A single 30min treadmill exercise session is suitable for ¢

Available online at www.sciencedirect.com

ARTICLE IN PRESS

www.elsevier.com/locate/nmd

Neuromuscular Disorders xxx (2011) xxx–xxx

A single 30 min treadmill exercise session is suitable for‘proof-of concept studies’ in adult mdx mice: A comparison of the

early consequences of two different treadmill protocols

Hannah Radley-Crabb a,⇑, Jessica Terrill a,b, Thea Shavlakadze a, Joanne Tonkin a,Peter Arthur b, Miranda Grounds a

a School of Anatomy and Human Biology, The University of Western Australia, Crawley, Australiab School of Biomedical, Biomolecular and Chemical Sciences, The University of Western Australia, Crawley, Australia

Received 11 March 2011; received in revised form 24 June 2011; accepted 11 July 2011

Abstract

The extent of muscle pathology in sedentary adult mdx mice is very low and treadmill exercise is often used to increase myofibrenecrosis; however, the early events in dystrophic muscle and blood in response to treadmill exercise (leading to myofibre necrosis) areunknown. This study describes in detail two standardised protocols for the treadmill exercise of mdx mice and profiles changes in molec-ular and cellular events after a single 30 min treadmill session (Protocol A) or after 4 weeks of (twice weekly) treadmill exercise (ProtocolB). Both treadmill protocols increased multiple markers of muscle damage. We conclude that a single 30 min treadmill exercise session isa sufficient and conveniently fast screening test and could be used in ‘proof-of-concept’ studies to evaluate the benefits of pre-clinicaldrugs in vivo. Myofibre necrosis, blood serum CK and oxidative stress (specifically the ratio of oxidised to reduced protein thiols) arereliable markers of muscle damage after exercise; many parameters demonstrated high biological variation including changes in mRNAlevels for key inflammatory cytokines in muscle. The sampling (sacrifice and tissue collection) time after exercise for these parameters iscritical. A more precise understanding of the changes in dystrophic muscle after exercise aims to identify biomarkers and new potentialtherapeutic drug targets for Duchenne Muscular Dystrophy.� 2011 Published by Elsevier B.V.

Keywords: Mdx mouse; Treadmill exercise; Skeletal muscle damage; Myofibre necrosis; Creatine kinase; Inflammation; Oxidative stress

1. Introduction

Duchenne Muscular Dystrophy (DMD) is an X-linked,lethal muscle wasting disorder that affects mainly boys[1,2]. Impaired function or absence of the sub-sarcolemmalprotein dystrophin, renders dystrophic myofibres suscepti-ble to sarcolemma damage in response to contraction [3–7].

0960-8966/$ - see front matter � 2011 Published by Elsevier B.V.

doi:10.1016/j.nmd.2011.07.008

⇑ Corresponding author. Address: Mail Delivery M309, School ofAnatomy and Human Biology, The University of Western Australia, 35Stirling Highway, Crawley, Western Australia 6009, Australia. Tel.: +61 86488 7127; fax: +61 8 6488 1051.

E-mail address: [email protected] (H. Radley-Crabb).

Please cite this article in press as: Radley-Crabb H et al., A single 30 min treadmice: A comparison of the early consequences of two different treadmill prot

This initial damage can progress to myofibre necrosis andsubsequent regeneration; repeated cycles of necrosis ulti-mately result in the replacement of myofibres with fatand/or fibrotic connective tissue [8]. A progressive loss ofmuscle mass and function in DMD leads to prematuredeath often due to respiratory or cardiac failure [9]. Whilethe genetic defect was identified over 20 years ago the spe-cific cause of myofibre necrosis is still unknown, althoughincreased levels (or dysregulation) of inflammation, oxida-tive stress and intracellular calcium are all heavily impli-cated [7,10–16].

Mdx mice (C57Bl/10ScSnmdx/mdx), which lack dystro-phin, are an animal model for DMD and are widely used

mill exercise session is suitable for ‘proof-of concept studies’ in adult mdxocols, Neuromuscul Disord (2011), doi:10.1016/j.nmd.2011.07.008

http://dx.doi.org/10.1016/j.nmd.2011.07.008

mailto:[email protected]



2 H. Radley-Crabb et al. / Neuromuscular Disorders xxx (2011) xxx–xxx

ARTICLE IN PRESS

in pre-clinical research [11,17]. In sedentary adult mdxmice, the extent of dystropathology is relatively mild, withusually <6% myofibre necrosis in the quadriceps muscle(expressed as % cross-sectional area CSA) and relativelylow serum creatine kinase (CK) activity [11], thus exerciseis routinely used to increase dystropathology [18–21]enabling potential therapeutic interventions to be more rig-orously evaluated in vivo [22–29].

The term ‘exercise’ is used broadly and can cover wholebody in vivo exercise such as voluntary wheel and treadmillrunning; it can also cover ex vivo stretching protocols orin vivo electrically stimulated eccentric contractions thatrequire surgical intervention. This paper is focused onin vivo treadmill exercise as this closely represents the phys-iological situation and is a technique widely accessible bymany research groups across the world.

In the past, our laboratory has used voluntary wheelexercise over 48 h to increase myofibre necrosis and histol-ogy to demonstrate the benefits of anti-inflammatory drugson dystrophic muscle in vivo [26,27,30]. Muscle necrosis isroughly doubled (�6% to 12% CSA) in quadriceps muscleafter 48 h of voluntary exercise, although other musclessuch as the tibialis anterior (TA) are barely affected by vol-untary wheel exercise [11,24,26].

A widely used alternative to voluntary wheel (usuallynocturnal) exercise is controlled treadmill running (experi-ments usually conducted during the day). This occurs at acontrolled speed for a pre-determined length of time, thuseliminating some of the behavioural variables experiencedwith voluntary exercise. A protocol of 30 min treadmillrunning on a horizontal treadmill at a speed of 12 m/min,twice a week for at least 4 weeks, causes a significantincrease in the dystropathology of adult mdx mice and iswidely used in pre-clinical research [20,22,25,31,32]. Thereis however some concerns regarding treadmill exercisebecause mdx mice can have problems coping and thus bereluctant to run, although a short warm-up period at aslower speed appears to help with treadmill running [23].

A single 30 min treadmill exercise session represents aprecise amount of controlled exercise that allows thetime-course of early cellular and molecular events to bemeasured. It is of fundamental interest to determine theextent of the initial skeletal muscle damage and associatedmolecular changes in response to a single 30 min exercisesession (Protocol A) in unexercised mdx mice, comparedwith mice exercised for 4 weeks (Protocol B) which is awidely used exercise regime [20,22,25,31,32]. In the presentstudy parameters measured included; histological quantifi-cation of myofibre necrosis, circulating blood CK activity,quadriceps muscle gene expression levels (mRNA) of thepro-inflammatory cytokines interleukin-1b (IL-1b), inter-leukin-6 (IL-6) and tumour necrosis factor (TNF) andquantification of oxidative stress (protein thiol oxidationratio and malondialdehyde (MDA) quantitation in thequadriceps muscle.

The aims of the present study were to: (1) develop a short(30 min) and repeatable in vivo treadmill protocol to increase


myofibre necrosis in adult mdx mice; (2) profile the timecourse of multiple indicators of muscle damage immediatelyafter a single exercise session; (3) compare these responsesafter a single treadmill exercise session to responses after4 weeks of treadmill exercise; (4) establish if a single 30 minexercise session is an appropriate protocol to increase muscledamage in adult mdx mice and thus be used in pre-clinical‘proof-of-concept’ studies; (5) identify key parameters withpotential as diagnostic biomarkers to rapidly monitor effi-cacy of pre-clinical drug treatments.

2. Materials and methods

2.1. Animal procedures

All experiments were carried out on 8- to 12-week-old(adult) male non-dystrophic control C57Bl/10 and dystro-phic mdx mice; mice were obtained from the AnimalResource Centre, Murdoch, Western Australia. They weremaintained at the University of Western Australia on a 12-h light/dark cycle, under standard conditions, with freeaccess to food and drinking water. Mice of each strain werecaged in groups of 3–4. All animal experiments were con-ducted in strict accordance with the guidelines of theNational Health and Medical Research Council Code ofpractice for the care and use of animals for scientific pur-poses (2004) and the Animal Welfare act of WesternAustralia (2002) and were approved by the Animal Ethicscommittee at the University of Western Australia.

