www.aging-us.com AGING 2021, Vol. 13, No. 14 Research Paper Beneficial effects of dietary supplementation with green tea catechins and cocoa flavanols on aging-related regressive changes in the mouse neuromuscular system Sílvia Gras1,*, Alba Blasco1,*, Guillem Mòdol-Caballero2, Olga Tarabal1, Anna Casanovas1, Lídia Piedrafita1, Alejandro Barranco3, Tapas Das4, Ricardo Rueda5, Suzette L. Pereira4, Xavier Navarro2, Josep E. Esquerda1, Jordi Calderó1 1 Unitat de Neurobiologia Cellular, Departament de Medicina Experimental, Facultat de Medicina, Universitat de Lleida and Institut de Recerca Biomèdica de Lleida (IRBLleida), Lleida, Spain 2 Grup de Neuroplasticitat i Regeneració, Institut de Neurociències, Departament de Biologia Cel·lular, Fisiologia i Immunologia, Universitat Autònoma de Barcelona and CIBERNED, Bellaterra, Spain 3 Department of Biochemistry and Molecular Biology II, School of Pharmacy, University of Granada, Granada, Spain 4 Abbott Nutrition, Research and Development, Columbus, OH 43215, USA 5 Abbott Nutrition, Research and Development, Granada, Spain *Equal contribution In memory of Ron Oppenheim (1938-2021), passionate scientist, great neurobiologist and better friend Correspondence to: Jordi Calderó; email: jordi.caldero@udl.cat Keywords: sarcopenia, green tea, neuromuscular system, aging, cocoa Received: April 29, 2021 Accepted: June 19, 2021 Published: July 28, 2021 Copyright: © 2021 Gras et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. ABSTRACT Besides skeletal muscle wasting, sarcopenia entails morphological and molecular changes in distinct components of the neuromuscular system, including spinal cord motoneurons (MNs) and neuromuscular junctions (NMJs); moreover, noticeable microgliosis has also been observed around aged MNs. Here we examined the impact of two flavonoid-enriched diets containing either green tea extract (GTE) catechins or cocoa flavanols on age-associated regressive changes in the neuromuscular system of C57BL/6J mice. Compared to control mice, GTE- and cocoa-supplementation significantly improved the survival rate of mice, reduced the proportion of fibers with lipofuscin aggregates and central nuclei, and increased the density of satellite cells in skeletal muscles. Additionally, both supplements significantly augmented the number of innervated NMJs and their degree of maturity compared to controls. GTE, but not cocoa, prominently increased the density of VAChT and VGluT2 afferent synapses on MNs, which were lost in control aged spinal cords; conversely, cocoa, but not GTE, significantly augmented the proportion of VGluT1 afferent synapses on aged MNs. Moreover, GTE, but not cocoa, reduced aging-associated microgliosis and increased the proportion of neuroprotective microglial phenotypes. Our data indicate that certain plant flavonoids may be beneficial in the nutritional management of age-related deterioration of the neuromuscular system. INTRODUCTION The compromised muscular function associated to sarcopenia has a negative impact on the life quality of Sarcopenia, the progressive loss of skeletal muscle mass older adults and increases the risk for adverse health and function with age, is considered the main causative outcomes including disability, fall-associated injuries, factor of the physical performance decline in the elderly. morbidity, and mortality [1]. All these aspects stress www.aging-us.com 18051 AGING healthcare systems, increasing the care needs of older low in calories, has been shown to attenuate aging people and their costs. With a prevalence of 8-40% in sarcopenia in various species by acting at different people over 60 years, depending on studies [2, 3], levels of the skeletal muscle. Thus, a line of evidence sarcopenia constitutes a critical challenge in our aging indicates that a moderate reduction in caloric intake, society. Starting in the third decade of life, humans alone or combined with exercise training, is able to experience a gradual deterioration of muscle power, and mitigate mitochondrial alterations and production of over the age of 60 a decline of about 3% of muscle mass reactive oxygen species, modulate autophagy, enhance per year has been reported to occur in most of individuals the endogenous myofiber repair and counteract the [3, 4]. Different mechanisms have been identified to increase in proapoptotic signals observed in aged contribute to the loss of muscle mass with age. In skeletal muscles [16–20]. Caloric restriction has also fact, sarcopenia in the elderly is presently considered a been reported to ameliorate age-related changes in complex multifactorial condition which involves intrinsic rodent NMJs and to prevent MN and motor axon and extrinsic causative factors (reviewed by [5]); among degeneration found to occur with aging [11, 21]. In a of them, chronic inflammation, metabolic and endocrine similar way, some dietary supplements have been alterations, poor nutrition, mitochondrial dysfunction, shown to counteract age-related changes that contribute oxidative damage, and neurogenic factors, appear to play to neuromuscular dysfunction (reviewed by [12]). an important role in sarcopenia. Thus, several nutritional interventions, involving supplementations with vitamin D, omega-3 fatty Accompanying the skeletal muscle wasting, a plethora acids, creatine, β-hydroxy-β-methybutyrate, or dietary of structural and functional changes has been found phospholipids, have been reported to be beneficial in at advanced ages in distinct components of the promoting healthy neuromuscular aging throughout neuromuscular system, including spinal cord different mechanisms; these include: a) myotrophic, motoneurons (MNs), dorsal root ganglion (DRG) neurotrophic and anti-inflammatory actions; b) sensory neurons, and neuromuscular junctions (NMJs) regulation of pathways involved in mitochondrial ([6], see also [5, 7–9] as reviews). Thus, studies function and reactive oxygen species production; c) performed in both animals and humans have reported inhibition of muscle proteolysis; and d) improvement of that, with aging, MNs become dysfunctional, show a cellular bioenergetics. marked depletion of afferent nerve terminals and, some of them, exhibit signs of degeneration. Prominent Plant flavonoids have gained particular attention as gliosis, involving both microglia and astroglia, have dietary compounds for keeping good health and also been found in the spinal cord around aged MNs. In preventing a number of diseases, particularly cardiac this vein, we have recently reported a marked increase disorders and cancer [22]. Several studies have reported in the microglial and astroglial pro-inflammatory the effects of polyphenolic-flavonoids-containing sup- phenotypes (M1 and A1, respectively) in the spinal cord plements in the attenuation of oxidative stress and muscle of aged mice to the detriment of anti-inflammatory and damage, particularly after intensive strength and neuroprotective (M2 and A2) glial subpopulations [6]. resistance exercise [23–28]. In this regard, previous Additionally, with age, NMJs appear to be remodeled research has shown the benefits of polyphenols contained and exhibit marked changes in their pre- and in green tea and its extracts on the endurance capacity by postsynaptic compartments in association with the preventing muscle impairment in patients with Duchenne loss of motor units, reactive nerve sprouting and cycles muscular dystrophy [29] and mouse models of this of muscle denervation/reinnervation. Aging is also disease [30–32], and also in aging [33, 34] and accompanied by alterations in DRG sensory neurons immobilization-induced muscle loss [35, 36]. Green tea in form of changes in the expression of different extract (GTE) is enriched in flavonoids known as neuropeptides and atrophy of both proprioceptive and catechins (including catechin, epicatechin, gallocatechin, nociceptive neuronal populations [6]. However, on the epigallocatechin, epicatechin gallate and, predominantly, whole, the primary contribution of such defects in the epigallocatechin gallate [EGCg]), which have strong skeletal muscle deterioration with aging is not yet well antioxidant and anti-inflammatory properties [37, 38]. A stablished. In this regard, the fact that the age-associated class of flavonoids known as flavanols are also the main NMJ disruption is a secondary process resulting from constituents of cocoa beans and their derivatives. intrinsic defects in the skeletal muscle cannot be Thus, cocoa products are high in non-esterified monomers excluded. of the flavanols catechin and epicatechin, and also proanthocyanidins [39]. Among of these compounds, Age-related decline in muscle function and motor epicatechin represents about 35% of the total content of activity is susceptible to be ameliorated by lifestyle polyphenols in cocoa beans, and appears to be the main interventions such as exercise and/or nutritional responsible for the beneficial effects of dietary cocoa. In strategies [10–15]. Caloric restriction, based on a diet fact, cocoa polyphenols have been shown to exert diverse www.aging-us.com 18052 AGING physiological activities as antioxidants, cardiovascular GTE or cocoa flavanols ameliorate the aging degree protectors, immunomodulators, anti-inflammatory agents and degeneration rate of skeletal muscle fibers of and potential tumor protectors [40–44]. Epicatechin has aged mice been reported to promote neuroprotection in vitro against oxidative stress [45]. Moreover, in vivo studies in mice We next examined the potential effects of the GTE- or have found that an epicatechin dietary supplementation cocoa-supplemented diets on the aging-associated enhances the oxidative capacity of the skeletal muscle and histological changes found in skeletal muscles in our improves the survival rate of animals and the physical previous study [6]. Compared to control group, no performance impairment occurring with aging [46, 47]. significant changes in tibialis anterior (TA) and soleus Despite the body of reported data on the beneficial actions (Sol) muscle-wet weights, either gross or normalized to of dietary polyphenols on physical activity and exercise body weight (muscle mass/body weight) or cross performance, there is little work examining the effects of section area, were found in animals fed with any of the these compounds on the structural and molecular two supplemented diets (Figure 2A–2C), indicating that organization of the neuromuscular system. In fact, most they are not able to prevent the moderate reduction of of the previous analyses have been focused on flavonoid muscle mass observed in the course of aging. Moreover, actions in diverse skeletal muscle parameters with very supplemented diets did not either provoke any little information on their effects on the neural elements significant change in the size and density of fibers or the of the neuromuscular system. content of connective tissue in aged control muscles (Figure 2D–2F). In the present study, we tested the impact of flavonoid- enriched diets containing either GTE catechins or cocoa We and others have reported that skeletal muscles of flavanols, on the age-related deterioration of the aged mice frequently display high proportions of fibers neuromuscular system. Our main goal was to explore exhibiting centrally located nuclei [6, 11]. This change the potential benefits of these flavonoids on age- appears to reflect a regenerative process in muscles in associated alterations of the spinal cord MNs and glia, an attempt to replace degenerated myofibers that are DRG sensory neurons, motor nerves, NMJs, and lost in the course of aging [48, 49]. We found that, skeletal muscles, and whether these interventions can compared to control muscles, those from animals of mitigate the age-related motor decline in a relevant either GTE or cocoa groups showed a significant mouse model [6]. reduction in the proportion myofibers displaying central nuclei (Figure 2G–2J). On the other hand, our and other RESULTS labs have described that aging entails a depletion of muscle satellite cells (SCs) [6, 50–53]. The effects of GTE and cocoa flavanols improve the survival rate the supplemented diets on SCs were then examined by of mice immunohistochemistry for paired box protein 7 (Pax7), a paired homeobox transcription factor involved in the Starting at age of 92 weeks (21 months), mice were proliferation of the muscle precursor cells, in fed with a standard (AIN-93M) diet (control group) or combination with DAPI staining. We found that TA and AIN-93M diet supplemented with either GTE (GTE Sol muscles from animals of GTE and cocoa groups had group) or cocoa flavanols (cocoa group). All diets significantly higher densities of SCs than those from were maintained until week 116 (27 months), the control mice (Figure 2K–2N3). selected endpoint in which animals were euthanized for tissue sampling. Daily food intake was 3 g/d and We wanted to analyze next the effects of the flavonoid- average body weight was ~30 g. At endpoint, animals enriched diets on the presence of lipofuscin in aged from GTE group and cocoa group exhibited a muscles. Lipofuscin is considered a pigment resulting significant higher survival rate (~66%, GTE group; from an oxidative stress process [54], and is widely seen ~73%, cocoa group, p < 0.05 and p < 0.01 as a hallmark of aging. Lipofuscin has been shown to respectively) compared to control mice (~33%; accumulate in fibers of aged and dystrophic muscles [6, Gehan-Breslow-Wilcoxon test) (Figure 1A). In 55, 56]. As expected from our previous results [6], comparison with control group mice, those from GTE control aged hindlimb muscles exhibited significant and cocoa groups had a reduction in food intake differences in lipofuscin content, with the slow-twitch between 105 and 109 weeks of age (Figure 1B). No Sol muscle displaying much lower proportions of the significant differences in body weight were observed pigment than the fast-twitch TA muscle. GTE- and between animals of control and cocoa groups; cocoa-supplemented diets significantly reduced the however, mice fed with GTE-enriched diet exhibited, amount of lipofuscin in muscles by decreasing both the at 111-115 weeks of age, significantly reduced body number of myofibers containing the pigment and the weights compared with control animals (Figure 1C). number of its aggregates (Figure 2O–2S). Overall, these www.aging-us.com 18053 AGING results suggest that flavonoid-enriched diets are able to amount of newly regenerated myofibers, and in the alleviate the process of skeletal muscle senescence, preservation of SC population due to the decline in the which is reflected in the reduction of myofiber number of these stem cells that engage their myogenic degeneration. This would result in a decrease in the program to replace compromised aged myofibers. Figure 1. GTE and cocoa supplementations improve mouse survival rate. (A) Survival rate of mice from control, GTE and cocoa groups; the standard (AIN-93M) diet supplemented with either cocoa or GTE significantly increased the percentage of mice that were alive at the endpoint of experiment compared with control animals. (B) Food intake (expressed as g of food per week and mouse) was reduced in animals fed with the GTE- or cocoa-supplemented diets, this reduction being statistically significant in a period ranging from 105-109 weeks of age. (C) Body weight (g) of animals from the three experimental groups; compared to control mice, no significant changes in body weight were found in animals from cocoa group; a significant reduction in weight was, however, observed in GTE-supplemented mice. Data are shown as the mean ± SEM (number of animals per group: 91 weeks of age, n = 15 in all groups; 107 weeks, control n = 8, GTE n = 13; cocoa n = 15; 114 weeks, control n = 6, GTE n = 12, cocoa n = 11); *p < 0.05 and **p < 0.01 vs. control (Gehan-Breslow-Wilcoxon test, in (A) and multiple t-test (Bonferroni correction), in (B, C). www.aging-us.com 18054 AGING Figure 2. Impact of GTE- and cocoa-supplemented diets on aging-associated changes in TA and Sol muscles of mice. (A–C) Muscle wet weight (in mg, A), muscle weight relative to body weight (expressed in mg/g, B) and muscle cross-sectional area (in µm2, C) are shown. (D) Average myofiber size (cross-sectional area in µm2). (E, F) Density of myofibers (E) and muscular content of connective tissue (F), expressed as the percentage of area occupied by either myofibers or connective tissue respect to the total cross-sectional muscle area). (G) Proportion of myofibers displaying central nuclei. (H–J) Representative images of transversal cryosections of TA muscles of control, GTE and cocoa groups (as indicated); sections were double labeled with an antibody against laminin (red) and DAPI (blue) for DNA; arrows in (H) indicate central nuclei. (K) Percentage of Pax7-immunostained cells (SCs) with respect to DAPI-positive nuclei. (L1–N3) Representative images of a combined immunolabeling for Pax7 (red) and laminin (green), and DAPI staining (blue) in transversal cryosections of Sol muscles from control, GTE and cocoa groups, as indicated. (O, P) Percentage of myofibers containing lipofuscin aggregates (O), and average number of lipofuscin granules per myofiber (P). (Q–S) Representative images of transversal cryosections of TA muscles from control, GTE and cocoa groups (as indicated) immunolabeled for laminin (green); lipofuscin autofluorescence (red) was excited using 510-560 nm excitation and 590 emission filters. Data in graphs are expressed as the mean ± SEM; the average values of adult muscles found in a previous study [6] were indicated by a red line in each graph for comparative purposes; ***p < 0.001 vs. control (Ctrl), two-way ANOVA, Bonferroni’s post hoc test; sample size per condition: (A–C) = 6-12 muscles and (D–F) = 3-5 muscles from different mice; (G, K) = 2500-4000 fibers and (O, P) = 1200- 2200 fibers per muscle from 3-5 animals. Scale bars: 40 µm in (J) (valid for H, I), in N3 (valid for L1–N2) and S (valid for Q, R). www.aging-us.com 18055 AGING GTE and cocoa flavanols prevent age-associate TA muscle, which is predominantly composed of fast alterations in skeletal muscle mitochondria myofibers (2A > 2B), which have a glycolytic metabolism and reduced mitochondrial content Peroxisome proliferator-activated receptor γ co- compared to slow (oxidative) myofibers [6, 63]. coactivator 1α (PGC-1α) is a master