37 G.S. Lynch (ed.), Sarcopenia – Age-Related Muscle Wasting and Weakness, DOI 10.1007/978-90-481-9713-2_3, © Springer Science+Business Media B.V. 2011 Abstract Remodeling of neuromuscular junctions (NMJs) and ensuing structural and functional plasticity occurs with aging. Age-related changes result from reductions in physical activity, loss of motor neurons, and decreased muscle fiber size (sarcopenia). The properties of motor neurons and muscle fibers are precisely matched. In addition, motor unit recruitment in a selective manner is a primary mechanism by which the nervous system controls muscle contraction. Thus, it is essential to consider motor unit (and muscle fiber) type in any age-related plasticity. The following chapter examines changes in motor unit properties associated with aging and how these affect structural and functional remodeling at NMJs. Keywords Aging • Morphological adaptations • Motor units • Muscle fiber type • Plasticity • Recruitment • Skeletal muscle 1 Introduction The neuromuscular junction provides the sole link between a motor neuron and muscle fibers. Within a motor unit (Fig. 1), the mechanical and biochemical proper- ties of muscle fibers are relatively uniform, and it is clear that the motor neuron plays an important role in influencing these properties through the neuromuscular junction. This influence is imparted either through activity levels or nerve-derived trophic factors (Mantilla and Sieck 2008; Delbono 2003). As a result, the mechanical and metabolic properties of muscle fibers and motor neurons are precisely matched (Burke et al. 1971; Sieck et al. 1989) – an essential feature of neuromotor control and functional performance of a skeletal muscle across a range of physiological behaviors. C.B. Mantilla and G.C. Sieck (*) Departments of Physiology and Biomedical Engineering and Anesthesiology, College of Medicine, Mayo Clinic, St. Marys Hospital, Joseph 4W-184, 200 First Street SW, Rochester, MN 55905, USA e-mail: mantilla.carlos@mayo.edu; sieck.gary@mayo.edu Age-Related Remodeling of Neuromuscular Junctions Carlos B. Mantilla and Gary C. Sieck 38 C.B. Mantilla and G.C. Sieck In most skeletal muscles, motor units exhibit considerable functional diversity in terms of size, mechanical and fatigue properties (Burke et al. 1971; Sieck et al. 1989). Accordingly, recruitment of specific motor unit types is a major mechanism in neural control of muscle force generation and fatigue resistance (Clamann 1993). 1.1 Synaptic Plasticity More than 60 years ago, Donald Hebb introduced a conceptual framework (Hebbian Theory) to describe the basic mechanisms for changes in synaptic efficacy (synaptic plasticity). Central to his theory was the observation that synaptic efficacy improves when the fidelity between pre- and post-synaptic activity increases. Conversely, when fidelity between pre- and postsynaptic activity is disrupted, synaptic transmis- sion worsens. Synaptic plasticity has both structural and functional correlates. For examples, structurally, there may be axonal terminal sprouting or retraction, changes in the size and distribution of synaptic vesicle pools, and/or changes in the FF Type IIb fibers MyHC 2B FInt Type IIx fibers MyHC 2X FR Type IIa fibers MyHC 2A S Type I fibers MyHC Slow Motor Unit Types Fig. 1 Motor units (i.e., a motor neuron and the muscle fibers it innervates) are classified based on the mechanical and fatigue properties of muscle fibers. Four types are commonly described: (1) slow-twitch, fatigue resistant (type S), (2) fast-twitch, fatigue resistant (type FR), (3) fast- twitch, fatigue-intermediate (type FInt), and (4) fast-twitch, fatigable (type FF), which generally correspond to the expression of specific myosin heavy chain (MyHC) isoforms in the muscle fibers (type I fibers - MyHC Slow , type IIa fibers - MyHC 2A , type IIx fibers - MyHC 2X and type IIb fibers - MyHC 2B ). Motor unit recruitment order is generally matched to their mechanical and fatigue properties; thus, type S and FR motor units are recruited first and more often than type FInt and FF units 39Age-Related Remodeling of Neuromuscular Junctions extent of pre- and postsynaptic apposition and overlap. Functionally, synaptic plas- ticity is reflected by enhanced evoked postsynaptic potentials, persistent changes in presynaptic neurotransmitter release or postsynaptic excitability (long-term facilita- tion or depression), and changes in safety factor for neurotransmission resulting in either improved neurotransmission fidelity or neurotransmission failure. 