Long-Term High-Level Exercise Promotes Muscle Reinnervation With Age

ORIGINAL ARTICLE

Long-Term High-Level Exercise Promotes MuscleReinnervation With Age

Simone Mosole, BS, Ugo Carraro, MD, Helmut Kern, MD, Stefan Loefler, Eng, Hannah Fruhmann, PT,Michael Vogelauer, MD, Samantha Burggraf, Eng, Winfried Mayr, PhD, Matthias Krenn, DI,Tatjana Paternostro-Sluga, PhD, Dusan Hamar, MD, Jan Cvecka, PhD, Milan Sedliak, PhD,Veronika Tirpakova, PhD, Nejc Sarabon, PhD, Antonio Musaro, PhD, Marco Sandri, MD,

Feliciano Protasi, PhD, Alessandra Nori, PhD, Amber Pond, PhD, and Sandra Zampieri, PhD

AbstractThe histologic features of aging muscle suggest that denervation

contributes to atrophy, that immobility accelerates the process, andthat routine exercise may protect against loss of motor units andmuscle tissue. Here, we compared muscle biopsies from sedentaryand physically active seniors and found that seniors with a longhistory of high-level recreational activity up to the time of musclebiopsy had 1) lower loss of muscle strength versus young men (32%loss in physically active vs 51% loss in sedentary seniors); 2) fewersmall angulated (denervated) myofibers; 3) a higher percentage offiber-type groups (reinnervated muscle fibers) that were almost ex-clusive of the slow type; and 4) sparse normal-size muscle fiberscoexpressing fast and slow myosin heavy chains, which is notcompatible with exercise-driven muscle-type transformation. Thebiopsies from the old physically active seniors varied from sparsefiber-type groupings to almost fully transformed muscle, suggestingthat coexpressing fibers appear to fill gaps. Altogether, the data showthat long-term physical activity promotes reinnervation of musclefibers and suggest that decades of high-level exercise allow the bodyto adapt to age-related denervation by saving otherwise lost musclefibers through selective recruitment to slow motor units. These ef-fects on size and structure of myofibers may delay functional declinein late aging.Trial registration: ClinicalTrials.gov (NCT01679977).

Key Words: Aging, Coexpression of fast and slow myosin heavychains, Denervation and reinnervation, Fiber-type grouping, Humanskeletal muscle, Recreational sport activity.

INTRODUCTIONAging is characterized by a gradual decline that impairs

cell homeostasis and functional reserves. Degeneration, apo-ptosis, and death (accompanied by loss of regenerativecapacity) of all cell types progressively accumulate and ulti-mately lead to organism death (1Y4). Histologic studies ofskeletal muscle have shown that denervation is among the nu-merous mechanisms that contribute to tissue atrophy and de-generation in aging (5, 6). The term ‘‘disseminated neurogenicatrophy’’ was coined to describe the progressive accumula-tion and clustering of small angulated fibers with aging (7Y10);there is also evidence of progressive loss of >-motoneurons(11, 12). Electrophysiologic studies have confirmed that thereis a decrease in the number of motor units with a concomi-tant increase in their size with age. These results suggest thatsome reinnervation events follow muscle fiber denervation(13). Further evidence supporting the occurrence of roundsof denervation and reinnervation includes the increased clus-tering of myofiber types in the motor units of rodents andother mammals as they age (11, 14). In adult humans, fiber

J Neuropathol Exp Neurol � Volume 73, Number 4, April 2014284

J Neuropathol Exp NeurolCopyright � 2014 by the American Association of Neuropathologists, Inc.

Vol. 73, No. 4April 2014

pp. 284Y294

From the Laboratory of Translation Myology, Department of BiomedicalSciences, University of Padua, Padua, Italy (SM, UC, AN, SZ); LudwigBoltzmann Institute of Electrical Stimulation and Physical Rehabilita-tion, Vienna, Austria (HK, SL, HF, SB, SZ); Department of PhysicalMedicine and Rehabilitation, Wilhelminenspital, Vienna, Austria (HK,MV); Center of Biomedical Engineering and Physics, Medical Uni-versity of Vienna, Vienna, Austria (WM, MK); Department of PhysicalMedicine and Rehabilitation, Medical University of Vienna, Vienna,Austria (TP-S); Faculty of Physical Education and Sport, ComeniusUniversity, Bratislava, Slovakia (DH, JC, MSedliak, VT); Science andResearch Centre, Institute for Kinesilogical Research, University ofPrimorska, Koper, Slovenia (NS); DAHFMO-Unit of Histology andMedical Embryology, IIM, Sapienza University of Rome, Rome, Italy(AM); Venetian Institute of Molecular Medicine, Dulbecco Telethon Insti-tute, and Department of Biomedical Science, University of Padua, Padua,Italy (MSandri); CeSI-Center for Research on Aging, and DNI-Departmentof Neuroscience and Imaging, University G. d’Annunzio of Chieti, Chieti,Italy (FP); and Anatomy Department, Southern Illinois University School ofMedicine, Carbondale, Illinois (AP).

Send correspondence and reprint requests to: Sandra Zampieri, PhD, Laboratory ofTranslational Myology, Department of Biomedical Sciences, University ofPadua, Viale G. Colombo, 3 I-35121 Padua, Italy; E-mail: [email protected]

Simone Mosole and Ugo Carraro contributed equally.This study was supported by the European Regional Development Fund-Cross

Border Cooperation Programme Slovakia-Austria 2007Y2013 (Interreg-Iva), project Mobilitat im Alter, MOBIL, N_00033 (partners: LudwigBoltzmann Institute of Electrical Stimulation and Physical Rehabilitation,Austria; Center for Medical Physics and Biomedical Engineering, MedicalUniversity of Vienna, Vienna, Austria; and Faculty of Physical Educationand Sports, Comenius University, Bratislava, Slovakia); Austrian nationalcofinancing of the Austrian Federal Ministry of Science and Research;Austrian national cofinancing of the Medical University of Vienna, Vienna,Austria; Ludwig Boltzmann Society, Vienna, Austria; and institutional fundsto Ugo Carraro, Antonio Musaro, Marco Sandri, and Feliciano Protasi ofMinistero dell’Istruzione, dell’Universita e della Ricerca, Italy.