2.1.1. Establishing the 30 min treadmill protocol

Based on previous research [20,22,31] and as per theTREAT-NMD recommended standard protocol “Use oftreadmill and wheel exercise for impact on mdx mice phe-notype M.2.1_001” http://www.treat-nmd.eu/research/pre-clinical/SOPs/ a treadmill exercise regime consisting of30 min treadmill running at a speed of 12 m/min was used.The rodent treadmill was an Exer 3/6 from ColumbusInstruments (USA).

Treadmill setup: Individual running lanes were separatedby clear Perspex dividers so that the mice could see eachother while exercising. The treadmill was horizontal (0�incline) and mdx mice were run in groups of 3 or 4 as itis time consuming and inefficient to run mdx miceindividually.

Exercise protocol: Groups of 3 or 4 mdx mice were all (1)settled for 2 min with the treadmill belt stationary, (2) thenacclimatized with gentle walking for 2 min at a speed of4 m/min, followed immediately by (3) a warm-up of8 min at 8 m/min and then (4) the main exercise sessionfor 30 min at 12 m/min. If during the 30 min exercise ses-sion a mouse fatigued and could no longer run, the proce-dure was as follows: turn the treadmill belt off and give allmice a 2 min rest, turn the belt on at 4 m/min for 2 min,increase the speed to 12 m/min and run for the remainderof the 30 min. Repeat this process if fatigue occurs again(up to 5 times for an individual mdx mouse).


http://www.treat-nmd.eu/research/preclinical/SOPs/



Table 1Summary of the animal groups and numbers used for both Protocol A (1x single 30 min exercise session) and Protocol B (4 weeks of treadmill exercise).

Protocol (A) 1 single 30 min treadmill session

Histological analysis (archival) Unexercised Mdx 60Exercised – 24 h post Mdx 25Exercised – 48 h post Mdx 43

Time-course analysis Unexercised Mdx and C57Bl/10 8Exercised – 0 min post Mdx and C57Bl/10 8Exercised – 2 h post Mdx 8Exercised – 24 h post Mdx 8

Protocol (B) 4 weeks treadmill exercise (8 sessions)

Time-course and histological analysis Unexercised Mdx 8Exercised – 0 min post Mdx 8Exercised – 24 h post Mdx 8Exercised – 96 h post Mdx 8

H. Radley-Crabb et al. / Neuromuscular Disorders xxx (2011) xxx–xxx 3

ARTICLE IN PRESS

2.1.2. Treadmill regime and animal sample groups

All experiments (exercise and sampling) were started at8am and completed by 11am each day.

Exercise Protocol A (a single 30 min treadmill exercise

session): 12-week-old (completely untrained) mdx and con-trol C57Bl/10 mice were exercised for a single session onthe rodent treadmill. All male mice were sampled at12 weeks of age. The following three groups of male mdxmice were used for histological analysis: (1) unexercised,(2) mice subjected to 30 min treadmill exercise and sampled24 h or (3) 48 h post exercise. Numerous treadmill exerciseexperiments have been conducted in our laboratory overthe last year using 12-week-old male mdx mice and the his-tological data from all experiments were pooled to providelarge group numbers (see Table 1). The time course studywas conducted on a total of 32 mdx and 16 C57Bl/10 12-week-old mice representing six different groups with n = 8for each group. The following groups were used: (1) unex-ercised C57Bl/10, (2) exercised C57Bl/10 sampled immedi-ately (0 min) post exercise, (3) unexercised mdx, (4)exercised mdx sampled immediately (0 min) or (5) 2 h exer-cise or (6) 24 h post exercise (see Table 1).

Exercise Protocol B (4 weeks of treadmill exercise): mdxmice were exercised on the treadmill twice a week for4 weeks, with a consistent 72 or 96 h break between eachexercise session. The treadmill exercise started when micewere 8 weeks old and therefore all mice were 12 weeksold at time of sampling. For consistency all mice had a72 h (3 day) break before the final (8th) exercise sessionand subsequent sampling. The 4 week treadmill exerciseprotocol was conducted on a total of 32 male mdx micerepresenting four different groups with n = 8 for eachgroup. The following groups were used: (1) unexercisedmdx, (2) 4 week exercised mdx sampled immediately(0 min) or (3) 24 h or (4) 96 h post exercise (see Table 1).C57Bl/10 mice were not included in Protocol B.

2.1.3. Forelimb grip strength

Mice from Protocol B were also assessed throughout thestudy for forelimb grip strength (measured 24 h prior to1st, 5th and 8th treadmill exercise session). Grip strengthwas measured using a Chatillon Digital Force Gauge


(DFE-002) and a triangle metal bar, as per the TREAT-NMD recommended standard protocol “Use of gripstrength metre to assess limb strength of mdx mice –M.2.2_001” http://www.treat-nmd.eu/research/preclinical/SOPs/. In brief, the mouse was placed on the front of thetriangle bar (attached to a force transducer) and pulledgently until release. Each mouse underwent five consecutivegrip-strength trials; the grip strength value for each mousewas recorded as the average of the three best efforts. Aver-age grip strength was then normalized for body weight[force (kg)/BW (g)]. Change in normalised grip strengthwas determined by subtracting normalised grip strength(8 weeks) from normalised grip strength (12 weeks) [25].

2.2. Tissue collection and image acquisition

All mice were sacrificed by cervical dislocation whileunder terminal anaesthesia (2%v/v Attane isofluraneBomac Australia). Various muscles were collected, somewere immediately snap frozen in liquid nitrogen for molec-ular analysis (quadriceps) and some were prepared for his-tology (quadriceps, triceps, gastrocnemius, diaphragm,tibialis anterior and extensor digitorum longus). Limbmuscles were fixed immediately in 10% BFS (ConfixAustralian Biostain AB1020) and remained in solutionfor at least 72 h. Tissues were placed into 70% ethanol, pro-cessed in a Shandon automatic tissue processor overnight,and paraffin embedded for sectioning. Transverse sections(5 lm) were cut through the mid-region of each muscle.Slides were routinely stained with Haematoxylin and Eosin(H&E) for morphological analysis of the histology. Non-overlapping tiled images of transverse muscle sections pro-vided a picture of the entire muscle cross-section. Imageswere acquired using a Leica DM RBE microscope, a per-sonal computer, a Hitachi HVC2OM digital camera, ImagePro Plus 4.5.1 software and Vexta stage movement soft-ware. Tiled images were taken at 10� magnification.

2.3. Histological image analysis

Histological analysis of muscle necrosis was carried outon whole muscle cross-sections. Muscle morphology was






ARTICLE IN PRESS

drawn manually by the researcher using Image Pro Plus4.5.1 software. The area occupied by necrotic myofibres(i.e. myofibres with fragmented sarcoplasm and/or areasof inflammatory cells) was measured as a percentage (area)of the whole muscle section. All section analysis was done‘blind’. Histological analysis was completed as per theTREAT-NMD recommended standard protocol “Histo-logical measurements of dystrophic muscle – M.1.2_007”

http://www.treat-nmd.eu/research/preclinical/SOPs/.

2.4. Blood collection and serum creatine kinase (CK) assay

While mice were under terminal anaesthesia, whole blood(approx. 0.5 ml) was collected via cardiac puncture using a27.5 gauge tuberculin syringe (Sigma Z192082), into a1.5 ml tube. Extensive experimentation revealed that storageof blood samples overnight at 4 �C to enable clotting leads toa false increase in serum CK levels and therefore blood sam-ples were immediately spun down in a refrigerated centri-fuged for 5 min (12,000g), serum was removed andaliquoted. Blood serum CK activity was determined induplicate using the CK-NAC kit (Randox Laboratories)and analysed kinetically using a BioTek Powerwave XSSpectrophotometer using the KC4 (v 3.4) program. A mini-mum of 10 ll serum is required to complete this assay.

