Available online at www.sciencedirect.com

08) 840–846www.elsevier.com/locate/lifescie

Life Sciences 82 (20

The effects of aerobic exercise training on the age-related lipid peroxidation,Schwann cell apoptosis and ultrastructural changes in the sciatic nerve of rats

Ghaffar Shokouhi a,c, R. Shane Tubbs b, Mohammadali M. Shoja c,⁎, Leila Roshangar d,Mehran Mesgari d, Amir Ghorbanihaghjo d, Naser Ahmadi d,

Farzam Sheikhzadeh e, Jafar S. Rad d

a Department of Neurosurgery, Tabriz University of Medical Sciences, Tabriz, Iranb Department of Cell Biology and Neurosurgery, University of Alabama at Birmingham, Alabama, USAc Tuberculosis and Lung Disease Research Center, Tabriz University of Medical Sciences, Tabriz, Iran

d Drug Applied Research Center, Tabriz University of Medical Sciences, Tabriz, Irane Animal Sciences Department, Faculty of Biology, Tabriz University, Tabriz, Iran

Received 2 October 2007; accepted 25 January 2008

Abstract

The potential role of exercise in preventing the age-related spontaneous peripheral neuropathy has not been studied.We examined the effects of long-term aerobic exercise training on lipid peroxidation, Schwann cell (SC) apoptosis and ultrastructural changes in the sciatic nerve of rats during aging.Three groups of 12-week old Wistar rats ran on a treadmill for 6, 9 and 12 months (exercise trained (ET) group, n=10 each) according to an exercisetraining program targeted at a speed of 22 m/min (at 7° incline), 60min/day, 6 days/week. Three corresponding groups of untrained rats were used as thecontrols (sedentary (SED) group). At the end of each period, sciatic nerve biopsies were performed, and processed for biochemical,immunohistochemical and ultrastructural analyses. The results showed that aging was associated with an increased level of nerve malondialdehyde(MDA,marker of lipid peroxidation) and a higher number of SC apoptosis in SED group. The SED group showed irregular nerve fibers with thinmyelinsheaths and areas of myelin-axon detachment. However, the ET group had significantly diminished nerve lipid peroxidation and SC apoptosis. In the ETgroup, nerve fibers had a thick myelin sheath with frequent folding. These findings suggest that aerobic exercise training protects peripheral nerves byattenuating oxidative reactions, and preserving SCs and myelin sheath from pathologic changes, which occur during normal aging.© 2008 Elsevier Inc. All rights reserved.

Keywords: Aging; Apoptosis; Exercise; Lipid peroxidation; Neuropathy; Sciatic nerve; Treadmill

Introduction

Aging is a natural phenomenon associated with structuraland functional alterations of the cells and tissues. The age-related functional decline in the biological systems is mainlyattributed to the increased production of free radicals; Harman'sfree radical theory of aging (Harman, 1992; Muller et al., 2007).The exposure to reactive oxygen species induces oxidation ofproteins, lipids and nucleic acids. The oxidation of amino acidside chains results in protein–protein cross linkage, protein

⁎ Corresponding author. Tuberculosis and Lung Disease Research Center,Tabriz University of Medical Sciences, Tabriz, Iran. Tel./fax: +98 411 4438523.

E-mail address: shoja.m@gmail.com (M.M. Shoja).

0024-3205/$ - see front matter © 2008 Elsevier Inc. All rights reserved.doi:10.1016/j.lfs.2008.01.018

fragmentation and heightened protein carbonyl levels leading togradual cell death (Bandyopadhyay et al., 1999). Oxidativestress is also one of the several mechanisms that induceapoptosis in cells (Atmaca, 2004).

Normal aging brings about several changes in the central andperipheral nervous systems (Melcangi et al., 2000). Theseinclude loss of neurons and oligodendrocytes (Tanaka et al.,2005), diminished synaptic connections (Burke and Barnes,2006), hippocampal atrophy (Lupien et al., 1998), white mattermyelin breakdown and loss of optic nerve fibers (Peters, 2002;Sandell and Peters, 2001). In the peripheral nerves, age-relateddecrease in the number of myelinated fibers (Mortelliti et al.,1990), impaired Schwann cell (SC)–axon interaction (Wang etal., 2007), splitting of myelin sheaths, redundant myelin and

Fig. 1. Sciatic nerve TUNEL-positive (TP) cells at 12 months; A) Brownishapoptotic SCs are frequently seen in SED group. B) No TP cells are found in asample from the ET group. (For interpretation of the references to colour in thisfigure legend, the reader is referred to the web version of this article.)