Supplemental digital content is available for this article. Direct URL citationsappear in the printed text and are provided in the HTML and PDF versionsof this article on the journal’s Web site (www.jneuropath.com).

Copyright © 2014 by the American Association of Neuropathologists, Inc. Unauthorized reproduction of this article is prohibited.

http://www.jneuropath.com

types appear randomly distributed across the muscle andbecome increasingly grouped with age (15). Therefore, it hasbeen proposed that, in addition to axonal disorders, apoptosisof >-motoneurons in the spinal cord, with subsequent incom-plete reinnervation of fibers by surviving motor neurons, maycontribute to loss of muscle strength and mass as people growolder (16). These rearrangement processes are generally ac-companied by a progressive increase in the proportion of slowmuscle fibers, although there is some evidence to the contrary(17). Some of the discrepancies have been dispelled by com-parisons of muscle from normally active and immobile olderpatients that show that muscle wasting in ‘‘normally active’’seniors is accompanied by a shift toward a slow-twitch phe-notype, whereas inactive seniors demonstrates a shift towardfast-twitch isoform expression. This latter case is common in‘‘unloaded’’ muscle undergoing atrophy, for example, duringlimb suspension, immobilization, paralysis, and spaceflight(18Y22). To complicate the situation further, conflicting re-sults regarding fast to slow myosin transition arise in endur-ance training studies using animal models and in clinical trialsof humans involving either voluntary exercise or electricalstimulationVboth directly to denervated muscle and indirectlyto muscle through nerve stimulation (19, 21, 23Y27). Further-more, increased exercise that is sustained for decades (e.g.training as performed by track and field masters athletes) pro-tects against age-related loss of motor units (28Y30) and,thereby, of lean muscle mass (31). However, the degree towhich denervation causes loss of myofibers is an open questionbecause reinnervation events may compensate for motor neu-ron loss during aging as well as with spinal cord injury and/oraxonal abnormalities of peripheral nerves (13Y15, 32Y34).Whether the aging-related shifts are under neural control orthe result of the direct influence of use/disuse on myogenicprocesses remains to be clarified.

In the present study, we analyzed muscle biopsies har-vested from the vastus lateralis of young men (aged 22Y33 years),sedentary seniors (aged 67Y77 years), and senior amateur ath-letes (aged 65Y79 years). The latter routinely practiced sportactivities usually more than 3 times per week up to the time ofbiopsy. In agreement with previous studies of masters athletes(35Y37), we show that long-term high-level physical activityconsiderably increases the percentage of slow-type myofibersand the number of muscle fiberYtype groupings. The latterprovides direct evidence that long-term cycles of denervation/reinnervation occurred. In recent interim reports (38Y40), weshowed, and here confirm, that muscle properties of these

senior recreational athletes are more similar to those of activeyoung men than to those of sedentary seniors.

By analyzing coexpression of fast and slow myosinheavy-chain (MHC) isoforms in the muscle biopsies, we showfor the first time that these events occur with recreationalphysical activity in seniors and that the changes may be re-lated to selective reinnervation events. Our study supportsthe concept that long-term high-level exercise has beneficialeffects on reinnervation of muscle fibers, resulting in preser-vation of muscle function, size, structure, and ultrastructure(41Y43) and thereby delaying mobility decline and loss ofindependence that are commonly seen in aging.

MATERIALS AND METHODS

Study SubjectsApproval from the national committee for medical

ethics was obtained before the study onset (EK08-102-0608).With the exception of 2 female subjects in the sedentarygroup, recruited subjects were male volunteers. All subjectsreceived detailed information on the study and gave informedconsent. Three groups were enrolled: young men (n = 5; aged22Y33 years; 10 biopsies); seniors with a sedentary lifestyle(n = 6; aged 67Y74 years; 10 biopsies), and seniors with a longhistory of high-level recreational sport activities (n = 7; aged 65to 79 years; 10 biopsies) (Table, Supplemental Digital Content 1,http://links.lww.com/NEN/A541). All subjects were healthyand declared not to have any specific mobility impairment ordisease. All of the subjects declared that they had no pre-scriptions for anti-inflammatory therapy related to neuropathiesor myopathies. The seniors were not outpatients of anyrehabilitation clinic but were located by newspaper advertise-ments that stressed that only healthy subjects without mobil-ity impairment would be enrolled. Sedentary seniors wereenrolled on the basis of their declaration that they had notperformed any routine physical activity/training during theprevious 10 years. On enrollment in the study, needle musclebiopsies were obtained from the vastus lateralis musclesthrough a small skin incision (6 mm); the biopsied tissue wasthen frozen for light microscopy or fixed for electron micros-copy, as described (21, 41Y43).

Light and Immunofluorescence MicroscopySerial cryosections (8-Km thick) from frozen muscle

biopsies were mounted on Polysine glass slides, air-dried, and

TABLE 1. Denervated Fibers in Young Men and in Sedentary and Physically Active SeniorsMyofiber Diameter

G30 Km G25 Km

Subjects No. Biopsies % ANOVA % ANOVA

Young men 10 0.4 T 0.5* Yes 0.2 T 0.5* Yes (vs sedentary)

Seniors

Sedentary 10 6.5 T 3.8† Yes 2.6 T 1.9† Yes (vs physically active)

Physically active 10 1.8 T 3.9* No 0.4 T 1.1* No (vs young men)

Occurrences of small fibers are expressed as percentages of the total muscle fibers. Parentheses apply to both columns.*†Within columns, groups with different footnote symbols are statistically different, p G 0.05. Yes and No refer to significant differences by ANOVA.

J Neuropathol Exp Neurol � Volume 73, Number 4, April 2014 High-Level Activity Promotes Muscle Reinnervation

� 2014 American Association of Neuropathologists, Inc. 285


http://links.lww.com/NEN/A541

stained with hematoxylin and eosin (42) or immunostainedfor either fast or slow MHC, laminin, or neural cell adhesionmolecule (N-CAM), as described below.