2.5. Measuring cytokine gene expression by RNA extraction

and RT-PCR

Levels of mRNA for three inflammatory cytokines (IL-1b, IL-6 and TNF) was measured in the quadriceps musclesince this appeared to have the greatest amount of exer-cise-induced muscle damage, as indicated by the extent ofmyofibre necrosis (Figs. 4 and 5). RNA was extracted fromone half of a snap frozen quadriceps muscles using Tri-reagent (Sigma T9424) and quantitated using a Nano DropSpectrophotometer (ND 1000) and ND 1000 software ver-sion 3.5.2. The RNA was DNAse treated using PromegaRQ1 RNAse free DNAse (M610A), RQ1 RNAse free 10�buffer (M198A) and RQ1 DNAse stop solution (M199A).RNA was reverse transcribed into cDNA using PromegaM-MLV Reverse Transcriptase (M3682), random primers(C1181) and 10 mM dNTPs (U1515) and the cDNA waspurified using a MoBiol Clean up kit (12500-250). RT-PCRswere run on a Corbett 3000 (Corbett Research) using QIA-GEN quantifast SYBR green PCR mix (204054) and QIA-GEN Quantitect Primer Assays for IL-1b (QT01048355),IL-6 (QT00098875) and TNF (QT00104006), and standard-ised to a house-keeping gene; ribosomal protein L-19(QT01779218) as per [33]. mRNA expression levels werecalculated and standardised using Rotor-gene 6.1 andMicrosoft Excel software.

2.6. Quantitation of oxidative stress

Oxidative stress in the quadriceps muscle was measuredin two different ways:


2.6.1. As a ratio of oxidised (di-sulphide) to reduced

(sulfhydryl) protein thiols – 2-Tag technique

Frozen quadriceps muscles were crushed under liquidnitrogen, before protein extraction with 20% trichloroaceticacid/acetone. Protein was washed with acetone and solubi-lised in 0.5% sodium dodecyl sulphate 0.5 M tris (SDS buf-fer), pH 7.3 and thiols were labelled with the fluorescentdye BODIPY FL-N-(2-aminoethyl) maleimide (FLM, Invit-rogen). Following removal of the unbound dye using etha-nol, protein was resolubilised in SDS buffer, pH 7 andoxidised thiols were reduced with tris(2-carboxyethyl)phos-phine (TCEP, Sigma) before the subsequent unlabelledreduced thiols were labelled with a second fluorescent dye,texas red maleimide (Invitrogen). The sample was washedin pure ethanol and resuspended in SDS buffer. Samples wereread using a fluorescent plate reader (Fluostar Optima) withwavelengths set at excitation 485, emission 520 for FLM andexcitation 595, emission 610 for texas red. A standard curvefor each dye was created using ovalubumin and all resultswere expressed per mg of protein, quantified using DetergentCompatible protein assay (BioRad), as per [34,35].

2.6.2. Malondialdehyde (MDA)A product formed via the decomposition of lipid perox-

idation products [36] was quantitated using High Perfor-mance Liquid Chromatography (HPLC). Quadricepsmuscles were ground under liquid nitrogen, homogenisedin 10 � 5% perchloric acid and 150 ll of supernatant mixedwith 150 ll of 40 mM Thiobarbituric acid. Samples werethen incubated at 50 �C for 90 min and cooled on ice for15 min. Butanol (250 ll) was added, before vortexing andcentrifuging for 5 min. Twenty microliters of the upperbutanol layer was injected into a C18 HPLC column(5 ll, 4.6 � 150 mm, Dionex) with an isocratic mobilephase of 60:40 50 mM KH2PO4:methanol. Samples wererun at a flow rate of 800 ll/min for 7 min; the retentiontime was approximately 4.5 min. Fluorescent detectionwas achieved using the bandpass filters of 515 for excita-tion and 553 for emission. Tetraethoxypropane (Sigma)was used as a standard for absolute calculation. Approxi-mately 30 mg of tissue was required for this assay.

2.7. Statistics

Statistical analysis was performed using Microsoft exceland SPSS 16.0. Data were checked for equal distributionand normality using Q–Q plots. Multiple variables wereanalysed by ANOVAs (one, two or three-way to accountfor exercise, sampling time and strain). All data areexpressed as mean ± SEM unless otherwise stated.

3. Results

3.1. Ability of mice to run on the treadmill

Protocol A: Preliminary studies revealed that approxi-mately 45% of 12-week-old untrained mdx mice could




Fig. 2. Forelimb grip strength (kg) and normalised strength change (kg/gbw); a comparison of unexercised mdx mice with mice subjected to4 weeks of treadmill exercise (Protocol B). Absolute forelimb strength issignificantly decreased (�) in 12-week-old treadmill exercised mice andnormalised change in forelimb strength is also decreased after 4 weeks oftreadmill exercise. Normalised change in forelimb strength was calculatedby normalising absolute force to bodyweight (kg force/g body weight) andby subtracting normalised forelimb strength of 12-week-old mice fromnormalised forelimb strength of 8-week-old mice. *Significant difference(P < 0.05), in exercised compared to unexercised mdx mice. N = 32, 8, 24,8, 24 mice, respectively. Normalised change in forelimb strength isincreased by a scale factor of 100 for graphical purposes only.


ARTICLE IN PRESS

not complete a full 30 min of treadmill exercise at 12 m/min(despite being rested 5 times during the 30 min exercise ses-sion), some exhibited severe fatigue after 10 min exercise.This inability to exercise on a treadmill is similar to previ-ous reports [23,31,37]. For this reason the additional‘warm-up’ period (8 min at 8 m/min) was included in thetreadmill protocol to help the mdx mice to complete thetreadmill exercise session, as per [23]. It is extremely impor-tant, especially when only conducting one single exercisesession, to minimise biological variation in the experimen-tal exercise protocol. Adding a ‘warm-up’ period signifi-cantly increased the ability of mdx mice (92%) tocomplete the treadmill exercise session. Out of the 24mdx mice exercised in Protocol A, 2 mice did not fullycomplete the 30 min treadmill exercise; these 2 mice eachcompleted approximately 26 min exercise but, since neitherproduced any outlying results, the data were included in allanalyses.

C57Bl/10 mice were all capable of completing the30 min treadmill exercise without any rests. Previous exper-iments in our laboratory indicate that C57Bl/10 mice arealso capable of completing 30 min of treadmill exercise at12 m/min without an extensive warm-up and can also runcomfortably for 60 min (Grounds et al., unpublished data).

Protocol B: The average number of rests required tocomplete each exercise session reduced throughout the4 week exercise period, particularly after the 3rd week, sug-gesting that mdx mice can improve their running abilitywith exercise training (Fig. 1). Untrained 12-week-old malemdx mice (Protocol A) required significantly (P < 0.01)more rests to complete a 30 min exercise session comparedto 12-week-old male mice that had been ‘trained’ for4 weeks (Protocol B) (Fig. 1). However, it must be notedthat the running protocol used in this study involved rest-ing all the mdx mice on the treadmill when only one wasexperiencing fatigue and this may have had an influence

Fig. 1. Running pattern of mdx mice over 4 weeks of treadmill exercise (8sessions) compared to mice subjected to a single exercise session. Theaverage number of rests required to complete each 30 min exercise sessionsis shown. Mice were 8 weeks old at the start of the Protocol B and all micewere sampled when 12 weeks old (Protocols A and B). The number of restsis significantly reduced (�) in 12-week-old mdx mice after 4 weeks oftreadmill exercise (Protocol B) in comparison to both untrained 12-week-old (Protocol A) and untrained 8-week-old mdx mice. P < 0.05, N = 24mice in each session. Bars represent standard error.


on mdx running ability overtime. There was no significantdifference in the running ability of unexercised 8-week-old(1st session of Protocol B) and unexercised 12-week-old(Protocol A) mdx mice (Fig. 1).

3.2. Forelimb grip strength (Protocol B only)

After 4 weeks of treadmill exercise (Protocol B) the fore-limb grip strength of 12-week-old mdx mice was signifi-cantly weaker than unexercised mice (Fig. 2) as shown bya significant (P < 0.01) decrease in both absolute forelimb

Fig. 3. Biological variation in myofibre necrosis in the quadriceps muscleof 12-week-old unexercised male mdx mice. Each bar represents theunexercised control group from a separate experiment conducted in ourlaboratory over a period of 12 months. When pooled together (n = 60muscles) the average amount of myofibre necrosis in the quadricepsmuscle of unexercised 12-week-old male mice is 6.12%. Bars representstandard deviation (to highlight range) and n = 4–8 muscles per group. (Aand B) Significant differences, groups with different letters are significantlydifferent from each other (P < 0.05).