841G. Shokouhi et al. / Life Sciences 82 (2008) 840–846

disruption of node of Ranvier (Hinman et al., 2006) have beenreported.Myelin basic protein and glycoprotein Po also have beenfound to decrease in the sciatic nerve of aged rats (Melcangi et al.,1998). In vitro studies have demonstrated that aged SCsunderwent chromosomal alterations, and had a reduced popula-tion doubling time (Funk et al., 2007). SCs also are the target ofvarious oxidative stress-mediated processes (Iida et al., 2004;Vincent et al., 2002).

The present study was performed to investigate the effects oflong-term, mild-to-moderate aerobic exercise on the peripheralnerve of rats during normal aging. The sciatic nerve lipidperoxidation, ultrastructural changes of neural fibers and SCapoptosis were analyzed in groups of rats that underwent long-term treadmill exercise and compared to the controls. Theterminal deoxynucleotidyl transferase-mediated dUTP-biotinnick end labeling (TUNEL) was used to detect DNAfragmentation, a characteristic of apoptotic cell death (Gavrieliet al., 1992; Li et al., 1995).

Materials and methods

Subjects

Sixty male 12-week old Wistar rats weighing 225 to 280 gwere used in this study according to the NIH Guide for the Careand Use of Laboratory Animals and local institutional guide-lines for human use of animals in research. The animals wereobtained from the Laboratory Animal House of Tabriz MedicalUniversity. They were housed three to five per cage andprovided with free access to compact food and water. This foodconsisted of all essential ingredients including vitamins andminerals. Animals were kept under similar laboratory condi-tions with regard to temperature, humidity and illumination, andwere regularly observed by two veterinary doctors throughoutthe study.

Group design and treadmill exercise

Rats were divided into three experimental groups thatunderwent treadmill exercise for 6, 9, and 12 months (exercisetrained (ET) group), and three corresponding control groupsthat were subjected to rest (sedentary (SED) group). Eachexperimental and control group consisted of 10 animals. Rats inthe experimental groups ran on a motorized 10-channeltreadmill (Danesh Yakhteh Co., Tabriz, Iran). Exercise startedbetween 11 a.m. and 2 p.m. The training program was begunwith a speed of 15 m/min at 7° incline for 10 min. The intensityand duration of treadmill exercise were then gradually increasedduring the initial 4 weeks to reach the target of 22 m/min at 7°incline for 60 min/day and six days per week. Animals ran awarm-up period of 5 min at 15 m/min each exercise day.Previously, exercise protocols with similar characteristics weredescribed as mild-to-moderate exercise protocols (Matsunoet al., 1999; Mottola et al., 1983; Taylor et al., 2003). Controlrats were placed on a treadmill that was in the off position for10 min a day. At the end of study periods, sciatic nerve biopsywas performed.

Sample preparation and determination of lipid peroxides

Under generalized anesthesia, rats underwent unilateralexposure of the left or right sciatic nerve at the upper border ofquadratus femoris. A one-centimeter segment of the sciatic nervewas harvested. Lipid peroxidation in the sciatic nerve of the ratwas measured as thiobarbituric acid reactive substance asdescribed previously (Shokouhi et al., 2008; Yagi, 1994). Briefly,tissue homogenates (10%W/V) were prepared by homogenizingsciatic nerve tissue in 50 mM cold potassium phosphate bufferusing a glass/glass homogenizer. Then, in a test tube, 0.5 ml oftissue homogenates was mixed with 3 ml of 1% orthophosphoricacid. After addition of 1 ml of 0.67% thiobarbituric acid, themixture was heated in boiling water for 45 min. The color formedwas extracted into 4 ml of n-butanol and the spectrophotometricabsorbancewasmeasured at 532 nmusing tetramethoxypropan asthe standard. Tissue malondialdehyde (MDA) levels werecalculated as nanomole per gram (nmol/gr) of wet tissue.