regulator of Compared to TA muscles of adult mice, those of aged mitochondrial biogenesis and function [57]. PGC-1α animals showed a significant (~25%) reduction of appears to play an important role in muscle by ATP5A immunostaining in myofibers, indicating that regulating myofiber metabolism, autophagy, and the an overt decrease in the number of mitochondria occurs expression of different components of NMJs [58, 59]. with aging in muscles (ATP5A-immunoreactivity Some studies have shown that PGC-1α content declines expressed as arbitrary units: adult TA, 94.01 ± 2.10; in muscle with normal aging in parallel with the aged TA, 72.59 ± 1.38; p < 0.001, Student’s t-test, 100 decrease in mitochondrial function [60–62]. To examine randomly selected myofibers were analyzed from 4 whether GTE and cocoa enhance skeletal muscle mice per condition). Muscles of mice fed with either metabolism via regulation of PGC-1α, its expression GTE- or cocoa-supplemented diets exhibited was analyzed by western blotting. Muscle extracts of significantly higher mitochondrial contents than those animals which received either of those supplements of controls (Figure 3C–3F). resulted in higher levels of PGC-1α protein compared to controls, but only the GTE group achieved statistical GTE- and cocoa-supplemented diets prevent muscle significance (Figure 3A, 3B). To examine whether the denervation and improve regressive NMJ alterations increased expression of PGC-1α in skeletal muscle occurring with aging induced by GTE and cocoa supplementation was linked to the augmented density of mitochondria in myofibers, Previous studies in rodents have reported that NMJs of we performed immunohistochemistry with an anti- skeletal muscles undergo structural and molecular ATP5A antibody, used as a mitochondrial marker. alterations with aging [6, 9, 11, 63–65]. Using the same ATP5A immunolabeling was analyzed in the fast-twitch mouse model, we observed that a significant number of Figure 3. GTE- and cocoa-supplemented diets partially prevent the decrease in PGC-1α expression and restore mitochondrial depletion occurring muscles in the course of age. (A) Densitometric analysis of changes in PGC-1α levels in muscles from the three experimental conditions; data were normalized to α-tubulin. Bars represent the values (mean ± SEM) of 3 mice per condition from 2 independent western blot analysis; *p < 0.05 vs. control, two-way ANOVA (Bonferroni’s post-hoc test); red lines in graph indicate PGC-1α levels found in adult muscles. (B) Representative western blots of PGC-1α and α-tubulin (as loading control) proteins in TA and Sol muscles from mice of control (Ctrl), GTE and cocoa groups. (C) Quantification of ATP5A-immunoreactivity in TA muscles of control, GTE and cocoa groups; bars represent the values (mean ± SEM) of 3 mice per condition; **p < 0.01 and ***p < 0.001 vs. control, one-way ANOVA (Bonferroni’s post-hoc test). (D–F) Representative images showing a combined immunolabeling for laminin (lam, green) and ATP5A (red) in transversal cryosections of TA muscles from three experimental groups used for quantification, as indicated. Note the overt increase in ATP5A immunostaining after GTE (E) and, particularly, cocoa (F) supplementation compared to control (D). Arrow in (D) points out lipofuscin deposition in a myofiber. Scale bar in (F): 50 µm (valid for D, E). www.aging-us.com 18056 AGING NMJs of aged hindlimb muscles exhibit overt signs of young mice, the levels of these synaptic molecules in denervation and polyinnervation, increased branching NMJs markedly decline in adult muscles, but are complexity and sprouting of nerve terminals; upregulated in those of aged animals [6]. The potential additionally, postsynaptic sites display a marked impact of GTE and cocoa supplementations in the fragmentation of endplates, with numerous scattered content of CGRP, agrin and GAP-43 in aged muscles acetylcholine receptor (AChR) spots which are was examined. In comparison with control NMJs, those associated to neurofilament 68 kDa (NF)-positive and from GTE and cocoa groups exhibited a modest, but vesicular acetylcholine transporter (VAChT)-positive non-significant increase in CGRP- and agrin- immunostaining of presynaptic structures. Overall, this immunoreactivity (Supplementary Figure 1A–1F3). On pattern differs from that we observed in NMJs from the other hand, no significant changes in GAP-43- mice at early postnatal ages, in which postsynaptic immunostaining at NMJs were found between the three sites display a more homogeneous and undivided groups (Supplementary Figure 1G–1I3), although the morphology, and nerve branching is highly pleomorphic GAP-43 content in the entire muscles determined by (not shown). All these changes are indicative of a western blotting showed a tendency to be decreased in reactive plasticity and neoformation of neuromuscular animals that received the dietary supplementations synapses in the course of aging [6]. We wanted to (Supplementary Figure 1J, 1K). analyze whether these age-associated changes could be prevented with either a GTE- or a cocoa-supplemented GTE or cocoa supplements do not prevent the diet. The architecture and innervation of NMJs were degenerative changes occurring in motor nerves with examined by double immunostaining for NF and aging synaptic vesicle protein 2 (SV2, for the presynaptic nerve terminal visualization) combined with α- As previously reported, aging is associated with motor bungarotoxin (α-Bgtx, for the postsynaptic site axon degeneration. This can be already noticed identification). The two tested diets significantly in middle-aged mice and include Wallerian-like reduced the proportion of denervated or partially degenerative changes, myelin spheroidal inclusions, innervated NMJs found in aged muscles, which was axonal swelling, and reduction of myelin thickness [6, accompanied by an increased amount of NMJs 72]. To analyze whether either GTE- or cocoa- exhibiting a higher degree of innervation (Figure 4A); supplemented diets were able to prevent motor nerve moreover, the proportion of monoinnervated NMJs was degeneration, we examined L4 ventral nerve roots significantly increased with polyinnervated NMJs (VRs), which are shaped by the most proximal part of exhibiting either no changes or a significant reduction in motor axons contributing to the sciatic nerve formation. number, except for TA muscle of GTE group mice Neither flavonoids promoted a substantial improvement (Figure 4B). In the same line, the amount of motor of structural age-associated motor nerve alterations. No axons showing terminal sprouting was reduced in significant changes in the cross-sectional area of VRs, muscles from GTE and cocoa groups compared to number and diameter (including myelin sheath) of controls (Figure 4C); this decline, however, did not axons, and the proportion of degenerating axons per attain statistical significance possibly due to sample size nerve root were found between the three experimental and variability between animals. Animals fed with groups (Supplementary Figure 2A–2F, 2I–2K). We either of the two dietary supplements also displayed even noticed that, compared with the control group, significantly decreased proportions of fragmented VRs of animals fed with GTE or cocoa supplements had endplates in muscles (Figure 4D). Additionally, muscles a higher, although not significant, increase in the from dietary supplemented animals globally exhibited number of axons displaying minor signs of degeneration more “healthy” NMJs, as indicated by the increased and myelin inclusions. However, roots from cocoa number of endplates exhibiting a well-defined “pretzel- group animals, showed a slightly increase in the number like” pattern, which contrasted with the less mature and of clusters of small myelinated axons (Supplementary altered appearance of numerous endplates typically Figure 2G–2K); these could represent immature axons found in control muscles (Figure 4E). Some examples formed by sprouting from healthy ones and suggest the of aged TA and Sol NMJs of control, GTE and cocoa generation of new nerve fibers aimed at replacing those groups are shown in Figure 4F1–4K3. previously degenerated. G-ratio value, as an indicator of the relationship between myelin thickness and axon In our previous work [6] we have shown that aging is diameter, did not show significant differences between associated with changes in the expression of different groups (Supplementary Figure 2L, 2M). However, proteins related to development, plasticity and compared to control VRs, those from GTE and cocoa maintenance of NMJs, such as calcitonin gene-related groups exhibited slightly smaller axonal diameters peptide (CGRP) [66–69], agrin [70], and growth (excluding the myelin sheath), with values closer to associated protein 43 (GAP-43) [71]. Compared to those found in adult mice (Supplementary Figure 2N). www.aging-us.com 18057 AGING Dietary supplementation with GTE or cocoa flavanols suggest that the degree of muscular wasting with age also ameliorates aging-associated alterations in would be associated not only to muscle location and proximal hindlimb muscles fiber type composition, but also to specific function and activity of muscles. We have also shown that gracilis There is evidence that aging affects differentially (Gra) muscle, which is proximally placed in the distinct muscle groups [73–75]. Our previous data [6] hindlimb, exhibited more prominent age-related Figure 4. GTE- and cocoa-supplemented diets prevent aging-associated muscle denervation and regressive morphological alterations in NMJs. (A) Proportion of TA and Sol NMJs displaying different degrees of innervation; quantification was based on the percentage of α-Bgtx-labeled postsynaptic site area covered by SV2-immunostained presynaptic terminals (see Materials and Methods, <15% innervation was considered as denervated). (B) Number of NMJs of TA and Sol muscles exhibiting single (mono.) or multiple (poly.) innervation expressed as fold change of control group (blue dashed line). (C–E) Percentage of NMJs showing terminal axonal sprouts (C), fragmented endplates (D) and postsynaptic sites exhibiting a pretzel-like appearance (E, indicative of high degree of synaptic maturity), in TA and Sol muscles of animals from different experimental groups. Bars in graphs represent the mean ± SEM; sample size: 30-58 (A), and 50-85 (B–E) NMJs per muscle from 3-5 animals per condition; *p < 0.05, **p < 0.01 and ***p < 0.001 vs. control, one or two-way ANOVA, Bonferroni’s post hoc test; red lines in (C–E) indicate values in adult mice previously reported [6]. (F1–K3) Representative maximal projections of confocal stacks of NMJs of TA (F1–H3) and Sol (I1–K3) from mice of control, GTE and cocoa groups (as indicated in panels); muscle sections were stained with antibodies against NF and SV2 (green, for presynaptic nerve terminals), and α-Bgtx (red, for postsynaptic AChR). Scale bar in K3 = 20 µm (valid for F1–K2). www.aging-us.com 18058 AGING alterations than distal leg muscles. The effects of labeled with antibodies against choline GTE- and cocoa-enriched diets were then examined in acetyltransferase (ChAT) or VAChT (Figure 6A1, this proximal muscle. We found that compared to 6A2). In lumbar spinal cord sections of adult and old control group, Gra muscles from GTE and cocoa mice we counted the number of VAChT- groups displayed significant increased density of immunolabeled V0C interneurons in an area of 0.1 Pax7-positive SCs, and reduced proportions of mm2 around the central canal. In comparison with myofibers exhibiting lipofuscin contents (GTE) and adult mice, aged animals exhibited a prominent decreased numbers of lipofuscin aggregates per tendency to reduce (~30%) the density of V0 C myofiber (GTE and cocoa) (Supplementary Figure interneurons (adult: 5.2 ± 0.4, old: 3.6 ± 0.6 [mean ± 3A–3G). Furthermore, Gra muscles from GTE and SEM]; n = 4 animals per group, p = 0.07, Student’s t- cocoa groups had a significant reduction in the test). The decline in the number of V0 C interneurons number of denervated and partially innervated NMJs found with aging correlates with the reduction in C- and an increase in the proportion of those showing a bouton density observed on aged MNs, suggesting single innervation with respect to control counterparts that cholinergic deafferentation of MNs is, at least in (Supplementary Figure 4A–E3). part, due to the loss of V0 C interneurons. Since GTE prevented the loss of C-boutons on MNs, we Effects of dietary supplementations with GTE and examined whether this supplement was able to cocoa flavanols on MN synaptic afferents of old mice prevent the aging-associated decline of V0C cells and found that mice fed with GTE-enriched diet exhibited We and others have reported that aging is accompanied significant higher density of such interneuron by a significant depletion of afferent inputs to MNs [6, population compared to control animals (Figure 6B– 76, 77]. Thus, with age, MNs suffer a loss of excitatory 6E2). Thus, the beneficial action of GTE on C-bouton cholinergic and glutamatergic inputs, with no overt synapse stabilization is in concordance with its changes in the density of inhibitory GABAergic afferent capacity to preclude V0 C interneuron loss. synapses. We examined then whether supplementations with GTE or cocoa flavanols were able to prevent the Impact of dietary supplementations with GTE and process of aging-associated MN deafferentation. Spinal cocoa flavanols on age-related gliosis in ventral horn cord sections from old mice were immunostained with spinal cord of old mice antibodies against VAChT, vesicular glutamate transporter 1 (VGluT1) and vesicular glutamate We have previously reported that, in parallel with transporter 2 (VGluT2) (for cholinergic [C-boutons] and central synapse loss, a prominent gliosis occurs in the glutamatergic excitatory synapses, respectively), and spinal cord ventral horn of old mice, suggestive that the number and size of labeled synaptic puncta on MN aging is accompanied by a neuroinflammatory somata were analyzed. We found that, in comparison response around MNs [6]. Glia, particularly microglia, with MNs of control mice, those of animals that appear to play a major role in synapse elimination received GTE-, but not cocoa-, supplementation during development and in pathological conditions; exhibited significantly higher density and larger area of moreover, glia can indirectly alter synaptic function cholinergic synaptic terminals (Figure 5A, 5B, 5G1– throughout the release of pro-inflammatory mediators 5I2), and significantly increased density of VGluT2- (recently reviewed by [79]). We wanted to examine positive synapses (Figure 5E, 5F, 5M1–5O2). whether a diet supplemented with either GTE or cocoa Conversely, a significant increase in the density of flavanols is able to prevent aging-associated VGluT1-positive synapses was observed on MNs of microgliosis in spinal cord. Ionized calcium-binding cocoa group animals; MNs of GTE group mice, adaptor molecule 1 (Iba1)-immunoreactivity, used as however, exhibited no significant changes in the a specific marker for microglia, was analyzed in number of VGluT1-contacting boutons with respect to ventral horn. Compared to control mice, a moderate, those on control MNs, although VGluT1 synapses on although significant, reduction in Iba1-immuno- the GTE group MNs showed a significant increased size labeling was noticed in spinal cords of GTE group (Figure 5C, 5D, 5J1–5L2). mice, but not in those of cocoa group (Figure 7A, 7B1–7D2). Aging-associated spinal cord neuro- Cholinergic C-bouton, VAChT-positive, inputs to inflammation is accompanied by structural changes MNs arise from small clusters of cholinergic in microglial cells, which exhibit increased interneurons which express the paired transcription morphological complexity and branching degree [6]. factor Pitx2 and are distributed in a longitudinal By using the skeleton approach [80], we analyzed and column near the central canal of lumbar spinal cord compare microglia morphology in aged animals of the [78]. Due to their cholinergic phenotype, these three experimental groups and found no differences interneurons, referred to as V0 C interneurons, can be between them. Thus, our flavonoid-enriched diets do www.aging-us.com 18059 AGING not to promote any overt changes in microglial phenotypes M1 and M2; whereas M1 are complexity (Supplementary Figure 5A, 5F3). proinflammatory cell types, M2 play a role as down- regulators of inflammation. We have shown that age- Microglia are plastic cells which can be activated to related reactive microgliosis in spinal cord entails an produce detrimental (cytotoxic) or beneficial imbalance in M1/M2 phenotypes, which leads to a (neuroprotective) effects on neurons [81, 82]. Thus, polarization to harmful activated M1 state [6]. To microglia can be categorized into two opposite examine whether the supplemented diets tested here Figure 5. Effects of GTE- and cocoa-supplemented diets on excitatory cholinergic (VAChT-positive) and glutamatergic (VGluT1 and VGluT2) synaptic inputs to aged spinal MNs. (A–F) Graphs show the average density (number of puncta per 100 μm of MN soma perimeter, A, C, E) and size (in μm2, B, D, F) of the different types of afferent synapses examined; the red dashed line in each graph indicates the mean value of the corresponding afferent synapse density or size found in adult mice [6]. (G1–O2) Representative confocal micrographs of VAChT, VGluT1 and VGluT2 nerve terminals contacting MN cell bodies of animals from control, GTE and cocoa groups, as indicated. Spinal cord sections were immunolabeled for either VAChT, VGluT1 or VGluT2 (green), and counterstained with fluorescent Nissl staining (blue) to visualize MN cell bodies, as indicated in panels. Data in the graphs are expressed as the mean ± SEM, **p < 0.01 and ***p < 0.001 vs. control (one-way ANOVA, Bonferroni’s post hoc test); 50-60 MNs were analyzed per animal (number of animals per group: control =3, GTE = 4; and cocoa = 5). Scale bar in O2 = 20 μm (valid for G1–O1). www.aging-us.com 18060 AGING were able to modulate this microglial phenotype Mac-2-immunolabeled (M1) microglia in GTE and activation, Iba1 immunohistochemistry was combined cocoa groups when compared with control group with either Mac-2 or macrophage mannose receptor (Figure 7E–7I4). However, in comparison with the (CD206) immunolabeling