesting to examine whether the regulation of MMPs and TIMPs during disuse is a tendon length-dependent process that may explain the protective role of stretching against tendon deterioration during immobilization.
IMPLICATIONS OF MYOTENDINOUS ADAPTATIONS ON THE FORCE-LENGTH RELATIONSHIP
One limitation of the studies on elderly populations is that older individuals are generally more sedentary than younger individuals. Nonetheless, the present theoretical considerations regarding the force-length relation of muscle are based on the common phenomenological adaptations of muscle and tendon to disuse and old age as discussed in the previous sections, irrespective of the probable interaction between aging and sedentary lifestyles.
Two major parameters that affect the length range over which a muscle can exert its maximal force are the tendon stiffness and the number of serial sarcomeres in the fibers of the muscle. All things being the same, the stiffer the tendon, the longer the average sarcomere in the muscle during maximal isometric contraction at any given muscle-tendon length. The resultant right shift in the sarcomeric force-length curve means that muscles attached to more compliant tendons, as is the case in disuse and old age, would theoretically be able to produce less force than nondisused and younger muscles at lengths corresponding to the ascending limb of the force-length relation. For example, the vastus lateralis muscle operates in the ascending limb of the force-length relation at knee joint angles between 10 and 60 degrees (0 degree being full knee extension) as recently shown by Reeves et al. (20). Conversely, the force should be enhanced when the muscle operates on the descending limb of its force-length relation (e.g., the knee joint flexed more than 60 degrees for the vastus lateralis muscle) because of greater myofilament overlap caused by the increased tendon compliance (Fig. 8).
In contrast, increasing the number of serial sarcomeres in the fibers of the muscle reduces the average length of the sarcomere during a contraction. The resultant left shift in the sarcomeric force-length curve means that disused and older muscles, which have shorter fascicles and therefore presumably also fewer serial sarcomeres, would be able to produce more force than nondisused and younger muscles operating on the ascending limb and less force on the descending limb of the force-length curve (Fig. 8).
Because the reductions in fascicle length and tendon stiffness shift the force-length curve in opposite directions, these adaptations may counteract each other to a degree that the relative contractile force at a given length is maintained in inactive and older muscles. Support for the notion of preservation of contractile force-length characteristics with altered mechanical environment has recently been obtained by Reeves et al. (20). These authors showed that the measured fascicle force-length and estimated sarcomere force-length characteristics of the human vastus lateralis muscle remained largely unaffected after 14 wk of knee extensor muscle strength training (Fig. 9), which increased the muscle fascicle length and stiffened the patellar tendon. Clearly, the above study used a change in mechanical loading (increased use model) in the opposite direction to that caused by disuse and aging; hence, actual experimental studies on the effects of disuse and aging on human contractile force-length characteristics are required to substantiate our theoretical considerations.
Muscles and tendons show considerable plasticity in response to chronic unloading and aging. Both conditions are characterized by a withdrawal of anabolic stimuli and an increase in catabolic processes. In old age, these myotendinous changes are further amplified by the effects of disuse. Recent evidence obtained from ultrasound imaging shows marked changes in muscle architecture, pointing to significant sarcomere remodeling in both unloading and aging. The mechanisms regulating these alterations are complex, but a key role seems played by costameres that are responsible for the structural and functional adhesion of muscle fibers to the surrounding connective tissue and adjacent muscle fibers. Both in vivo and in vitro studies have shown that mechanical stimulation of costameric integrins results in activation of the major signaling pathway leading to an increase in protein synthesis, whereas the removal of mechanical loading and aging have the opposite effect. The alterations in tendon mechanical properties associated with aging and in medium-term disuse (months) seem to be mainly because of changes in tendon material properties; whereas in long-term disuse (years), changes in tendon dimensions also seem to play a role. Increasing evidence supports the contention that the functional repercussions of the above changes are the result of the close interaction between the muscular and tendinous alterations associated with aging and disuse.
The authors thank Professors David Jones and Martin Flück for advice on some of the issues discussed in this review.
1. Andersen, J.L. Muscle fibre type adaptation in the elderly human muscle. Scand. J. Med. Sci. Sports
2. Butler, D.L., E.S. Grood, F.R. Noyes, and R.F. Zernicke. Biomechanics of ligaments and tendons. Exerc. Sport Sci. Rev.