TUNEL assay

To detect apoptotic SCs in the sciatic nerve, TUNEL assaywas performed using a commercial kit (In Situ Cell DeathDetection Kit; Roche, Mannheim, Germany). Samples from all

Fig. 3. Changes in the sciatic nerve SC apoptosis during aging.

842 G. Shokouhi et al. / Life Sciences 82 (2008) 840–846

animals underwent this immunohistochemical staining. In brief,sections were first fixed in 10% neutral buffered formaldehydefor 72 h and then pretreated with 100 mg/ml proteinase K for30 min at 37 °C. Subsequently, the sections of 3–4 µm thicknesswere mounted on a gelatinized glass slide, incubated in TUNELreaction mixture for 1 h at 37 °C, and reacted with converted-POD. For staining, DAB solution was used. Mayer's hematox-ylin (DAKO, Glostrup, Denmark) was used for counterstainingbefore dehydration and mounting onto gelatinized glass slide.For assessment of apoptotic cells, the stained sections wereviewed under a bright field microscope and TUNEL-positivecells were detected based on their brownish color (Fig. 1).

Ultrastructural studies

For transmission electron microscopy, the sciatic nervesamples (from three animals in each group) were cut into piecesof 2×2 mm and fixed in 2% glutaraldehyde in a 0.1 Mphosphate buffer (Thuringowa, Australia) and postfixed in 1%aqueous osmium tetroxide (TAAB, England). The pieces werethen dehydrated through graded concentration of ethanol, andembedded in resin. One micron semi-thin sections were stainedwith toluidine blue. Ultra-thin sections were stained with uranylacetate and lead citrate, and observed with a LEO 906transmission electron microscope (Zeiss LEO 906, Germany).

Data analysis

Data are expressed in mean±standard deviation. Analysis wasperformed by SPSS version 13.0 using ANOVA with LSD forPost-Hoc to compare the sciatic nerve MDA levels and apoptoticcell counts between groups. Pearson correlation test was used toestimate the correlation between the sciatic nerve MDA level and

Fig. 2. Changes in the sciatic nerve MDA levels during aging.

SC apoptosis in exercise and control groups. A P value less than0.05 was considered to indicate statistical significance.

Results

Lipid peroxidation and SC apoptosis of the sciatic nerve duringnormal aging (SED group)

In the SED group, the mean levels of sciatic nerve MDAwere0.84±0.16, 1.0±0.21 and 1.26±0.42 nmol/gr of tissue at 6, 9 and12 months (6 vs 9 months, P=0.181 and 6 vs 12 months,P=0.001) showing that sciatic nerve lipid peroxidation increasedwith aging (Fig. 2). The number of sciatic nerve apoptotic figures(TUNEL-positive SCs) increased from 3.2±1.3 at 6 months to8.6±2.5 at 9 months (Pb0.01) and to 9.0±2.6 at 12 months

Fig. 4. Scatterplot showing the association between sciatic nerve MDA levelsand SC apoptosis in the SED group.

Table 1Comparison of sciatic nerve MDA levels and SC apoptosis between exerciseand control groups at 6, 9 and 12 months

6 months (n=10) 9 months (n=10) 12 months (n=10)

MDA(nmol/gr)

Apoptosis(n)

MDA(nmol/gr)

Apoptosis(n)

MDA(nmol/gr)

Apoptosis(n)

Exercisegroup

0.63±0.12 1.8±1.3 0.74±0.18 0.3±0.5 1.01±0.24 1.0±1.1

Controlgroup

0.84±0.16 3.2±1.3 1.00±0.21 8.6±2.5 1.26±0.42 9.0±2.6

P value 0.082 0.076 0.034 b0.01 0.026 b0.01

843G. Shokouhi et al. / Life Sciences 82 (2008) 840–846

(Pb0.01) (Fig. 3). A significant correlation was found betweenthe sciatic nerve MDA level and SC apoptosis in SED group(r=0.408, P=0.031) (Fig. 4).