ARTICLE IN PRESS

strength (0.168 kg ± 0.007 unexercised vs 0.124 kg ± 0.005exercised) and a significant (P < 0.01) decrease in norma-lised change (normalised strength at 12 weeks minus nor-malised strength at 8 weeks) in forelimb strength(0.002 kg/g ± 0.0004 unexercised vs 0.00038 kg/g ± 0.0002exercised) (Fig. 2). This decrease in forelimb grip strengthis similar to previous reports [31,32] and falls within theexpected general range discussed in the TREAT-NMD rec-ommended standard protocol “Use of grip strength metreto assess limb strength of mdx mice – M.2.2_001” http://www.treat-nmd.eu/research/preclinical/SOPs/.

3.3. Histological analysis of muscle necrosis

3.3.1. Biological variation

Protocol A: One striking feature of histological analysiswas the high variation in the amount of myofibre necrosis

Fig. 4. Myofibre necrosis (% CSA) in the quadriceps, triceps andgastrocnemius muscles of 12-week-old male mdx mice: a comparison ofunexercised mice with mice subjected to a single 30 min exercise session(Protocol A). Myofibre necrosis in the quadriceps muscle is significantlyincreased in treadmill exercised mice when sampled both 24 h (n = 25) and48 h (n = 43) after exercise in comparison to unexercised mice (n = 60).Bars represent standard error. N = 28, 19, 12, respectively, for triceps and11, 15, 20, respectively, for gastrocnemius muscle. *Significant difference(P < 0.05), in exercised quadriceps compared to unexercised quadriceps.

Fig. 5. Myofibre necrosis (% CSA) in the quadriceps, triceps, gastrocnemius,comparison of unexercised mdx mice with mdx mice sampled after 4 weeks of tmuscle in comparison to unexercised muscle (same muscle only). #Significant de(same muscle only). For unexercised mice n = 60, 28, 11, 8, 8 muscles, respectivedifferences were determined by P < 0.05.


(fragmented sarcoplasm and inflammatory cell infiltration)from both sedentary and exercised (age, sex and musclematched) mdx mice. The variation in myofibre necrosis(% area) in quadriceps muscle from 12-week-old sedentarymale mdx mice is demonstrated in Fig. 3 and ranges from1.04% to 23.1% for each individual quadriceps muscle and2.97–17.15% average per experimental group. When resultsfrom nine separate experiments were pooled together(n = 60 quadriceps) the average amount (% CSA) of myo-fibre necrosis in the quadriceps muscle of an unexercised12-week-old male mice is 6.12%. Pooled histological datawere also used for myofibre necrosis in unexercised tricepsmuscle – 8.5% (n = 28) and unexercised gastrocnemiusmuscle – 6.89% (n = 11).

3.3.2. Exercise induced myofibre necrosis

Protocol A: High variation in myofibre necrosis wasagain seen in response to a single 30 min treadmill exercisesession and pooled histological data were also used for thevarious muscles sampled 24 h after exercise from fourexperiments (n = 15–25) and at 48 h after exercise fromsix experiments (n = 13–43) (Fig. 4). The quadriceps mus-cle showed the highest level of myofibre necrosis(15.06 ± 6.01%) after a single bout of treadmill exercisewhen sampled 24 h, compared with triceps, gastrocnemius,(Fig. 4) tibialis anterior (TA) and extensor digitorum lon-gus (EDL) (data not shown). Necrosis was significantlyincreased in treadmill exercised (compared with unexer-cised) quadriceps muscles when sampled at either 24 h(P < 0.01) or 48 h (P = 0.04) post exercise. Both the TAand the EDL muscle from 12-week-old mdx mice had avery low level of background myofibre necrosis (aver-age < 3%) and both appeared unaffected by the singletreadmill exercise session with no consistent or significantincrease in myofibre necrosis (data not shown), in accor-dance with previous reports [24,26,27].

Unexercised C57Bl/10 mice show no myofibre necrosisin their skeletal muscles (0%). The C57Bl/10 mice from

diaphragm and tibialis anterior muscles of 12-week-old male mdx mice: areadmill exercise (Protocol B). *Significant increase in necrosis in exercisedcrease in necrosis in exercised muscle in comparison to unexercised musclely. N = 8 for all other groups. Bars represent standard error and significant





Fig. 6. Blood serum CK in 12-week-old male C57 and mdx mice,comparing unexercised mice with mice subjected to a single 30 mintreadmill session (Protocol A) and 4 weeks of treadmill exercise (ProtocolB). There is no change in serum CK in C57 mice after treadmill exercise.Serum CK is significantly increased after a single 30 min exercise session(Protocol A) when mdx mice are sampled 0 min and 2 h post exercise incomparison to unexercised mdx mice. Serum CK is significantly elevatedin mdx mice subject to 4 weeks of treadmill exercise (Protocol B) whensampled 24 h post exercise in comparison to unexercised mdx mice. N = 8mice per group and bars represent standard error. (A–C) Significantdifferences, groups with different letters are significantly different fromeach other.


ARTICLE IN PRESS

Protocol A were sampled 0 min after treadmill exercise andthus do not show any myofibre necrosis (nor did we expectthem to within 30 min of the start of the exercise). In orderfor precise comparison with mdx mice we examined C57Bl/10 mice sampled at 24 h after treadmill exercise fromanother experiment conducted within our laboratory(Radley-Crabb et al. unpublished data), again theseC57Bl/10 mice show 0% myofibre necrosis after horizontaltreadmill exercise.

Protocol B: Myofibre necrosis was significantly elevatedin the quadriceps (P = 0.04), triceps (P = 0.05), diaphragm(P = 0.04) and TA (P = 0.05) muscles after 4 weeks oftreadmill exercise training when mdx mice were sampled24 h after the last exercise session. Necrosis was also signif-icantly elevated in the diaphragm muscle (P = 0.02) whensampled immediately (0 min) after the last exercise session(Fig. 5), suggesting prolonged myofibre necrosis after thepenultimate exercise session or a particular sensitivity toexercise induced damage in the diaphragm.

Necrosis was most elevated in the quadriceps (2� fold)and diaphragm (3� fold) muscle at 24 h after the last exer-cise session. The consistently elevated necrosis in the quad-riceps is similar to that seen after a single exercise session(Protocol A). No significant increase in myofibre necrosiswas seen in the exercised gastrocnemius muscle possiblydue to high level of variation in this parameter (Fig. 5).Myofibre necrosis (fragmented sarcoplasm and inflamma-tion) returned to unexercised levels (or below) in all mus-cles, within 96 h after exercise (i.e. when the next exercisesession would be due). This indicates that dystrophic myo-fibres can regenerate muscle to replace necrotic sarcoplasmand inflammation in between each treadmill exercise ses-sion. This also emphasises the importance of sampling timewhen quantitating myofibre necrosis after treadmillexercise.

3.4. Blood serum CK levels as a measure of muscle leakiness

Protocol A: Serum CK levels are very low in controlC57Bl/10 mice and no significant change was seen after30 min of treadmill exercise in control C57Bl/10 mice sam-pled 0 min after exercise (unexercised 183.3U/L ± 53.8 vs0 min post exercise 267.7U/L ± 98.7) (Fig. 6). Additionaldata from another experiment conducted in our laboratoryalso show no change in serum CK levels at 24 h after tread-mill exercise in C57Bl/10 mice (unexercised 183.3U/L ± 53.8 vs 24 h post exercise 167.9U/L ± 54.9). However,serum CK levels were rapidly elevated in response to tread-mill exercise in mdx mice and were significantly higher(P = 0.01) when blood was collected immediately (0 min)after exercise. The exercise induced increase in serum CKwas transient and CK levels dropped rapidly down to unex-ercised level within 24 h (Fig. 6).

Protocol B: In mdx mice subject to 4 weeks treadmillexercise, blood serum CK levels were significantly(P < 0.01) elevated at 24 h after the last exercise sessionand decreased to unexercised level within 96 h (Fig. 6). This


is in marked contrast to the rapid elevation of CK in mdxmice subjected to a single exercise session (Protocol A).

