2014 cochran, intermittent and continuous high-intensity exercise training induce similar acute but...

Expe

rimen

talP

hysi

olog

y782 Exp Physiol 99.5 (2014) pp 782791

Research PaperResearch Paper

Intermittent and continuous high-intensity exercisetraining induce similar acute but different chronicmuscle adaptations

Andrew J. R. Cochran1, Michael E. Percival1, Steven Tricarico1, Jonathan P. Little1, Naomi Cermak1,Jenna B. Gillen1, Mark A. Tarnopolsky2 and Martin J. Gibala1

1Exercise Metabolism Research Group, Department of Kinesiology, McMaster University, Hamilton, Ontario, Canada2Department of Pediatrics and Medicine, Division of Neuromuscular and Neurometabolic Disorders, McMaster University, McMaster UniversityMedical Centre, Hamilton, Ontario, Canada

New Findings! What is the central question of this study?How important is the interval in high-intensity interval training (HIIT)?! What is the main finding and its importance?The intermittent nature of HIIT is important for maximizing skeletal muscle adaptations tothis type of exercise, at least when a relatively small total volume of work is performed in anall-out manner. The protein signalling responses to an acute bout of HIIT were generallynot predictive of training-induced outcomes. Nonetheless, a single session of exercise lasting

Exp Physiol 99.5 (2014) pp 782791 783Muscle adaptations to high-intensity exercise training

adaptations, our data show that a training programme based on a brief bout of high-intensityexercise, which lasted

784 Exp Physiol 99.5 (2014) pp 782791A. J. R. Cochran and others

Table 1. Subject characteristics for those completing acute INTversus CONT high-intensity exercise, and those completing6 weeks of CONT-based training

Variable Acute study Chronic study

Participants 8 men; 0 women 5 men; 4 womenAge (years) 22 1 22 2Weight (kg) 78 8 78 11Height (cm) 181 5 173 9Peak oxygen uptake

(ml kg1 min1)48 7 47 5

Values are means SD. Abbreviations: CONT, continuous trial;and INT, intermittent trial.

and provided written informed consent prior to studyparticipation.

Subjects

A total of 17 subjects volunteered to participate inthe two studies (Table 1). Eight subjects took partin the acute investigation (study 1), which involved arepeated-measures design to evaluate the skeletal musclemetabolic response to an acute bout of high-intensityexercise matched for total work but performed in anintermittent (INT) or continuous (CONT) manner. Ninesubjects took part in the training study (study 2), whichexamined skeletal muscle remodelling in response to6 weeks of training using the CONT protocol. All subjectswere young healthy individuals who were habitually activebut not specifically trained in any sport.

Study 1: acute investigation

Pre-experimental procedures. The VO2peak and peakaerobic power output were initially determined duringa ramp protocol to volitional fatigue on an electromag-netically braked cycle ergometer (Lode Excalibur Sport;Lode BV, Groningen, The Netherlands) using an online gascollection system (Moxus modular oxygen uptake system;AEI Technologies Inc., Pittsburgh, PA, USA) as we havepreviously described (Cochran et al. 2010). Specifically,participants began cycling for 2 min at 50 W, followed bya progressive increase in power demand at the rate of 1 Wevery 2 s.

Thereafter, subjects participated in a minimum of twofamiliarization trials on separate days using the sameelectronically braked cycle ergometer employed during themain phase of the study (Velotron; RacerMate Inc., Seattle,WA, USA) in order to become acquainted with the exerciseprotocols. Due to the nature of the experimental design,all subjects performed the INT exercise protocol duringtheir first familiarization visit. This was necessary in orderto determine the total amount of work that was requiredto be performed during the CONT exercise protocol for agiven subject.

The INT protocol consisted of four 30 s all-out sprints,performed against a resistance equivalent to 7.5% ofbody mass (i.e. repeated Wingate tests), interspersedwith 4 min of recovery, as we have previously described(Burgomaster et al. 2005). A computer with appropriatesoftware (Velotron Wingate Software v1.0) was interfacedwith the ergometer and permitted the appropriate loadto be applied for each subject. Total work output, peakpower and mean power were calculated and recorded byan online data acquisition system.

For the CONT protocol, subjects performed the sametotal volume of work as in the INT exercise session, but asa single, continuous, all-out effort. The ergometer wasinterfaced with software (Velotron Coaching Softwarev1.5) that linked power output directly to pedallingcadence, while quantifying total work done in real time.Subjects were instructed to complete their designatedamount of work as quickly as possible by maintainingthe highest pedalling cadence possible. Between 50 and100 r.p.m., power output corresponded to a range of75500 W. Cycling was terminated immediately uponcompletion of the designated amount of work.