Lipid peroxidation and SC apoptosis of the sciatic nerve in theET group

In the ET group, the mean levels of sciatic nerve MDAwere0.63±0.12 at 6 months, 0.74±0.18 at 9 months (P=0.346) and

Fig. 5. Semi-thin sections stainedwith toluidine blue (original magnification, 200–400×). A) SED group: irregular nerve fibers and thin myelin. B) ET group: thickmyelin sheath with frequent folding. (For interpretation of the references to colourin this figure legend, the reader is referred to the web version of this article.)

1.01±0.24 nmol/gr of tissue at 12 months (P=0.002 whencompared to that of 6 months) (Fig. 2). The number of sciaticnerve TUNEL-positive cells was 1.8±1.3, 0.3±0.5 and 1.0±1.1 at 6, 9 and 12 months, respectively (6 versus 9 months,P=0.058 and 6 versus 12 months, P=0.306) (Fig. 3). Noassociation was found between the sciatic nerve MDA level andSC apoptosis in the ET group (r=−0.127, P=0.520).

Comparisons of age-related changes in sciatic nerve lipidperoxidation and SC apoptosis between exercise and controlgroups (Table 1)

Although sciatic nerve MDA level at 6 months were lower inET than in SED groups, this trend did not reach statistical

Fig. 6. Electron micrographs of the sciatic nerve samples (Uranyl acetate and leadcitrate; original magnification, 3750×). A) 9-month SED group: Myelin layer isirregular and thin. Submyelin blebs (arrows) are seen. B) 9-month ET group: Thereare circular nerve fibers, thickenedmyelin layer (MF), and protrusion of myelin intothe axoplasm (axonal bodies, AB). C, collagen fibers.

844 G. Shokouhi et al. / Life Sciences 82 (2008) 840–846

significance (P=0.082). However, at 9 and 12 months, nerveMDA levels in the ET group were significantly lower than thoseof the SED group (P=0.034 and 0.026, respectively). SCapoptosis was more prominent in the SED than in ET group at 9and 12 months (Pb0.01; Fig. 1).

Ultrastructural findings

Semi-thin sections of the sciatic nerve revealed that neuralfibers were smaller in diameter and had irregular contours in theSED group (Fig. 5A) while those of the ET group were circularand larger (Fig. 5B). The nerve fibers were separated by aconnective tissue rich in collagen fibers. In the ET group, themyelin layer was thickened and protruded into the axoplasm,which could be seen as axonal bodies (Fig. 6A). Folding ofmyelin layers with frequent outward protrusions was present.Contrary to this, in the SED group, myelin appeared as darklystained (condensed) thin layers and the axoplasm washomogenous containing vesicle-like structures. Axonal shrink-age and occasional sub myelin bleb formation were also notedin the SED group (Fig. 6B). The density of collagen fibersamong nerve fibers was comparable between groups.

Discussion

The results of this study show that aging is associated withincreased lipid peroxidation, apoptosis of SCs, and axon/myelinultrastructural alterations in the rat sciatic nerve. However, long-term aerobic exercise training significantly attenuated nervelipid peroxidation and decreased SC apoptosis. These findingsindicate that long-term aerobic exercise could afford someneuroprotection to the peripheral nerves by preserving SCs anddiminishing oxidative stress-mediated injuries, which occurduring normal aging.

Aging is associated with a decreased compound muscle actionpotential on nerve conduction studies (Kurokawa et al., 1999).Verghese et al. (2001) noted an incidence of 39% for idiopathicperipheral neuropathy among elderly patients. Also occasionallyreferred to as spontaneous, senile, or age-associated peripheralneuropathy, the underlying etiologies of such a phenomenon arenot fully understood. Robertson et al. (1993) observed a highdegree of demyelination and axonal atrophy in the sciatic nervesof aged mice together with a decreased motor nerve conductionvelocity and decreased F-wave latency. In the present study, wefound a high level of TUNEL-positive cells in the sciatic nerve ofaged rats with a sedentary life. In this group, lipid peroxidationwas correlated with SC apoptosis. Although MDA levels of thesciatic nerve only mildly and insignificantly increased from 6 to9 months, SC apoptosis increased by ∼60%. We concluded thatSCs are highly vulnerable to oxidative stress or oxidation by-products in the aging process so that a minor increase in oxidativereactions substantially enhances SC apoptosis.