for M1 or M2 microglia, latter, we found that spinal cords of the GTE group respectively [82–84]. No overt changes were found in exhibited a significant higher density of beneficial Figure 6. GTE-supplemented diet prevents the age-related loss of V0C interneurons. (A1, A2) A general view of a spinal cord hemisection of an adult mouse immunolabeled for VAChT (red) and counterstained with fluorescent Nissl (blue) for neuron identification; the arrow points to a V0C interneuron cluster located near the central canal (delimited by a dotted line); note also the different VAChT-positive MN pools in the ventral horn. (B) Density of V0C interneurons in spinal cords of aged animals from control and GTE groups; bars represent the mean ± SEM of 3-5 animals (20-28 images) per condition; *p < 0.05 vs. control (Student’s t-test); the red line indicates the mean value in adult mice found in [6]. (C1–F2) Representative confocal micrographs of VAChT-positive V0C interneurons (red) in the spinal cords of an adult mouse (C1, C2) and of old animals fed with the control (D1, D2) and GTE-supplemented diet (E1, E2); sections were counterstained with fluorescent Nissl (blue) for neuron visualization; the central canal is delimited by dotted lines. Scale bars: in A2 = 200 µm (valid for A1), and in E2 = 100 µm (valid for C1–E1). www.aging-us.com 18061 AGING Figure 7. Impact of GTE- and cocoa-supplemented diets on the aging-associated microgliosis and imbalance in M1/M2 microglial phenotypes found in the ventral horn spinal cord of old mice. Lumbar spinal cord sections were immunostained for the microglial marker Iba1 (red), and either Mac-2 or CD206 (green), for M1 or M2 microglia, respectively; fluorescent Nissl staining (blue) was used for MN visualization. (A) Quantification of microglia expressed as the percentage of ventral horn occupied by Iba1-positive profiles. (B1–D2) Representative confocal images showing Iba1-staining around spinal cord MNs of animals from control, GTE and cocoa groups as indicated in panels. (E, F, J, K) Quantification of Mac-2-positive (E) and CD206-positive (J) profiles surrounding MNs shown as the percentage of ventral horn area occupied by the immunostained profiles; the proportion of microglial profiles expressing both Iba1 and either Mac-2 (F) or CD206 (K) is also shown. The average values of these parameters in adult mice from our previous study [6] are indicated in each graph (red lines) for comparison purposes. (G1–I4, L1–N4) Representative confocal micrographs used for data analysis showing Mac-2 (G1–I4) and CD206 (L1–N4) in combination with Iba1 and fluorescent Nissl staining, as indicated in panels. Data in the graphs are expressed as the mean ± SEM; a total of 45-50 images per experimental group were analyzed (number of animals per group: control [Ctrl] = 3, GTE = 4, cocoa = 5); *p < 0.05 and ***p < 0.001 vs. Ctrl (one- way ANOVA, Bonferroni's post hoc test). Scale bar in N4 = 50 μm (valid for B1–D1, G1–I4, L1–N3). www.aging-us.com 18062 AGING CD206-positive, M2, microglia around spinal MNs; this microgliosis observed in the spinal cord in the course of change was not found in animals that received a dietary aging is associated with a prominent increase in supplementation with cocoa flavanols (Figure 7J–7N4). activated phagocytic microglial cells. Whereas no To further study the impact of our flavonoid- changes in the density of CD68-positive microglial cells supplemented diets on modulating microglia, we were found in ventral horns of GTE-supplemented aged examine the expression of the G-protein-coupled mice, the number of those profiles surrounding MNs purinergic receptor P2Y12R in ventral horn. P2Y12R is was found significantly reduced in animals from cocoa exclusively expressed in microglia and allows to group compared to controls (Figure 9A–9E4). This distinguish these cells from infiltrating monocytes and indicates that cocoa-, but not GTE-, flavonoids are able other peripheral macrophages [85–87]. P2Y12R has to prevent the activation of microglia and their been shown to be increased during anti-inflammatory increased phagocytic activity in the course of aging. conditions and is considered as a marker for M2-like microglia [88, 89]. Spinal cord sections were double Changes in age-related alterations of DRG sensory immunolabeled for Iba1 and P2Y12R, and the number neurons induced by either GTE- or cocoa- and density of Iba1-positive profiles exhibiting P2Y12R supplementation intake positivity were analyzed in ventral horn. Compared to control spinal cords, those of animals fed with the GTE- We have reported that aging is associated with changes , but not cocoa-, supplemented diet displayed a in nociceptive (CGRP-positive and Bandeiraea significantly higher amount of Iba1-positive profiles simplicifolia lectin [isolectin B4, IB4]-positive) and also expressing P2Y12R (Figure 8A, 8C1–8E4). proprioceptive (parvalbumin [PV]-positive) DRG Additionally, we noticed that whereas in the vast neurons [6]. Thus, we have shown that L4 DRGs of majority of control microglial profiles P2Y12R aged mice exhibited increased proportions of sensory exhibited a cytoplasmic expression, in those of GTE neurons expressing CGRP and PV, and stained with group, P2Y12R-immunoreactivity is mainly located in IB4. These changes have been found to be accompanied the nucleus (Figure 8B, 8F1, 8F2), as previously by a marked atrophy of cell bodies of all those neuronal reported to occur for M2 phenotype [90]. Overall, these populations. Here, we analyzed the effects of GTE and findings indicate that age-associated reactive cocoa-flavanols on these age-related sensory neuron microgliosis can be potentially modulated to M2 neuro- alterations. We found that, compared to control group, protective phenotype by GTE. DRGs of GTE- or cocoa-supplemented animals displayed a significant decrease in the number of Activated microglia undergo changes in the expression CGRP-positive and IB4-positive neurons; additionally, of the cluster differentiation 68 (CD68) protein, cocoa, but not GTE, promoted a significant enlargement a member of the lysosomal/endosomal-associated of soma size of those neuronal populations. Moreover, membrane glycoprotein (LAMP) that regulates cocoa, but not GTE intake, also significantly reduced phagocytosis in macrophage lineage cells. CD68 is the number, although increased soma size of PV- present in lysosomes and is commonly considered a positive neurons (Supplementary Figure 6A–6O2). marker of activated phagocytic microglia [91]. Resting mouse microglia exhibit some basal expression of GTE and cocoa flavanols do not impact locomotor CD68, which is prominently upregulated after activity decline or electromyographical changes activation [92]. Double immunohistochemistry for found to occur with aging CD68 and Iba1 in spinal cords of adult and old mice revealed that virtually all CD68-positive profiles are The effects of GTE or cocoa supplementation on aging- located inside Iba1-immunolabeled cells. Interestingly, related motor decline were examined using open-field compared with adult spinal cords, those of aged mice and pen tests. In a previous study [6], we have shown exhibited a significant increase in CD68-expression in that aging results in a gradual reduction in the microglial cells located around MNs, as revealed by the locomotor activity, as measured after the automated percentage of ventral horn area occupied by CD68- recording of mouse movements using the open-field positive profiles (adult mice: 3.12 ± 0.63 %, n = 4; old test. We found that neither GTE nor cocoa mice: 18.85 ± 1.58 %, n = 5; p < 0.0001, Student’s t- supplementation improved the impaired motor test). Moreover, individual Iba1-immunolabeled performance of old mice assessed with this test; in fact, microglial cells of aged mice exhibited significant motor activity, measured by open-field test, was even larger CD68-positive puncta than those of adult animals noticed to be slightly (although not significantly) (area of CD-68-positive profiles: adult mice, 2.85 ± 0.24 reduced at the end of the experimental period in animals µm2, n = 4; old mice, 10.71 ± 0.69 µm2, n = 5; p < fed with any of the two tested supplemented diets 0.0001, Student’s t-test; 60 CD68-stained profiles per (Supplementary Figure 7A, 7B). Additionally, animal were analyzed). These data indicate that compared to control diet, neither GTE nor cocoa www.aging-us.com 18063 AGING Figure 8. Impact of GTE- and cocoa-supplemented diets on P2Y12R expression in spinal cord microglia of old mice. Sections of lumbar spinal cords from mice of different experimental groups were double immunostained for Iba1 and P2Y12R. (A, B) Quantification of Iba1- positive profiles also exhibiting P2Y12R immunoreactivity (A) and of those displaying nuclear P2Y12R expression (B). (C1–F2) Representative confocal micrographs used for data analysis showing P2Y12R (green) in combination with Iba1 (red) and fluorescent Nissl staining (blue, for MN visualization), as indicated in panels. A higher magnification of area delimited by the dashed square in D4 is shown in (F1, F2). Note the nuclear expression of P2Y12R in Iba1-positive microglial cells in close contact with a MN. Data in the graphs are expressed as the mean ± SEM; a total of 40-50 images per experimental group were analyzed (number of animals per group: control [Ctrl] = 3, GTE = 4, cocoa = 5); ***p < 0.001 vs. Ctrl (one-way ANOVA, Bonferroni's post hoc test). Scale bar: 50 μm in (E4) (valid for C1–E3) and 10 µm in (F2) (valid for F1). www.aging-us.com 18064 AGING supplementation modified the parameters of pen test excitability of the H reflex circuit was not significantly performance observed in old mice (not shown). different, although the cocoa group had a lower H/M amplitude ratio closer to normal young adult values. It We have previously demonstrated that aging in mice is of interest to note that the motor unit number is accompanied by noticeable electrophysiological estimation (MUNE) test gave a slight increase in the changes in form of significant reduction in nerve number of motor units in the TA muscle of both conduction velocity and compound muscle action supplement groups compared with the control old potential (CMAP) amplitude and, also, by a mild group, which is in agreement with the results of NMJ decrease in functional motor unit number [6]. We innervation detailed above. examined whether GTE or cocoa supplemented diets were able to improve the age-related alterations in such DISCUSSION electrophysiological parameters. Neither green tea extract nor cocoa intake induced any significant change To evaluate nutritional ingredients that can mitigate in either the CMAP latency or CMAP amplitude found sarcopenia progression, we tested the effects of two- in old animals (Supplementary Table 1). The flavonoid enriched plant extracts, green tea catechins or Figure 9. Impact of GTE- and cocoa-supplemented diets on microglial activation in ventral horn spinal cord of old mice. Sections of lumbar spinal cords from mice of different experimental groups were double immunostained for Iba1 and CD68, a marker of activated phagocytic microglia. (A) Quantification of CD68-positive profiles around MNs in control, GTE and cocoa groups. (B–E4) Representative confocal micrographs used for data analysis showing CD68 (green) in combination with Iba1 (red) and fluorescent Nissl staining (blue, for MN visualization), as indicated in panels. A higher magnification of area delimited by the dashed square in C4 is shown in (B). Data in the graph are expressed as the mean ± SEM; a total of 40-50 images per experimental group were analyzed (number of animals per group: control [Ctrl] = 3, GTE = 4, cocoa = 5). *p < 0.05 vs. Ctrl (one-way ANOVA, Bonferroni's post hoc test). Scale bar: 10 μm in (C) and 50 µm in (E4) (valid for C1–E3). www.aging-us.com 18065 AGING cocoa flavanols. Besides their potential benefits on aged and density, or connective tissue content of aged muscles. muscles, the impact of these dietary supplements on Nevertheless, both dietary supplements significantly: a) aging-associated alterations in other constituents of the prevented the age-associated decline of SC numbers [6, neuromuscular system was also examined. Our results 93, 94]; b) reduced the density of fibers with central indicate that GTE- and cocoa-flavonoid intake may nuclei (taken as an indicator of a degenerative/ ameliorate some aging-related neuromuscular regressive regenerative process of muscles [6, 11, 49, 95]); and c) phenomena occurring in mice by: a) preventing MN attenuated the accumulation of the age pigment deafferentation; b) elevating the proportion of anti- lipofuscin in myofibers [6]. SCs are essential for skeletal inflammatory microglial phenotypes, found to be muscle regeneration [96–98], and the age-associated reduced in the aged spinal cord; c) improving the degree decline in number and function of SCs negatively affects of muscle innervation and NMJ maturity; and d) the capacity to replace and/or repair damaged myofibers delaying some potential harmful aging-related changes (reviewed by [53, 99]). However, although SCs are in skeletal muscles. crucial for the genesis of new fibers during the regenerative process of the aged muscle, they do not Although GTE or cocoa supplementations are able to seem to have any impact on fiber size during aging [100]. increase the survival rate of mice, we did not notice a This is in agreement with the absence of significant significant impact on the motor activity decline or on the changes in the cross-sectional area of myofibers we explored electrophysiological alterations associated with found in flavonoid-supplemented animals, despite the senescence. These results contrast with those from increased proportion of SCs noticed in muscles. Previous previous studies showing benefits of GTE catechins or studies have shown that caloric restriction is able to cocoa flavanols on physical activity and muscle function preclude the age-related loss of myofibers [11] and to in rat and mouse models of muscle dysfunction. In aged enhance SC function and muscle repair [20]. This rats, GTE has been reported to improve muscle function indicates that metabolic factors are critical in the after reloading following disuse [36]. In a murine model regulation of myofiber turnover and myogenic activity of of accelerating aging, GTE catechins along with habitual SCs. The beneficial effects of caloric restriction on aged exercise have been shown to suppress the aging-related muscles appear to be associated to the enhancement of decline of physical performance [34]. Other studies have mitochondrial biogenesis and oxygen consumption revealed that GTE and its major catechin EGCg improve resulting from this dietary intervention [20]. On the other muscle function and endurance capacity in mouse models hand, EGCg has been reported to significantly promote of Duchenne muscular dystrophy [30, 31]. The disparity SC proliferation and differentiation in muscles of between these and our results could be explained by senescent rats after muscle disuse recovery [35, 36]. This differences in the animal models, and tests used to action, which appears to be mainly mediated by oxidative evaluate motor behavior and muscle function. Moreover, stress inhibition, is able to attenuate the muscle mass loss it is reasonable to assume that the tests used in the present but fails to improve muscle recovery during reloading study to assess motor phenotype of mice, although allow following inactivity [36]. EGCg has also been shown to us to detect an age-related decline in muscle function [6] promote SC activation in aged mice and humans, are not sensitive enough to detect subtle improvements in although this effect has been mainly ascribed to the motor activity promoted by the dietary intervention. On reduction of myostatin levels induced by the compound the other hand, a previous study has shown that the cocoa [101]. Here, we confirm the efficacy of GTE in flavanol-epicatechin could improve the age-associated maintaining the SC population in aged mice and also decline of physical activity in mice [46]. However, this provide evidence that cocoa flavanols have a similar intervention was carried out for a longer length of time (~ impact on preventing age-related SC loss. The benefits 9 months) than our study, and the dose of epicatechin is of these flavonoids on aged muscles could be a much higher than that delivered in the present work. It is consequence of the enhancement and acceleration of possible that higher doses of cocoa flavanols and longer myofiber turnover and repair, which would be intervention periods may be needed to impact motor facilitated by the increased SC proliferation, as reported output. Nevertheless, our study identified positive previously for green tea catechins [99]. Nevertheless, structural changes in the neuromuscular system as well as the possibility that our dietary supplements promote a in muscle regenerative and energy pathways, a harbinger beneficial action by slowing down the aging process of of positive impact of our tested compounds on the muscle. muscles and, in this way, diminishing the number of SCs used for muscle regeneration, cannot be excluded. Effects of green tea- and cocoa-enriched diets in Indeed, aged muscles of animals that received the skeletal muscles of aged mice flavonoids, exhibited a significant increase in the proportion of apparent younger myofibers, with reduced In the present study, we did not observe an effect of GTE lipofuscin deposition and decreased frequency of or cocoa flavanol supplementations on mass, fiber size centrally located nuclei. www.aging-us.com 18066 AGING Due to the reported antioxidant activities of flavonoids Effects of green tea- and cocoa-enriched diets in [22, 43, 102], it is tempting to ascribe the muscular neuromuscular junctions of aged mice benefits of GTE and cocoa flavanols to their protective effects against reactive oxygen species. Reactive Aging is accompanied by structural and molecular oxygen species are by-products of mitochondrial changes at NMJs (reviewed by [5]). Although it is not oxidative phosphorylation, which are increased in aging yet clear whether these are a consequence of muscles due to mitochondrial dysfunction [60]. On the progressive aged-related alterations in either MNs or other hand, mitochondria play a crucial role in the muscle