1.2 Aging and Synaptic Plasticity With aging and senescence, there is a decrease in muscle activity often accompanied by unloading of limb muscle fibers. However, inactivity alone may not drive synaptic plasticity at the neuromuscular junction if fidelity of neuromuscular transmission (i.e., extent of correlation between pre- and postsynaptic activity) is maintained. Other age-related changes may drive synaptic plasticity. For example, an age-related loss of motor neurons amounts to denervation of some muscle fibers, consequently there may be axonal sprouting of spared motor neurons and re- innervation of muscle fibers and an increase in motor unit innervation ratio (Gordon et al. 2004; Balice-Gordon 1997). Age-related muscle fiber atrophy (i.e., sarcopenia) is also associated with concomitant changes in neuromuscular junction morphology, which may relate to removal of shared trophic influences (Vandervoort 2002; Delbono 2003). The effects of age-related inactivity, motor neuron loss and sarcopenia all depend on motor unit and/or muscle fiber type (Macaluso and De Vito 2004). Thus, it is likely that synaptic plasticity is a part of the normal aging process necessary to maintain muscle performance. 2 Motor Unit Properties and Recruitment The concept of the motor unit was introduced by Charles Sherrington in 1925 and forms the cornerstone of neuromotor control. A motor unit comprises a motor neu- ron and the group of muscle fibers it innervates (Fig. 1). In adult mammals, each muscle fiber is innervated by only a single motor neuron, while each motor neuron can innervate multiple muscle fibers. The number of muscle fibers innervated by a motor neuron (innervation ratio) varies widely from very small innervation ratios in hand and eye muscles (<10 fibers per motor neuron) to very large innervation ratios in trunk and proximal limb muscles (>500 fibers per motor neuron). Innervation ratio is inversely related to the fine control of force gradation with motor unit recruitment. Together with average muscle fiber cross-sectional area, innervation ratio determines the size of a motor unit and maximal force contributed by the motor unit. The level of force contributed by a motor unit is also dependent on the frequency of motor neuron discharge rate (frequency coding of force). Force-frequency properties of muscle fibers comprising motor units vary depending on contractile protein composition, which forms the basis of muscle fiber type clas- sification (Fig. 1; see below). 40 C.B. Mantilla and G.C. Sieck 2.1 Motor Unit and Muscle Fiber Type Classification Motor unit and muscle fiber type classification are concordant since they both relate to the mechanical and fatigue properties of muscle fibers. Different muscle fiber type classification schemes have been proposed, but the most commonly accepted scheme is based on the expression of different myosin heavy chain (MyHC) isoforms. Accordingly, in adult mammals, four muscle fiber types are classified: (1) type I (fibers expressing MyHC Slow ), (2) type IIa (fibers expressing MyHC 2A ), (3) type IIx (fibers expressing MyHC 2X ) and (4) type IIb (fibers express- ing MyHC 2B ). In single fiber studies, MyHC isoform expression has been shown to correlate with maximum isometric force, Ca 2+ sensitivity (related to force at submaximal activation underlying the force-frequency relationship), maximum velocity of shortening, cross-bridge cycling rate, ATP consumption rate, mito- chondrial volume density, and fatigue resistance (Geiger et al. 1999, 2000; Han et al. 2001, 2003; Sieck et al. 2003). Since motor units comprise a relatively homogenous group of muscle fibers, classification of four motor unit types is based on the mechanical and fatigue properties of their constituent muscle fibers: (1) slow-twitch, fatigue resistant (type S; comprising type I fibers), (2) fast-twitch, fatigue resistant (type FR; comprising type IIa fibers), (3) fast-twitch, fatigue-intermediate (type FInt; com- prising type IIx fibers), and (4) fast-twitch, fatigable (type FF; comprising type IIb fibers) (Fig. 1). As mentioned above, innervation ratio varies across muscles, but within a muscle, innervation ratio is generally greater for type FInt and FF motor units compared to type S and FR units. Muscle fiber size also varies across muscles, but within a muscle type IIx and IIb fibers are generally larger than type I and IIa fibers. Thus, there are differences in motor unit size across muscles and within a muscle, but generally type FInt and FF motor units are larger than type S and FR motor units. There are also differences in specific force (i.e., force per unit cross-sectional area) of different muscle fiber and motor unit types. Generally, type IIx and IIb fibers (type FInt and FF motor units) have greater specific force than type I and IIa fibers (type S and FR motor units). Consequently, because of their greater innervation ratio, larger fiber size and greater specific force, type FInt and FF motor units contribute greater forces than type S and FR units. 