Experimental trials. The main experiment consisted oftwo trial days separated by at least 1 week. Trials wereconducted in a randomized, counterbalanced manner,with half the subjects starting with the INT protocol andthe other half with the CONT protocol. Subjects wereinstructed to refrain from exercise for 48 h prior to eachexperimental trial and to avoid caffeine and alcohol for atleast 12 h before the trials. Subjects maintained individualfood diaries for the 24 h period preceding the first trialand replicated their diet during the second trial.

On the day of each trial, subjects arrived at thelaboratory in the morning, 6090 min after ingestingtheir habitual breakfast. Food records were collected, andsubjects then changed into athletic apparel and restedquietly until trial commencement. A resting needle musclebiopsy sample was obtained from the vastus lateralisof one thigh under local anaesthesia (1% xylocaine) aspreviously described (Gibala et al. 2006). The musclesample was immediately frozen in liquid nitrogen andstored at 80C until further analyses. After resting foranother 10 min, the subjects moved to the cycle ergometerand completed a standardized warm-up that consistedof 2 min of unloaded cycling followed by 5 min of rest.Subjects then performed the designated exercise protocol.A second muscle biopsy was obtained immediately uponcessation of cycling, and subjects were asked to providea rating of perceived exertion for the overall exerciseprotocol, using the Borg scale (Borg, 1974). Subjects thenrested quietly in the laboratory for 3 h, at which pointa third muscle biopsy was taken. The three biopsies fora given trial were obtained from the same leg throughseparate incisions >2 cm apart.

C 2014 The Authors. Experimental Physiology C 2014 The Physiological Society


Study 2: training study

Pre-experimental procedures. Subjects initially perfo-rmed baseline VO2peak testing as described for study 1.Thereafter, subjects undertook a series of familiarizationsessions in order to become accustomed to the testingand training procedures. These sessions included a 250 kJsimulated cycling time trial, a 60 min steady-state sessionat !65% VO2peak and a practice training session, whichwas modelled after the CONT protocol employed instudy 1. Time trial familiarizations were repeated at1 week intervals until participants could not improvefurther beyond their previous session. Consistency inperformance during familiarizations were verified byStudents paired t test (P = 0.3), and the latter oftwo similar results were taken as baseline time trialperformance. Subjects completed 24 h diet records priorto each of these tests, and diets were replicated over the24 h period preceding post-training tests.

All chronic study participants were instructed tocomplete, as quickly as possible, a simulated timetrial consisting of 250 kJ of total work. This test wasperformed on the same electromagnetically braked cycleergometer (Velotron, RacerMate Inc.) interfaced withsoftware (Velotron Coaching Software v1.5) as trainingat a standardized gearing. Again, the cycle ergometer wasprogrammed such that power outputs between 75 and500 W were directly associated with pedalling rates, andsubjects were instructed to maintain the highest pedallingcadence possible. No feedback was given during the rides,with the exception of work remaining, and the test wasterminated immediately upon the completion of 250 kJ.

Subjects cycled continuously for 60 min at an intensitydesigned to elicit 65% of their VO2peak. The steady-stateride was conducted on the same cycle ergometer as theVO2peak measurement (Lode Excalibur), and respiratorymeasurements were made at specific 5 min intervalsthroughout exercise using the same metabolic cart systemdescribed previously (Moxus oxygen uptake system; AEITechnologies Inc.).

A resting skeletal muscle biopsy was taken !1 weekfollowing performance testing as described for study 1.Subjects were instructed to record their diet for the24 h preceding the biopsy, while refraining from exercisefor a minimum of 48 h and abstaining from caffeineand alcohol for a minimum of 12 h prebiopsy. Musclesamples were immediately frozen under liquid nitrogenand, subsequently, stored at 80C until further analysis.Diets were replicated post-training, and a second restingbiopsy was taken 72 h following the last exercise trainingsession.

Exercise training. Training was performed 3 days perweek for 6 weeks, for a total of 18 sessions, to aligndirectly with our previous 6 week INT study schedule.