fibers, it has been unambiguously demonstrated regulation of apoptosis in the senescent muscle [103, that an important proportion of NMJs in murine models 104]. Thus, age-related mitochondrial dysregulation of aging exhibit signs of partial or complete appears to be a critical factor contributing to muscle denervation, and changes compatible with structural wasting in the elderly [105, 106]. PGC-1α is considered remodeling [6, 9, 11, 63–65]. Here we found that some an essential transcription factor in the mitochondrial alterations in aged NMJs of mice could be mitigated by biogenesis and function [57], and its overexpression has any of the two flavonoids-enriched diets tested. Indeed, been reported to increase mitochondrial mass [107]. compared to controls, muscles from animals fed with Additionally, PGC-1α is also critical in regulating either GTE- or cocoa flavanol-supplements displayed a numerous genes involved in muscle morphology and significant reduction in the proportion of denervated or function throughout its interactions with signaling partially innervated NMJs. Moreover, we also noticed pathways that control mitochondria homeostasis and that muscles of GTE and cocoa groups showed a protein balance [108]. PGC-1α content and activity have significant decline in the number of fragmented been shown to be reduced in aged muscles, although endplates, and increased amounts of them exhibiting a these findings are not free from certain controversy pretzel-like appearance, indicative of a higher degree of (reviewed by [108]). Redox regulation is an important NMJ maturity. Nevertheless, it is necessary to take into mechanism in controlling PGC-1α activity. Indeed, account that, compared to mouse NMJs, those from reactive oxygen species could negatively regulate PGC- humans are structurally more stable across the entire 1α in aged skeletal muscle, as reported in cardiac adult lifespan, showing scarce signs of degeneration or myocytes [109]. On the other hand, there is evidence remodeling with age [116]. This aspect must be taken that PGC-1α is able to activate the expression of into consideration regarding the applicability to our endogenous antioxidant proteins (including superoxide results to humans. dismutase and glutathione peroxidase), exert anti- inflammatory effects, promote neovascularization, and These improvements promoted by the dietary protect the senescent muscle from increased degradative supplements in age-related NMJ alterations were not processes such as proteolysis, autophagy and apoptosis accompanied by significant changes in the expression of [110–112]. Chronic physical training increases the CGRP, GAP-43, or agrin, three developmentally levels of PGC-1α in the skeletal muscle [113–115]. regulated molecules involved in NMJ development, Here, we found that animals fed with either GTE- or maintenance and plasticity. All them are synthetized in cocoa flavanols exhibited increased muscular content of MN cell bodies, axonally transported and released at PGC-1α compared to controls. The augmented levels of nerve terminals. CGRP is a neuropeptide which is PGC-1α after either of the two diets tested were highly expressed in NMJs during development, observed in both TA (fast) and Sol (slow) muscles, decreases its content with synaptic maturation, but is although the effects on this transcription factor upregulated during nerve sprouting and muscle expression were slightly more prominent with GTE than reinnervation [66–69]. Agrin, throughout the clustering with cocoa supplementation. In fact, we did not find of AChRs, plays an essential role in the initiation of statistically significant changes in the muscular content NMJ formation, and later in the maintenance and of PGC-1α between control and cocoa group, although regeneration of adult NMJs [70]. GAP-43 is expressed latter exhibited a remarkable increased expression of the in growth cones, and its content is increased during transcription factor in both TA and Sol muscles. As motor axonal growth and plasticity, and in regeneration aforementioned, reductions in PGC-1α expression following axonal injury [71]. We have previously impairs mitochondrial dynamics, structure and function reported that the content of these synaptic proteins in muscles [105]. Although a number of metabolic decreases in the course of the postnatal development, regulators besides PGC-1α could be involved in the with NMJs of adult mice exhibiting significantly lower maintenance of mitochondrial homeostasis, our results protein levels than those of young animals; suggest that the beneficial effect of GTE and cocoa nevertheless, their expression increases again in aged supplementation on mitochondrial content in aged NMJs, suggesting a role for these molecules in the muscles could be due, at least in part, to their capacity promotion of compensatory nerve sprouting and of restoring PGC-1α levels. reinnervation of denervated muscles in aging [6]. We www.aging-us.com 18067 AGING observed that, GTE- and cocoa flavanols only promoted degree of intrinsic excitability which confers to slow minor increases in the levels of CGRP and agrin MNs a higher resistance to degeneration compared to regularly found in aged NMJs. Additionally, no fast fatigable ones. Our results indicate a differential significant differences were observed in GAP-43 effect of GTE and cocoa supplementation in the expression at the NMJs of flavonoid-supplemented preservation of excitatory synaptic afferents on aged animals, although moderate reductions of this protein MNs. GTE-, but not cocoa-, flavonoids promoted were noticed when determined by western blot in the protection against cholinergic C-bouton- and total muscle. Overall, these results indicate that, at least glutamatergic VGluT2 synapse-deafferentation. at the concentrations and dietary regimens used here, However cocoa flavanols, but not GTE, reduced the our supplements are not able to significantly modulate age-associated decline of glutamatergic VGluT1 the expression of these proteins linked to NMJ synaptic puncta on MNs. These results suggest the need maintenance. The modest decline in GAP-43 content to combine distinct flavonoids in order to achieve a found in muscles of GTE and cocoa groups correlates in wider effectiveness on MN afferent preservation during some way with the reduction of nerve terminal aging. sprouting observed in NMJs following dietary supplementations and suggest that, in those conditions In an attempt to explore further the cellular targets of in which aged NMJs are more stable, intramuscular these flavonoids linked to their preventive benefits on nerves have a diminished need to sprout for maintaining age-associated MN deafferentation, we examined proper levels of innervation. potential changes in the neuronal sources of C-boutons, VGluT1 and VGluT2 synapses contacting spinal MNs. In our knowledge, this is the first study reporting the GTE significantly increased the number of V0C attenuation of age-associated alterations in NMJs interneurons (the source of motoneuronal C-boutons), promoted by flavonoid-enriched diets. which were depleted in control aged spinal cords. This suggests that the attenuation of aging-related C-bouton Effects of green tea- and cocoa-enriched diets in loss is, in part, a consequence of the flavonoid action on preserving afferent synaptic inputs to motoneurons the maintenance of V0C interneurons during senescence. On the other hand, a substantial number of VGluT1 An interesting finding of our study is the prevention of synaptic terminals on MNs arise from DRG age-associated MN deafferentation [6, 76, 77] observed, proprioceptive (PV-positive) sensory neurons that particularly, following GTE intake. As we have peripherally innervate muscle spindles, stretch-sensitive previously reported, aging is accompanied by a mechanoreceptors embedded in skeletal muscles. These prominent loss of excitatory cholinergic (C-boutons) VGluT1 proprioceptive terminals, named Ia afferents, and glutamatergic synapses on MNs [6], a pathological establish highly selective excitatory monosynaptic change also observed in different murine models of MN connections with spinal MNs, and are essential for the diseases, such as spinal muscular atrophy (SMA) [117– muscle stretch reflex and, consequently for posture and 121] and amyotrophic lateral sclerosis (ALS) [117, motor coordination [133]. Our finding that cocoa, but 122–125], and after nerve injury [126–128]. It is well not GTE, are able to remarkably ameliorate aging- known that MN excitability is accurately regulated by related changes found in proprioceptive PV-positive afferent synaptic inputs which mediate the genesis of DRG neurons correlates with the effects of this coordinated firing patterns resulting in the appropriate flavonoid on preventing VGluT1 afferent synaptic loss muscle contraction during motor activities [129]. Since on aged MNs. This suggests that the improvement the control of MN excitability appears to be crucial provided by cocoa flavanols on aging-associated for motor function, it seems clear that age-associated glutamatergic MN deafferentation could be mediated, at MN deafferentation can markedly alter the normal least in part, by their actions on DRG sensory neurons. electrophysiological patterns and impair motor skills of aged mice, as we previously reported [6]. In this regard, Despite the preventive action of these flavonoids on cholinergic C-bouton synapses appear to be particularly age-related MN deafferentation, this effect was not important in the control of motor activity during associated to any noticeable improvement in motor locomotion, being involved in the task dependent axon degeneration that occurs with age [6, 72]. This regulation of motor outputs [130]. C-boutons also could contribute to explain the lack of improvement of determine the existing differences in intrinsic cellular the electromyographical alterations observed in aged excitability between MN subtypes innervating slow- or animals [6]. Our results also suggest that preventing fast-twitch muscles [131, 132]. Thus, MNs innervating MN deafferentiation is not enough to achieve the slow-twitch muscles (slow MNs) have a higher appropriate morphofunctional state of MNs necessary excitability and slower firing than MNs innervating for maintaining or ameliorating the impaired motor fast-twitch muscles (fast-fatigable MNs), and it is that activity. On the other hand, more studies are required to www.aging-us.com 18068 AGING examine whether the preserved synaptic afferents are [128]. We do not know, however, whether microglial able to mediate a proper neuromodulatory action for an activation in aging is a primary event resulting in the accurate MN firing and adjust muscle activation to elimination of normal functional nerve afferents on demands of different motor tasks. MNs with the consequent neuronal impairment, or a secondary process occurring in response to a previous Effects of green tea- and cocoa-enriched diets in altered synaptic function. Nevertheless, as we show preventing age-associated microgliosis here, aging-associated imbalance in M1/M2 phenotypes may be partially modulated by a dietary Interestingly, GTE supplementation induced intervention based on GTE. In fact, we found that a conspicuous changes in aging-associated spinal cord diet enriched with GTE, but not cocoa flavanols, not microgliosis. We have previously shown that age is only significantly reduces the density of microglial accompanied by a neuroinflammatory reaction in profiles, but also increases the proportion of ventral horn, which is particularly prominent around neuroprotective (M2) microglia in the ventral horn MNs [6]. Neuroinflammation in aged spinal cord spinal cord of aged mice. This was placed in evidence could be considered as a part of a more general by the increased proportion of microglial profiles process chronically affecting different regions of the displaying CD206 and P2Y12R, two molecules central nervous system, as reported in healthy elderly associated with the M2 phenotype, [82, 88, 89], in [134, 135]. The fact that neuroinflammation has been spinal cords of GTE-supplemented mice. The elevated found to be also associated to a number of presence of P2Y12R-positive cells with GTE neurodegenerative disorders, including the MN supplementation is particularly interesting, since the diseases ALS and SMA [118–120, 136–139], suggests purinergic receptor P2Y12 has been shown to be that age-related gliosis can enhance the neuronal exclusively expressed in microglia and not in other vulnerability to dysfunction and degeneration. Gliosis infiltrating myeloid cells, and appears to be critical for in aged spinal cord entails a marked increase in microglial chemotaxis and extension of processes in harmful microglial (M1) and astroglial (A1) response to injury [141]. In normal conditions, phenotypes, and a depletion of neuroprotective M2 P2Y12R-positive microglia is a relative stable cell microglia and A2 astroglia [6]. Activation of population throughout the lifespan, and this stability proinflammatory M1 and A1 cells in spinal cord seems to be important to ensure the homeostatic during aging could exert detrimental actions on MNs functions of central nervous system during aging that may negatively affect their structure and function, [142]. P2Y12R expression has been reported to with the subsequent impairment of NMJs. In this decline in hippocampal demyelinating lesions of aged regard, it has been recently shown that pharma- mice [143] and in multiple sclerosis patients [142], cological reduction of microglia is able to maintain but increases in human brain after anti-inflammatory the NMJ integrity during normal aging [140]. This stimuli [88, 89]. In line with these results and in indicates that microglial activation is a critical step in agreement with the anti-inflammatory and anti- the pathogenic changes leading to neuromuscular oxidative actions of GTE, we found that P2Y12R- decline during aging. However, the cellular positive, neuroprotective M2, microglia significantly mechanisms by which microglia exert their increases after chronic intake of GTE and translocate detrimental actions in the aged neuromuscular system, P2Y12R to the nucleus. The nuclear expression of this and how we can modulate the microglial reaction in purinergic receptor in microglia following GTE aging are not clarified. A line of evidence assigns to supplementation is difficult to conciliate with the role microglia an active role in synapse elimination during of a G-protein coupled receptor such as P2Y12R, neurodegeneration, as occurs in ALS, and after considered as a cell-surface receptor whose primary peripheral nerve injury [79, 128]. Therefore, age- function is to transduce extracellular stimuli into related changes in the microglial cell state could be intracellular signals [144]. Nevertheless, our results responsible for the synaptic loss observed on aged are in concordance with previous data on microglial MNs. Thus, it is tempting to assume that aging entails activation in gliomas, in which it has been reported a an altered microenvironment in the spinal cord, which shift from cytoplasmic to nuclear expression of triggers a microglial reaction leading to central P2Y12R linked to the acquisition of the M2 synapse elimination. Indeed, in a model of sciatic phenotype [90]. The functional consequences of this nerve axotomy, we have recently demonstrated that nuclear translocation of P2Y12R are not unraveled microglia are recruited by injured MNs and displayed yet. In any case, the increase of neuroprotective M2 positive chemotaxis to afferent synapses; activated microglia we found correlates with the prevention of microglia stablish close contacts with degenerating MN deafferentation observed in aged animals afferent terminals, suggesting a role for those cells in subjected to dietary supplementation. However, more the disintegration of defective synapses on MNs work is necessary to analyze whether GTE- intake www.aging-us.com 18069 AGING results in a direct action on microglia, dampening “youthful” cellular phenotype of myofibers, the their anti-inflammatory phenotype or, conversely, apparent reduction of myofiber degeneration/ flavonoids have their primary effect stabilizing MN regeneration cycles, the preservation of the myogenic afferent inputs, this resulting in the preclusion of SC population, and the increased expression of PGC-1α. microglial reaction. Additionally, GTE- and cocoa-flavonoids differentially promoted the stability of excitatory synaptic inputs to The effects of supplementation with cocoa flavanols on MN, alleviate microgliosis and modulate the balance age-associated microglia activation found here are between proinflammatory (M1) and neuroprotective striking. Although cocoa did not reduce microgliosis in (M2) microglial phenotypes. The differential effects of ventral horn of aged animals, it significantly decreased flavonoids from GTE and cocoa on changes occurring the density of activated/phagocytic, CD68-positve, in the spinal cord with age, suggest that there might be a microglia located close to MNs. Microglia activation benefit to combine different flavonoids to counteract the appears to be necessary for the removal of VGluT1 negative effects of aging in the neuromuscular system. synapses on MNs, as reported following peripheral Furthermore, the association of flavonoid supplemen- nerve injury [145]. In this regard, we have previously tations with regular physical exercise must be taken in shown that microglia expressing CD68 are involved in consideration in the therapy of sarcopenia. Future the phagocytosis of VGluT1 nerve terminals in ventral research is needed to investigate whether higher doses horn of SMA mice [118]. Mechanisms underlying the of flavonoids are needed and/or longer-term elimination of afferent nerve terminals contacting MNs interventions can help restore proper motor function. by microglia in pathological conditions are still not well In any case, the results shown here provide insight into clarified; it seems likely, however, that not all types of the use of flavonoids as nutritional interventions for synaptic contacts on MNs are eliminated in the same the treatment of aging-related neuromuscular way by microglia. Thus, it is tempting to speculate that, alterations that occur with sarcopenia. Our data during the neuroinflammatory response