Although sciatic nerve lipid peroxidationwas generally lowerin the ET than in the SED group, the former showed a relativeincrease in lipid peroxidation from 6 to 12 months. Despite thisincrease in the ET group, SC apoptosis at 6 and 12 months werecomparable. Moreover, the association between nerve MDA

level and SC apoptosis, which was found in the SED group, wasnot present in the ET group. There are several potentialexplanations for these findings. First, even a slight reductionof oxidative reactions could substantially improve SC survival.Second, exercise might abolish cytotoxic mechanisms other thanoxidative reactions. Previously, glutamate-induced neural celldeath has been shown to increase with aging (Parihar andBrewer, 2007). Excitotoxicity afforded by N-methyl-D-aspartateadministration also has been shown to promote spinalmotoneuron death and subsequently the peripheral SC apoptosis(Winseck et al., 2002). Carro et al. (2001) revealed that lowintensity treadmill exercise protected neurons from excitotoxicdamages. Guezennec et al. (1998) also found that treadmillexercise down-regulated glutamate receptors. However, the roleof excitotoxic mechanisms in the age-related peripheral neuro-pathy remains to be identified. Third, the induction ofendogenous neuroprotective factors by exercise might counter-balance toxic effects of oxidation by-products. For example,treadmill exercise is known to increase insulin-like growthfactor-1 (Karatay et al., 2007; Turgut et al., 2006), a potentneurotropic factor, which enhances myelin formation bystimulating SC differentiation, free fatty acid synthesis andglycoprotein Po expression (Cheng et al., 1999; Liang et al.,2007; Russell et al., 2000). Moreover, the expression of nervegrowth factor, which counterbalances pro-apoptotic signals byactivating nuclear factor kappaB in SCs, is increased withchronic exercise (Dishman et al., 2006; Gentry et al., 2000;Stummer et al., 1994). Nerve growth factor has been shown toenhance the activity of free-radical scavengers and to protectagainst excitotoxicity by reducing intracellular calcium overload(Ang et al., 2003). In one study, physical exercise reduced theexpression of apoptosis-associated genes including Bcl-xh andDP5 (Tong et al., 2001).

In the present study, myelin condensation and thinning,axonal shrinkage with irregular nerve fibers, and foci of submyelin bleb formation indicating detachment of the myelinsheath from the axolemma were found in aged rats of the SEDgroup. Loss of contact between paranodal SC loops/myelin andaxolemma has been reported with aging (Hinman et al., 2006;Sugiyama et al., 2002) and other pathologic conditions (Changet al., 1994; Saida et al., 1984; Ho and Yu, 1989). However,exercise training abolished these effects and in fact, the myelinsheath was thickened and had frequent folding with outward andinward protrusions in the ET group. These changes might besecondary to a relative excess of myelin circumference to theaxon periphery, which could result from enhanced myelinsynthesis or decreased breakdown in the ET group. Increasedsecretion of growth factors as well as improved SC survivalmight have contributed to these observed effects.

The relations between aerobic exercise and cellular apoptosisas suggested previously are not predictable. Exercise-inducedapoptosis is known to occur in lymphocytes and skeletal musclecells (Carraro and Franceschi, 1997; Rosolowska-Huszcz,1998). These instances have been attributed to the increasedsecretion of glucocorticoids, intracellular calcium concentra-tions, and reactive oxygen species produced during strenuousexercises. Concordet and Ferry (1993) noticed an increase in

845G. Shokouhi et al. / Life Sciences 82 (2008) 840–846

apoptosis of thymocytes in rats exhausted with treadmillrunning. In another study, no significant changes in theapoptotic neurons of rat dentate gyrus were found followingmild or moderate exercise (Kim et al., 2002). In contrast,Hoffman-Goetz et al. (1999) found that mice subjected tosubmaximal treadmill exercise for a duration of 90 min had alower percentage of apoptotic thymocytes, despite the elevationof plasma corticosteroids. It seems that exercise intensity andduration could be potentially important in determining the anti-apoptotic or pro-apoptotic properties of an exercise protocol andthat most of the exercise regimens known to induce apoptosisare generally acute and exhaustive.