For MHC, sections were washed with PBS and perme-abilized with 0.1% Triton (Sigma-Aldrich, St. Louis, MO) inPBS for 15 minutes. After a PBS wash, nonspecific proteininteractions were blocked by incubation with 10% fetal bo-vine serum in PBS for 30 minutes at room temperature (RT).The sections were then incubated for 1 hour at RT in primarymouse monoclonal anti-MHC fast or anti-MHC slow anti-body (Novocastra, Milan, Italy) diluted 1:10 in PBS. Thesections were subsequently washed in PBS and incubatedwith antiYmouse Cy3 secondary antibody (1:100; Sigma-Aldrich) for MHC slow and with antiYmouse FITC (1:100;Sigma-Aldrich) for MHC fast for 1 hour. The sections werewashed again in PBS, and coverslips were mounted onto theglass slides using ProLong Gold antifade reagent with DAPI(Life Technologies, Carlsbad, CA). Images were acquiredusing a Zeiss microscope connected to a Leica DC 300Fcamera.

For detection of N-CAMYexpressing myofibers, sec-tions were fixed in methanol for 15 minutes at 20-C andthen labeled for 1 hour at RT using rabbit polyclonal anti-body directed against N-CAM (cat. no. AB5032; Chemicon,Millipore, Milan, Italy) diluted 1:200 in PBS (44, 45). Sec-tions were rinsed 3� 5 minutes in PBS and then incubated for1 hour at RT with Cy3-labeled conjugate directed againstrabbit IgG (Chemicon, Millipore) diluted 1:200 in 10% goatserum in PBS. Negative controls were performed by omit-ting the primary antibodies from sample incubations. Afterwashes, nuclei were counterstained for 5 minutes at RT withHoechst 33258 (Sigma-Aldrich); sections were then cover-slipped using mounting medium (Dako, Glostrup, Denmark)and observed under a Zeiss microscope connected to a LeicaDC 300F camera.

For coimmunolocalization of fast and slow MHC insingle sections, the sections were washed, permeabilized,washed again, and incubated with blocking solution as pre-viously described. They were then incubated for 1 hour at RTwith mouse antiYMHC slow primary monoclonal antibody(Sigma-Aldrich) diluted 1:10 in PBS and secondarily withrabbit anti-laminin (Sigma-Aldrich) diluted 1:100 in PBS.Next, the sections were washed in PBS and incubated with theanti-mouseYAlexa 594 (1:200; Life Technologies) secondaryantibody and the antiYrabbit FITC antibody (1:200; Sigma-Aldrich) for 1 hour. After a PBS wash, the sections wereincubated for 1 hour with an antiYMHC fast primary mono-clonal antibody produced in mouse (1:10; Novocastra). Thesections were then washed with PBS and incubated for an-other hour with an anti-mouseYAlexa 488 secondary antibody(1:200; Life Technologies). After another wash with PBS,coverslips were mounted onto the glass slides using ProLongGold antifade reagent with DAPI (Life Technologies).

Morphometric AnalysesMorphometric analyses of the fiber diameter and of the

fiber-type distribution were performed on cryosections usingScion Image for Windows version Beta 4.0.2 (2000 ScionCorporation), as previously described (42Y47).

Slow Fiber CorrelationsTo determine whether there was a correlation between

slow fibers and the type of training undertaken by the physicallyactive seniors (i.e. endurance, strength, or mixed training), per-centages of slow muscle fibers were plotted against the proba-bility that the value for each biopsy fell within the area under thecurve. Excel equation DISTRIB.NORM(X;Mean;Dev_standard;Cumulative):

f ðxÞ ¼ 1ffiffiffiffiffiffiffiffiffiffiffi

2PR2p e

ðx� KÞ2

2R2

Statistical AnalysisAnalysis of variance (ANOVA) tests were performed

with statistical algorithms of Origin (OriginLab Corp.,Northampton, MA). The level of statistical significance wasset at p G 0.05.

RESULTS

Demography and Clinical CharacteristicsIndicated That the Enrolled Subjects WereHealthy and Mobile

Detailed demographic and clinical characteristics of theenrolled subjects are described in (Table, Supplemental Dig-ital Content 1, http://links.lww.com/NEN/A541). All subjectswere healthy and declared not to have any specific mobil-ity impairment or disease. Nonetheless, clinical and functionalevaluations, in addition to electromyographic analyses, wereperformed in a few physically active seniors, resulting in de-tection of some neuropathic or myopathic features (Table, Sup-plemental Digital Content 1, http://links.lww.com/NEN/A541).

Amount of Weekly Physical Activity and KneeExtension Strength Are Significantly Higher inPhysically Active Versus Sedentary Seniors

Amounts and types of physical exercise performed by theactive subjects are detailed in Table, Supplemental DigitalContent 2, http://links.lww.com/NEN/A542. Four of the 5 youngmen performed strength training; the fifth did endurance ac-tivity. They declared that they exercised 4.0 to 7.5 hours perweek during the previous 5 years. The physically active seniorstrained 4.5 to 24.0 hours per week. Thus, a few seniors spentmore time training than the young men; however, the apparentlarge difference between the mean times (11.7 T 7.3 vs 5.3 T1.4 hours per week T SD, respectively) was not significant.Sedentary seniors had not performed any exercise abovenormal everyday living activities throughout the previous20 years.

To permit a comparison of specific muscle strengthamong the groups, quadriceps force was measured (Table, Sup-plemental Digital Content 2, http://links.lww.com/NEN/A542).The mean (TSD) knee contraction strength in young men was3.21 T 0.55 Nm/kg; it was 2.17 T 0.42 in the physically activeseniors (a decrease of 32% relative to the young men); it was1.57 T 0.39 in sedentary seniors (a decrease of 51% vs theyoung men). The performances in a battery of functional mo-bility tests of the physically active seniors were more similar to

Mosole et al J Neuropathol Exp Neurol � Volume 73, Number 4, April 2014

� 2014 American Association of Neuropathologists, Inc.286






those of young men than to those of the sedentary co-agedgroup (Kern et al, unpublished data). This is sound evidencethat the enrolled physically active seniors are a highly activegroup and are likely comparable to masters athletes (35Y37).