occurring in the should be also considered in the context of the dietary ventral horn with aging, the removal of VGluT1 nerve management of patients affected by muscular and MN terminals contacting MNs requires specific mechanisms diseases. different from those operating to eliminate VAChT or VGluT2 synapses. In fact, in a previous work, we have MATERIALS AND METHODS reported that after peripheral nerve transection, activated microglial cells are recruited to injured MNs, Animals and dietary regimens with a predominant positive chemoattraction to VGluT1 and less to VAChT synapses, but with minor pre- Male C57BL/6JRj mice were obtained from Janvier ferences for contacting other synapse types, such Labs (Saint Berthevin, France), housed in a standard as GABAergic and serotoninergic boutons [146]. cage (3 animals per cage) containing a nest box Accordingly, it is plausible that activated microglia (Smart-House, ZOONLAB GmbH, Castrop-Rauxel, remove VGluT1 synapses more easily and in a higher Germany), and maintained in a 12-h light/dark cycle, extent than VAChT and VGluT2 boutons. This would and controlled room temperature (20 ± 2° C) and explain the differential effect on preventing MN humidity (50%). A total of 45 mice (22 month-old) deafferentation observed with GTE- and cocoa- were randomly distributed in 3 groups (n = 15 per supplementation and why dampening microglial group) and fed with either: a standard AIN-93M diet activation (indicated by the reduced density of CD68- (control group), the AIN-93M diet supplemented with positive profiles) in areas surrounding MNs following 0.5% GTE (Suphenon 90D, Taiyo, GTE group), or the cocoa supplementation precludes VGluT1, but not AIN-93M diet supplemented with 0.133% high- VAChT and VGluT2, synapse loss. flavanol cocoa powder (Acticoa, Barry Callebaut, cocoa group). The GTE group was designed to receive CONCLUSIONS an average dose of ~200 mg EGCg/Kg body weight/d. The cocoa group was designed to receive an average In summary, dietary intake of flavonoids from green tea dose was ~13.33 mg flavanols/kg body weight/d. or cocoa was able to significantly increase the survival Diets started at week 92 (21 months) and were rate of aged mice and to prevent some regressive maintained until week 116 (27 months). Daily food structural changes occurring with senescence in distinct intake was measured across groups. Animals were cellular components of the neuromuscular system allowed ad libitum access to drinking water and to (Figure 10). Both diets clearly preserved NMJ either the control, or GTE- or cocoa-supplemented innervation and maturity, delayed the senescence diets. Feed intake was estimated by subtracting the process of the skeletal muscle, and enhanced its weight of the remaining food from that of the food regenerative capacity, as inferred from the more administered the week before. For some analyses, www.aging-us.com 18070 AGING Figure 10. Overview of main benefits promoted by GTE and cocoa dietary supplementations on aging-associated changes in the neuromuscular system of C57BL/6JRj mice. The hallmark neuromuscular alterations occurring in these mice in the course of aging [6] are also summarized. (A) NMJs of aged mice display signs of either denervation or polyinnervation, and endplate fragmentation, suggesting an active process of NMJ remodeling and muscle reinnervation. Additionally, aged muscles show increased fibrosis, abundant fibers with lipofuscin accumulation and centrally located nuclei (indicative of muscle regeneration), and a marked reduction in the proportion of SCs. Aged mouse spinal cords exhibit reactive gliosis in ventral horn with increased proportion of harmful M1 microglia and significant loss of excitatory cholinergic (C-boutons) and glutamatergic synapses on MNs; atrophy of sensory proprioceptive (PV-positive) DRG neurons was also seen. (B) GTE and cocoa supplementations significantly decrease muscle denervation and signs of NMJ degeneration; both supplements augment the proportion of NMJs exhibiting single innervation, reduce fragmentation of endplates and increase the number of them exhibiting a healthier, “pretzel-like”, appearance. Furthermore, GTE- and cocoa-enriched diets increase the density of satellite cells, and reduce lipofuscin deposition in myofibers and the proportion of them displaying central nuclei. PGC-1α, a key regulatory factor of mitochondrial biogenesis, shows increased muscular levels in animals fed with GTE and cocoa-supplemented diets. GTE-, but not cocoa-, supplementation prevents the aging-associated loss of cholinergic (C-bouton) and VGluT2-positive glutamatergic synapses on lumbar spinal cord MNs; cocoa, but not GTE, increases the density of VGluT1-positive glutamatergic nerve terminals contacting MNs. The prevention of age-related C-bouton loss promoted by GTE is associated with increased numbers of V0C interneurons, the neuronal origin of cholinergic C- www.aging-us.com 18071 AGING bouton inputs to MNs. Additionally, the prevention of aging-associated loss of VGluT1-positive MN-afferents by cocoa is accompanied by the increased body size of PV-positive proprioceptive DRG neurons, the source of Ia VGluT1 afferents to MNs. Moreover, GTE-supplementation improves age-related reactive microgliosis in the spinal cord and increases the proportion of neuroprotective M2 microglial cells around MNs, indicating that the imbalance of M1/M2 microglia found to occur with aging can be potentially modulated by GTE. Created with BioRender.com. adult (4 month-old) C57BL/6JRj mice, feed with the and plantar interossei (PL) muscles by using standard diet, were also used. microneedle electrodes [150]. The H/M ratio was calculated as the quotient of the maximal H wave and All animal handling and experimentation procedures maximal M wave amplitude for each recorded muscle. were conducted in accordance with the ethical standards For sensory nerve conduction tests, the recording laid down in the 1964 Declaration of Helsinki and its electrodes were placed near the digital nerves of the later amendments, and the European Communities fourth toe, recording the compound sensory nerve Council Directive 2010/63/EU for the care and use of action potential (SNAP) following stimulation of the laboratory animals and the norms established by the sciatic nerve as above. Signals were amplified and Generalitat de Catalunya (published in the Diari Oficial displayed on a digital oscilloscope (Tektronix 450S, de la Generalitat de Catalunya [DOGC] 2073, 1995). Tektronix, Beaverton, OR, USA) at appropriate settings All experiments were previously evaluated and to measure the amplitude from baseline to the maximal approved by the Committee for Animal Care and Use of negative peak, and the latency from stimulus to the the Universitat de Lleida, and the Committee for Ethics maximal negative peak. To ensure reproducibility, the on Experimental Animal and Human Research of the recording needles were placed under microscope guided Universitat Autònoma de Barcelona. All efforts were by anatomical landmarks to secure the same placement made to minimize suffering and reduce the number of on all animals. During the tests, the mice body animals in agreement with the European Communities temperature was maintained constant between 34-36° C Council Directive (November 24, 1986; 86/609/EEC). by means of a thermostat-controlled heating pad. According to previous defined criteria [147, 148], mice displaying tumors, physical abnormalities and/or MUNE test was performed in the TA muscle, using the evidence of disease were excluded from the study and same setting as for the motor nerve conduction test. The euthanized by an overdose of pentobarbital (30 mg, protocol used was based on the incremental technique intraperitoneally). [151]. Starting from subthreshold intensity, the sciatic nerve was stimulated with single pulses of gradual Motor behavior tests increased intensity until the first CMAP response appeared, representing the first motor unit recruited. To evaluate the impact of different diets on motor With the next stimuli, quantal increases in the response skills, mice were periodically weighed and carefully were recorded. Increments >50 μV were considered as examined. Open-field (for assessing locomotor the recruitment of an additional motor unit. The activity), and pen (balance beam, for assessing motor amplitude of a single motor unit action potential coordination and balance) tests were subsequently (SMUA) was calculated as the mean of more than 15 performed every 4 weeks according to previously consistent increases. The estimated number of motor described guidelines [149]. Tests were conducted by the units resulted from the equation: MUNE = CMAP same investigator in a blinded manner. In the pen test, maximal amplitude/mean amplitude of SMUA. every mouse was evaluated 3 times, with a 15-minutes recovery period between them; the value obtained in the Tissue sample collection and histological analysis best test performance was used as the final score. Open- field test was performed by the automated recording of Mice were anaesthetized and transcardially perfused mouse movements using Smart Video Tracking with 4% paraformaldehyde (PFA) in 0.1 M phosphate software (v2.5.21, Panlab Harvard Apparatus, Holliston, buffer (PB), pH 7.4. After animal perfusion, spinal MA, USA); different parameters such as time, distance, cords, L4 DRGs, L4 VRs, and entire individual entries in zones, and average speed were measured. hindlimb skeletal muscles were rapidly dissected from each mouse. VRs were then immersed in 1% PFA and Electrophysiology tests 1% glutaraldehyde in 0.1 M PB (pH 7.4) for 24 hours. For motor nerve conduction tests, the sciatic nerve was Muscle samples were processed for histology and stimulated percutaneously by means of single pulses of further morphometry according to previously described 25 µs duration delivered through a pair of needle procedures [152]. Muscles examined were the TA, a electrodes placed at the sciatic notch. The CMAP (M fast-twitch muscle, which helps with dorsiflexion and wave) and the H-reflex wave were recorded from TA inversion of the foot, and the Sol, a relative slow-twitch www.aging-us.com 18072 AGING muscle in mice, which plays an important function in were incubated with Alexa Fluor 555-conjugated α- maintaining the standing posture and is also used for Bgtx (1:500; Molecular Probes) to identify postsynaptic plantarflexion during walking [153]; to examine acetylcholine receptors. Some sections were also whether some changes also occurred in a more proximal stained with 4’,6-diamidino-2-phenylindole dihydro- hindlimb muscle, the fast-twitch Gra muscle, which is chloride (DAPI, 50 ng/ml, Molecular Probes) for DNA mainly involved in the adduction of the limb and knee staining. To avoid lipofuscin-like autofluorescence, flexion [153], was also included in some experiments. slides with sections of spinal cord from old mice were Muscles were cleaned, cleared of the excess of treated with the Autofluorescent Eliminator Reagent connective tissue, blotted dry and weighed. After this, (Merck, Kenilworth, NJ, USA), following incubation they were post-fixed in 4% PFA in 0.1 M PB (pH 7.4), with the secondary antibodies. Some slides with DRG cryoprotected with 30% sucrose in 0.1 M PB, embedded were stained with FITC-conjugated IB4 (1:25; Sigma- in Tissue Freezing Medium (Triangle Biomedical Aldrich). Sciences, Durham, NC) and frozen. Some cryostat transverse sections (16-µm thick), obtained from the After washing, slides were coverslipped using an anti- mid-belly of the muscle, were stained with hematoxylin fading medium containing 0.1 M Tris-HCl buffer (pH and eosin for histological examination. 8.5), 20% glycerol, 10% Mowiol, and 0.1% 1,4- diazabicyclo[2,2,2]octane. Slides were examined with VRs were post-fixed in 1% osmium tetroxide, and an Olympus BX51 epifluorescence microscope embedded in Embed 812 (Electron Microscopy (Olympus, Hamburg, Germany) equipped with a Sciences, Fort Washington, PA) epoxy resin, following DP30BW camera, or with FluoView 500 or FluoView standard procedures. Semithin transversal sections (1- 1000 Olympus laser scanning confocal microscopes. μm thick) were stained with methylene blue and imaged For comparisons, slides from different animals and using an Olympus 60x/1.4NA PlanApo oil immersion experimental conditions were processed in parallel for objective (Olympus) and a DMX 1200 Nikon (Tokyo, immunohistochemistry and subsequent imaging. The Japan) digital camera. same scanning parameters were used for the acquisition of images corresponding to different experimental Immunohistochemistry and imaging groups. For spinal cord immunohistochemical studies, digital images of the entire lumbar region were obtained For immunohistochemical analysis, lumbar spinal from every 30th section. cords, L4 DRGs and skeletal muscles were post-fixed by immersion in 4% PFA in 0.1M PB (pH 7.4), either Image analysis and morphometry overnight (for spinal cords and DRGs) or for 2 hours (for muscles), and cryoprotected. Tissue samples were Image and morphometric examinations were performed embedded in tissue freezing medium and frozen. by two independent investigators. Digital images were Longitudinal (16-µm thick, for muscles) and transverse analyzed with ImageJ software (US National Institutes (16-µm thick, for muscles, and 14-µm thick, for spinal of Health, Bethesda, MD, USA). cords and DRGs) serial cryostat sections were obtained and stored at -80° C. Myofiber cell counts and size measurements were performed in a single random image per muscle taken Tissue sections were sequentially rinsed in phosphate- from its mid-belly and immunostained for laminin. The buffered saline (PBS) containing 0.1% Triton X-100 for proportion of myofibers showing central nuclei was also 30 min, blocked in normal goat serum or normal horse counted on images from DAPI-stained sections. The serum, and subsequently incubated with the chosen percentage of myofibers showing lipofuscin aggregates, primary antibody. The primary antibodies used are the number and the area of these aggregates, and the indicated in Table 1. area and intensity of ATP5A immunostaining, used as mitochondrial marker, were also measured by ImageJ. After incubation with the primary antibody, sections The number of DAPI-stained nuclei exhibiting Pax7- were washed with PBS and incubated for 1 hour at positive immunolabeling, as a marker of satellite cells, room temperature (RT) with the appropriate secondary was also counted in confocal images of muscle sections. antibodies labeled with one of the following fluorochromes (1:500): Alexa Fluor 488, Alexa Fluor The cytoarchitecture of the NMJs was assessed in 546, (Molecular Probes, Eugene, OR, USA), Cy3, or longitudinal immunostained sections of muscles. Z- Cy5 (Jackson Immuno Research Laboratories, West stack optical sections (1-µm thick) were obtained with Grove, PA, USA). The spinal cord sections were finally the confocal microscope and projected to reconstruct counterstained with blue fluorescent NeuroTrace Nissl NMJs; maximum intensity projections of stacks were staining (1:100; Molecular Probes). Muscle sections created using the microscope software. Three-five www.aging-us.com 18073 AGING Table 1. Primary antibodies used for immunohistochemistry. Target Abbreviation Host species Source /catalog no. Dilution Millipore (Temecula, CA) / Agrin Agrin mouse monoclonal 1:100 MAB5204 Abcam (Cambridge, UK) / ATP5A ATP5A mouse monoclonal 1:100 ab14748 Sigma-Aldrich (Madrid, Calcitonin gene-related peptide CGRP rabbit polyclonal 1:1000 Spain) / C8198 R&D Systems (Minneapolis, Macrophage Mannose Receptor MMR/CD206 goat polyclonal 1:100 MN) / AF2535 Bio-Rad Laboratories Inc. Cluster differentiation 68 CD68 rat monoclonal 1:100 (Hercules, CA) / MCA1957 Santa Cruz Biotechnology Growth associated protein 43 (H-100) GAP-43 rabbit polyclonal 1:50 (Dallas, TX) / sc-10786 Wako Pure Chemical Ionized calcium-binding adaptor molecule 1 Iba 1 rabbit polyclonal Industries Ltd. (Osaka, Japan) 1:500 / 019-19741 Ionized calcium-binding adaptor molecule 1 Iba 1 goat polyclonal Abcam / ab5076 1:500 Laminin 2 Laminin 2 rat monoclonal Sigma-Aldrich / L0663 1:100 Cedarlane (Burlington, Mac-2/Galectin 3 Mac-2 rat monoclonal 1:800 Canada) / CL8942AP Neurofilament 68 KDa NF chicken polyclonal Abcam / ab72997 1:1000 Biolegend (San Diego, CA) / G-protein-coupled purinergic receptor P2Y12R P2Y12R rat monoclonal 1:100 848001 Paired box protein 7 Pax7 mouse monoclonal R&D Systems / MAB1675 1:50 Swant (Marly, Switzerland) / Parvalbumin PV mouse monoclonal 1:2000 PV235 Synaptic vesicle protein 2 SV2 mouse monoclonal DSHB / SV2 1:1000 Synaptic Systems (Gottingen, Vesicular acetylcholine transporter VAChT guinea pig polyclonal 1:500 Germany / #139105 Vesicular glutamate transporter 1 VGluT1 rabbit polyclonal Synaptic Systems / #135302 1:1000 Vesicular glutamate transporter 2 VGluT2 guinea pig polyclonal Synaptic Systems / #135404 1:200 sections were examined for each muscle, in which (secondary) structure ranging from immature to mature, NMJs from different randomly selected visual fields according to previously described criteria [154]. An were analyzed. NMJ size was measured by determining NMJ was defined as fragmented when its postsynaptic the area of manually outlined α-Bgtx-labeled site displayed a discontinued appearance, with 5 or postsynaptic site. Full or partial denervation was more AChR-islands stained with α-Bgtx, and/or considered when the postsynaptic site was not unusually small and irregular clusters of AChR. completely apposed by the nerve terminal immuno- Immunostaining for CGRP, GAP-43 and agrin was stained for both NF and SV2, as presynaptic marker: a estimated based on pixel intensity after background NMJ was considered to be denervated when the subtraction in NMJs which postsynaptic site was percentage of α-Bgtx-labeled postsynaptic surface visualized with α-Bgtx. An average of 50 NMJs per covered by SV2-stained presynaptic terminal was muscle was examined to assess the different <15%. Single or poly-innervation was estimated by morphometric parameters and the intensity of counting the number of axon terminals, stained with immunostaining. Any NMJ that was difficult to anti-NF+SV2 antibodies, entering a single postsynaptic examine due to its location and/or orientation was site: an NMJ was defined as polyinnervated