Conclusions

The present study revealed that aging is associated withsignificantly elevated peripheral nerve lipid peroxidation and SCapoptosis, and that aerobic exercise training with mild-to-moderate intensity can diminish these effects over the long-term. The presence of an association between nerve lipidperoxidation and SC apoptosis in the aged resting group and thelack of such a relationship in trained rats indicate that in addition toa diminished oxidative reactions, some endogenous exercisemediated protection of peripheral nerves might exist. Moreover,our study shows that aerobic exercise training is associated withmyelin thickening and folding. The potential functional correlatesof these observations remain to be identified in the future.

Acknowledgment

This study was supported by a grant from the Drug AppliedResearch Center, Tabriz University of Medical Sciences, Tabriz,Iran.

References

Ang, E.T., Wong, P.T., Moochhala, S., Ng, Y.K., 2003. Neuroprotectionassociated with running: is it a result of increased endogenous neurotrophicfactors? Neuroscience 118, 335–345.

Atmaca, G., 2004. Antioxidant effects of sulfur-containing amino acids. YonseiMedical Journal 45, 776–788.

Bandyopadhyay, U., Das, D., Banerjee, R.K., 1999. Reactive oxygen species,oxidative stress and pathogenesis. Current Science 77, 658–666.

Burke, S.N., Barnes, C.A., 2006. Neural plasticity in the ageing brain. NatureReviews Neuroscience 7, 30–40.

Carro, E., Trejo, J.L., Busiguina, S., Torres-Aleman, I., 2001. Circulatinginsulinlike growth factor I mediates the protective effects of physicalexercise against brain insults of different etiology and anatomy. Journal ofNeuroscience 21, 5678–5684.

Carraro, U., Franceschi, C., 1997. Apoptosis of skeletal and cardiac muscles andphysical exercise. Aging Clinical and Experimental Research 9, 919–934.

Chang, K.Y., Ho, S.T., Yu, H.S., 1994. Vibration induced neurophysiologicaland electron microscopical changes in rat peripheral nerves. Occupationaland Environmental Medicine 51, 130–135.

Cheng, H.L., Russell, J.W., Feldman, E.L., 1999. IGF-I promotes peripheralnervous system myelination. Annals of the New York Academy of Sciences883, 124–130.

Concordet, J.P., Ferry, A., 1993. Physiological programmed cell death inthymocytes is induced by physical stress (exercise). American Journal ofPhysiology 265, C626–C629.

Funk, D., Fricke, C., Schlosshauer, B., 2007. Aging SCs in vitro. EuropeanJournal of Cell Biology 86, 207–219.

Dishman, R.K., Berthoud, H.R., Booth, F.W., Cotman, C.W., Edgerton, V.R.,Fleshner, M.R., Gandevia, S.C., Gomez-Pinilla, F., Greenwood, B.N.,Hillman, C.H., Kramer, A.F., Levin, B.E.,Moran, T.H., Russo-Neustadt, A.A.,Salamone, J.D., VanHoomissen, J.D.,Wade, C.E., York, D.A., Zigmond,M.J.,2006. Neurobiology of exercise. Obesity 14, 345–356.

Gavrieli, Y., Sherman, Y., Ben-Sasson, S.A., 1992. Identification ofprogrammed cell death in situ via specific labeling of nuclear DNAfragmentation. Journal of Cell Biology 119, 493–501.

Gentry, J.J., Casaccia-Bonnefil, P., Carter, B.D., 2000. Nerve growth factoractivation of nuclear factor kappaB through its p75 receptor is an anti-apoptotic signal in RN22 schwannoma cells. Journal of BiologicalChemistry 275, 7558–7565.

Guezennec,C.Y.,Abdelmalki,A., Serrurier,B.,Merino,D.,Bigard,X.,Berthelot,M.,Pierard, C., Peres,M., 1998. Effects of prolonged exercise on brain ammonia andamino acids. International Journal of Sports Medicine 19, 323–327.

Harman, D., 1992. Role of free radicals in ageing and disease. Annals of theNew York Academy of Sciences 673, 598–620.