3.5. Inflammatory cytokine gene expression in the quadriceps

muscle

Protocol A: IL-6 mRNA levels were significantlyincreased in the quadriceps muscle from C57Bl/10 miceafter a single 30 min treadmill session (Fig. 7ii); althoughthere was no significant change in IL-1b or TNF mRNA.Expression of all three inflammatory cytokines was signif-icantly higher in unexercised mdx mice compared to unex-ercised C57Bl/10 mice (Fig. 7i–iii). In mdx mice, mRNAlevels for both IL-1b (2 h post exercise P = 0.03) and IL-6 (0 min post exercise, P = 0.05 and 2 h post exerciseP = 0.05) were significantly elevated after a single 30 minexercise session compared to unexercised mice (Fig. 7iand ii). This rapid increase in gene expression returned tothe unexercised level for both genes within 24 h. In con-trast, levels of mRNA for TNF were significantly reducedafter exercise (0 min post exercise P < 0.01, 2 h post exer-cise P = 0.02 and 24 h post exercise P < 0.01) comparedto unexercised mdx mice (Fig. 7iii).

Protocol B: In mdx mice there was no change in mRNAfor IL-1b or IL-6 immediately after 4 weeks of treadmillexercise, although mRNA for both IL-1b and IL-6 was sig-nificantly (P = 0.05) decreased for muscles sampled 96 hpost exercise, compared to unexercised mdx mice (Fig. 8iand ii). TNF mRNA was significantly decreased in musclessampled immediately (0 min) after exercise compared tounexercised mdx mice (Fig. 8iii), but returned to unexer-cised level within 24 h.



Fig. 7. Gene expression (mRNA) changes in the quadriceps muscle ofnon-dystrophic C57Bl/10 mice and dystrophic mdx mice in response to asingle 30 min exercise session (Protocol A) for (i) IL-1b, (ii) IL-6 and (iii)TNF. IL-1b mRNA is significantly increased in mdx mice at 2 h postexercise compared to unexercised mdx mice. IL-6 is significantly increasedin both C57Bl/10 and mdx mice immediately after treadmill exercise. TNFis significantly decreased after treadmill exercise at all times in mdx mice.Bars represent standard error. N = 7–8 mice per group. (A–C) Significantdifferences, groups with different letters are significantly different fromeach other (P < 0.05).

Fig. 8. Gene expression (mRNA) changes in the quadriceps muscle ofmdx mice in response to 4 weeks treadmill exercise (Protocol B) for (i) IL-1b, (ii) IL-6 and (iii) TNF. Both IL-1b and IL-6 mRNA are significantlydecreased after 4 weeks of treadmill exercise when mdx mice are sampled96 h post exercise. TNF mRNA is significantly decreased at 0 min afterexercise compared to unexercised mdx mice. Bars represent standarderror. N = 8 mice per group. (A–C) Significant differences, groups withdifferent letters are significantly different from each other (P < 0.05).


ARTICLE IN PRESS

3.6. Oxidative stress measurement in the quadriceps muscle

3.6.1. Ratio of oxidised (di-sulphide) to reduced

(sulfhydryl) protein thiols

Protocol A: Oxidative stress in the quadriceps muscle,specifically the ratio of oxidised to reduced protein thiolsas measured by the novel 2-tag technique, was significantly(P = 0.01) higher in unexercised mdx compared to unexer-cised C57Bl/10 12-week-old male mice (Fig. 9i). Proteinthiol oxidation was also significantly elevated in mdx mice


0 min (P = 0.015) and 2 h (P = 0.04) after a single treadmillsession compared to unexercised mdx mice (Fig. 9i). Proteinthiol oxidation returned to unexercised level within 24 h.

Protocol B: Similarly, protein thiol oxidation was signif-icantly increased (P = 0.02) in mdx mice after 4 weeks oftreadmill exercise when sampled immediately (0 min) afterexercise (Fig. 9ii) and returned to unexercised level within24 h.

3.7. Quantitation of malondialdehyde (MDA)

Protocol A: Oxidative stress in the quadriceps musclewith respect to irreversible lipid peroxidation, measured




ARTICLE IN PRESS

as concentration of MDA, showed no significant differencebetween C57Bl/10 and mdx mice, and was unaffected byexercise even in mdx mice (Fig. 10). MDA levels werenot measured for Protocol B.

Fig. 9. Ratio of oxidised to reduced protein thiols in the quadricepsmuscle of non-dystrophic C57Bl/10 and dystrophic mdx mice in response(i) a single 30 min treadmill exercise session (Protocol A) and (ii) 4 weeksof treadmill exercise (Protocol B). Oxidative stress is significantly higher in12-week-old male mdx mice compared to C57Bl/10 mice. It is furtherelevated in mdx mice at both 0 min and 2 h after a single treadmill exercisesession (Protocol A) and at 0 min after 4 weeks of treadmill exercise(Protocol B). (A–C) Significant differences, groups with different letters aresignificantly different from each other (P < 0.05).

Fig. 10. Malondialdehyde (MDA) concentration in the quadriceps muscleof C57Bl/10 and mdx mice in response to a single 30 min treadmill exercisesession (Protocol A). There is no difference in the MDA level between non-dystrophic C57Bl/10 and mdx muscle from both unexercised mice andmice sampled immediately after exercise (0 min). (A and B) Significantdifferences, groups with different letters are significantly different fromeach other (P < 0.05).


4. Discussion

Treadmill exercise is widely used in pre-clinical experi-ments to increase the extent of dystropathology in mdxmice, yet the cellular consequences of a single 30 min tread-mill exercise session (Protocol A) have not been describedpreviously. The present study analysed the time-course ofmolecular and cellular changes after a single standardised30 min treadmill exercise session (Protocol A). These datawere compared to data from age matched mdx mice sub-jected to 4 weeks of twice weekly treadmill exercise (Proto-col B) a protocol currently widely used for pre-clinical drugscreening in mdx mice [11,20,22,25].

4.1. Running ability of mdx mice

Adding a short warm-up period for 8 min at a slowerspeed (8 m/min) produced much more consistent runningby the mdx mice in both treadmill exercise protocols. Withonly a single exercise session it is important that all mice com-plete the exercise protocol to reduce variation. It is not rec-ommended to remove the ‘non-running’ mice from samplegroups as these may represent mice with the most severe dyst-ropathology; in comparison to a full range of dystropathol-ogy being represented in unexercised control mice.Preliminary experiments showed that adding a warm-up per-iod significantly increased the ability of mdx mice (from 45%up to 92%) to complete the 30 min treadmill exercise session.

The intermittent running pattern of mdx mice andreduced capacity for exercise (compared to C57Bl/10 mice)on a voluntary exercise wheel is well documented[11,38,39]. With such voluntary running mdx mice can stopand start as they wish, yet they still manage to run a con-siderable total amount e.g. up to 14 km over 48 h and upto 435 km over 8 weeks for untreated adult mdx mice[26,27,30]. In contrast, treadmill exercise requires continu-ous running (for at least 30 min) and mdx mice can strugglewith this type of exercise (indicating fatigue) [23,31,32,37].Thus, during long-term studies with repeat (often twiceweekly) treadmill sessions it is important to note the num-ber of times that a mouse stops running during each tread-mill exercise session as this provides some insight intorunning ability and thus muscle condition.

In the present study, 4 weeks of treadmill exercise train-ing (Protocol B) significantly improved the running abilityof mdx mice (Fig. 1). Despite exhibiting a significant reduc-tion in both absolute forelimb strength (kg) and change innormalised forelimb strength (kg/g bw) compared to unex-ercised mdx mice (Fig. 2), exercised mdx mice show a smallincrease in normalised forelimb strength after 4 weeks oftreadmill exercise (Protocol B). This increase in normalisedforelimb strength along with behavioural adaptation toexercise training may account for the improvement intreadmill running ability. However, it must be noted thatduring the treadmill exercise protocol mice were run ingroups of 3 or 4 and if one mouse fatigued during the30 min protocol all mice on the treadmill were rested and




ARTICLE IN PRESS

this may have impacted the results. Some studies documentan improvement in the voluntary wheel exercise ability(distance run) of mdx mice over time, especially when exer-cise is started at a young age (reviewed in [40]). However,an improvement in running capacity on a treadmill over-time has not been previously reported for mdx mice.

