Role of exercise in maintaining the integrity of the neuromuscular junction

Abstract

Physical activity plays an important role in preventing chronic disease in adults and the elderly. Exercise has beneficial effects on the nervous system, including at the neuromuscular junction (NMJ). Exercise causes hypertrophy of NMJs and improves recovery from peripheral nerve injuries, whereas decreased physical activity causes degenerative changes in NMJs. Recent studies have begun to elucidate molecular mechanisms underlying the beneficial effects of exercise. These mechanisms involve Bassoon, neuregulin-1, peroxisome proliferator-activated receptor gamma coactivator 1α, Insulin-like growth factor-1, glial cell line-derived neurotrophic factor, neurotrophin 4, Homer, and nuclear factor of activated T cells c1. For example, NMJ denervation and active zone decreases have been observed in aged NMJs, but these age-dependent degenerative changes can be ameliorated by exercise. This review will discuss the effects of exercise on the maintenance and regeneration of NMJs and will highlight recent insights into the molecular mechanisms underlying these exercise effects.

Introduction

In recent years, the importance of physical activity in preventing the development of chronic diseases, such as cardiovascular, metabolic, musculoskeletal, and neurological disorders, has gained increasing recognition¹. The forms of exercise training that increase physical activity can be divided into endurance and resistance. The effects of these exercise types on the nervous system have been studied in humans and rodents. Endurance training is any exercise that increases the functional capacity of the aerobic system. The effects of endurance training on laboratory animals have been studied using a treadmill, running wheel, and swimming ^2–5. Resistance training is any exercise that increases muscle contraction strength and anaerobic endurance. The effects of resistance training on laboratory animals have been studied using ladder climbing ⁶ and isometric resistance training ^7–9.

Exercise is known to have beneficial effects on the nervous system, including the peripheral nervous system and NMJs ^1,10. Training improves the recovery from peripheral nerve injury or degenerative changes ^1,10–14. For example, increased physical activity is beneficial in recovery from the disruptive structural changes that occur in NMJs due to exposure to zero gravity during space flight ^12,15. It is important to keep in mind that adult mammalian NMJs are not rigid structures. They show some degree of remodeling, including addition or reduction of presynaptic branches and postsynaptic receptors within one NMJ, and the rate of remodeling differs depending on the type of muscle ^16–18. NMJs in soleus and pectineus muscles show small additions and deletions to parts of the NMJs ^16,18, but sternocleidomastoid muscle shows only synapse size growth that appears to be related directly to increased muscle fiber diameter ¹⁷. Another example of the plastic nature of NMJs is seen at the synaptic vesicle release sites known as active zones. NMJ active zones are not stable structures. They are quickly altered by extracellular stimuli ^19,20 and degenerate during aging ^9,21. Interestingly, exercise can ameliorate degenerative changes in the presynaptic active zones of NMJs in aged rats,⁹ as will be described later in this review. The beneficial effects of exercise on maintenance and recovery of NMJs have been summarized previously ^1,10–13,22. However, important findings elucidating the cellular and molecular mechanisms have been reported recently. The goal of this review is to highlight recent insights into the role of exercise in maintaining the integrity of NMJs.

Adult NMJs and exercise

Endurance training increases the synapse size of NMJs in adult mice ² and rats ^3,4 (Table 1). Exercise causes hypertrophy of the NMJs of extensor digitorum longus and gluteus maximus muscles in adult mice and rats ^2,3,23. Soleus muscle, however, has been used more widely as a model to study NMJ adaptations to exercise, partly because of its homogeneous muscle fiber type composition and its antigravitational function ¹². Exercise generally has been reported to cause hypertrophy of the NMJs of soleus muscles in adult mice and rats ^2,4,24, and to increase the branching and complexity of presynaptic nerve terminals, ⁴ although 1 study reported that there is no effect in adult rats ³. Furthermore, different types of exercise, endurance versus resistance, produced slightly different results in soleus muscle in studies conducted by the same research group ^4,6. Endurance training induced significant hypertrophy of the NMJs of soleus muscles, but resistance exercise induced only a trend of increase (not significant) of AChR cluster area. The result of resistance training in the soleus muscle is similar to the results for resistance training of genioglossus muscle in rats ⁹ and to the results of rat hypoglossal nerve stimulation to mimic resistance training of the tongue ²⁵, neither of which causes an increase in the AChR cluster area. These results suggest that the effects of exercise on NMJs depend on the type of exercise performed.