when it was excluded from the analysis. An average of 35 muscle occupied by two or more axons. Terminal sprouting was sections from 3-5 animals per condition were analyzed. quantified by counting the number of NF+SV2-stained axonal processes coming from a nerve terminal that The analysis of afferent synapses on MNs and glia in escaped from a α-Bgtx-labeled postsynaptic site; the the ventral horn was performed on confocal images number of NMJs showing sprouts was referred as a taken from every 30th section of the entire lumbar spinal percentage of the total number of NMJs examined per cord. The number and area of VAChT-, VGluT1- and muscle. On the basis of the morphological appearance VGluT2-immunoreactive synaptic boutons on MN of the postsynaptic site, the degree of NMJ maturity was somata were counted and measured by delineating their classified as plaque, folds, perforations, or pretzel-like periphery on the screen; only boutons that were in close www.aging-us.com 18074 AGING contact with MNs displaying a large nucleus and visible protein concentrations of supernatants were determined nucleolus were included in the counts, and numbers by BIO-RAD Micro DC protein assay (BIO-RAD, were normalized to the perimeter of MN soma. Levels Laboratories Inc., Hercules, CA, USA). Loading buffer of immunoreactivity for Iba1, Mac2, CD206, P2Y12R 4 x SS (20% sucrose and 0.05% bromophenol blue, and CD68 in the spinal cord were quantified by 0.1% sodium azide) containing 5-10% β-mercapto- analyzing pixel intensity following background ethanol (Sigma-Aldrich) and 15-50 µg of protein were subtraction. Spinal cord microglia morphology was loaded in either 7.5% or 10% polyacrylamide quantified in confocal images from Iba1 immunostained electrophoresis gel. Proteins were electrotransferred to sections by using ImageJ software for skeleton analysis polyvinyldifluoride membranes (ImmobilonTM-P, (AnalyzeSkeleton [2D/3D] from http://imageJ.net/ Millipore) in Tris-glycine-methanol-buffered solution. AnalyzeSkeleton). Analysis was performed according to Membranes were blocked with 5% dried skim milk in a previously described procedure [80]. For this, digital 0.1% Tween 20 and Tris-buffered saline pH 8 (TBST) photomicrographs were transformed to 8-bit grayscale, for 1 hour at RT, and then extensively washed in TBST. and then binarized to obtain a black and white image by Immunodetection was performed by incubating the means of a formerly established threshold. Every image membranes overnight at 4° C with rabbit polyclonal was manually edited to obtain a continuous set of pixels anti-PGC-1α (1:500; Santa Cruz Biotechnology, cat. # and gaps between processes belonging to neighboring sc-13067), rabbit polyclonal anti-GAP-43 (diluted cells. The image was then saved and the plugin 1:1000; Santa Cruz Biotechnology cat # sc-10786), AnalyzeSkeleton(2D/3D) run. The accuracy of mouse monoclonal anti-α-tubulin (diluted 1:1000; skeletonized images was assessed, by creating an Sigma-Aldrich; cat. # T5168), and mouse monoclonal overlay of the obtained skeleton with the corresponding anti-glyceraldehyde 3-phosphatase dehydrogenase original image. (GAPDH, diluted 1:10,000; Abcam, cat. # ab8245) antibodies, the two latter used for loading controls. The The density of V0C neurons in lumbar spinal cord was membranes were washed in TBST, incubated with the measured by counting on digital images the number of appropriate peroxidase-conjugated secondary antibodies VAChT-immunolabeled interneurons in a 0.1 mm2 area (1:10,000; Amersham Biosciences, Buckinghamshire, around the central canal where these cells are placed. UK) for 60 minutes at RT, washed in TBST and visualized using the ECL Prime Western Blotting The number and soma area of CGRP, IB4 and PV- Detection Reagent detection kit (GE Healthcare, immunolabeled sensory neurons in DRGs were also Buckinghamshire, UK), as described by the measured on digital images. The number and diameter manufacturer. The quantification of band densities was of healthy and degenerating myelinated axons were performed by using Chemi-Doc MP Imaging System determined on 60x images taken from VR semithin (BIO-RAD Laboratories Inc.). cross-sections stained with methylene blue. The images were joined together to obtain a whole picture of an Statistical analysis entire nerve root transverse section. The diameter of the myelinated axons was measured by delineating the Data are shown as the mean ± standard error of the outer profile of the myelin sheath. Axon diameter and g- mean (SEM). The statistical analysis was performed ratio measurements were performed on micrographs of by either one-way or two-way analysis of variance VRs; ImageJ g-ratio plug-in was used to obtain (ANOVA) followed by post hoc Bonferroni’s test or semiautomated measurements of randomly selected two-tailed Student’s t-test when only two nerve fibers; the axon diameter was divided by the outer different groups were compared. Gehan-Breslow- diameter of the myelin sheath. At least 100 myelinated Wilcoxon test was used for comparing survival curves axons per mouse (4-5 mice per condition) were between groups. The level of significance was measured. established at p ≤ 0.05. GraphPad Prism 6 software was used for statistical analysis and graph Western blotting presentations of data. Frozen hindlimb muscles were fragmented and AUTHOR CONTRIBUTIONS homogenized using an electric homogenizer with ice- cold RIPA lysis buffer (50 mM Tris-HCl [pH 7.4], 150 JC, JEE, SLP, ABa, TD and RR contributed to mM NaCl, 1 mM EDTA, 1% NP-40, 1% Na- conception of the research, and JC and JEE designed deoxycholate, 0.1% SDS) supplemented with protease the study. AB, SG, GM-C, OT, AC and LP performed inhibitor (Sigma-Aldrich, cat. # P8340) and PhosSTOP the research. AB, SG, OT, XN and JC analyzed the (Roche, Laval, Canada). The homogenized samples data. JC, JEE, XN and SLP contributed to writing the were centrifuged at 13,000 rpm for 20 min at 4° C. The paper. www.aging-us.com 18075 AGING ACKNOWLEDGMENTS https://doi.org/10.1152/physrev.00061.2017 PMID:30427277 We thank Sara Salvany, Sara Hernández, Alaó Gatius, 6. Blasco A, Gras S, Mòdol-Caballero G, Tarabal O, Mònica Esteve, Maria Beltrán, Clara Paniagua and Pol Casanovas A, Piedrafita L, Barranco A, Das T, Pereira SL, Garcia for collaborating in some experiments, Anaïs Navarro X, Rueda R, Esquerda JE, Calderó J. Panosa and Alba Garcia from the SCT of Microscopy of Motoneuron deafferentation and gliosis occur in the Universitat de Lleida for technical support with association with neuromuscular regressive changes confocal microscopy, and the staff from the SCT during ageing in mice. J Cachexia Sarcopenia Muscle. Animal Facility of the Universitat de Lleida for mouse 2020; 11:1628–60. care and housing. https://doi.org/10.1002/jcsm.12599 PMID:32691534 CONFLICTS OF INTEREST 7. Tintignac LA, Brenner HR, Rüegg MA. Mechanisms Regulating Neuromuscular Junction Development and SLP, TD and RR are employees and stock holders of Function and Causes of Muscle Wasting. Physiol Rev. Abbott. ABa was employee and stock holder of Abbott 2015; 95:809–52. during initiation of study. The other authors declare that https://doi.org/10.1152/physrev.00033.2014 they have no conflict of interest. PMID:26109340 8. Li L, Xiong WC, Mei L. Neuromuscular Junction FUNDING Formation, Aging, and Disorders. Annu Rev Physiol. 2018; 80:159–88. This work was supported by Abbott and a grant from https://doi.org/10.1146/annurev-physiol-022516- the Spanish Ministerio de Ciencia, Innovación y 034255 PMID:29195055 Universidades cofinanced by Fondo Europeo de Desarrollo Regional (RTI2018-099278-B-I00 to JC and 9. Willadt S, Nash M, Slater C. Age-related changes in the JE). structure and function of mammalian neuromuscular junctions. Ann N Y Acad Sci. 2018; 1412:41–53. https://doi.org/10.1111/nyas.13521 PMID:29291259 REFERENCES 10. Pedersen BK, Febbraio MA. Muscles, exercise and 1. Janssen I, Baumgartner RN, Ross R, Rosenberg IH, obesity: skeletal muscle as a secretory organ. Nat Rev Roubenoff R. Skeletal muscle cutpoints associated with Endocrinol. 2012; 8:457–65. elevated physical disability risk in older men and https://doi.org/10.1038/nrendo.2012.49 women. Am J Epidemiol. 2004; 159:413–21. PMID:22473333 https://doi.org/10.1093/aje/kwh058 11. Valdez G, Tapia JC, Kang H, Clemenson GD Jr, Gage FH, PMID:14769646 Lichtman JW, Sanes JR. Attenuation of age-related 2. Abellan van Kan G. Epidemiology and consequences of changes in mouse neuromuscular synapses by caloric sarcopenia. J Nutr Health Aging. 2009; 13:708–12. restriction and exercise. Proc Natl Acad Sci USA. 2010; https://doi.org/10.1007/s12603-009-0201-z 107:14863–68. PMID:19657554 https://doi.org/10.1073/pnas.1002220107 PMID:20679195 3. von Haehling S, Morley JE, Anker SD. An overview of sarcopenia: facts and numbers on prevalence and 12. Kougias DG, Das T, Perez AB, Pereira SL. A role for clinical impact. J Cachexia Sarcopenia Muscle. 2010; nutritional intervention in addressing the aging 1:129–33. neuromuscular junction. Nutr Res. 2018; 53:1–14. https://doi.org/10.1007/s13539-010-0014-2 https://doi.org/10.1016/j.nutres.2018.02.006 PMID:21475695 PMID:29804584 4. Gava P, Kern H, Carraro U. Age-associated power 13. Haramizu S, Mori T, Yano M, Ota N, Hashizume K, decline from running, jumping, and throwing male Otsuka A, Hase T, Shimotoyodome A. Habitual exercise masters world records. Exp Aging Res. 2015; 41: plus dietary supplementation with milk fat globule 115–35. membrane improves muscle function deficits via https://doi.org/10.1080/0361073X.2015.1001648 neuromuscular development in senescence- PMID:25724012 accelerated mice. Springerplus. 2014; 3:339. https://doi.org/10.1186/2193-1801-3-339 5. Larsson L, Degens H, Li M, Salviati L, Lee YI, Thompson PMID:25110626 W, Kirkland JL, Sandri M. Sarcopenia: Aging-Related Loss of Muscle Mass and Function. Physiol Rev. 2019; 14. Yano M, Haramizu S, Ota N, Minegishi Y, 99:427–511. Shimotoyodome A. Continuous Supplementation of www.aging-us.com 18076 AGING Milk Fat Globule Membrane with Habitual Exercise PMID:27259307 from a Young Age Improves Motor Coordination and 23. Arent SM, Senso M, Golem DL, McKeever KH. The Skeletal Muscle Function in Aged Mice. J Nutr Sci effects of theaflavin-enriched black tea extract on Vitaminol (Tokyo). 2019; 65:405–13. muscle soreness, oxidative stress, inflammation, and https://doi.org/10.3177/jnsv.65.405 PMID:31666477 endocrine responses to acute anaerobic interval 15. Nishimune H, Stanford JA, Mori Y. Role of exercise in training: a randomized, double-blind, crossover study. J maintaining the integrity of the neuromuscular Int Soc Sports Nutr. 2010; 7:11. junction. Muscle Nerve. 2014; 49:315–24. https://doi.org/10.1186/1550-2783-7-11 https://doi.org/10.1002/mus.24095 PMID:20178632 PMID:24122772 24. Bowtell JL, Sumners DP, Dyer A, Fox P, Mileva KN. 16. Marzetti E, Lees HA, Wohlgemuth SE, Leeuwenburgh C. Montmorency cherry juice reduces muscle damage Sarcopenia of aging: underlying cellular mechanisms caused by intensive strength exercise. Med Sci Sports and protection by calorie restriction. Biofactors. 2009; Exerc. 2011; 43:1544–51. 35:28–35. https://doi.org/10.1249/MSS.0b013e31820e5adc https://doi.org/10.1002/biof.5 PMID:19319843 PMID:21233776 17. Marzetti E, Lawler JM, Hiona A, Manini T, Seo AY, 25. Herrlinger KA, Chirouzes DM, Ceddia MA. Leeuwenburgh C. Modulation of age-induced Supplementation with a polyphenolic blend improves apoptotic signaling and cellular remodeling by exercise post-exercise strength recovery and muscle soreness. and calorie restriction in skeletal muscle. Free Radic Food Nutr Res. 2015; 59:30034. Biol Med. 2008; 44:160–68. https://doi.org/10.3402/fnr.v59.30034 https://doi.org/10.1016/j.freeradbiomed.2007.05.028 PMID:26689317 PMID:18191752 26. Trombold JR, Reinfeld AS, Casler JR, Coyle EF. The 18. Faitg J, Leduc-Gaudet JP, Reynaud O, Ferland G, effect of pomegranate juice supplementation on Gaudreau P, Gouspillou G. Effects of Aging and Caloric strength and soreness after eccentric exercise. J Restriction on Fiber Type Composition, Mitochondrial Strength Cond Res. 2011; 25:1782–88. Morphology and Dynamics in Rat Oxidative and https://doi.org/10.1519/JSC.0b013e318220d992 Glycolytic Muscles. Front Physiol. 2019; 10:420. PMID:21659887 https://doi.org/10.3389/fphys.2019.00420 27. Machin DR, Christmas KM, Chou TH, Hill SC, Van PMID:31114501 Pelt DW, Trombold JR, Coyle EF. Effects of Differing 19. Jang YC, Liu Y, Hayworth CR, Bhattacharya A, Dosages of Pomegranate Juice Supplementation Lustgarten MS, Muller FL, Chaudhuri A, Qi W, Li Y, after Eccentric Exercise. Physiology Journal. 2014; Huang JY, Verdin E, Richardson A, Van Remmen H. 2014:1–7. Dietary restriction attenuates age-associated muscle https://doi.org/10.1155/2014/271959 atrophy by lowering oxidative stress in mice even in 28. Beyer KS, Stout JR, Fukuda DH, Jajtner AR, Townsend complete absence of CuZnSOD. Aging Cell. 2012; JR, Church DD, Wang R, Riffe JJ, Muddle TW, Herrlinger 11:770–82. KA, Hoffman JR. Impact of Polyphenol https://doi.org/10.1111/j.1474-9726.2012.00843.x Supplementation on Acute and Chronic Response to PMID:22672615 Resistance Training. J Strength Cond Res. 2017; 20. Cerletti M, Jang YC, Finley LW, Haigis MC, Wagers AJ. 31:2945–54. Short-term calorie restriction enhances skeletal muscle https://doi.org/10.1519/JSC.0000000000002104 stem cell function. Cell Stem Cell. 2012; 10:515–19. PMID:29068862 https://doi.org/10.1016/j.stem.2012.04.002 29. Davoodi J, Markert CD, Voelker KA, Hutson SM, Grange PMID:22560075 RW. Nutrition strategies to improve physical 21. Chopek JW, Gardiner PF. Life-long caloric restriction: capabilities in Duchenne muscular dystrophy. Phys Effect on age-related changes in motoneuron Med Rehabil Clin N Am. 2012; 23:187–99. numbers, sizes and apoptotic markers. Mech Ageing https://doi.org/10.1016/j.pmr.2011.11.010 Dev. 2010; 131:650–59. PMID:22239883 https://doi.org/10.1016/j.mad.2010.09.001 30. Call JA, Voelker KA, Wolff AV, McMillan RP, Evans NP, PMID:20851712 Hulver MW, Talmadge RJ, Grange RW. Endurance 22. Zhang PY. Polyphenols in Health and Disease. Cell capacity in maturing mdx mice is markedly enhanced Biochem Biophys. 2015; 73:649–64. by combined voluntary wheel running and green tea https://doi.org/10.1007/s12013-015-0558-z extract. J Appl Physiol (1985). 2008; 105:923–32. www.aging-us.com 18077 AGING https://doi.org/10.1152/japplphysiol.00028.2008 radicals. Chem Res Toxicol. 2000; 13:801–10. PMID:18583385 https://doi.org/10.1021/tx000080k PMID:10995252 31. Dorchies OM, Wagner S, Vuadens O, Waldhauser K, 39. Wollgast J, Anklam E. Review on polyphenols in Buetler TM, Kucera P, Ruegg UT. Green tea extract and Theobroma cacao: changes in composition during the its major polyphenol (-)-epigallocatechin gallate manufacture of chocolate and methodology for improve muscle function in a mouse model for identification and quantification. Food Research Duchenne muscular dystrophy. Am J Physiol Cell International. 2000; 33:423–47. Physiol. 2006; 290:C616–25. https://doi.org/10.1016/S0963-9969(00)00068-5 https://doi.org/10.1152/ajpcell.00425.2005 40. Lamuela-Raventós RM, Romero-Pérez AI, Andrés- PMID:16403950 Lacueva C, Tornero A. Review: Health Effects of Cocoa 32. Buetler TM, Renard M, Offord EA, Schneider H, Ruegg Flavonoids. Food Science and Technology UT. Green tea extract decreases muscle necrosis in International. 2016; 11:159–76. mdx mice and protects against reactive oxygen species. https://doi.org/10.1177/1082013205054498 Am J Clin Nutr. 2002; 75:749–53. 41. Fraga CG, Litterio MC, Prince PD, Calabró V, https://doi.org/10.1093/ajcn/75.4.749 Piotrkowski B, Galleano M. Cocoa flavanols: effects on PMID:11916763 vascular nitric oxide and blood pressure. J Clin Biochem 33. Meador BM, Mirza KA, Tian M, Skelding MB, Reaves Nutr. 2011; 48:63–67. LA, Edens NK, Tisdale MJ, Pereira SL. The Green Tea https://doi.org/10.3164/jcbn.11-010FR Polyphenol Epigallocatechin-3-Gallate (EGCg) PMID:21297914 Attenuates Skeletal Muscle Atrophy in a Rat Model of 42. Munguia L, Rubio-Gayosso I, Ramirez-Sanchez I, Ortiz Sarcopenia. J Frailty Aging. 2015; 4:209–15. A, Hidalgo I, Gonzalez C, Meaney E, Villarreal F, Najera https://doi.org/10.14283/jfa.2015.58 PMID:27031020 N, Ceballos G. High Flavonoid Cocoa Supplement 34. Murase T, Haramizu S, Ota N, Hase T. Tea catechin Ameliorates Plasma Oxidative Stress and Inflammation ingestion combined with habitual exercise suppresses Levels While Improving Mobility and Quality of Life in the aging-associated decline in physical performance in Older Subjects: A Double-Blind Randomized Clinical senescence-accelerated mice. Am J Physiol Regul Trial. J Gerontol A Biol Sci Med Sci. 2019; 74:1620–27. Integr Comp Physiol. 2008; 295:R281–89. https://doi.org/10.1093/gerona/glz107 https://doi.org/10.1152/ajpregu.00880.2007 PMID:31056655 PMID:18480242 43. Magrone T, Russo MA, Jirillo E. Cocoa and Dark 35. Alway SE, Bennett BT, Wilson JC, Edens NK, Pereira SL. Chocolate Polyphenols: From Biology to Clinical Epigallocatechin-3-gallate improves plantaris muscle Applications. Front Immunol. 2017; 8:677. recovery after disuse in aged rats. Exp Gerontol. 2014; https://doi.org/10.3389/fimmu.2017.00677 50:82–94. PMID:28649251 https://doi.org/10.1016/j.exger.2013.11.011 44. Massaro M, Scoditti E, Carluccio MA, Kaltsatou A, PMID:24316035 Cicchella A. Effect of Cocoa Products and Its 36. Alway SE, Bennett BT, Wilson JC, Sperringer J, Polyphenolic Constituents on Exercise Performance Mohamed JS, Edens NK, Pereira SL. Green tea extract and Exercise-Induced Muscle Damage and attenuates muscle loss and improves muscle function Inflammation: A Review of Clinical Trials. Nutrients. during disuse, but fails to improve muscle recovery 2019; 11:1471. following unloading in aged rats. J Appl Physiol (1985). https://doi.org/10.3390/nu11071471 2015; 118:319–30. PMID:31261645 https://doi.org/10.1152/japplphysiol.00674.2014 45. Ramiro-Puig E, Casadesús G, Lee HG, Zhu X, McShea A, PMID:25414242 Perry G, Pérez-Cano FJ, Smith MA, Castell M. 37. Meador BM, Pereira SL, Edens NK. Green Tea Neuroprotective effect of cocoa flavonoids on in vitro Supplementation: Current Research, Literature Gaps, oxidative stress. Eur J Nutr. 2009; 48:54–61. and Product Safety. Journal of Food and Nutritional https://doi.org/10.1007/s00394-008-0761-4 Disorders. 