2.2 Motor Unit Recruitment In muscles of heterogeneous muscle fiber type composition, motor unit recruitment order is generally matched to their mechanical and fatigue properties; thus, type S and FR motor units are recruited first followed by type FInt and FF units. In models 41Age-Related Remodeling of Neuromuscular Junctions where this recruitment order was assumed and where the force contributed by each motor unit type was known, it was predicted that the forces required during most sustained motor behaviors (e.g., standing in the medial gastrocnemius (Walmsley et al. 1978) or quiet breathing in the diaphragm muscle (Sieck and Fournier 1989)) could be accomplished by recruitment of only type S and FR motor units (Fig. 2). In these models, recruitment of type FInt and FF motor units was required only during high force, short duration motor behaviors (e.g., jumping in the medial gas- trocnemius and coughing/sneezing in the diaphragm). Airway occlusion Fictive sneezing Eupnea Hypercapnia & Hypoxia 0 10 20 30 40 50 60 70 80 90 100 Recruitment of motor unit pool (%) Force (%) Type S Type FR Type FInt Type FF 0102030405060708090 100 Fig. 2 Model of motor unit recruitment for the rat diaphragm muscle. Motor units were assumed to be recruited in order: type S ® type FR ® type FInt ® type FF with complete activation of one motor unit type before the next type is recruited. Data is derived from previous studies reporting diaphragm muscle fiber type composition, force generated by type-identified fibers, and innervation ratio in adult male rats (Miyata et al. 1995; Zhan et al. 1997; Geiger et al. 2000; Sieck 1994). The relative force developed during different ventilatory (e.g., eupnea and hyper- capnia & hypoxia) and non-ventilatory tasks (e.g., airway occlusion and fictive sneezing). Based on the model, the inspiratory effort necessary to accomplish ventilatory demands imposed during eupnea requires recruitment of all of the type S motor units and some of the type FR motor units, while chemical airway irritation (i.e., fictive sneezing) would result in recruitment of most diaphragm motor units 42 C.B. Mantilla and G.C. Sieck 2.3 Aging Effects on Motor Unit Properties Clearly, age affects the mechanical properties of muscle fibers and consequently motor units. Generally muscles become weaker with age and this effect may reflect changes in muscle fiber cross-sectional area, MyHC content per half-sarcomere, and/or specific force. The cross-sectional area of type IIx and/or IIb fibers decreases with age (Maxwell et al. 1973). This may be the result of motor neuron loss and consequent denervation-induced atrophy (Xie et al. 2003). It may also reflect decreased neuromus- cular activity, mechanical unloading or altered trophic influences (Delbono 2003). MyHC content per half-sarcomere varies across muscle fiber types (Geiger et al. 2003, 2000), but does not appear to be affected by aging (Lowe et al. 2004b). However, with aging there is an increase in the proportion of fibers co-expressing MyHC isoforms, something that is relatively rare in young adults (Andersen et al. 1999). Specific force decreases with age, and this effect is especially pronounced at type IIx and IIb muscle fibers (i.e., type FInt and FF motor units) (Gosselin et al. 1994). Thus, muscle fiber weakness appears to reflect the combined influence of decreased fiber cross-sectional area and specific force. With respect to other mechanical properties of muscle fibers, converging evidence indicates that maximum velocity of shortening, cross-bridge cycling rate and ATP consumption rate are unaffected by aging across fiber types, but there may be differences across muscles (Lowe et al. 2004a). Importantly, there appears to be no age-related change in fatigability across muscle fiber types (Gonzalez and Delbono 2001), although maximum oxidative capacity is reduced in type II fibers of aged individuals (Proctor et al. 1995). 2.4 Aging Effects on Motor Unit Recruitment Based on converging indirect evidence it appears that with aging, there is a decrease in the number of type FInt and FF motor units due to the specific loss of these motor neurons (Hashizume et al. 1988; Caccia et al. 1979; Ishihara et al. 1987; Hashizume and Kanda 1995). This conclusion is based on the observation of a reduction in the number of retrogradely labeled motor neurons which appears to be most pronounced in fast-twitch hind limb muscles (Ishihara et al. 1987; Hashizume and Kanda 1995). In the same studies, it was observed that there were fewer type II fibers (no distinction was made between type IIa, IIx or IIb fibers) in hind limb muscles showing fewer motor neurons. In separate studies that did not estimate the number of motor neurons, selective reduction in the proportion of type IIx and IIb fibers was observed (Caccia et al. 1979). Selective loss of type FF and FInt motor units is also indirectly supported by the observation of an age-related increase in the proportion of type S and FR motor units in the rat plantaris muscle (Pettigrew and Gardiner 1987; Pettigrew and Noble 1991). The underlying basis for a selective loss of motor neurons