Table 1.

Effects of exercise type and aging on neuromuscular junction size

Exercise type	Muscle	Adult NMJ	Aged NMJ
Endurance	Extensor digitorum longus	Increase (mouse [2], rat [3])	Decrease (mouse [2])
	Gluteus maximus	Increase (mouse [23])	Decrease (mouse [23])
	Soleus	Increase NMJ size (mouse [2], rat [4, 24]), No change (rat [3])	No change (mouse [2], rat [24])
Resistance	Soleus	No change (rat [6])	Unknown
	Genioglossus	* No change (rat [25])	No change (rat [9]), * decrease (rat [25])

NMJ = neuromuscular junction.

^*In reference 25, chronic electrical stimulation was applied to the nerve to mimic a clinical exercise paradigm ¹⁴⁴.

In contrast to increased physical activity causing hypertrophy of adult NMJs, decreased physical activity results in degenerative changes and nerve terminal sprouting in adult NMJs (reviewed in ¹²). For example, degenerative changes in NMJs occur upon exposure to zero gravity during space flight, during extended bed rest, or when the limbs of laboratory animals are fixed artificially to prevent use ¹². These observations suggest that daily physical activity is required for maintenance of adult mammalian NMJs and prevention of degeneration.

Aged NMJs and exercise

Exercise has beneficial effects also on aged NMJs, even though the effects are different from those observed in younger adults (Table 1). Endurance training decreases AChR cluster size in the extensor digitorum longus muscle of aged mice and the gluteus maximus muscle of aged rat, which is the opposite effect of endurance training in younger adult animals ^2,23. Furthermore, endurance training does not alter synapse size in the soleus muscle of aged mice ² and rats ²⁴. Similarly, isometric force training of the genioglossus muscle in aged rats does not alter synapse size, although it appears to have a beneficial effect on presynaptic active zones ⁹. Furthermore, endurance training in aged mice reduces age-related morphological alterations and denervation of NMJs in other hind limb muscles: tibialis anterior, gracilis, and gastrocnemius muscles ²⁶. Importantly, time-lapse analysis in vivo reveals that endurance training of aged mice partially reverses the age-related alteration in NMJs. ²⁶

These effects of exercise observed in aged animals should be interpreted with caution, because aged NMJs exhibit various degrees of denervation in different muscles ²⁷. In young adult muscles, all NMJs are innervated, so analyses are limited to the adaptive changes in existing NMJs in response to increased activity. However, aged animals exhibit denervation in subpopulations of NMJs. Therefore, analyses of NMJs include the adaptive changes that occur in the existing NMJs and the reinnervation of the NMJs that were denervated prior to exercise. If analyses use aged muscles without denervation, for example the genioglossus muscle ⁹ or extraocular muscles ²⁷, then the adaptive changes in aged NMJs would be the focus of analysis ⁹. These 2 approaches allow investigation of the potential therapeutic effects of exercise in maintaining and/or regenerating aged NMJs depending on the type of aged muscle.