Small Angular Muscle Fibers in Both YoungMen and Seniors (Sedentary and Sportsmen)Had the Size and Morphology of DenervatedMuscle Fibers

Based on our experience with muscle biopsies fromspinal cordYinjured paraplegic patients with either disuse at-rophy resulting from lesions of the central motoneuron orextreme atrophy caused by lack of innervation secondary tocomplete peripheral motoneuron lesions, we identify musclefibers with a diameter less than 30 Km as denervated (20, 21,43Y47). Our interpretation of these myofibers as denervated isstrengthened by the facts that half of these small fibers actuallyhave diameters less than 25Km and that several have angulatedshapes. In the present study, serial sections of muscle biopsiesfrom the young men reveal that myofibers having a diameterless than 30 Km are infrequent (0.4%; Table 1) and that thosewith a diameter less than 25 Km are even less abundant (0.2%;Table 1); the vast majority of the muscle fibers in these sectionsare predominantly round (Fig. 1A, B). The muscle sectionsfrom the sedentary seniors (Figs. 1C, D; 2, 3) and physicallyactive seniors (Figs. 1E, F; 3) reveal more abundant musclefibers having diameters less than 30 and 25 Km (Table 1),and some of these have distinct angulation (Fig. 1, whitearrowheads). In addition, Figure 2 shows that, in sedentary

seniors, the small angular myofibers have appreciable expres-sion of N-CAM, an accepted marker of denervation (48). Thebiopsies taken from the sedentary seniors contain the highestpercentage of denervated muscle fibers having a diameter lessthan 30 Km (6.5%) and of those having a diameter less than25 Km (2.6%). These percentages are significantly higher thanthose in the other 2 groups, whereas the percentages of dener-vated fibers are not significantly different between the youngmen and physically active seniors (Table 1). Furthermore, theN-CAM staining was much less abundant in biopsies from theactive seniors and even less so in sections from the young men(not shown). Indeed, in a previous study, we observed only 1N-CAMYpositive muscle fiber among the 10,000 analyzed inthe biopsies from young men (49).

Percentages of Fast and Slow Myofibers inPhysically Active Seniors Showed a SignificantShift Toward Slow Fibers Relative to the YoungMen and the Sedentary Seniors

Immunofluorescence revealed both fast and slow MHCproteins in serial sections of muscle biopsies from the youngmen (Fig. 1A, B), with the fast fibers being slightly moreabundant than the slow fibers in these muscle sections (Table 2).Interestingly, this pattern of fiber-type distribution was notsignificantly different from that observed in the matched mus-cle of sedentary seniors, although the latter was slightly shiftedtoward the slow type relative to the young men (Fig. 1C, D;Table 2). However, in the active seniors (Fig. 1E, F), slowfibers were most numerous (68.5%; Fig. 1E, F); the increase

FIGURE 1. (AYF) Immunofluorescence staining for fast (A, C, E) and slow (B, D, F) myosin heavy chain (MHC) proteins in serialsections of biopsies from young men (A, B), sedentary seniors (C, D), and physically active seniors (E, F). White arrows point tosmall angulated fibers; white circles show fiber-type groups. Physically active seniors have fewer small fibers and more numerousslow fiberYtype clusters than sedentary seniors. Fiber sizes in the biopsies of physically active seniors are comparable to those of theyoung men.

FIGURE 2. Neural cell adhesion molecule (N-CAM), a marker of denervation and aneural regeneration, is expressed in myofibers ofbiopsies from seniors. (AYC) Small angulated N-CAMYpositive myofibers (red) are detected in muscle biopsies from sedentaryseniors. Based on their distinct morphologic aspects, size, and N-CAM expression, these myofibers are denervated. Myonuclei arecounterstained in blue. Scale bar = 100 Km.




was significant versus both the young men (42% slow fibers)and the sedentary seniors (46% slow fibers) (Table 2).

Fiber-type grouping was almost absent in the youngmen; however, grouping was greater in the old subjects, withthe physically active seniors having the greatest number; mostof these were of the slow type, whereas those in sedentaryseniors were mainly of the fast type. Fiber-type groupings areidentified on the basis that at least 1 muscle fiber is completelysurrounded by fibers of the same phenotype. Percentages offiber-type groupings are determined by counting the numberof muscle fibers in the biopsy that are surrounded by fibers ofthe same type and then dividing this number by the totalnumber of fibers. To avoid problems related to the existenceof many different fast MHC isoforms, we used an antiYfastMHC antibody that does not discriminate among the fastisoforms; therefore, we describe fiber-type clusters as eitheronly ‘‘slow’’ or only ‘‘fast.’’ Muscle sections from the youngmen had few fiber-type groupings, and those that weredetected were mainly of the fast type (1%; Fig. 1; Table 3). Inthe sedentary seniors, although both fast- and slow-typegroupings were present, the fast type (3.0%) were more nu-merous than the slow type (0.5%) (Table 3). Most notable isthe fact that biopsies taken from physically active seniors hadthe highest percentage of slow-type fiber groupings, with amean of 7.9%, reaching almost 25% in extreme cases, in which93% of total myofibers were of the slow type (Table 3). It isworth stressing that the physically active seniors with the mostsevere neuropathic or myopathic electromyographic results werenot the subjects with the higher content of slow-type fibers andslow fiberYtype groupings (Table, Supplemental Digital Content1, http://links.lww.com/NEN/A541 and Table, SupplementalDigital Content 3, http://links.lww.com/NEN/A543).

Muscle Fibers Coexpressing Fast and Slow MHCMuscle fibers coexpressing fast and slow MHCs were

sparse in all of the analyzed biopsies, and there were no sig-nificant differences among these groups in terms of this pa-rameter (Table 4). However, the serial sections from sedentaryseniors that did show MHC coexpression were often smallangular (denervated) muscle fibers (Fig. 1C, D, white arrows);therefore, we suggest that these are slow myofibers coexpressingfast isoforms of MHC by default myogenic programs. In con-trast, the muscle fibers in the physically active seniors that were

positive for both fast (green) and slow (red) MHC proteins weresimilar in size to the pure fast or pure slow myofibers (Fig. 4C).In Figure 4C, it is interesting to note that, although some fibersare green and others are red, not all of the fibers have the samecolor intensity. This might indicate that these fibers contain somevariable combinations of fast and slow MHCs but not enoughof both to produce the orange color of coexpression.