is not yet resolved, but such an effect would definitely impact the ability to accomplish motor behaviors that require generation of greater forces (Fig. 2). As a result of motor neuron loss, some type IIx and IIb fibers would be denervated, and with subsequent reinnervation 43Age-Related Remodeling of Neuromuscular Junctions by remaining motor axons (mostly those of type S and FR motor units), there may be fiber type conversion as reflected by an increase in the proportion of fibers co-expressing different MyHC isoforms (Larsson et al. 1991). Sprouting and rein- nervation of adjacent muscle fibers should lead to an increase in motor unit innerva- tion ratios. Indirect evidence for such an increase in innervation ratios stems from analysis of changes in EMG during incremental force steps relative to the maximum evoked EMG response (M-wave) (Galea 1996). Age-related changes in the specific force of type IIx and IIb fibers together with the decrease in the overall proportion of these motor unit types would tend to decrease the diversity of motor unit properties within a muscle. An increase in the innervation ratio of type S and FR motor units would result in increased force production by these units, but it is unclear whether this increased force is required for the normal recruitment of these motor unit types (e.g., standing or quiet breathing). It is possible that an age-related increase in force genera- tion by type S and FR motor units partially offsets any age-related effects on type FInt and FF motor units, but it is unlikely that recruitment of type S and FR motor units can completely compensate for the forces required during high-force generating behaviors (e.g., jumping or coughing/sneezing). With aging, there appears to be a selective preservation of mechanical properties of motor units required for low force, sustained motor behaviors. In some cases, the advantage of such preservation is quite obvious, e.g., recruitment of type S and FR motor units in the diaphragm muscle to sustain ventilation or a similar recruitment of motor units in anti-gravity muscles to sustain posture. 3 Structural Properties of Neuromuscular Junctions The structural properties of neuromuscular junctions are matched to the functional demands of muscle fibers such that within a motor unit type the structure of neuromus- cular junctions is relatively uniform but there is considerable variability across differ- ent muscle fiber types (Fig. 3). The matching of pre- and post-synaptic specializations at the neuromuscular junction also depends on muscle fiber type. For example, presyn- aptically, there are differences in the distribution and size of synaptic vesicle pools and terminal surface area. Postsynaptically, there are differences in the number and depth of junctional folds and apposition of subcellular organelles such as mitochondria. Finally, the overlap of pre- and post-synaptic structures varies across fiber types. 3.1 Fiber Type Differences in Neuromuscular Junction Structure Within a muscle, neuromuscular junctions at type I and IIa fibers are smaller with less complex branching patterns than those at type IIx and/or IIb fibers (Prakash and Sieck 1998; Mantilla et al. 2004; Prakash et al. 1995, 1996b; Sieck and Prakash 1997). 44 C.B. Mantilla and G.C. Sieck However, it is difficult to extrapolate across muscles since neuromuscular junctions at type I fibers in the soleus muscle are larger and more complex than neuromuscular junctions at type IIx and/or IIb fibers in the extensor digitorum longus muscle (Reid et al. 2003). Within a muscle, fiber size is an important determinant of neuromuscular junction area and complexity. For example, in the rat diaphragm muscle, the area of neuromuscular junctions among type I fibers varies directly with fiber cross-sectional area (Prakash and Sieck 1998; Sieck and Prakash 1997). Fiber type dependent differences in gross structural properties of neuromuscu- lar junctions are also reflected at pre- and post-synaptic elements. For example, both axon terminal and motor end-plate surface areas are ~75–90% greater at type IIx and/or IIb fibers than at type I and IIa fibers in the rat diaphragm (Sieck and Prakash 1997; Prakash et al. 1996b; Rowley et al. 2007; Mantilla et al. 2004). At all muscle fibers, the surface area of axon terminals is smaller than their corre- sponding motor end-plate and the extent of this difference varies across muscle fiber types (Prakash et al. 1996b). For example, at type I diaphragm fibers, the surface area of the presynaptic terminal more closely approximates that of the motor end-plate, with nearly 95% overlap. By comparison, at type IIb fibers, the presynaptic terminal only overlaps ~70% of the motor end-plate. These differ- ences in the extent of overlap may reflect phenotypic differences in the ability of nerve terminal branches to invade motor end-plate gutters during development (Prakash et al. 1995) or remodeling (Prakash et al. 1996a, 1999). It is also possible that the increased fragmentation of neuromuscular junctions at muscle fibers of greater size results in greater branch termination limiting invagination of the axon terminal into motor end-plate gutters. In either case, these differences in extent of overlap may have significant physiological implications, impacting neuromuscular transmission. Fig. 3 Structural characteristics of a neuromuscular junction (NMJ) vary across muscle fiber types. Pre-synaptic terminals and motor end-plates at the diaphragm muscle of young (6 months) and old rats (24 months) were labeled with the neuronal ubiquitin decarboxylase PGP9.5 and a-bungarotoxin, respectively (Prakash and Sieck 1998). Note the differences in size and complex- ity (number and length of branches) across fiber types, with NMJs present at type I or IIa fibers being smaller and less complex than those at type IIx and/or IIb fibers. With aging there is con- siderable fragmentation and expansion of both pre- and post-synaptic elements 45Age-Related Remodeling of Neuromuscular Junctions Fiber type differences in neuromuscular junction remodeling vary depending on a number of factors including hormonal environment and activity. For example, the areas of both pre- and postsynaptic elements of neuromuscular junctions at type I diaphragm fibers decreased after 3 weeks of hypothyroidism induced by propylthiouracil (Prakash et al. 1996a). In contrast, after 2 weeks of diaphragm inactivity induced by either tetrodotoxin phrenic nerve blockade or spinal cord hemisection at C 2 the areas of both pre- and postsynaptic elements of neuromuscular junctions at type IIx and/or IIb diaphragm fibers increased while those at type I fibers decreased (Prakash et al. 1999). At type IIx and/or IIb fibers, the extent of overlap between pre- and postsynaptic elements of the neuromuscular junction increased to ~90% after 2 weeks of diaphragm inactiv- ity induced by tetrodotoxin phrenic nerve blockade or spinal cord hemisection at C 2 . Surprisingly, the similar structural changes induced by tetrodotoxin phrenic nerve blockade and spinal cord hemisection at C 2 yielded markedly dif- ferent effects on neuromuscular transmission. Following inactivity induced by spinal cord hemisection at C 2 neuromuscular transmission with repetitive acti- vation was markedly improved, whereas there was substantially greater neuro- muscular transmission failure following tetrodotoxin phrenic nerve blockade. These functional differences are closely related to ultrastructural differences at the neuromuscular junction that form the basis of neuromuscular transmission (see below). 3.2 Ultrastructural Properties of Presynaptic Terminals The total number of synaptic vesicles undergoing repeated cycles of endo- and exocytosis (i.e., cycling) is greater at type IIx and/or IIb fibers compared to type I and IIa fibers (Mantilla et al. 2004, 2007; Rowley et al. 2007). Ultrastructurally, synaptic vesicles at presynaptic terminals segregate into a pool of vesicles docked at specialized sites for neurotransmitter release – active zones – i.e., readily releas- able, a pool immediately adjacent to active zones (within 200 nm) and a more distant, reserve pool (Sudhof 2004). Consistent with greater overall size of the cycling synaptic vesicle pool size, the densities of synaptic vesicles in both the immediately adjacent pool and the reserve pool are greater at presynaptic termi- nals of type I and IIa fibers compared to type IIx and/or IIb fibers. The size (length) and distribution of individual active zones does not vary across presynap- tic terminals at the different fibers types (Fig. 4). Similarly, the number of synaptic vesicles docked at each active zone (i.e., readily releasable) is consistent across fiber types (Rowley et al. 2007). However, fiber type differences in presynaptic terminal surface area yield greater total number of active zones per presynaptic terminal at type IIx and/or IIb fibers than at type I and IIa fibers, and thus, a greater total number of synaptic vesicles in the readily releasable pool at type IIx and/or IIb fibers compared to type I and IIa fibers (Mantilla et al. 2004; Rowley et al. 2007). Consistent with these ultrastructural properties, quantal release at type IIx . influences (Vandervoort 2002; Delbono 2003). The effects of age-related inactivity, motor neuron loss and sarcopenia all depend on motor unit and/ or muscle fiber type (Macaluso and De Vito. type FInt and FF motor units compared to type S and FR units. Muscle fiber size also varies across muscles, but within a muscle type IIx and IIb fibers are generally larger than type I and IIa. (ed.), Sarcopenia – Age-Related Muscle Wasting and Weakness, DOI 10.1007/978-90-481-9713-2_3, © Springer Science+Business Media B.V. 2011 Abstract Remodeling of neuromuscular junctions (NMJs) and