In a recent study using aged genioglossus muscles, resistance training showed a beneficial effect on the presynaptic active zones in aged NMJs ⁹. Active zones are cytosolic structures needed for synaptic vesicle release ^28–31. The active zone-specific protein Bassoon is absent in many NMJs of aged mice and rats, while nerve terminals still fully innervate the endplates ^9,21. A loss of Bassoon can be seen prior to denervation of aged NMJs, which suggests that a loss of active zones may play a role in age-dependent degenerative changes in NMJs, including denervation. This is because NMJ denervation has been observed in young humans and mice that exhibit a loss of active zones resulting from gene mutations ^31–33. This proposal is further supported by the finding that less active motor nerve terminals withdraw from NMJs when NMJ synaptic transmissions are attenuated under experimental conditions ^34,35. Thus, these findings also suggest that NMJ synaptic activity and innervation maintenance are closely related. Importantly, 2 months of isometric force training reduces the active zone protein loss in the genioglossus muscle of aged rats ⁹. In trained-aged rats, the average level of Bassoon protein at the nerve terminals increases to the level observed in young adult NMJs, and the number of NMJs that do not contain a detectable level of Bassoon signal decreases significantly. This additional Bassoon protein in the trained-aged NMJs distributes to active zones, which may aid in functional recovery of NMJs.

Muscle fibers and exercise

Are these adaptive changes of NMJs a secondary effect of muscle fiber hypertrophy induced by exercise? Exercise causes muscle hypertrophy, and exercise-induced genes, such as peroxisome proliferator-activated receptor gamma, coactivator 1α (PGC-1α), have been shown to increase muscle fiber diameter ³⁶. The role that these exercise-induced genes play in muscle hypertrophy has been reviewed in detail^10,37. The NMJ size is coupled, to a certain degree, to muscle fiber size ^17,38. However, aged NMJs respond differently than muscle fibers to exercise. Exercise increases muscle fiber size in the soleus muscle of aged rats ^24,39, but the NMJ size does not increase ^2,24. Furthermore, some of the exercise-based changes of NMJs cannot be explained merely by changes in muscle fiber diameter. Examples include increased active zone protein level ^9,40, reduced denervation rates, reduced age-related morphological alterations in aged NMJs ²⁶, and a reduction of AChR cluster size in extensor digitorum longus and gluteus maximus muscles of aged mice ^2,23. Therefore, exercise induces NMJ hypertrophy partly by increasing muscle fiber diameter, but also by direct modification of NMJs.

Molecular mechanism underlying the exercise effect on NMJs

The beneficial effects of exercise are observed predominantly in trained muscles, and a systemic effect in non-trained muscles has not been observed ^26,41 (also see ⁴²). These results suggest that NMJ hypertrophy in exercised young adults and the increase in active zone protein levels in exercised aged NMJs are localized effects within the exercised muscles and motor neurons innervating exercised muscles. The molecular mechanisms through which exercise produces beneficial effects on NMJs have not been elucidated fully.

In human muscles, resistance training upregulates mRNA and protein expression levels of extracellular matrix molecules ^41,43, including laminin β2 ⁴¹ (Figure 1, Table 2). The exercise induced upregulation of laminin β2 expression may play a role in the effect of exercise that increases the level of active zone protein Bassoon in aged NMJs ⁹. The link between these 2 studies will be discussed below.

Figure 1.

A schematic diagram showing genes and proteins controlled by exercise at the vertebrate neuromuscular junction. Solid arrows represent an upregulation or enhancement of RNAs or proteins by exercise. T-shaped arrows represent a suppression of RNAs or proteins by exercise. See text and tables for the modification mechanisms and functions of these RNAs and proteins by exercise. A motor neuron and its presynaptic terminal are indicated in green. The green arrow indicates hypertrophy or sprouting of the motor nerve terminal induced by exercise. Synaptic vesicles and voltage-dependent calcium channels accumulate near the active zone indicated using a black triangle, which depicts the electron dense material of the active zones detected by electron microscopy. A muscle fiber is indicated in pink with acetylcholine receptors indicated in red and synaptic extracellular matrix indicated in orange at the synaptic cleft. A junctional fold is indicated as a trough on the postsynaptic membrane. The relative size of the structures in this diagram is not in scale.

Table 2.

Effects of exercise on synaptic genes, NMJ morphology, and NMJ function.