4.2. Myofibre necrosis

Increased myofibre necrosis after both treadmill exerciseprotocols is transient (Figs. 4 and 5) and muscles must besampled between 24 and 48 h after exercise to visualisethe increase in this histological parameter. While dystrophicskeletal muscles appear fully capable of regeneratingbetween repeat exercise sessions (Fig. 5) it must be notedthat grip strength is significantly reduced in mdx mice sub-jected to repeated bouts of treadmill exercise compared withage-matched unexercised mdx mice. Skeletal muscle fibrosiswas not measured in this study although it is likely that, dueto exercise induced cycles of myofibre necrosis associatedwith inflammation (and regeneration), fibrosis is indeedprogressively increased after 4 weeks of treadmill exerciseand may impact negatively on forelimb grip strength.

Similar to histological results seen after voluntary wheelexercise, treadmill exercise induces a large amount of dam-age in the quadriceps muscle [24,26,27]. Large variation inthe extent of myofibre necrosis is seen for most muscles;with the quadriceps muscle showing the highest increasein necrosis after a single 30 min exercise session (ProtocolA – Fig. 4) and the quadriceps, triceps and diaphragm mus-cles all showing increased necrosis after 4 weeks of tread-mill exercise (Protocol B – Fig. 5).

Four weeks of treadmill exercise (Protocol B) induced amore consistent increase in myofibre necrosis in many mus-cles (excluding the gastrocnemius) compared to a single30 min exercise session (Protocol A). These data emphasisethat large groups of mdx mice (at least 8 mice) are requiredfor histological analysis, due to the notorious variation inmdx mice (reviewed in detail [11,17]) and that the choiceof muscle is critical to assess the impact of exercise induceddamage.

There is no myofibre necrosis in unexercised C57Bl/10control mice nor does horizontal treadmill exercise producenecrosis in these mice. The resistance of normal (non-dys-trophic) muscle to treadmill exercise-induced necrosis isnot surprising, especially since we have previously reportedthat even much more damaging eccentric contractionsin vivo do not result in muscle necrosis in control C57Bl/10 mice (in marked contrast with mdx mice) [28]. Similarly,Roche et al. clearly showed no significant increase in myo-fibre necrosis after eccentric contractions (large straininjury) in normal control (A/WySnJ) mice [41].

4.3. Blood serum creatine kinase level

Serum CK activity is widely used as an indirect measureof muscle damage (sarcolemma leakiness) and levels are


consistently increased in mdx mice after exercise[22,26,30,31,42]. There is no absolute correlation betweenthe extent of dystropathology in an individual mdx mouseand CK activity [30], with many factors including stressand muscle mass influencing CK activity (reviewed in[11]). CK is an enzyme and has a short circulating half lifeof approximately 12 h [43], thus there is considerable inter-est in understanding the kinetics of CK release from dys-trophic muscle after treadmill exercise.

Serum CK was strikingly increased immediately (0 min)after a single 30 min exercise session (Protocol A) andreturned to baseline within 24 h post exercise (Fig. 6). Thiselevated level is similar to the transient elevation in mdxserum CK reported at 1 h after eccentric exercise (16�downhill, 10 m/min for 5 min) [19] and to the increased(10� fold) serum CK after 8 weeks of voluntary wheelexercise [26]. This immediate increase in serum CK afterexercise indicates that Protocol A was sufficiently strenuousto render myofibres ‘leaky’ and allow the release of CKinto circulation; however it was not damaging enough toinduce widespread myofibre necrosis (Fig. 4) in many mus-cles with the exception of the quadriceps. This is in markedcontrast to the blood serum CK levels in control C57Bl/10mice which are very low and do not increase after treadmillexercise (Fig. 6).

Serum CK activity is a measure of muscle damagethrough-out the whole body, unlike histological analysisof myofibre necrosis which exclusively examines sectionsof an individual muscle. In addition to the results pre-sented, some mice received Evans Blue Dye injections24 h prior to the single 30 min exercise session to enablehistological quantitation of ‘leaky’ myofibres; however thisdid not produce any consistent results (data not shown).

In contrast to a single exercise session (Protocol A), CKlevels in mdx mice subjected to 4 weeks treadmill training(Protocol B) were significantly increased at 24 h after exer-cise, but not immediately after exercise (Fig. 6). This sug-gests either a delayed or sustained release of CK fromleaky myofibres to account for the prolonged elevation ofblood CK activity. Many factors including training andtype of exercise can affect the level and duration of CKincrease; for example in humans a single session of highintensity resistance exercise immediately increases bloodserum CK which peaks at 24 h and begins to decline within48 h [44]. Our result emphasises the importance of docu-menting the timing of such events after different exerciseregimes in order to determine the optimal time for sam-pling (sacrifice and tissue collection) after exercise whenmeasuring specific parameters.

4.4. Gene expression of inflammatory cytokines

IL-1b, IL-6 and TNF are three major pro-inflammatorycytokines; however it is also recognised that IL-6 is a myo-kine, with many important anti-inflammatory and meta-bolic effects, produced by and released from contractingmyofibres in vivo (reviewed in [45,46]). Accordingly, a




ARTICLE IN PRESS

significant increase in IL-6 mRNA was seen in the quadri-ceps muscle from both C57Bl/10 and mdx mice after a sin-gle 30 min treadmill session (Fig. 7ii). The increased IL-6mRNA in C57Bl/10 mice after treadmill exercise does notcoincide with any inflammation (no change in the expres-sion level of IL1b or TNF – Fig. 7i and iii) or with muscledamage (no increase in blood serum CK level (Fig. 6) ormuscle necrosis) and further demonstrates the capacity ofnormal muscle contraction to increase IL-6 production.The pronounced increase in IL-6 mRNA in mdx mice aftera single 30 min exercise session (Protocol A) is presumablycaused by both a response to myofibre contraction andexercise induced muscle damage. There was no significantincrease in IL-6 mRNA in mdx mice after 4 weeks of tread-mill exercise (Protocol B); this may reflect an adaptation(training) in response to treadmill exercise over time.

The level of TNF mRNA was significantly reduced afterboth treadmill exercise protocols in mdx mice. While thereis strong evidence that TNF plays a major role in the dyst-ropathology of mdx mice and blockade of TNF can reducemyofibre necrosis [26,30,42,47], increased IL-6 after exer-cise can inhibit TNF [48–50] and this may explain the tran-sient reduction in TNF mRNA seen in response to exercisein the present study.

While there were transient changes in TNF, IL1b andIL-6 mRNA expression after exercise, changes and bio-availability at the protein level were not measured. This isin part because of issues with sensitivity of quantification[51]. Indeed, attempts were made to measure TNF proteinlevels in blood serum using a standard ELISA (Invitrogen,USA), however serum TNF levels for all mice were belowthe lowest standard (15.6 qg/ml) and therefore results wereuninformative (data not shown). Another important factorto consider is that high levels of TNF (and other cytokine)protein are already present, yet sequestered, within residentand invading inflammatory cells (e.g. mast cells, neutrophilsand macrophages) in dystrophic muscle [13,27,30,52–55].Protein quantification makes no comment on the bioavail-ability or the re-distribution of these cytokines: for examplethese proteins are rapidly released from activated mast cellsand other inflammatory cells (e.g. neutrophils and macro-phages) that accumulate after myofibre damage. Thus it islikely that, despite a reduction in TNF mRNA after tread-mill exercise, localised bioavailable TNF protein increasesrapidly independent of gene transcription [56].

4.5. Oxidative stress

A significant increase in oxidative stress in unexercisedmdx compared to unexercised C57Bl/10 quadriceps musclewas demonstrated by the increased ratio of oxidised toreduced protein thiols (reversible oxidative modification)as measured by the novel 2-tag technique [34,35]. Thiselevated oxidative stress in dystrophic muscle supportsearlier reports that measured oxidative stress in dystrophicmuscle (from both DMD patients and mdx mice)[16,25,57–60].