Slow Fiber CorrelationsThere was a strong correlation between the slow fiber

percentages and slow fiberYtype groupings in the physicallyactive seniors (R2 = 0.82; Fig. 5), but there was no correlationbetween the percentages of slow fibers and the prevalent kindsof training undertaken by the physically active seniors (Fig. 6).Indeed, the marks indicating whether the subjects had performedmainly strength training, endurance training, or a mixture of bothstrength and endurance trainings (mixed training) were randomlydistributed both among scarcely or highly transformed musclebiopsies.

The ages of the seniors were not correlated with per-centages of slow fibers and slow fiberYtype groupings in ei-ther group of seniors as follows: 1) Sedentary seniors, ageversus percentage of slow fibers, R2 = 0.12; age versus per-centage of slow-type groupings, R2 = 0.02; and 2) Physicallyactive seniors, age versus percentage of slow fibers, R2 = 0.25;age versus percentage of slow-type groupings, R2 = 0.27.There was no correlation even when the 3 groups were pooled:age versus percentage of slow fibers, R2 = 0.15; age versuspercentage of slow-type groupings, R2 = 0.56.

DICUSSIONIn the present study, we compared muscle parameters

from a group of young men (training by weight lifting) withthose from 2 groups of healthy mobile seniors (Table, Sup-plemental Digital Content 1, http://links.lww.com/NEN/A541):one composed of people leading a sedentary lifestyle and asecond made up of recreational sportsmen. The muscle fromthe physically active seniors more closely resembled that of theyounger men in terms of force generation and fiber size thanthat of the muscle of the sedentary seniors, indicating that long-term exercise aids in preservation of muscle health. More in-terestingly, the study reveals some unique characteristics of the

FIGURE 3. (AYD) Immunofluorescence staining for fast (A, C) and slow (B, D) myosin heavy chain (MHC) proteins in serialsections of biopsies from sedentary seniors (A, B) and physically active seniors (C, D). White arrows point to small angulatedmuscle fibers; white circles surround the central fibers that delineate fiber-type groupings. Note that the clustered fibers in thebiopsies of the sedentary seniors are of the fast type, whereas those of the physically active seniors are of the slow type.

TABLE 2. Percentages of Fast and Slow Myofibers in Young Men and in Sedentary and Physically Active SeniorsMyosin Heavy ChainYPositive Fibers, %

Subjects No. Biopsies Fast Slow ANOVA

Young men 10 58.1 T 4.3* 42.0 T 4.3* No (vs sedentary)

Seniors

Sedentary 10 53.8 T 13.2* 46.2 T 13.2* Yes (vs physically active)

Physically active 10 31.5 T 14.1† 68.5 T 14.1† Yes (vs young men)

Proportions of fast and slow fibers are expressed as percentages of total fibers in each muscle biopsy.*†Within columns, groups with different footnote symbols are significantly different, p G 0.05. Yes and No refer to significant differences by ANOVA.







muscle of the active seniors in terms of fiber type and fiber-typegroupings that suggest that denervation/reinnervation plays arole in the maintenance of muscle health.

The knee contraction strength in the active seniors wassignificantly greater than that of the sedentary seniors and notsignificantly different from that of the younger men (Table, Sup-plemental Digital Content 2, http://links.lww.com/NEN/A542).Although the physically active seniors generated 32% lessforce than the young men on knee contraction (despite the factthat the groups had dedicated similar time to training), this isnot surprising because it is well documented within the worldsporting records of masters athletes that the young outperformthe old (35Y37, 50).

To explore the mechanisms that delayed deteriorationin the muscle of the physically active seniors, we analyzedimmunolabeled muscle biopsies taken from our groups andcompared their relative amounts of 1) small angular myofibers(i.e. denervated muscle fibers); 2) molecular markers of fastand slow muscle fiber types (a measure of residual muscleplasticity); and 3) fiber-type grouping (representing dener-vated/reinnervated muscle fibers). We found that 1) biopsiesfrom young men seldom contain denervated, reinnervated,or grouped muscle fibers; 2) biopsies from sedentary seniorscontained both denervated and a few reinnervated clusteredmyofibers of the fast type; and 3) physically active seniors hada larger percentage of healthy slow myofibers, up to 90%,which appeared mainly clustered in slow fiberYtype groups.The finding that physically active seniors have a significantlyhigher percentage of slow-type fibers and slow-type fibergroupings is consistent with our previous results using histo-chemical myosin ATPase staining in muscle biopsies fromsenior sportsmen (n = 15) (only 2 of those subjects are alsopart of the present study); 27 out of 28 biopsies had slow-type

myofiber groupings (38). The increased slow-type fiber con-tent and percentage of slow-type fiber groupings reflect thefact that immobilization drives muscle fibers toward atrophyand fast-type transformation. Because these parameters weresignificantly higher in the physically active versus the seden-tary seniors, this is not simply a function of age. Furthermore,the lack of correlation between the kind of training and thepercentages of slow-type fibers demonstrates that the type ofactivity is not the main determinant factor.

Interestingly, muscle fibers coexpressing fast and slowMCH proteins were seldom detected in the biopsies of any ofour groups, and no statistical differences were found amongthe groups with respect to this parameter. It is possible that theobserved coexpression is scant either because it is actually arare event or because the denervation that occurs is promptlyfollowed by reinnervation so that obvious coexpression of theMHCs is short lived. To our knowledge, this is the first evi-dence that fiber transformation cannot be the direct conse-quence of decades of high-level activity. It also furthersupports the hypothesis that these infrequent denervationevents are not easily detectable with standard clinical elec-tromyography. When these coexpressing fibers were found inthe sedentary senior muscles, they were small (G30 Km) andoften had the distinct angulation noted after experimental orclinical denervation; some were also positive with the antiYN-CAM antibody (Fig. 2), an accepted marker of denervation(48, 49). This type of myofiber is common in unloaded mus-cle (i.e. resulting from spaceflight, limb suspension, or im-mobilization) and with spinal cord injury and peripheraldenervation (18Y27, 51Y53). Thus, we consider these to bedenervated muscle fibers. Furthermore, because it is knownthat slow-type muscle fibers revert to the fast isotype whendenervated during development and adulthood (18Y27, 51Y53),

TABLE 3. Percentages of Fast and Slow FiberYType Groupings in Young Men and in Sedentary and Physically Active SeniorsFiber-Type Groupings (% of Central Fibers in Clustered Areas vs Total Fibers)

Subjects No. Biopsies Fast ANOVA Slow ANOVA

Young men 10 1.0 T 2.0* No G0.1 T 0.1* No (vs sedentary)

Seniors

Sedentary 10 3.0 T 4.7* No 0.5 T 0.6* Yes (vs physically active)

Physically active 10 0.1 T 0.1* No 7.9 T 7.4† Yes (vs young)

Contents of reinnervated muscle fibers are expressed as the percentage of central muscle fibers surrounded by muscle fibers of the same phenotype relative to the total number ofmuscle fibers in each biopsy.