Signaling pathways, genes, and proteins	Effects of exercise	Effects on NMJ morphology and function
[Presynaptic active zone organizer]
Laminin β2	Increases the level of extracellular matrix molecules, including laminin β2 (mRNA) in humans [41].	Organizes presynaptic active zones [19]. Lack of laminin β2 decreases synaptic transmission efficiency [31, 74]
Bassoon	Reverts the Bassoon protein level at the presynaptic terminal of aged NMJs to young adult level [9].	Organizes presynaptic active zones [9, 48]. Lack of Bassoon decreases the function of presynaptic calcium channel [9].
[Neuregulin/PGC-1α]
Neuregulin-1	Increases the phosphorylation and proteolytic processing of neuregulin-1 in skeletal muscle [78].	Induces the expression of synaptic genes [80].
PGC-1α	Increases the expression level of PGC-1α in humans [83] and rats [82].	Increases the transcription of synaptic genes in muscles [81]. Neuregulin-1 mediated upregulation of synaptic gene expression in muscles requires PGC-1α, phosphorylation of PGC-1α, and interaction of PGC- 1α and GA-binding proteins [81].
AMPK	Increases AMPK activation in human [86,87].	AMPK activates PGC-1α by directly phosphorylating it [88].
[Neurotrophic factors]
IGF-1	Increases the expression level of IGF-1 in human [90, 91].	Preserves NMJs from degeneration in motor neuron disease model mice [92]. Inhibits the expression of microRNA (miR)-378, which negatively regulates the expression level of IGF-1 receptor [99].
miR-206	Decreases the expression level of miR-206 in human [98].	Increases the expression level of IGF-1, which is negatively regulated by miR-206 in the skeletal muscles of fish [93]. miR-206 promotes the regeneration of NMJs in motor neuron disease model mice [100].
NT-4	Increases NT-4 mRNA level by electrical stimulation of sciatic nerves [105].	Induces sprouting of NMJs [105]. Enhances neuromuscular transmission [120].
BDNF	Increases the expression level of BDNF [102].	Increases EPP amplitude and enhances neuromuscular transmission [120,121]. Increases the survival of motor neurons [106–116].
GDNF	Increases the expression level of GDNF [102]. Increases the protein level of GDNF in the rat soleus muscle, but decreases in the extensor digitorum longus muscle [103].	Increases the number of motor units and induces the continuous synaptic remodeling of adult NMJs [122] Increase the survival of motor neurons [106–116].
[Homer-NFATc1]
Homer	Increases protein level of Homer at postsynaptic side of NMJs [125].	Homer2 binds directly to the transcription factor nuclear factor of activated T cells (NFATc1) and activated NFATc1 moves from the cytoplasm to the nucleus in exercised muscle [125]. Activated NFATc1 upregulates the expression levels of synaptic genes, acetylcholinesterase and utrophin [126, 127].

Laminin β2 is an extracellular matrix protein that is secreted by muscles and is concentrated specifically in the synaptic cleft of NMJs ^44,45. Laminin β2 binds directly and specifically to P/Q- and N-type voltage-dependent calcium channels (VDCCs) ¹⁹. These VDCC pore-forming subunits bind to synaptic laminins that contain laminin β2 and do not bind to non-synaptic laminins, which contain laminin β1 ¹⁹. Furthermore, synaptic laminins will bind to VDCCs that are highly concentrated at presynaptic terminals in NMJs (e.g., P/Q- and N-types) and not to other VDCCs [e.g., R- and L-type VDCCs (Cav1.2)] ^19,46,47, which suggests that laminin β2 is an extracellular ligand of synaptic VDCCs. Interactions between laminin β2 and VDCCs lead to the clustering of VDCCs and presynaptic components in cultured motor neurons ¹⁹. In vivo studies provide compelling evidence that this extracellular interaction between laminin and VDCCs organizes the NMJ active zones. The number of active zones decreased when the interaction between the VDCCs and laminin β2 was perturbed in wild-type mice by infusing an inhibitor of this interaction ¹⁹. Moreover, P/Q- and N-type VDCCs double knockout mice exhibit specific defects in the number of active zones and docked synaptic vesicles, which were twice as severe as the defects observed in the P/Q- and N-type VDCC single knockout mice ^19,48. Humans who carry laminin β2 mutations that result in active zone loss and denervation develop Pierson syndrome, an autosomal recessive movement disorder associated with microcoria, and nephrotic syndrome^32,49. These data suggest that laminin β2 binds to synaptic VDCCs to organize the active zones.