Our second measurement of oxidative stress, using aHPLC method to quantitate MDA and irreversible perox-idation of membrane lipids, showed no significant differ-ence in oxidative stress between C57Bl/10 and mdxquadriceps muscle and MDA levels were not increased inmdx muscle after 30 min treadmill exercise (Protocol A).While previous studies have reported increased MDA inthe hind limbs of young (<20 days) mdx mice [60] andthe gastrocnemius muscle of 90 day old mdx mice [57],compared to age-matched C57Bl/10 mice, there are numer-ous reasons for this discrepancy in MDA results, includinganimal age and specific muscle examined. It is also impor-tant to note that many commonly used methods for mea-suring MDA have specificity issues and only HPLCbased methods are recommended (reviewed in [61]).

Despite no evidence of increased lipid peroxidation inadult mdx quadriceps muscle, reversible changes in oxida-tive state were evident by the ratio of oxidised to reducedprotein thiols in mdx muscle. Importantly this 2-tagmethod identified rapid and significant increases in proteinthiol oxidation immediately (0 min) and 2 h after a single30 min treadmill exercise session, emphasising the speedof such changes in vivo and the sensitivity of this specificassay for oxidative stress.

There is an increasing need for standardised experimentsinvolving mdx mice in order to readily compare databetween laboratory groups globally; this is the subject ofrecent reviews [11,17] that aim to establish a set of recom-mendations for pre-clinical mdx drug trials. The presentstudy provides strong support for a single 30 min exercisesession in adult mdx mice as an appropriate fast protocolto conduct preliminary ‘proof-of-concept’ testing of poten-tial therapeutic drugs to reduce the severity of myofibrenecrosis associated with muscular dystrophy. We have suc-cessfully conducted in vivo studies in adult mdx mice exam-ining the potential benefits of on N-acetylcysteine (NAC)using the 30 min treadmill exercise protocol established inthis manuscript (Terrill et al., under review).

A single 30 min exercise session (Protocol A) results in asimilar level of muscle damage (muscle necrosis, serum CK,oxidative stress) as 4 weeks of treadmill exercise (ProtocolB) and thus the single exercise session appears suitable asa high through-put screening test. However a short termprotocol does not allow for monitoring of running patternor changes in normalised grip strength over time. It isnoted that potential therapeutic drugs for DMD (identifiedin pre-clinical proof-of-concept short studies) should betested chronically to examine efficiency, toxicity and alsopossible negative side effects e.g. on heart function.

We conclude that a single 30 min treadmill exercise ses-sion is a suitable screening protocol for assessing therapeu-tic interventions in adult mdx mice (proof-of-conceptstudies), that serum CK level, myofibre necrosis and oxida-tive stress in the quadriceps muscle are key endpointswhich should be monitored when assessing the efficacy ofdrug treatments in combination with treadmill exercise(summarised in Table 2), and emphasise the importance



Table 2Summary of reliable indicators of muscle damage after treadmill exercise in dystrophic mdx mice.

Indicator of muscle damage Reliable indicator of muscle damage – Yes/No

1 � 30 min treadmill 4 weeks treadmill exercise

Histology – quadriceps Yes (large sample group critical) Yes (timing critical)Histology – triceps, TA, gastrocnemius, diaphragm No Yes (timing critical)Blood serum CK Yes (timing critical) Yes (timing critical)IL-1b mRNA level – quadriceps Noa NoIL-16 mRNA level – quadriceps Noa NoTNF mRNA level – quadriceps Noa NoProtein thiol redox state – Quadriceps Yes (timing critical) Yes (timing critical)Malondialdehyde concentration – quadriceps No Not done

a An extremely high level of high biological variation (large standard error) thus making the parameter unreliable as an indicator ofmuscle damage.


ARTICLE IN PRESS

of specific muscle and sampling time (sacrifice and tissuecollection). It is hoped that this efficient single exercise pro-tocol will help accelerate pre-clinical drug trials in mdxmice and that further insight into the very early events thatlead to myofibre necrosis will identify more precise and bet-ter targets for drug interventions to reduce the severity ofthe dystropathology.

Acknowledgements

The authors thank Greg Cozens and Griffin Groundsfor excellent technical assistance. Research funding fromthe Australian National Health and Medical ResearchCouncil (M.G., T.S. and P.A.) and Australian Postgradu-ate Award Scholarships (H.R.-C. and J.T.) are gratefullyacknowledged.

References

[1] Emery EH. The muscular dystrophies. Lancet 2002;359:687–95.[2] Manzur AY, Muntoni F. Diagnosis and new treatments in muscular

dystrophies. J Neurol Neurosurg Psychiatry 2009;80:706–14.[3] Watchko JF, O’Day TL, Hoffman EP. Functional characteristics of

dystrophic skeletal muscle: insights from animal models. J ApplPhysiol 2002;93:407–17.

[4] Reed P, Bloch RJ. Postnatal changes in sarcolemmal organization inthe mdx mouse. Neuromuscul Disord 2005;15:552–61.

[5] Petrof BJ, Shrager JB, Stedman HH, et al. Dystrophin protects thesarcolemma from stresses developed during muscle contraction. ProcNatl Acad Sci USA 1993;90:3710–4.

[6] Ohlendieck K, Campbell KP. Dystrophin-associated proteins aregreatly reduced in skeletal muscle from mdx mice. J Cell Biol1991;115:1685–94.

[7] Gailly P. New aspects of calcium signaling in skeletal muscle cells:implications in Duchenne muscular dystrophy. Biochim Biophys Acta2002;1600:38–44.

[8] Blake DJ, Weir A, Newey SE, et al. Function and genetics ofdystrophin and dystrophin-related proteins in muscle. Physiol Rev2002;82:291–329.

[9] Biggar WD. Duchenne muscular dystrophy. Pediatr Rev2006;27:83–8.

[10] Grounds MD, Radley HG, Gebski BL, et al. Implications of cross-talk between tumour necrosis factor and insulin-like growth factor-1signalling in skeletal muscle. Clin Exp Pharmacol Physiol2008;35:846–51.

[11] Grounds MD, Radley HG, Lynch GS, et al. Towards developingstandard operating procedures for pre-clinical testing in the mdx


mouse model of Duchenne muscular dystrophy. Neurobiol Dis2008;31:1–19.

[12] Whitehead NP, Yeung EW, Allen DG. Muscle damage in mdx(dystrophic) mice. role of calcium and reactive oxygen species. ClinExp Pharmacol Physiol 2006;33:657–62.

[13] Tidball JG, Wehling-Henricks M. Damage and inflammation inmuscular dystrophy: potential implications and relationships withautoimmune myositis. Curr Opin Rheumatol 2005;17:707–13.

[14] Evans NP, Misyak SA, Robertson JL, et al. Dysregulated intracel-lular signaling and inflammatory gene expression during initialdisease onset in Duchenne muscular dystrophy. Am J Phys MedRehabil 2009;88:502–22.

[15] Radley HG, De Luca A, Lynch GS, et al. Duchenne musculardystrophy: focus on pharmaceutical and nutritional interventions. IntJ Biochem Cell Biol 2007;39:469–77.

[16] Tidball JG, Wehling-Henricks M. The role of free radicals in thepathophysiology of muscular dystrophy. J Appl Physiol2007;102:1677–86.

[17] Spurney CF, Gordish-Dressman H, Guerron AD, et al. Preclinicaldrug trials in the mdx mouse: assessment of reliable and sensitiveoutcome measures. Muscle Nerve 2009;39:591–602.

[18] Brussee V, Tardif F, Tremblay JP. Muscle fibers of mdx mice aremore vulnerable to exercise than those of normal mice. NeuromusculDisord 1997;7:487–92.

[19] Vilquin JT, Brussee V, Asselin I, et al. Evidence of mdx mouseskeletal muscle fragility in vivo by eccentric running exercise. MuscleNerve 1998;21:567–76.

[20] De Luca A, Pierno S, Liantonio A, et al. Enhanced dystrophicprogression in mdx mice by exercise and beneficial effects of taurineand insulin-like growth factor-1. J Pharmacol Exp Ther2003;304:453–63.

[21] Okano T, Yoshida K, Nakamura A, et al. Chronic exerciseaccelerates the degeneration–regeneration cycle and downregulatesinsulin-like growth factor-1 in muscle of mdx mice. Muscle Nerve2005;32:191–9.