2013; 02. PMID:19043659 https://doi.org/10.4172/2324-9323.1000126 46. Si H, Wang X, Zhang L, Parnell LD, Admed B, LeRoith T, 38. Valcic S, Burr JA, Timmermann BN, Liebler DC. Ansah TA, Zhang L, Li J, Ordovás JM, Si H, Liu D, Lai CQ. Antioxidant chemistry of green tea catechins. New Dietary epicatechin improves survival and delays oxidation products of (-)-epigallocatechin gallate and skeletal muscle degeneration in aged mice. FASEB J. (-)-epigallocatechin from their reactions with peroxyl 2019; 33:965–77. www.aging-us.com 18078 AGING https://doi.org/10.1096/fj.201800554RR 56. Nakae Y, Stoward PJ. The high correlation between PMID:30096038 counts and area fractions of lipofuscin granules, a biomarker of oxidative stress in muscular dystrophies. 47. Nogueira L, Ramirez-Sanchez I, Perkins GA, Murphy A, Histochem Cell Biol. 2016; 146:627–34. Taub PR, Ceballos G, Villarreal FJ, Hogan MC, Malek MH. (-)-Epicatechin enhances fatigue resistance and https://doi.org/10.1007/s00418-016-1462-3 PMID:27394427 oxidative capacity in mouse muscle. J Physiol. 2011; 589:4615–31. 57. Lin J, Wu H, Tarr PT, Zhang CY, Wu Z, Boss O, Michael https://doi.org/10.1113/jphysiol.2011.209924 LF, Puigserver P, Isotani E, Olson EN, Lowell BB, Bassel- PMID:21788351 Duby R, Spiegelman BM. Transcriptional co-activator PGC-1 alpha drives the formation of slow-twitch 48. Carlson BM. The regeneration of skeletal muscle. A review. Am J Anat. 1973; 137:119–49. muscle fibres. Nature. 2002; 418:797–801. https://doi.org/10.1038/nature00904 PMID:12181572 https://doi.org/10.1002/aja.1001370202 PMID:4350147 58. Sandri M, Lin J, Handschin C, Yang W, Arany ZP, Lecker SH, Goldberg AL, Spiegelman BM. PGC-1alpha protects 49. Folker ES, Baylies MK. Nuclear positioning in muscle skeletal muscle from atrophy by suppressing FoxO3 development and disease. Front Physiol. 2013; 4:363. action and atrophy-specific gene transcription. Proc https://doi.org/10.3389/fphys.2013.00363 PMID:24376424 Natl Acad Sci USA. 2006; 103:16260–65. https://doi.org/10.1073/pnas.0607795103 50. Day K, Shefer G, Shearer A, Yablonka-Reuveni Z. The PMID:17053067 depletion of skeletal muscle satellite cells with age is concomitant with reduced capacity of single 59. Handschin C, Kobayashi YM, Chin S, Seale P, Campbell KP, Spiegelman BM. PGC-1alpha regulates the progenitors to produce reserve progeny. Dev Biol. neuromuscular junction program and ameliorates 2010; 340:330–43. https://doi.org/10.1016/j.ydbio.2010.01.006 Duchenne muscular dystrophy. Genes Dev. 2007; PMID:20079729 21:770–83. https://doi.org/10.1101/gad.1525107 51. Gopinath SD, Rando TA. Stem cell review series: aging PMID:17403779 of the skeletal muscle stem cell niche. Aging Cell. 2008; 7:590–98. 60. Chabi B, Ljubicic V, Menzies KJ, Huang JH, Saleem A, https://doi.org/10.1111/j.1474-9726.2008.00399.x Hood DA. Mitochondrial function and apoptotic PMID:18462272 susceptibility in aging skeletal muscle. Aging Cell. 2008; 7:2–12. 52. Snow MH. The effects of aging on satellite cells in https://doi.org/10.1111/j.1474-9726.2007.00347.x skeletal muscles of mice and rats. Cell Tissue Res. 1977; PMID:18028258 185:399–408. https://doi.org/10.1007/BF00220299 61. Reznick RM, Zong H, Li J, Morino K, Moore IK, Yu HJ, Liu PMID:597854 ZX, Dong J, Mustard KJ, Hawley SA, Befroy D, Pypaert M, Hardie DG, et al. Aging-associated reductions in 53. Sousa-Victor P, García-Prat L, Serrano AL, Perdiguero E, AMP-activated protein kinase activity and Muñoz-Cánoves P. Muscle stem cell aging: regulation mitochondrial biogenesis. Cell Metab. 2007; 5:151–56. and rejuvenation. Trends Endocrinol Metab. 2015; https://doi.org/10.1016/j.cmet.2007.01.008 26:287–96. PMID:17276357 https://doi.org/10.1016/j.tem.2015.03.006 PMID:25869211 62. Baker DJ, Betik AC, Krause DJ, Hepple RT. No decline in skeletal muscle oxidative capacity with aging in long- 54. Gray DA, Woulfe J. Lipofuscin and aging: a matter of term calorically restricted rats: effects are independent toxic waste. Sci Aging Knowledge Environ. 2005; of mitochondrial DNA integrity. J Gerontol A Biol Sci 2005:re1. Med Sci. 2006; 61:675–84. https://doi.org/10.1126/sageke.2005.5.re1 https://doi.org/10.1093/gerona/61.7.675 PMID:15689603 PMID:16870628 55. Nakae Y, Stoward PJ, Kashiyama T, Shono M, Akagi A, 63. Valdez G, Tapia JC, Lichtman JW, Fox MA, Sanes JR. Matsuzaki T, Nonaka I. Early onset of lipofuscin Shared resistance to aging and ALS in neuromuscular accumulation in dystrophin-deficient skeletal muscles junctions of specific muscles. PLoS One. 2012; of DMD patients and mdx mice. J Mol Histol. 2004; 7:e34640. 35:489–99. https://doi.org/10.1371/journal.pone.0034640 https://doi.org/10.1023/b:hijo.0000045947.83628.a7 PMID:22485182 PMID:15571326 www.aging-us.com 18079 AGING 64. Chai RJ, Vukovic J, Dunlop S, Grounds MD, Shavlakadze associated skeletal muscle decline. Muscle Nerve. T. Striking denervation of neuromuscular junctions 2017; 55:894–901. without lumbar motoneuron loss in geriatric mouse https://doi.org/10.1002/mus.25267 muscle. PLoS One. 2011; 6:e28090. PMID:27464347 https://doi.org/10.1371/journal.pone.0028090 73. Sheard PW, Anderson RD. Age-related loss of muscle PMID:22164231 fibres is highly variable amongst mouse skeletal 65. Willadt S, Nash M, Slater CR. Age-related muscles. Biogerontology. 2012; 13:157–67. fragmentation of the motor endplate is not associated https://doi.org/10.1007/s10522-011-9365-0 with impaired neuromuscular transmission in the PMID:22048817 mouse diaphragm. Sci Rep. 2016; 6:24849. 74. Hashizume K, Kanda K. Differential effects of aging on https://doi.org/10.1038/srep24849 PMID:27094316 motoneurons and peripheral nerves innervating the 66. Sala C, Andreose JS, Fumagalli G, Lømo T. Calcitonin hindlimb and forelimb muscles of rats. Neurosci Res. gene-related peptide: possible role in formation and 1995; 22:189–96. maintenance of neuromuscular junctions. J Neurosci. https://doi.org/10.1016/0168-0102(95)00889-3 1995; 15:520–28. PMID:7566699 https://doi.org/10.1523/JNEUROSCI.15-01-00520.1995 75. Oertel G. Changes in human skeletal muscles due to PMID:7823160 ageing. Histological and histochemical observations on 67. Tarabal O, Calderó J, Esquerda JE. Intramuscular nerve autopsy material. Acta Neuropathol. 1986; 69:309–13. sprouting induced by CNTF is associated with increases https://doi.org/10.1007/BF00688309 PMID:3962608 in CGRP content in mouse motor nerve terminals. 76. Maxwell N, Castro RW, Sutherland NM, Vaughan KL, Neurosci Lett. 1996; 219:60–64. Szarowicz MD, de Cabo R, Mattison JA, Valdez G. α- https://doi.org/10.1016/s0304-3940(96)13174-8 Motor neurons are spared from aging while their PMID:8961304 synaptic inputs degenerate in monkeys and mice. 68. Tarabal O, Calderó J, Ribera J, Sorribas A, López R, Aging Cell. 2018; 17:e12726. Molgó J, Esquerda JE. Regulation of motoneuronal https://doi.org/10.1111/acel.12726 PMID:29397579 calcitonin gene-related peptide (CGRP) during axonal 77. Krishnan VS, Shavlakadze T, Grounds MD, Hodgetts SI, growth and neuromuscular synaptic plasticity induced Harvey AR. Age-related loss of VGLUT1 excitatory, but by botulinum toxin in rats. Eur J Neurosci. 1996; 8: not VGAT inhibitory, immunoreactive terminals on 829–36. motor neurons in spinal cords of old sarcopenic male https://doi.org/10.1111/j.1460-9568.1996.tb01269.x mice. Biogerontology. 2018; 19:385–99. PMID:9081635 https://doi.org/10.1007/s10522-018-9765-5 69. Matteoli M, Balbi S, Sala C, Chini B, Cimino M, Vitadello PMID:30084046 M, Fumagalli G. Developmentally regulated expression 78. Zagoraiou L, Akay T, Martin JF, Brownstone RM, Jessell of calcitonin gene-related peptide at mammalian TM, Miles GB. A cluster of cholinergic premotor neuromuscular junction. J Mol Neurosci. 1990; 2: interneurons modulates mouse locomotor activity. 175–84. Neuron. 2009; 64:645–62. https://doi.org/10.1007/BF02896842 https://doi.org/10.1016/j.neuron.2009.10.017 PMID:2275848 PMID:20005822 70. Samuel MA, Valdez G, Tapia JC, Lichtman JW, Sanes JR. 79. Henstridge CM, Tzioras M, Paolicelli RC. Glial Agrin and synaptic laminin are required to maintain Contribution to Excitatory and Inhibitory Synapse Loss adult neuromuscular junctions. PLoS One. 2012; in Neurodegeneration. Front Cell Neurosci. 2019; 7:e46663. 13:63. https://doi.org/10.1371/journal.pone.0046663 https://doi.org/10.3389/fncel.2019.00063 PMID:23056392 PMID:30863284 71. Holahan MR. A Shift from a Pivotal to Supporting Role 80. Young K, Morrison H. Quantifying Microglia for the Growth-Associated Protein (GAP-43) in the Morphology from Photomicrographs of Coordination of Axonal Structural and Functional Immunohistochemistry Prepared Tissue Using ImageJ. J Plasticity. Front Cell Neurosci. 2017; 11:266. Vis Exp. 2018; 57648. https://doi.org/10.3389/fncel.2017.00266 https://doi.org/10.3791/57648 PMID:28912688 PMID:29939190 72. Chung T, Park JS, Kim S, Montes N, Walston J, Höke A. 81. Cherry JD, Olschowka JA, O’Banion MK. Evidence for dying-back axonal degeneration in age- Neuroinflammation and M2 microglia: the good, the www.aging-us.com 18080 AGING bad, and the inflamed. J Neuroinflammation. 2014; https://doi.org/10.1186/s12974-017-1034-z 11:98. PMID:29273052 https://doi.org/10.1186/1742-2094-11-98 90. Zhu C, Kros JM, van der Weiden M, Zheng P, Cheng C, PMID:24889886 Mustafa DA. Expression site of P2RY12 in residential 82. Tang Y, Le W. Differential Roles of M1 and M2 microglial cells in astrocytomas correlates with M1 and Microglia in Neurodegenerative Diseases. Mol M2 marker expression and tumor grade. Acta Neurobiol. 2016; 53:1181–94. Neuropathol Commun. 2017; 5:4. https://doi.org/10.1007/s12035-014-9070-5 https://doi.org/10.1186/s40478-016-0405-5 PMID:25598354 PMID:28073370 83. Kigerl KA, Gensel JC, Ankeny DP, Alexander JK, 91. Walker DG, Lue LF. Immune phenotypes of microglia in Donnelly DJ, Popovich PG. Identification of two distinct human neurodegenerative disease: challenges to macrophage subsets with divergent effects causing detecting microglial polarization in human brains. either neurotoxicity or regeneration in the injured Alzheimers Res Ther. 2015; 7:56. mouse spinal cord. J Neurosci. 2009; 29:13435–44. https://doi.org/10.1186/s13195-015-0139-9 https://doi.org/10.1523/JNEUROSCI.3257-09.2009 PMID:26286145 PMID:19864556 92. Rotterman TM, Alvarez FJ. Microglia Dynamics and 84. Franco R, Fernández-Suárez D. Alternatively activated Interactions with Motoneurons Axotomized After microglia and macrophages in the central nervous Nerve Injuries Revealed By Two-Photon Imaging. Sci system. Prog Neurobiol. 2015; 131:65–86. Rep. 2020; 10:8648. https://doi.org/10.1016/j.pneurobio.2015.05.003 https://doi.org/10.1038/s41598-020-65363-9 PMID:26067058 PMID:32457369 85. Sasaki Y, Hoshi M, Akazawa C, Nakamura Y, Tsuzuki H, 93. Brack AS, Bildsoe H, Hughes SM. Evidence that satellite Inoue K, Kohsaka S. Selective expression of Gi/o- cell decrement contributes to preferential decline in coupled ATP receptor P2Y12 in microglia in rat brain. nuclear number from large fibres during murine age- Glia. 2003; 44:242–50. related muscle atrophy. J Cell Sci. 2005; 118:4813–21. https://doi.org/10.1002/glia.10293 PMID:14603465 https://doi.org/10.1242/jcs.02602 PMID:16219688 86. Bennett ML, Bennett FC, Liddelow SA, Ajami B, 94. Chakkalakal JV, Jones KM, Basson MA, Brack AS. The Zamanian JL, Fernhoff NB, Mulinyawe SB, Bohlen CJ, aged niche disrupts muscle stem cell quiescence. Adil A, Tucker A, Weissman IL, Chang EF, Li G, et al. Nature. 2012; 490:355–60. New tools for studying microglia in the mouse and https://doi.org/10.1038/nature11438 PMID:23023126 human CNS. Proc Natl Acad Sci USA. 2016; 113: 95. Sacco A, Puri PL. Regulation of Muscle Satellite Cell E1738–46. Function in Tissue Homeostasis and Aging. Cell Stem https://doi.org/10.1073/pnas.1525528113 Cell. 2015; 16:585–87. PMID:26884166 https://doi.org/10.1016/j.stem.2015.05.007 PMID:26046757 87. Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G, Koeglsperger T, Dake B, Wu PM, 96. Lepper C, Partridge TA, Fan CM. An absolute Doykan CE, Fanek Z, Liu L, Chen Z, et al. Identification requirement for Pax7-positive satellite cells in acute of a unique TGF-β-dependent molecular and functional injury-induced skeletal muscle regeneration. signature in microglia. Nat Neurosci. 2014; 17:131–43. Development. 2011; 138:3639–46. https://doi.org/10.1038/nn.3599 PMID:24316888 https://doi.org/10.1242/dev.067595 PMID:21828092 88. Moore CS, Ase AR, Kinsara A, Rao VT, Michell-Robinson M, Leong SY, Butovsky O, Ludwin SK, Séguéla P, Bar-Or 97. Murphy MM, Lawson JA, Mathew SJ, Hutcheson DA, A, Antel JP. P2Y12 expression and function in Kardon G. Satellite cells, connective tissue fibroblasts alternatively activated human microglia. Neurol and their interactions are crucial for muscle Neuroimmunol Neuroinflamm. 2015; 2:e80. regeneration. Development. 2011; 138:3625–37. https://doi.org/10.1212/NXI.0000000000000080 https://doi.org/10.1242/dev.064162 PMID:25821842 PMID:21828091 89. Beaino W, Janssen B, Kooij G, van der Pol SM, van Het 98. Sambasivan R, Yao R, Kissenpfennig A, Van Hof B, van Horssen J, Windhorst AD, de Vries HE. Wittenberghe L, Paldi A, Gayraud-Morel B, Guenou H, Purinergic receptors P2Y12R and P2X7R: potential Malissen B, Tajbakhsh S, Galy A. Pax7-expressing targets for PET imaging of microglia phenotypes in satellite cells are indispensable for adult skeletal multiple sclerosis. J Neuroinflammation. 2017; 14:259. muscle regeneration. Development. 2011; 138: www.aging-us.com 18081 AGING 3647–56. 107. Wu Z, Puigserver P, Andersson U, Zhang C, Adelmant https://doi.org/10.1242/dev.067587 G, Mootha V, Troy A, Cinti S, Lowell B, Scarpulla PMID:21828093 RC, Spiegelman BM. Mechanisms controlling mitochondrial biogenesis and respiration through the 99. Alway SE, Myers MJ, Mohamed JS. Regulation of satellite cell function in sarcopenia. Front Aging thermogenic coactivator PGC-1. Cell. 1999; 98: 115–24. Neurosci. 2014; 6:246. https://doi.org/10.1016/S0092-8674(00)80611-X https://doi.org/10.3389/fnagi.2014.00246 PMID:25295003 PMID:10412986 100. Fry CS, Lee JD, Mula J, Kirby TJ, Jackson JR, Liu F, Yang 108. Ji LL, Kang C. Role of PGC-1α in sarcopenia: etiology L, Mendias CL, Dupont-Versteegden EE, McCarthy JJ, and potential intervention - a mini-review. Peterson CA. Inducible depletion of satellite cells in Gerontology. 2015; 61:139–48. adult, sedentary mice impairs muscle regenerative https://doi.org/10.1159/000365947 PMID:25502801 capacity without affecting sarcopenia. Nat Med. 2015; 109. Alvarez-Guardia D, Palomer X, Coll T, Davidson MM, 21:76–80. Chan TO, Feldman AM, Laguna JC, Vázquez-Carrera M. https://doi.org/10.1038/nm.3710 PMID:25501907 The p65 subunit of NF-kappaB binds to PGC-1alpha, 101. Gutierrez-Salmean G, Ciaraldi TP, Nogueira L, Barboza linking inflammation and metabolic disturbances in J, Taub PR, Hogan MC, Henry RR, Meaney E, Villarreal cardiac cells. Cardiovasc Res. 2010; 87:449–58. F, Ceballos G, Ramirez-Sanchez I. Effects of (-)- https://doi.org/10.1093/cvr/cvq080 PMID:20211864 epicatechin on molecular modulators of skeletal 110. St-Pierre J, Drori S, Uldry M, Silvaggi JM, Rhee J, Jäger muscle growth and differentiation. J Nutr Biochem. S, Handschin C, Zheng K, Lin J, Yang W, Simon DK, 2014; 25:91–94. Bachoo R, Spiegelman BM. Suppression of reactive https://doi.org/10.1016/j.jnutbio.2013.09.007 oxygen species and neurodegeneration by the PGC-1 PMID:24314870 transcriptional coactivators. Cell. 2006; 127:397–408. 102. Malaguti M, Angeloni C, Hrelia S. Polyphenols in https://doi.org/10.1016/j.cell.2006.09.024 exercise performance and prevention of exercise- PMID:17055439 induced muscle damage. Oxid Med Cell Longev. 2013; 111. Arany Z, Foo SY, Ma Y, Ruas JL, Bommi-Reddy A, 2013:825928. Girnun G, Cooper M, Laznik D, Chinsomboon J, https://doi.org/10.1155/2013/825928 Rangwala SM, Baek KH, Rosenzweig A, Spiegelman PMID:23983900 BM. HIF-independent regulation of VEGF and 103. Dirks A, Leeuwenburgh C. Apoptosis in skeletal muscle angiogenesis by the transcriptional coactivator PGC- with aging. Am J Physiol Regul Integr Comp Physiol. 1alpha. Nature. 2008; 451:1008–12. 2002; 282:R519–27. https://doi.org/10.1038/nature06613 https://doi.org/10.1152/ajpregu.00458.2001 PMID:18288196 PMID:11792662 112. Wenz T, Rossi SG, Rotundo RL, Spiegelman BM, 104. Marzetti E, Wohlgemuth SE, Lees HA, Chung HY, Moraes CT. Increased muscle PGC-1alpha expression Giovannini S, Leeuwenburgh C. Age-related activation protects from sarcopenia and metabolic disease of mitochondrial caspase-independent apoptotic during aging. Proc Natl Acad Sci USA. 2009; signaling in rat gastrocnemius muscle. Mech Ageing 106:20405–10. Dev. 2008; 129:542–49. https://doi.org/10.1073/pnas.0911570106 https://doi.org/10.1016/j.mad.2008.05.005 PMID:19918075 PMID:18579179 113. Ruas JL, White JP, Rao RR, Kleiner S, Brannan KT, 105. Hood DA, Memme JM, Oliveira AN, Triolo M. Harrison BC, Greene NP, Wu J, Estall JL, Irving BA, Maintenance of Skeletal Muscle Mitochondria in Lanza IR, Rasbach KA, Okutsu M, et al. A PGC-1α Health, Exercise, and Aging. Annu Rev Physiol. 2019; isoform induced by resistance training regulates 81:19–41. skeletal muscle hypertrophy. Cell. 2012; 151:1319–31. https://doi.org/10.1146/annurev-physiol-020518- https://doi.org/10.1016/j.cell.2012.10.050 114310 PMID:30216742 PMID:23217713 106. Anagnostou ME, Hepple RT. Mitochondrial 114. Derbré F, Gomez-Cabrera MC, Nascimento AL, Mechanisms of Neuromuscular Junction Degeneration Sanchis-Gomar F, Martinez-Bello VE, Tresguerres JA, with Aging. Cells. 2020; 9:197. Fuentes T, Gratas-Delamarche A, Monsalve M, Viña J. https://doi.org/10.3390/cells9010197 Age associated low mitochondrial biogenesis may be PMID:31941062 explained by lack of response of PGC-1α to exercise www.aging-us.com 18082 AGING training. Age (Dordr). 2012; 34:669–79. 122. Milan L, Courtand G, Cardoit L, Masmejean F, Barrière https://doi.org/10.1007/s11357-011-9264-y G, Cazalets JR, Garret M, Bertrand SS. Age-Related PMID:21590341 Changes in Pre- and Postsynaptic Partners of the Cholinergic C-Boutons in Wild-Type and SOD1G93A 115. Feng H, Kang C, Dickman JR, Koenig R, Awoyinka I, Lumbar Motoneurons. PLoS One. 2015; 10:e0135525. Zhang Y, Ji LL. Training-induced mitochondrial https://doi.org/10.1371/journal.pone.0135525 adaptation: role of peroxisome proliferator-activated PMID:26305672 receptor γ coactivator-1α, nuclear factor-κB and β- blockade. Exp Physiol. 2013; 98:784–95. 