P/Q-type VDCCs are distributed in a discrete punctate pattern within NMJs and preferentially co-localize with Bassoon⁹. It has been predicted that the VDCCs that trigger synaptic vesicle release are located at or in close proximity to active zones ^50–55. The three-dimensional alignment of the P/Q-type VDCCs and Bassoon immunohistochemistry signals suggests that these co-localization spots are discrete active zones within NMJs ⁹. This proposal is also supported by rodent NMJ active zone studies that have used freeze fracture electron microscopy ⁵⁶, electron microscope tomography ^57,58, and electrophysiology ⁵⁹. Importantly, this co-localization pattern of P/Q-type VDCCs and Bassoon in the NMJs is consistent with a report that identified direct binding between VDCCs and Bassoon ⁴⁸. In addition, the active zone specific proteins Bassoon, CAST/Erc2, ELKS, and RIMs interact with the VDCC β subunits ^48,60–62, which forms a tight complex with the pore-forming α subunits of the P/Q- and N-type VDCCs ⁶³. VDCC α subunits also interact with the active zone proteins RIMs and Piccolo ^64–66. These active zone proteins most likely form a large protein complex ^67–70. Therefore, presynaptic VDCCs tether active zone proteins to the presynaptic membrane and form electron-dense material in the NMJ active zones. Taken together with the previous paragraph, these findings show that the muscle-derived laminin β2 organizes NMJ active zones from the extracellular side by anchoring the VDCC subunits and active zone protein complex.

NMJ active zones are maintained at a constant density as the NMJ matures but are degraded in aged animals ²¹. The level of Bassoon decreases in the NMJs of aged mice and rats ^9,21. A lack of Bassoon is known to impair synaptic vesicle trafficking to presynaptic membranes in the central nervous system and sensory neurons ^71–73. Furthermore, a lack of Bassoon decreases VDCC Ca²⁺ influx and weakens synaptic transmission, because the direct interaction between VDCCs and Bassoon enhances the P/Q-type VDCC function ⁹. This modification to VDCCs by Bassoon is similar to the effect of another active zone protein, RIM1, on VDCCs ^60,61. These findings are consistent with the Bassoon-dependent increase in Ca²⁺ influx through L-type VDCCs in the inner hair cells of the auditory system ⁷³. Furthermore, impaired synaptic transmission at NMJs is a known characteristic in human diseases and knockout mice associated with a decreased number of active zones ^31,55,74. Therefore, the reduced Bassoon protein level in aged NMJs most likely weakens synaptic transmission. This hypothesis is supported by the observation that synaptic function is attenuated in aged NMJs compared with young adult NMJs, including stronger synaptic depression during repeated stimulation ⁷⁵, reduction in the end-plate potential amplitude at the plateau level after repetitive stimulation ⁷⁶, and a reduction in the frequency of miniature end-plate potentials ^76,77. Taken together, these findings suggest that active zone protein loss may be a part of the molecular mechanism that causes the deterioration observed in aged NMJs.

Active zone deterioration in aged NMJs is ameliorated by muscle exercise. Two months of isometric force training rescued the loss of Bassoon in aged NMJs in the genioglossus muscle of two-year-old rats ⁹. Exercise training did not alter NMJ size, which suggests that the increase in the Bassoon immunohistochemistry signal in aged NMJs reflects an increase in the protein quantity in each nerve terminal ⁹. The mean intensity of the Bassoon immunohistochemistry signal in the NMJs of exercised aged rats is similar to the mean intensity observed in the young adult rats. Importantly, the improvement in Bassoon protein level observed in exercised aged NMJs is consistent with improvements observed using electrophysiology in NMJ function after endurance training in aged mice ²³. In summary, exercise-induced upregulation of laminin β2 may play a role in preservation of active zones in aged NMJs, which most likely exerts a positive effect on NMJ synaptic transmission.