[22] Granchelli JA, Pollina C, Hudecki MS. Pre-clinical screening of drugsusing the mdx mouse. Neuromusclul Disord 2000;10:235–9.

[23] Payne ET, Yasuda N, Bourgeois JM, et al. Nutritional therapyimproves function and complements corticosteroid intervention inmdx mice. Muscle Nerve 2006;33:66–77.

[24] Archer JD, Vargas CC, Anderson JE. Persistent and improvedfunctional gain in mdx dystrophic mice after treatment with L-arginine and deflazacort. FASEB J 2006;20:738–40.

[25] Burdi R, Rolland JF, Fraysse B, et al. Multiple pathological events inexercised dystrophic mdx mice are targeted by pentoxifylline:outcome of a large array of in vivo and ex vivo tests. J Appl Physiol2009;106:1311–24.

[26] Radley HG, Davies MJ, Grounds MD. Reduced muscle necrosis andlong-term benefits in dystrophic mdx mice after cV1q (blockade ofTNF) treatment. Neuromuscul Disord 2008;18:227–38.




ARTICLE IN PRESS

[27] Radley HG, Grounds MD. Cromolyn administration (to block mastcell degranulation) reduces necrosis of dystrophic muscle in mdxmice. Neurobiol Dis 2006;23:387–97.

[28] Piers AT, Lavin T, Radley-Crabb HG, et al. Blockade of TNF in vivousing cV1q antibody reduces contractile dysfunction of skeletalmuscle in response to eccentric exercise in dystrophic mdx and normalmice. Neuromuscul Disord 2011;21:132–41.

[29] Blaauw B, Mammucari C, Toniolo L, et al. Akt activation preventsthe force drop induced by eccentric contractions in dystrophin-deficient skeletal muscle. Hum Mol Genet 2008;17:3686–96.

[30] Hodgetts S, Radley H, Davies M, et al. Reduced necrosis ofdystrophic muscle by depletion of host neutrophils, or blockingTNFalpha function with Etanercept in mdx mice. NeuromusculDisord 2006;16:591–602.

[31] De Luca A, Nico B, Liantonio A, et al. A multidisciplinary evaluationof the effectiveness of cyclosporine a in dystrophic mdx mice. Am JPathol 2005;166:477–89.

[32] De Luca A, Nico B, Rolland JF, et al. Gentamicin treatment inexercised mdx mice. Identification of dystrophin-sensitive pathwaysand evaluation of efficacy in work-loaded dystrophic muscle. Neu-robiol Dis 2008;32:243–53.

[33] Shavlakadze T, Chai J, Maley K, et al. A growth stimulus is neededfor IGF-1 to induce skeletal muscle hypertrophy in vivo. J Cell Sci2010;123:960–71.

[34] Armstrong AE, Zerbes R, Fournier PA, et al. A fluorescent duallabeling technique for the quantitative measurement of reduced andoxidized protein thiols in tissue samples. Free Radic Biol Med2010;15:510–7.

[35] Lui JK, Lipscombe R, Arthur PG. Detecting changes in the thiolredox state of proteins following a decrease in oxygen concentrationusing a dual labeling technique. J Proteome Res 2010;9:383–92.

[36] Davies KJ. Oxidative stress, antioxidant defenses, and damageremoval, repair, and replacement systems. IUBMB Life 2000;50:279–89.

[37] Hudecki MS, Pollina CM, Granchelli JA, et al. Strength andendurance in the therapeutic evaluation of prednisolone-treatedMDX mice. Res Commun Chem Pathol Pharmacol 1993;79:45–60.

[38] Wineinger MA, Abresch RT, Walsh SA, et al. Effects of aging andvoluntary exercise on the function of dystrophic muscle from mdxmice. Am J Physiol Med Rehabil 1998;77:20–7.

[39] Hara H, Nolan PM, Scott MO, et al. Running endurance abnormal-ity in mdx mice. Muscle Nerve 2002;25:207–11.

[40] Grange RW, Call JA. Recommendations to determine exerciseprescription for Duchenne Muscular Dystrophy. Sport Sci Rev2007;35:12–7.

[41] Roche JA, Lovering RM, Bloch RJ. Impaired recovery of dysferlin-null skeletal muscle after contraction-induced injury in vivo. Neuro-report 2008;19:1579–84.

[42] Pierno S, Nico B, Burdi R, et al. Role of tumour necrosis factoralpha, but not of cyclo-oxygenase-2-derived eicosanoids, on func-tional and morphological indices of dystrophic progression in mdxmice: a pharmacological approach. Neuropathol Appl Neurobiol2007;33:344–59.


[43] Lang H, Wurzburg U. Creatine kinase, an enzyme of many forms.Clin Chem 1982;28:1439–47.

[44] McBride JM, Kraemer WJ, Triplett-McBride T, et al. Effect ofresistance exercise on free radical production. Med Sci Sports Exerc1998;30:67–72.

[45] Pedersen BK, Fischer CP. Physiological roles of muscle-derivedinterleukin-6 in response to exercise. Curr Opin Clin Nutr MetabCare 2007;10:265–71.

[46] Pedersen BK, Fischer CP. Beneficial health effects of exercise – therole of IL-6 as a myokine. Trends Pharmacol Sci 2007;28:152–6.

[47] Grounds MD, Torrisi J. Anti-TNFa (Remicade) therapy protectsdystrophic skeletal muscle from necrosis. FASEB J 2004;18:676–82.

[48] Pedersen BK. IL-6 signalling in exercise and disease. Biochem SocTrans 2007;35:1295–7.

[49] Starkie R, Ostrowski SR, Jauffred S, et al. Exercise and IL-6 infusioninhibit endotoxin-induced TNF-alpha production in humans. FASEBJ 2003;17:884–6.

[50] Pedersen BK, Ostrowski K, Rohde T, et al. The cytokine response tostrenuous exercise. Can J Physiol Pharmacol 1998;76:505–11.

[51] Saito K, Kobayashi D, Komatsu M, et al. A sensitive assay of tumornecrosis factor alpha in sera from Duchenne muscular dystrophypatients. Clin Chem. 2000;46:1703–4.

[52] Lefaucheur J-P, Gjata B, Sebille A. Factors inducing mast cellaccumulation in skeletal muscle. Neuropathol Appl Neurobiol1996;22:248–55.

[53] Gorospe JRM, Nishikawa BK, Hoffman EP. Recruitment of mastcells to muscle after mild damage. J Neurol Sci 1996;135:10–7.

[54] Gorospe J, Tharp M, Hinckley J, et al. A role for mast cells in theprogression of Duchenne muscular dystrophy: correlations in dys-trophin-deficient humans, dogs, and mice. J Neurol Sci 1994;122:44–56.

[55] Granchelli JA, Avosso DL, Hudecki MS, et al. Cromolyn increasesstrength in exercised mdx mice. Res Commun Mol Pathol Pharmacol1996;91:287–96.

[56] Clark IA. How TNF was recognized as a key mechanism of disease.Cytokine Growth Factor Rev 2007;18:335–43.

[57] Kaczor JJ, Hall JE, Payne E, et al.. Low intensity training decreasesmarkers of oxidative stress in skeletal muscle of mdx mice. Free RadicBiol Med 2007;43:145–54.

[58] Whitehead NP, Pham C, Gervasio OL, et al. N-Acetylcysteineameliorates skeletal muscle pathophysiology in mdx mice. J Physiol2008;586:2003–14.

[59] Dudley RW, Khairallah M, Mohammed S, et al. Dynamic responsesof the glutathione system to acute oxidative stress in dystrophicmouse (mdx) muscles. Am J Physiol Regul Integr Comp Physiol2006;291:R704–10.

[60] Disatnik M-H, Dhawan J, Yu Y, et al. Evidence of oxidative stress inmdx mouse muscle: studies of the pre-necrotic state. J Neurol Sci1998;161:77–84.

[61] Halliwell B, Whiteman M. Measuring reactive species and oxidativedamage in vivo and in cell culture: how should you do it and what dothe results mean? Br J Pharmacol 2004;142:231–55.



Documents

A single 30min treadmill exercise session is suitable for ¢