*†Within columns, groups with different footnote symbols are significantly different, p G 0.05. Yes or No refer to significant differences by ANOVA.

TABLE 4. Percentages of Myofibers Coexpressing Fast and Slow MHCs in Young Men and in Sedentary or Physically Active SeniorsMyofibers Coexpressing Fast and Slow MHCs

Subjects No. Biopsies % ANOVA

Young men 10 0.5 T 0.6* No (vs sedentary)

Seniors

Sedentary 10 1.8 T 1.7* No (vs physically active)

Physically active 10 0.6 T 0.6* No (vs young)

Content of muscle fibers coexpressing fast and slow muscle fibers are expressed as percentages of total muscle fibers.*Groups with different footnote symbols are significantly different, p G 0.05. No refers to not significantly different by ANOVA.





we suggest that these fibers are denervated slow-type myofibersreexpressing fast MHC through default myogenic programs.

In contrast, when myofibers coexpressing fast and slowMHCs were detected in the muscles of physically active se-niors, they were similar in size and shape to the pure-typefibers in the sections and, therefore, cannot lack innervation(Fig. 4). Furthermore, their low density in these muscle sec-tions is not in agreement with the concept that they belong to amotor unit that is undergoing exercise-driven, slow-typetransformation of MHCs, a mechanism that is well known tooccur in cross-reinnervation models (19, 53) but is morepresumed than demonstrated in humans performing voluntaryexercise (17). It is likely that the transforming myofibers (i.e.those coexpressing fast and slow MHC proteins) contribute tothe increase in slow fiberYtype groupings (Fig. 4); this issupported by the positive correlation between the increasingpercentages of slow-type fiber number and slow-type fibergroupings in sections from the physically active seniors. Thus,we suggest that these normal-sized coexpressing fibers arevery likely previously denervated fast-type fibers that havebeen reinnervated by sprouts from slow axons and are nowtemporarily coexpressing fast and slow MHC isoforms beforethey will finally exclusively express slow MHC. This issupported by the law of ‘‘recruitment order,’’ which dictatesthat slow motor units (and thus, muscle fibers) are activatedmore frequently than the fast motor units (54). In fact, themost active of the >-motoneurons are the slow type, and itmay be this higher level of activity that most likely maintainsmotoneurons, muscle fibers, and their MHC content. There-fore, the increase in slow-type fiber groupings is evidencethat some muscle plasticity still exists in both sedentary and(especially) physically active seniors. In addition, the rein-nervation process may be more extensive than is obvious herebecause if the temporarily denervated myofibers were of theslow type, they would continue their current gene expressionwhen reinnervated by slow-type >-motoneuron axon termi-nals and, thereby, escape our detection.

It is our opinion that the lack of fast-type groupings insenior sportsmen is direct evidence of selective reinnerva-tion of denervated myofibers from slow-type >-motoneurons.In summary, our working hypothesis is that muscle fiberscoexpressing fast and slow MHCs are either denervated slowmyofibers also expressing fast MHC isoforms by the defaultmyogenic program (Fig. 3A, B) (22, 24, 27) or are denervatedfast fibers reinnervated by axons sprouting from slow motorneurons (19, 52, 53).

FIGURE 4. Coimmunofluorescence staining of fast and slowmyosin heavy chain (MHC) proteins in a single section of vastuslateralis from a physically active senior reveals sparse MHC-coexpressing myofibers. (AYC) Fast fiber MHC proteins andlaminin are labeled in green (A); slow fiber MHC proteins arelabeled in red (B). Muscle fibers coexpressing both fast andslow MHCs are labeled with various levels of yellow/orange(within white circles in C) in proportion to the predominanceof either fast or slow MHCs, respectively. The amount ofcoexpressing muscle fibers is far below the large number ofmuscle fibers (in the hundreds) belonging to 1 motor unit.




This speculation needs further study, in particular,in situ MHC expression analyses, in more numerous sub-jects and different muscle types. Indeed, our study has manypotential confounding factors. These include the use of thefiber typeYheterogeneous vastus lateralis muscle, the smallsize of the specimens because of the sampling method (needlebiopsy), the low number of study subjects, the inherently var-iable genetic backgrounds of the individual subjects, and dif-ferences in the type and extent of physical activities of thephysically active seniors. Moreover, muscle biopsies from thatgroup ranged from those with scarce fiber-type transformationand grouping to those with almost fully transformed muscles.Despite these limitations, the clinical significance of our ob-servations is confirmed by the fact that the muscle propertiesof the physically active senior group are more similar to thoseof the active young men than to those of sedentary seniors.Specifically, relative to their sedentary counterparts, the phys-ically active seniors had greater muscle maximal isometricforce, along with better-preserved muscle morphology andmobility (39, 40).

Taken together, our results suggest that, beyond thedirect effects of aging on the structure and function of musclefibers, changes occurring in the muscle tissue of the sedentarygroup seem to be in part a result of sparse incremental de-nervation. In physically active seniors, the increase in thepercentage of ‘‘slow fiber groupings’’ is likely the result ofthe positive effect of long-term physical activity on the mo-toneuron pool, which, conceivably, has mainly spared theslow motoneurons from age-related lesion/death, thereby in-creasing the chance that peripheral reinnervation occurs be-cause of sprouting of slow axons.