3. Chockalingam, P.S., R. Cholera, S.A. Oak, Y. Zheng, H.W. Jarrett, and D.B. Thomason. Dystrophin-glycoprotein complex and Ras and Rho GTPase signaling are altered in muscle atrophy. Am. J. Physiol. Cell Physiol.
4. Fluck, M., A. Ziemiecki, R. Billeter, and M. Muntener. Fibre-type specific concentration of focal adhesion kinase at the sarcolemma: influence of fibre innervation and regeneration. J. Exp. Biol.
5. Fluck, M., J.A. Carson, S.E. Gordon, A. Ziemiecki, and F.W. Booth. Focal adhesion proteins FAK and paxillin increase in hypertrophied skeletal muscle. Am. J. Physiol.
6. Glass, D.J. Signalling pathways that mediate skeletal muscle hypertrophy and atrophy. Nat. Cell Biol.
7. Goldspink, G. Malleability of the motor system: a comparative approach. J. Exp. Biol.
8. Huijing, P.A., and R.T. Jaspers. Adaptation of muscle size and myofascial force transmission: a review and some new experimental results. Scand. J. Med. Sci. Sports
9. Kjaer, M. Role of extracellular matrix in adaptation of tendon and skeletal muscle to mechanical loading. Physiol. Rev.
10. Maganaris, C.N., and J.P. Paul. In vivo
human tendon mechanical properties. J. Physiol.
521(Pt. 1):307-313, 1999.
11. Maganaris, C.N., N.D. Reeves, J. Rittweger, A.J. Sargeant, D.A. Jones, K. Gerrits, and A. De Haan. Adaptive response of human tendon to paralysis. Muscle Nerve
12. Magnusson, S.P., N. Beyer, H. Abrahamsen, P. Aargaard, K. Neergaard, and M. Kjaer. Increased cross-sectional area and reduced tensile stress of the Achilles tendon in elderly compared with young women. J. Gerontol. A Biol. Sci. Med. Sci.
13. Majima, T., K. Yasuda, T. Fujii, N. Yamamoto, K. Hayashi, and K. Kaneda. Biomechanical effects of stress shielding of the rabbit patellar tendon depend on the degree of stress reduction. J. Orthop. Res.
14. Mansour, H., P.P. de Tombe, A.M. Samarel, and B. Russell. Restoration of resting sarcomere length after uniaxial static strain is regulated by protein kinase Cepsilon and focal adhesion kinase. Circ. Res.
15. Morse, C.I., J.M. Thom, N.D. Reeves, K.M. Birch, and M.V. Narici. In vivo
physiological cross-sectional area and specific force are reduced in the gastrocnemius of elderly men. J. Appl. Physiol.
16. Narici, M.V., and C.N. Maganaris. Adaptability of elderly human muscles and tendons to increased loading. J. Anat.
17. Narici, M.V., C.N. Maganaris, N.D. Reeves, and P. Capodaglio. Effect of aging on human muscle architecture. J. Appl. Physiol.
18. Onambele, G.L., M.V. Narici, and C.N. Maganaris. Calf muscle-tendon properties and postural balance in old age. J. Appl. Physiol.
19. Reeves, N.D., C.N. Maganaris, G. Ferretti, and M.V. Narici. Influence of 90-day simulated microgravity on human tendon mechanical properties and the effect of resistive countermeasures. J. Appl. Physiol.
20. Reeves, N.D., M.V. Narici, and C.N. Maganaris. In vivo
human muscle structure and function: adaptations to resistance training in old age. Exp. Physiol.
21. Reeves, N.J., C.N. Maganaris, G. Ferretti, and M.V. Narici. Influence of simulated microgravity on human skeletal muscle architecture and function. J. Gravit. Physiol.
22. Shah, S.B., D. Peters, K.A. Jordan, D.J. Milner, J. Fridan, Y. Capetanaki, and R.L. Lieber. Sarcomere number regulation maintained after immobilization in desmin-null mouse skeletal muscle. J. Exp. Biol.
23. Thom, J.M., C.I. Morse, K.M. Birch, and M.V. Narici. Triceps surae muscle power, volume, and quality in older versus younger healthy men. J. Gerontol. A Biol. Sci. Med. Sci.