Other molecular mechanisms involved in the exercise effect on NMJs

Signaling cascades involving neuregulins and PGC-1α play a role in the beneficial effects of exercise (Figure 1, Table 2). Exercise increases the phosphorylation and proteolytic processing of a transmembrane isoform of the signaling molecule neuregulin-1 in skeletal muscle ⁷⁸. Proteolytic cleavage of neuregulin-1 can also be induced by increasing the neuronal activity in cultured neurons ⁷⁹. Neuregulin induces expression of synaptic genes in muscles, and this finding was confirmed in vivo using neuregulin1 heterozygote mice ⁸⁰. Neuregulin-1 mediated upregulation of synaptic gene expression in muscles requires the transcription coactivator PGC-1α, phosphorylation of PGC-1α, and interaction of PGC-1α and GA-binding proteins ⁸¹. Furthermore, exercise increases the expression level of PGC-17α in rodents and humans ^82,83, and elevated levels of PGC-1α increase the transcription of synaptic genes in cultured primary muscle cells ⁸¹. Additionally, AMP-activated protein kinase (AMPK) may play a role in activating PGC-1α in response to exercise ⁸⁴. AMPK is an energy-sensing enzyme ⁸⁵ that is activated in skeletal muscles during exercise ^86,87. AMPK phosphorylates PGC-1α directly, which is required for the PGC-1α-dependent induction of the PGC-1α promoter ⁸⁸. Interestingly, knockout mice for the neuregulin1 isoform highly expressed in motor neurons (CRD-NRG-1) exhibit presynaptic defects in developing NMJs ⁸⁹, which suggests that neuregulin signaling also plays a role on the presynaptic side of NMJs. In summary, these signaling cascades most likely play a role in the exercise-induced modification of NMJs.

Exercise induces expression of insulin-like growth factor-1 splice variants (a.k.a. mechano growth factor) in skeletal muscles ^90,91 (Table 2). Insulin-like growth factor-1 (IGF-1) preserves NMJs in a motor neuron disease mouse model ⁹², which suggests that IGF-1 also plays a role in the beneficial effects of exercise on NMJs. Interestingly, the expression level of IGF-1 is regulated negatively by microRNA (miR)-206 in the skeletal muscles of fish ⁹³. Several miRs are highly expressed in skeletal muscles and are considered to be muscle specific, including miR-206 ⁹⁴. In human skeletal muscles, many miRs are regulated by exercise ^95–97. For example, endurance training for 12 weeks significantly decreases the expression level of miR-206 ⁹⁸, which may increase the expression level of IGF-1. Additionally, IGF-1 inhibits expression of miR-378, and miR-378 negatively regulates the expression level of insulin-like growth factor 1 receptor (IGF1R) in mouse cardiomyocytes ⁹⁹. Together, IGF-1 and muscle-specific miRs may play a role in the beneficial effects of exercise by enhancing the IGF-1 signaling pathway. This interesting potential role for the IGF-1 signaling cascade awaits further study in human skeletal muscles. It has also been reported that miR-206 promotes regeneration of NMJs in a motor neuron disease mouse model ¹⁰⁰. The discrepancy between this positive role of miR-206 and the finding that exercise downregulates miR-206 is due potentially to the difference between motor neuron disease animals and healthy animals. Further investigation is needed to elucidate the role of miRs in exercise.