Certainly, numerous mechanisms contribute to long-term muscle health or deterioration, yet our study suggeststhat long-term exercise would allow the body to adapt tothe consequences of age-related denervation and to preservemuscle structure and function by saving otherwise lost mus-cle fibers through recruitment of muscle fibers to different,mainly slow, motor units. We further speculate that high-levelactivity either by voluntary exercise or functional electricalstimulation may be applied at any age to save neurons from

disorders secondary to inactivity (55) and to counteract mus-cle atrophy in other neuromuscular or metabolic diseases (56).

Although the subjects of this study were not mas-ters athletes, the intensity of recreational training reported issomething that the general population may achieve, particu-larly if properly motivated by specialists in the field. Weshow that recreational levels of activity are very effective indriving seniors toward improved functional performance andrearrangement of muscle fiber type. In particular, these levelsof exercise seem to have beneficial effects on reinnervationof muscle fibers, resulting in preservation of muscle function,size, and structure, thereby delaying the functional declineand loss of independence that are common in late aging.

ACKNOWLEDGMENTThe expert assistance of Christian Hofer, PhD, is

gratefully acknowledged.

REFERENCES1. Morrison JH, Baxter MG. The ageing cortical synapse: Hallmarks and

implications for cognitive decline. Nat Rev Neurosci 2012;13:240Y502. Small SA, Schobel SA, Buxton RB, et al. A pathophysiological frame-

work of hippocampal dysfunction in ageing and disease. Nat RevNeurosci 2011;12:585Y601

3. Balistreri CR, Candore G, Accardi G, et al. NF-JB pathway activatorsas potential ageing biomarkers: Targets for new therapeutic strategies.Immun Ageing 2013;10:24

4. Erickson KI, Voss MW, Prakash RS, et al. Exercise training increasessize of hippocampus and improves memory. Proc Natl Acad Sci USA2011;108:3017Y22

5. Aagaard P, Suetta C, Caserotti P, et al. Role of the nervous system insarcopenia and muscle atrophy with aging: Strength training as a coun-termeasure. Scand J Med Sci Sports 2010;20:49Y64

6. Ohlendieck K. Proteomic profiling of fast-to-slow muscle transitionsduring aging. Front Physiol 2011;2:105

7. Gutmann E, Hanzlikova V. Motor unit in old age. Nature 1966;209:921Y228. Tomlinson BE, Walton JN, Rebeiz JJ. The effects of ageing and of

cachexia upon skeletal muscle. A histopathological study. J Neurol Sci1969;9:321Y46

FIGURE 5. The percentages of slow muscle fibers and slowfiberYtype groupings in muscle biopsies of physically activeseniors are strongly correlated (R2 = 0.82).

FIGURE 6. The percentages of slow fibers in muscle biopsiesharvested from physically active seniors are independent of theprevalent kind of training (E, endurance; S, strength; or M,mixed) undertaken by the individuals. Data from each form ofexercise are randomly distributed within the normal distribu-tion produced when the percentage of slow muscle fibers areplotted against the probability that each datum falls within thearea under the curve.




9. Scelsi R, Marchetti C, Poggi P. Histochemical and ultrastructural aspectsof m. vastus lateralis in sedentary old people (aged 65Y89 years). ActaNeuropathol 1980;51:99Y105

10. Urbanchek MG, Picken EB, Kalliainen LK, et al. Specific force deficitin skeletal muscles of old rats is partially explained by the existence ofdenervated muscle fibers. J Gerontol A Biol Sci Med Sci 2001;56:B191Y97

11. Rowan SL, Rygiel K, Purves-Smith FM, et al. Denervation causes fiberatrophy and myosin heavy chain co-expression in senescent skeletalmuscle. PLoS ONE 2012;7:e29082

12. Tomlinson BE, Irving D. The numbers of limb motor neurons in thehuman lumbosacral cord throughout life. J Neurol Sci 1977;34:213Y19

13. Doherty TJ, Vandervoort AA, Taylor AW, et al. Effects of motor unitlosses on strength in older men and women. J Appl Physiol 1993;74:868Y74

14. Larsson LX. Motor units: Remodeling in aged animals. J Gerontol ABiol Sci Med Sci 1993;50:91Y95

15. Andersen JL. Muscle fibre type adaptation in the elderly human muscle.Scand J Med Sci Sports 2003;13:40Y47

16. Luff AR. Age-associated changes in the innervation of muscle fibers andchanges in the mechanical properties of motor units. Ann N Y Acad Sci1998;854:92Y101

17. Mitchell WK, Williams J, Atherton P, et al. Sarcopenia, dynapenia, andthe impact of advancing age on human skeletal muscle size and strength:A quantitative review. Front Physiol 2012;3:260

18. Zhou MY, Klitgaard H, Saltin B, et al. Myosin heavy chain isoformsof human muscle after short-term spaceflight. J Appl Physiol 1995;78:1740Y44

19. Schiaffino S, Reggiani C. Fiber types in mammalian skeletal muscles.Physiol Rev 2011;91:1447Y531

20. Kern H, Hofer C, Modlin M, et al. Stable muscle atrophy in long-termparaplegics with complete upper motor neuron lesion from 3- to 20-yearSCI. Spinal Cord 2008;46:293Y304

21. Kern H, Carraro U, Adami N, et al. Home-based functional electricalstimulation rescues permanently denervated muscles in paraplegic pa-tients with complete lower motor neuron lesion. Neurorehabil NeuralRepair 2010;24:709Y21

22. D’Antona G, Pellegrino MA, Adami R, et al. The effect of ageing andimmobilization on structure and function of human skeletal muscle fibres.J Physiol (Lond) 2003;552:499Y511

23. Carraro U, Catani C, Belluco S, et al. Slow-like electrostimulationswitches on slow myosin in denervated fast muscle. Exp Neurol 1986;94:537Y53

24. Carraro U. Modulation of trophism and fiber type expression of dener-vated muscle by different patterns of electrical stimulation. Basic ApplMyol 2002;12:263Y73

25. Mayne CN, Mokrusch T, Jarvis JC, et al. Stimulation-induced expres-sion of slow muscle myosin in a fast muscle of the rat. Evidence of anunrestricted adaptive capacity. FEBS Lett 1993;327:297Y300