Hinman, J.D., Peters, A., Cabral, H., Rosene, D.L., Hollander, W., Rasband,M.N.,Abraham, C.R., 2006. Age-related molecular reorganization at the node ofRanvier. Journal of Comparative Neurology 495, 351–362.

Ho, S.T., Yu, H.S., 1989. Ultrastructural changes of the peripheral nerve inducedby vibration: an experimental study. British Journal of Industrial Medicine46, 157–164.

Hoffman-Goetz, L., Zajchowski, S., Aldred, A., 1999. Impact of treadmillexercise on early apoptotic cells in mouse thymus and spleen. Life Sciences64, 191–200.

Iida, H., Schmeichel, A.M., Wang, Y., Schmelzer, J.D., Low, P.A., 2004. SC is atarget in ischemia-reperfusion injury to peripheral nerve. Muscle & Nerve30, 761–766.

Karatay, S., Yildirim, K., Melikoglu, M.A., Akcay, F., Senel, K., 2007. Effectsof dynamic exercise on circulating IGF-1 and IGFBP-3 levels in patientswith rheumatoid arthritis or ankylosing spondylitis. Clinical Rheumatology26, 1635–1639.

Kim, S.H., Kim, H.B., Jang, M.H., Lim, B.V., Kim, Y.J., Kim, Y.P., Kim, S.S.,Kim, E.H., Kim, C.J., 2002. Treadmill exercise increases cell proliferationwithout altering of apoptosis in dentate gyrus of Sprague–Dawley rats. LifeSciences 71, 1331–1340.

Kurokawa, K., Mimori, Y., Tanaka, E., Kohriyama, T., Nakamura, S., 1999.Age-related change in peripheral nerve conduction: compound muscleaction potential duration and dispersion. Gerontology 45, 168–173.

Li, Y., Chopp, M., Jiang, N., Zhang, Z.G., Zaloga, C., 1995. Induction of DNAfragmentation after 10 to 120 minutes of focal cerebral ischemia in rats.Stroke 26, 1252–1257.

Liang, G., Cline, G.W., Macica, C.M., 2007. IGF-1 stimulates de novo fatty acidbiosynthesis by SCs during myelination. Glia 55, 632–641.

Lupien, S.J., de Leon, M., de Santi, S., Convit, A., Tarshish, C., Nair, N.P.,Thakur, M., McEwen, B.S., Hauger, R.L., Meaney, M.J., 1998. Cortisollevels during human aging predict hippocampal atrophy and memorydeficits. Nature Neuroscience 1, 69–73.

Matsuno, A.Y., Esrey, K.L., Perrault, H., Koski, K.G., 1999. Low intensityexercise and varying proportions of dietary glucose and fat modify milk andmammary gland compositions and pup growth. Journal of Nutrition 129,1167–1175.

Melcangi, R.C., Magnaghi, V., Cavarretta, I., Martini, L., Piva, F., 1998. Age-induced decrease of glycoprotein Po and myelin basic protein gene expressionin the rat sciatic nerve. Repair by steroid derivatives. Neuroscience 85,569–578.

Melcangi, R.C., Magnaghi, V., Martini, L., 2000. Aging in peripheral nerves:regulation of myelin protein genes by steroid hormones. Progress inNeurobiology 60, 291–308.

Mottola, M., Bagnall, K.M., McFadden, K.D., 1983. The effects of maternalexercise on developing rat fetuses. British Journal of Sports Medicine 17,117–1121.

Mortelliti, A.J., Malmgren, L.T., Gacek, R.R., 1990. Ultrastructural changeswith age in the human superior laryngeal nerve. Archives of OtolaryngologyHead & Neck Surgery 116, 1062–1069.

846 G. Shokouhi et al. / Life Sciences 82 (2008) 840–846

Muller, F.L., Lustgarten, M.S., Jang, Y., Richardson, A., Van Remmen, H.,2007. Trends in oxidative aging theories. Free Radical Biology & Medicine43, 477–503.

Parihar, M.S., Brewer, G.J., 2007. Simultaneous age-related depolarization ofmitochondrial membrane potential and increased mitochondrial reactiveoxygen species production correlate with age-related glutamate excitotoxicityin rat hippocampal neurons. Journal ofNeuroscience Research 85, 1018–1032.