123. Sunico CR, Domínguez G, García-Verdugo JM, Osta R, https://doi.org/10.1113/expphysiol.2012.069286 Montero F, Moreno-López B. Reduction in the PMID:23104933 motoneuron inhibitory/excitatory synaptic ratio in an early-symptomatic mouse model of amyotrophic 116. Jones RA, Harrison C, Eaton SL, Llavero Hurtado M, Graham LC, Alkhammash L, Oladiran OA, Gale A, lateral sclerosis. Brain Pathol. 2011; 21:1–15. https://doi.org/10.1111/j.1750-3639.2010.00417.x Lamont DJ, Simpson H, Simmen MW, Soeller C, PMID:20653686 Wishart TM, Gillingwater TH. Cellular and Molecular Anatomy of the Human Neuromuscular Junction. Cell 124. Saxena S, Roselli F, Singh K, Leptien K, Julien JP, Gros- Rep. 2017; 21:2348–56. Louis F, Caroni P. Neuroprotection through excitability https://doi.org/10.1016/j.celrep.2017.11.008 and mTOR required in ALS motoneurons to delay PMID:29186674 disease and extend survival. Neuron. 2013; 80:80–96. 117. Gallart-Palau X, Tarabal O, Casanovas A, Sábado J, https://doi.org/10.1016/j.neuron.2013.07.027 Correa FJ, Hereu M, Piedrafita L, Calderó J, Esquerda PMID:24094105 JE. Neuregulin-1 is concentrated in the postsynaptic 125. Pullen AH, Athanasiou D. Increase in presynaptic subsurface cistern of C-bouton inputs to α- territory of C-terminals on lumbar motoneurons of motoneurons and altered during motoneuron G93A SOD1 mice during disease progression. Eur J diseases. FASEB J. 2014; 28:3618–32. Neurosci. 2009; 29:551–61. https://doi.org/10.1096/fj.13-248583 PMID:24803543 https://doi.org/10.1111/j.1460-9568.2008.06602.x 118. Tarabal O, Caraballo-Miralles V, Cardona-Rossinyol A, PMID:19187267 Correa FJ, Olmos G, Lladó J, Esquerda JE, Calderó J. 126. Alvarez FJ, Titus-Mitchell HE, Bullinger KL, Kraszpulski Mechanisms involved in spinal cord central synapse M, Nardelli P, Cope TC. Permanent central synaptic loss in a mouse model of spinal muscular atrophy. J disconnection of proprioceptors after nerve injury and Neuropathol Exp Neurol. 2014; 73:519–35. regeneration. I. Loss of VGLUT1/IA synapses on https://doi.org/10.1097/NEN.0000000000000074 motoneurons. J Neurophysiol. 2011; 106:2450–70. PMID:24806302 https://doi.org/10.1152/jn.01095.2010 119. Cerveró C, Blasco A, Tarabal O, Casanovas A, PMID:21832035 Piedrafita L, Navarro X, Esquerda JE, Calderó J. Glial 127. Rotterman TM, Nardelli P, Cope TC, Alvarez FJ. Activation and Central Synapse Loss, but Not Normal distribution of VGLUT1 synapses on spinal Motoneuron Degeneration, Are Prevented by the motoneuron dendrites and their reorganization after Sigma-1 Receptor Agonist PRE-084 in the Smn2B/- nerve injury. J Neurosci. 2014; 34:3475–92. Mouse Model of Spinal Muscular Atrophy. J https://doi.org/10.1523/JNEUROSCI.4768-13.2014 Neuropathol Exp Neurol. 2018; 77:577–97. PMID:24599449 https://doi.org/10.1093/jnen/nly033 PMID:29767748 128. Salvany S, Casanovas A, Tarabal O, Piedrafita L, 120. Ling KK, Lin MY, Zingg B, Feng Z, Ko CP. Synaptic Hernández S, Santafé M, Soto-Bernardini MC, Calderó defects in the spinal and neuromuscular circuitry in a J, Schwab MH, Esquerda JE. Localization and dynamic mouse model of spinal muscular atrophy. PLoS One. changes of neuregulin-1 at C-type synaptic boutons in 2010; 5:e15457. association with motor neuron injury and repair. https://doi.org/10.1371/journal.pone.0015457 FASEB J. 2019; 33:7833–51. PMID:21085654 https://doi.org/10.1096/fj.201802329R 121. Mentis GZ, Blivis D, Liu W, Drobac E, Crowder ME, PMID:30912977 Kong L, Alvarez FJ, Sumner CJ, O’Donovan MJ. Early 129. Rekling JC, Funk GD, Bayliss DA, Dong XW, Feldman JL. functional impairment of sensory-motor connectivity Synaptic control of motoneuronal excitability. Physiol in a mouse model of spinal muscular atrophy. Neuron. Rev. 2000; 80:767–852. 2011; 69:453–67. https://doi.org/10.1152/physrev.2000.80.2.767 https://doi.org/10.1016/j.neuron.2010.12.032 PMID:10747207 PMID:21315257 www.aging-us.com 18083 AGING 130. Deardorff AS, Romer SH, Sonner PM, Fyffe RE. https://doi.org/10.1038/s41593-018-0242-x Swimming against the tide: investigations of the C- PMID:30258234 bouton synapse. Front Neural Circuits. 2014; 8:106. 140. Giorgetti E, Panesar M, Zhang Y, Joller S, Ronco M, https://doi.org/10.3389/fncir.2014.00106 Obrecht M, Lambert C, Accart N, Beckmann N, PMID:25278842 Doelemeyer A, Perrot L, Fruh I, Mueller M, et al. 131. Roselli F, Caroni P. From intrinsic firing properties to Modulation of Microglia by Voluntary Exercise or selective neuronal vulnerability in neurodegenerative CSF1R Inhibition Prevents Age-Related Loss of diseases. Neuron. 2015; 85:901–10. Functional Motor Units. Cell Rep. 2019; 29: https://doi.org/10.1016/j.neuron.2014.12.063 1539–54.e7. PMID:25741719 https://doi.org/10.1016/j.celrep.2019.10.003 PMID:31693894 132. Nijssen J, Comley LH, Hedlund E. Motor neuron vulnerability and resistance in amyotrophic lateral 141. Haynes SE, Hollopeter G, Yang G, Kurpius D, Dailey sclerosis. Acta Neuropathol. 2017; 133:863–85. ME, Gan WB, Julius D. The P2Y12 receptor regulates https://doi.org/10.1007/s00401-017-1708-8 microglial activation by extracellular nucleotides. Nat PMID:28409282 Neurosci. 2006; 9:1512–19. https://doi.org/10.1038/nn1805 PMID:17115040 133. Chen HH, Hippenmeyer S, Arber S, Frank E. Development of the monosynaptic stretch reflex 142. Mildner A, Huang H, Radke J, Stenzel W, Priller J. P2Y12 circuit. Curr Opin Neurobiol. 2003; 13:96–102. receptor is expressed on human microglia under https://doi.org/10.1016/s0959-4388(03)00006-0 physiological conditions throughout development and PMID:12593987 is sensitive to neuroinflammatory diseases. Glia. 2017; 65:375–87. 134. Rodríguez-Arellano JJ, Parpura V, Zorec R, https://doi.org/10.1002/glia.23097 PMID:27862351 Verkhratsky A. Astrocytes in physiological aging and Alzheimer’s disease. Neuroscience. 2016; 323: 143. Klein B, Mrowetz H, Barker CM, Lange S, Rivera FJ, 170–82. Aigner L. Age Influences Microglial Activation After https://doi.org/10.1016/j.neuroscience.2015.01.007 Cuprizone-Induced Demyelination. Front Aging PMID:25595973 Neurosci. 2018; 10:278. https://doi.org/10.3389/fnagi.2018.00278 135. Spittau B. Aging Microglia-Phenotypes, Functions and PMID:30297998 Implications for Age-Related Neurodegenerative Diseases. Front Aging Neurosci. 2017; 9:194. 144. Odoemelam CS, Percival B, Wallis H, Chang M-W, https://doi.org/10.3389/fnagi.2017.00194 Ahmad Z, Scholey D, Burton E, Williams IH, Kamerlin PMID:28659790 CL, Wilson PB. G-Protein coupled receptors: structure 136. Philips T, Robberecht W. Neuroinflammation in and function in drug discovery. RSC Advances. 2020; amyotrophic lateral sclerosis: role of glial activation in 10:36337–48. motor neuron disease. Lancet Neurol. 2011; 10: https://doi.org/10.1039/D0RA08003A 253–63. 145. Rotterman TM, Akhter ET, Lane AR, MacPherson KP, https://doi.org/10.1016/S1474-4422(11)70015-1 García VV, Tansey MG, Alvarez FJ. Spinal Motor Circuit PMID:21349440 Synaptic Plasticity after Peripheral Nerve Injury 137. Papadimitriou D, Le Verche V, Jacquier A, Ikiz B, Depends on Microglia Activation and a CCR2 Przedborski S, Re DB. Inflammation in ALS and SMA: Mechanism. J Neurosci. 2019; 39:3412–33. sorting out the good from the evil. Neurobiol Dis. https://doi.org/10.1523/JNEUROSCI.2945-17.2019 2010; 37:493–502. PMID:30833511 https://doi.org/10.1016/j.nbd.2009.10.005 146. Casanovas A, Salvany S, Lahoz V, Tarabal O, Piedrafita PMID:19833209 L, Sabater R, Hernández S, Calderó J, Esquerda JE. 138. Colonna M, Butovsky O. Microglia Function in the Neuregulin 1-ErbB module in C-bouton synapses on Central Nervous System During Health and somatic motor neurons: molecular compartmentation Neurodegeneration. Annu Rev Immunol. 2017; and response to peripheral nerve injury. Sci Rep. 35:441–68. 2017; 7:40155. https://doi.org/10.1146/annurev-immunol-051116- https://doi.org/10.1038/srep40155 052358 PMID:28226226 PMID:28065942 139. Hickman S, Izzy S, Sen P, Morsett L, El Khoury J. 147. Miller RA, Nadon NL. Principles of animal use for Microglia in neurodegeneration. Nat Neurosci. 2018; gerontological research. J Gerontol A Biol Sci Med Sci. 21:1359–69. 2000; 55:B117–23. www.aging-us.com 18084 AGING https://doi.org/10.1093/gerona/55.3.b117 https://doi.org/10.1016/j.clinph.2011.01.045 PMID:10795715 PMID:21354365 148. Burkholder T, Foltz C, Karlsson E, Linton CG, Smith JM. 152. Bonetto A, Andersson DC, Waning DL. Assessment of Health Evaluation of Experimental Laboratory Mice. muscle mass and strength in mice. Bonekey Rep. Curr Protoc Mouse Biol. 2012; 2:145–65. 2015; 4:732. https://doi.org/10.1002/9780470942390.mo110217 https://doi.org/10.1038/bonekey.2015.101 PMID:22822473 PMID:26331011 149. Brooks SP, Dunnett SB. Tests to assess motor 153. Hebel R, Stromberg MW. Myology. Anatomy of the phenotype in mice: a user’s guide. Nat Rev Neurosci. laboratory rat. Baltimore, U.S.A.: Williams and 2009; 10:519–29. Wilkings Company, pp.18–42. https://doi.org/10.1038/nrn2652 PMID:19513088 154. Cerveró C, Montull N, Tarabal O, Piedrafita L, 150. Navarro X, Udina E, Ceballos D, Gold BG. Effects of Esquerda JE, Calderó J. Chronic Treatment with the FK506 on nerve regeneration and reinnervation after AMPK Agonist AICAR Prevents Skeletal Muscle graft or tube repair of long nerve gaps. Muscle Nerve. Pathology but Fails to Improve Clinical Outcome in a 2001; 24:905–15. Mouse Model of Severe Spinal Muscular Atrophy. https://doi.org/10.1002/mus.1088 Neurotherapeutics. 2016; 13:198–216. PMID:11410918 https://doi.org/10.1007/s13311-015-0399-x PMID:26582176 151. Mancuso R, Santos-Nogueira E, Osta R, Navarro X. Electrophysiological analysis of a murine model of motoneuron disease. Clin Neurophysiol. 2011; 122:1660–70. www.aging-us.com 18085 AGING SUPPLEMENTARY MATERIALS Supplementary Figures Supplementary Figure 1. Effects of GTE- and cocoa-supplemented diets on CGRP, agrin and GAP-43 expression in NMJs of aged muscles. The quantification of CGRP- (A) agrin- (D) and GAP-43- (G) immunoreactivity (IR, arbitrary units, [a.u.]) in NMJs of TA and Sol muscles from control, GTE and cocoa groups is shown. Data in the graphs are expressed as the mean ± SEM, and shown as the percentage of change compared with controls; two-way ANOVA, Bonferroni’s; post hoc test 40-70 NMJs per muscle type were examined; one TA and Sol muscle per animal (3, control; 4 GTE; and 5 cocoa) were used for analysis. (B1–C3, E1–F3, H1–I3) Representative confocal micrographs of NMJs from Sol and TA muscle sections immunolabeled for either, CGRP, agrin or GAP-43 (all in green), as indicated; sections were also stained with α-Bgtx (red) for endplate visualization. Note the slightly, although non-significant increase in agrin-immunoreactivity at NMJs, particularly of Sol muscles, of animals fed with the GTE-supplemented diet. (J) Densitometric analysis of GAP-43 content in TA and Sol muscles of animals belonging to the three experimental groups; data were normalized to GAPDH. Bars represent the values (mean ± SEM) of 3 mice per condition from 2 independent western blot analysis; two-way ANOVA (Bonferroni’s post hoc test). (K) Representative western blots of GAP-43 and GAPDH (as loading control) proteins in TA and Sol muscles from mice of control (Ctrl), GTE and cocoa groups. Scale bar in I3 = 20 μm (valid for B1–C3, E1–F3, H1–I2). www.aging-us.com 18086 AGING Supplementary Figure 2. Effects of GTE- and cocoa-supplementations in motor axons of aged mice. Analysis was performed in L4 VRs. (A–C) VR size (in µm2, A), total number of axons (B) and axon diameter (including myelin sheath, in µm, [C]) in VRs. (D–F) Proportion of healthy myelinated axons (D), axons exhibiting a degenerate appearance (E), and those with myelin inclusions (F). (G, H) Number of small (≤6 µm diameter) axon clusters (≥3 axons) (G) and number of small axons per cluster (H) in VRs. (I–K) Representative images of methylene blue-stained semithin cross-sections of VRs from mice belonging to control (I), GTE (J) and cocoa (K) groups; arrows indicate: degenerating axons (red), axons with myelin inclusions (green) and clusters of small axons (orange). (L, M) Quantification of myelin thickness by g-ratio analysis (L) and scatter plot depicting g-rations in relation to axon diameter in (M) VRs. (N) Relative frequency of myelinated axon diameter (in µm); note the slight displacement to left of GTE and cocoa curves, indicative of a diameter reduction of motor axons of animals subjected to dietary flavonoid intake. Data in the graphs are expressed as the mean ± SEM of 3 VRs from different mice per condition. Scale bar in (K) = 25 µm (valid for [I] and [J]). www.aging-us.com 18087 AGING Supplementary Figure 3. Effects of GTE- and cocoa-supplemented diets on aging-associated changes in the proximal hindlimb Gra muscle. (A) Proportion of myofibers displaying central nuclei. (B) Percentage of Pax7-immunostained cells (SCs) with respect to DAPI-positive nuclei. Representative images of a combined immunolabeling for Pax7 (red) and laminin (green), and DAPI staining (blue) in transversal cryosections of Gra muscles from control, GTE and cocoa groups, as indicated. (F, G) Percentage of myofibers containing lipofuscin aggregates (F), and average number of lipofuscin granules per myofiber (G). Data in graphs are expressed as the mean ± SEM. The average values of different parameters analyzed in adult Gra muscles in a previous study [6] were indicated by a red line in each graph for comparative purposes. ***p < 0.001 vs. control (Ctrl), two-way ANOVA, Bonferroni’s post hoc test; sample size (myofibers): (A) = 1500-2500, (B) = 1700-2200, (F, G) = 1200-1700, from 3-5 mice per condition. Scale bars: 40 µm in E3 (valid for C1–E2). www.aging-us.com 18088 AGING Supplementary Figure 4. Changes in the innervation degree of NMJs of Gra muscle in animals fed with GTE- or cocoa- supplemented diets. (A) Proportion of NMJs displaying different degrees of innervation; quantification was based on the percentage of α- Bgtx-labeled postsynaptic site area covered by SV2-immunostained presynaptic terminals (see Materials and Methods, <15% innervation was considered as denervated). (B) Number of NMJs exhibiting single (mono.) or multiple (poly.) innervation expressed as the percentage of control (blue dashed line). (C1–E3) Representative maximal projections of confocal stacks of NMJs from mice of control, GTE and cocoa groups (as indicated in panels); muscle sections were stained with antibodies against NF and SV2 (green, for presynaptic nerve terminals), and α-Bgtx (red, for postsynaptic AChR). Data in graphs are expressed as the mean ± SEM, ***p < 0.001 vs. control, one or two-way ANOVA, Bonferroni’s post hoc test; sample size: 40-64 NMJs per muscle from 3-5 animals per condition. Scale bar in (E3) = 20 µm (valid for C1–E2). www.aging-us.com 18089 AGING Supplementary Figure 5. Morphometric analysis of microglial complexity by using the skeleton approach (see Materials and Methods). The average branch length (in µm) (A), number of endpoints (B) and triple points (C) are shown in graphs. (D1–F3) Representative images showing Iba1-positive profiles (D1–F1), the skeletonized images (D2–F2), and the overlay of the skeleton and the original image of microglia (D3–F3) in control, GTE and cocoa groups as indicated in panels. Data in the graphs are expressed as the mean ± SEM; spinal cords of 3 control (Ctrl)-, 4 GTE- and 5 cocoa-group animals were used for analysis; a total of 50-130 images per experimental group were examined; the average values of the different parameters analyzed in adult animals [6] are shown in graphs as a red line for comparison. Scale bar in F1 = 50 μm (valid for D1, E1). www.aging-us.com 18090 AGING Supplementary Figure 6. Impact of GTE- and cocoa-supplemented diets in age-related changes in DRG sensory neurons. Proportion (A, F, K) and soma area (in µm2, B, G, L) of neurons expressing PV (A, B) or CGRP (F, G), or stained with IB4 (K, L). (C1–E2, H1–J2, M1–O2) Representative micrographs obtained from DRG cryostat sections processed for PV, CGRP and IB4 staining (green) as indicated in panels; sections were also counterstained with fluorescent Nissl (red) for neuron visualization. Data in graphs are expressed as the mean ± SEM, ***p < 0.001 vs. control (Ctrl) (one-way ANOVA, Bonferroni’s post hoc test); 30-80 images for each condition were analyzed; number of DRG examined: Ctrl =2, GTE = 3, cocoa = 5 from different animals; the red lines in graphs indicate the mean values in adult mice [6]. Scale bar: in O2 = 100 µm (valid for C1–E2, H1–J2, M1–O1). www.aging-us.com 18091 AGING Supplementary Figure 7. Motor behavior analysis. Motor performance shown as the distance (in cm, A) and average speed (cm/s, B) assessed by the open-field test in mice from the three groups during the experimental period; the age-related decline in motor performance observed in the animals subjected to the standard (control) diet was not prevented by either GTE or cocoa supplementations; conversely, open-field parameters were even worsened in mice subjected to these dietary supplementations. Data are shown as the mean ± SEM (number of animals per group: 91 weeks of age, n = 15 in all groups; 107 weeks, control n = 8, GTE n = 13; cocoa n = 15; 114 weeks, control n = 6, GTE n = 12, cocoa n = 11); *p < 0.05 and **p < 0.01 vs. control; multiple t-test. www.aging-us.com 18092 AGING Supplementary Table Supplementary Table 1. Results from motor and sensory nerve conduction tests performed in old mice (24 months) fed with a standard (AIN-93M) diet (control), or the AIN-93M diet supplemented with either GTE- or cocoa flavanols. Parameter Old control (n=7) Old GTE (n=5) Old cocoa (n=5) TA Lat peak (ms) 1.55 ± 0.08 1.52 ± 0.03 1.48 ± 0.02 TA CMAP (mV) 50.5 ± 2.7 47.0 ± 4.9 53.4 ± 4.7 SMUA (mV) 1.20 ± 0.08 1.01 ± 0.04 1.18 ± 0.12 MUNE 40.8 ± 1.5 44.5 ± 2.6 44.9 ± 0.7 PL Lat peak (ms) 2.44 ± 0.19 2.64 ± 0.07 2.34 ± 0.10 PL CMAP (mV) 4.60 ± 0.55 4.22 ± 0.75 4.80 ± 0.58 PL H wave Lat (ms) 6.92 ± 0.40 7.51 ± 0.36 6.52 ± 0.21 PL H/M ratio (%) 32.1 ± 3.6 28.0 ± 8.8 17.8 ± 3.3 DgN Lat (ms) 1.39 ± 0.06 1.29 ± 0.07 1.32 ± 0.05 DgN CNAP (µV) 29.1 ± 1.8 31.0 ± 5.4 32.0 ± 4.7 Weight (g) 29.7 ± 0.7 25.1 ± 1.3 27.8 ± 0.9 TA, tibialis anterior muscle; PL, plantar interosseus muscle; DgN, 4th digital nerve; Lat, latency at the peak of the wave; CMAP, compound muscle action potential; CNAP, compound nerve action potential; MUNE, motor unit number estimation; SMUA, singer motor unit amplitude. Statistical comparisons between groups were performed with one-way ANOVA (Bonferroni’s post hoc test); no significant differences were found. www.aging-us.com 18093 AGING