26. Salmons S. Exercise, stimulation and type transformation of skeletalmuscle. Int J Sports Med 1994;15:136Y41

27. Midrio M. The denervated muscle: Facts and hypotheses. A historicalreview. Eur J Appl Physiol 2006;98:1Y21

28. Meltzer DE. Age dependence of Olympic weightlifting ability. Med SciSports Exerc 1994;26:1053Y67

29. McNeil CJ, Doherty TJ, Stashuk DW. Motor unit number estimates in thetibialis anterior muscle of young, old, and very old men. Muscle Nerve2005;31:461Y67

30. Leyk D, Ruther T, Wunderlich M, et al. Physical performance in middleage and old age: Good news for our sedentary and aging society. DtschArztebl Int 2010;107:809Y16

31. Wroblewski AP, Amati F, Smiley MA, et al. Chronic exercise preserveslean muscle mass in masters athletes. Phys Sportsmed 2011;39:172Y78

32. Lexell J, Downham DY. The occurrence of fiber-type grouping in healthyhuman muscle: A quantitative study of cross-sections of whole vastuslateralis from men between 15 and 83 years. Acta Neuropathol (Berl)1991;81:377Y81

33. Sunnerhagen KS, Grimby G. Muscular effects in late polio. Acta PhysiolScand 2001;171:335Y40

34. Vetter C, Reichmann H, Pette D. Microphotometric determination ofenzyme activities in type-grouped fibres of reinnervated rat muscle.Histochemistry 1984;80:347Y51

35. Coggan AR, Spina RJ, Rogers MA, et al. Histochemical and enzymaticcharacteristics of skeletal muscle in master athletes. J Appl Physiol 1990;68:1896Y901

36. Trappe S. Master athletes. Int J Sport Nutr Exerc Metab 2001;11:S196Y207

37. Wright VJ, Perricelli BC. Age-related rates of decline in performanceamong elite senior athletes. Am J Sports Med 2008;36:443Y50

38. Mosole S, Rossini K, Kern H, et al. Significant increase of vastus lateralisreinnervation in 70-year sportsmen with a lifelong history of high-levelexercise. Eur J Transl Myol Basic Appl Myol 2013;23:117Y22

39. Zampieri S, Rossini K, Carrao U, et al. Morphometry of skeletal musclein sedentary elderly and senior sportsmen. Eur J Transl Myol Basic ApplMyol 2012;22:13

40. Kern H, Loefler S, Burggraf S, et al. Electrical stimulation counteractsmuscle atrophy associated with aging in humans. Eur J Transl MyolBasic Appl Myol 2013;23:105Y8

41. Boncompagni S, d’Amelio L, Fulle S, et al. Progressive disorganiza-tion of the excitation-contraction coupling apparatus in aging humanskeletal muscle as revealed by electron microscopy: A possible role in thedecline of muscle performance. J Gerontol A Biol Sci Med Sci 2006;61:995Y1008

42. Rossini K, Zanin ME, Podhorska-Okolow M, et al. To stage and quan-tify regenerative myogenesis in human long-term permanent denervatedmuscle. Basic Appl Myol 2002;12:277Y86

43. Boncompagni S, Kern H, Rossini K, et al. Structural differentiation ofskeletal muscle fibers in the absence of innervation in humans. Proc NatlAcad Sci USA 2007;104:19339Y44

44. Zampieri S, Valente M, Adami N, et al. Polymyositis, dermatomyo-sitis and malignancy: A further intriguing link. Autoimmun Rev 2010;9:449Y53

45. Zampieri S, Doria A, Adami N, et al. Subclinical myopathy in patientsaffected with newly diagnosed colorectal cancer at clinical onset of dis-ease: Evidence from skeletal muscle biopsies. Neurol Res 2010;32:20Y25

46. Kern H, Boncompagni S, Rossini K, et al. Long-term denervation inhumans causes degeneration of both contractile and excitation-contractioncoupling apparatus that can be reversed by functional electrical stimu-lation (FES). A role for myofiber regeneration? J Neuropathol Exp Neurol2004;63:919Y31

47. Kern H, Carraro U, Adami N, et al. One year of home-based daily FES incomplete lower motor neuron paraplegia: Recovery of tetanic contractil-ity drives the structural improvements of denervated muscle. Neurol Res2010;32:5Y12

48. Dickson G, Gower HJ, Barton CH, et al. Human muscle neural cell ad-hesion molecule (N-CAM): Identification of a muscle-specific sequencein the extracellular domain. Cell 1987;50:1119Y30

49. Kern H, Pelosi L, Coletto L, et al. Atrophy/hypertrophy cell signaling inmuscles of young athletes trained with vibrational-proprioceptive stimu-lation. Neurol Res 2011;33:998Y1009

50. Gava P, Kern H, Carraro U. Age-related decline of muscle power in trackand field master athletes indicates a lifespan of 110 years. Eur J TranslMyol Basic Appl Myol 2013;23:45

51. Carraro U, Catani C, Biral D. Selective maintenance of neurotrophicallyregulated proteins in denervated rat diaphragm. Exp Neurol 1979;63:468Y75

52. Carraro U, Catani C, Dalla Libera L. Myosin light and heavy chains inrat gastrocnemius and diaphragm muscles after chronic denervation orreinnervation. Exp Neurol 1981;72:401Y12

53. Pette D, Vrbova G. What does chronic electrical stimulation teach usabout muscle plasticity? Muscle Nerve 1999;22:666Y77

54. Henneman E, Somjen G, Carpenter DO. Functional significance of cellsize in spinal motoneurons. J Neurophysiol 1965;28:560Y80

55. Herrera-Rincon C, Torets C, Sanchez-Jimenez A, et al. Chronic electricalstimulation of transected peripheral nerves preserves anatomy and func-tion in the primary somatosensory cortex. Eur J Neurosci 2012;36:3679Y90

56. Wang XH, Mitch WE. Muscle wasting from kidney failureVA model forcatabolic conditions. Int J Biochem Cell Biol 2013;45:2230Y38




Long-Term High-Level Exercise Promotes Muscle Reinnervation With Age

Documents

Transcript of Long-Term High-Level Exercise Promotes Muscle Reinnervation With Age