Peters, A., 2002. The effects of normal aging on myelin and nerve fibers: areview. Journal of Neurocytology 31, 581–593.

Robertson, A., Day, B., Pollock, M., Collier, P., 1993. The neuropathy of elderlymice. Acta Neuropathologica 86, 163–171.

Rosolowska-Huszcz, D., 1998. The effect of exercise training intensity on thyroidactivity at rest. Journal of Physiology and Pharmacology 49, 457–466.

Russell, J.W., Cheng, H.L., Golovoy, D., 2000. Insulin-like growth factor-Ipromotes myelination of peripheral sensory axons. Journal of Neuropathol-ogy and Experimental Neurology 59, 575–584.

Saida, K., Saida, T., Kayama, H., Nishitani, H., 1984. Rapid alterations of theaxon membrane in antibody-mediated demyelination. Annals of Neurology15, 581–589.

Sandell, J.H., Peters, A., 2001. Effects of age on nerve fibers in the rhesusmonkey optic nerve. Journal of Comparative Neurology 429, 541–553.

Shokouhi, G., Tubbs, R.S., Shoja, M.M., Hadidchi, S., Ghorbanihaghjo, A.,Roshangar, L., Farahani, R.M., Mesgari, M., Oakes, W.J., 2008.Neuroprotective effects of high-dose vs low dose melatonin after bluntsciatic nerve injury. Child's Nervous System 24, 111–117.

Stummer, W., Weber, K., Tranmer, B., Baethmann, A., Kempski, O., 1994.Reduced mortality and brain damage after locomotor activity in gerbilforebrain ischemia. Stroke 25, 1862–1869.

Sugiyama, I., Tanaka, K., Akita, M., Yoshida, K., Kawase, T., Asou, H., 2002.Ultrastructural analysis of the paranodal junction of myelinated fibers in 31-month-old-rats. Journal of Neuroscience Research 70, 309–317.

Tanaka, J., Okuma, Y., Tomobe, K., Nomura, Y., 2005. The age-relateddegeneration of oligodendrocytes in the hippocampus of the senescence-accelerated mouse (SAM) P8: a quantitative immunohistochemical study.Biological & Pharmaceutical Bulletin 28, 615–618.

Taylor, R.P., Ciccolo, J.T., Starnes, J.W., 2003. Effect of exercise training on theability of the rat heart to tolerate hydrogen peroxide. CardiovascularResearch 58, 575–581.

Tong, L., Shen, H., Perreau, V.M., Balazs, R., Cotman, C.W., 2001. Effects ofexercise on gene-expression profile in the rat hippocampus. Neurobiology ofDisease 8, 1046–1056.

Turgut, S., Kaptanoglu, B., Emmungil, G., Turgut, G., 2006. Increased plasmalevels of growth hormone, insulin-like growth factor (IGF)-I and IGF-binding protein 3 in pregnant rats with exercise. Tohoku Journal ofExperimental Medicine 208, 75–81.

Verghese, J., Bieri, P.L., Gellido, C., Schaumburg, H.H., Herskovitz, S., 2001.Peripheral neuropathy in young-old and old-old patients. Muscle & Nerve24, 1476–1481.

Vincent, A.M., Brownlee, M., Russell, J.W., 2002. Oxidative stress andprogrammed cell death in diabetic neuropathy. Annals of the New YorkAcademy of Sciences 959, 368–383.

Wang, Y.J., Zhou, C.J., Shi, Q., Smith, N., Li, T.F., 2007. Aging delays theregeneration process following sciatic nerve injury in rats. Journal ofNeurotrauma 24, 885–894.

Winseck, A.K., Caldero, J., Ciutat, D., Prevette, D., Scott, S.A., Wang, G.,Esquerda, J.E., Oppenheim, R.W., 2002. In vivo analysis of Schwann cellprogrammed cell death in the embryonic chick: regulation by axons and glialgrowth factor. Journal of Neuroscience 22, 4509–4521.

Yagi, K., 1994. Lipid peroxides in hepatic, gastrointestinal, and pancreaticdiseases. Advances in Experimental Medicine and Biology 366, 165–169.