Exercise also upregulates transcription of several other neurotrophic factors ^101–104 which have beneficial effects on NMJs (Table 2). The expression level of neurotrophin 4 (NT-4) is activity-dependent, and electrical stimulation of the sciatic nerve increases NT-4 expression in mouse skeletal muscles ¹⁰⁵. Importantly, NT-4 receptor (TrkB) is expressed in motor neurons, and NT-4 induces the sprouting of motor nerve terminals in adult rats in vivo ¹⁰⁵. These results demonstrate that increased NT-4 can explain, at least in part, the beneficial effects of exercise on NMJs. However, the role NT-4 plays in exercise-induced hypertrophy of the NMJs is unknown, because NMJ size was not measured in this study. Similarly, brain-derived neurotrophic factor (BDNF) and glial cell line-derived neurotrophic factor (GDNF) are upregulated in the rat soleus muscle by exercise ^102,103. Therefore, the expression level of these factors in skeletal muscles is activity-dependent. These neurotrophic factors increase the survival of motor neurons ^106–116. Furthermore, neurotrophins modulate the synaptic transmission efficiency of the NMJs in embryos and adults ^117–121. In addition, GDNF increases the number of motor units and induces continuous synaptic remodeling of adult NMJs ¹²². These findings suggest that exercise induces expression of GDNF and BDNF, which increase the survival of the innervating motor neurons and have beneficial effects on NMJs in exercised muscles. Interestingly, the GDNF protein level is not upregulated uniformly by exercise and seems to be muscle specific, because it has been shown that GDNF is downregulated in the extensor digitorum longus muscle after exercise ¹⁰³.

Other signaling proteins are also regulated by exercise. Homer is an adaptor protein that has a role in controlling the Transient Receptor Potential (TRP) Channel activity in skeletal muscles ^123,124 and accumulates at postsynaptic sites in skeletal muscles ¹²⁵. In human skeletal muscles, postsynaptic protein levels of Homer increase with exercise and decrease with bed rest,¹²⁵ (Table 2). Homer2 binds directly to the transcription factor nuclear factor of activated T cells (NFATc1), and activated NFATc1 moves from the cytoplasm to the nucleus in exercised muscle ¹²⁵. NFAT1c1 upregulates expression levels of synaptic genes, acetylcholinesterase, and utrophin, in skeletal muscles ^126,127. Utrophin is a large cytoskeletal protein that accumulates preferentially at NMJs and participates in the maturation of the postsynaptic site ^128–131. These findings suggest that Homer plays a role in the NFATc1-dependent signaling pathway to increase transcription of synaptic gene in exercised muscle.

Implications for treatment

The molecular mechanisms underlying the beneficial effects of exercise on NMJs are being elucidated, but knowledge gaps currently prevent complete reproduction of these benefits by pharmacological treatments. This review covers a subset of the genes, proteins, and signaling pathways that are modified by exercise, and transcriptome analyses have recently revealed a large number of genes that are controlled in response to increased physical activity in the spinal cord ^132–134 and skeletal muscles ^{41–43,133,135–140}. Tissue-specific and age-dependent differences in these responses are being revealed within skeletal muscle tissues, and thus, further mechanistic investigations are needed. As summarized in this review, exercise clearly has beneficial effects on the maintenance and regeneration of NMJs. The molecular mechanisms underlying these beneficial effects provide potential new therapeutic targets for motor neuron diseases, neuromuscular junction diseases, musculoskeletal diseases, and age-dependent degeneration of NMJs. Currently, many studies that manipulate signaling pathways related to exercise have focused their analyses mostly on muscles but not NMJs ^84,141–143. The effects of exercise mimetics as pharmacological treatments of neuromuscular diseases have not yet been determined successfully. Therefore, it is anticipated that further research will yield methods to pharmacologically mimic, enhance, or modify these cellular and molecular mechanisms of exercise in order to enhance exercise-based interventions or replace them in individuals whose disabilities preclude exercise intervention.

Abbreviations

AChR

acetylcholine receptor

AMPK

AMP-activated protein kinase

BDNF

brain-derived neurotrophic factor

GDNF

glial cell line-derived neurotrophic factor

IGF-1

insulin-like growth factor-1

IGF1R

insulin-like growth factor 1 receptor

miR

microRNA

NFATc1

nuclear factor of activated T-cells

NMJs

neuromuscular junctions

NT-4

neurotrophin 4

PGC-1α

peroxisome proliferator-activated receptor gamma coactivator 1α

VDCCs

voltage dependent calcium channels