The training intervention was modelled after the CONTprotocol employed in study 1, and each session consisted ofa single bout of high-intensity cycling completed as quicklyas possible. Based on our acute investigation and otherpilot work, the mean power produced over the course offour Wingate tests interspersed with 4 min of recoveryin recreationally active subjects averaged !1.0 kJ (kgbody mass)1. Subjects were therefore assigned an initialexercise training load that corresponded to 1.0 kJ (kg bodyweight)1. Training load was subsequently increased to1.25 kJ (kg body weight)1 during the second half of the6 week intervention in order to provide progression andmaintain the duration of the training session. Workloadwas self-selected and varied over the training session basedon pedalling cadence, with a range of 50100 r.p.m.corresponding to !75500 W. During each trainingsession, heart rate was monitored and ratings of perceivedexertion scores were obtained based on the Borg scale(Borg, 1974).

Post-training testing and procedures. Post-trainingprocedures were identical in all respects to those conductedprior to training onset, with the exception of order.Subjects first underwent a second resting skeletal musclebiopsy !72 h post-training. This time point was chosento evaluate training-induced changes in resting muscle.Steady-state, time trial and VO2peak tests took place at48 h intervals thereafter. Due to scheduling difficultiesand travel conflicts, however, we could only obtainpost-training VO2peak measures on six of our nine subjects.All subjects adhered to previously recorded diet recordsfor the 24 h preceding each of the biopsies and testingprocedures.

Muscle analysis

Western blotting. Whole-cell lysates were prepared byadding!30 mg wet muscle to ice-cold RIPA buffer (50 mMHCl, 150 mM NaCl, 1 mM phenylmethylsulfonyl fluoride,1% nonyl phenoxypolyethoxylethanol (NP-40), 0.5%sodium deoxycholate and 0.1% SDS) containing protease(Complete Mini R; Roche Applied Science, Laval, Quebec,Canada) and phosphatase inhibitors (PhosSTOP R; RocheApplied Science). Samples were minced and homogenizedon ice (Pro 250; Pro Scientific, Oxford, CT, USA),sonicated, and agitated end-over-end for 15 min at 4C.Samples were then centrifuged at 15,000g for 5 min at 4C.The pellet was then resuspended and, following a secondcentrifugation at 15,000g for 10 min, the supernatant wascollected for subsequent analysis.

Homogenate protein concentrations were determinedusing a commercial, detergent-compatible, colorimetricassay (BCA protein assay; Pierce, Rockford, IL, USA).Equal amounts of protein (520 g, depending on the



protein of interest) were then loaded onto 7.512.5%SDS-PAGE gels and separated by electrophoresis for22.5 h at 100 V. Proteins were transferred to nitrocellulosemembranes for 1 h at 100 V. Ponceau S staining wasperformed following the transfer and was used to controlfor equal loading and transfer between lanes. Membraneswere blocked using a 5% fat-free milk or bovine serumalbumin solution in Tris-Buffered Saline with Tween 20(TBS-T) at room temperature, and incubated overnightwith the appropriate primary antibodies diluted in a 3%fat-free milk or bovine serum albumin in TBS-T.

For study 1, primary antibodies targeting phospho-p38mitogen-activated protein kinase (MAPK), total-p38MAPK and phospho-acetyl-CoA carboxylase (ACC) werepurchased from Cell Signaling Technology (Beverly, MA,USA). For study 2, primary antibodies were targetedagainst five separate mitochondrial protein markers,namely NDUFA9 (MS111; Mitosciences, Eugene, OR,USA), Complex II 70 kDa subunit (MS204; Mitosciences),Complex III Core 2 protein (MS304; Mitosciences),cytochrome c oxidase subunit IV (COXIV; MS408;Mitosciences) and the ATP synthase subunit (MS507;Mitosciences). We also probed nitrocellulose membranesagainst glucose transporter 4 (GLUT4; AB1345; Millipore,Billerica, MA, USA), and monocarboxylate transporters1 and 4 (MCT1 and MCT4; AB3538 and AB3316,respectively; Millipore). Blots were incubated in theappropriate secondary antibodies for 1 h at roomtemperature and visualized by chemiluminescence(Supersignal R West Dura; Pierce). Signal quantificationwas performed using NIH ImageJ software.

Real-time RT-PCR. Frozen wet muscle samples (!20 mg)were homogenized in TRIzol R reagent (Invitrogen,Carlsbad, CA, USA). Total RNA was isolated using theRNeasy Mini Kit in conjunction with the RNase-FreeDNase Set DNA digestion (Qiagen, Mississauga, Ontario,Canada). The RNA was then reverse transcribed usingthe High-Capacity cDNA Reverse Transcription Kit fromApplied Biosystems (Carlsbad, CA, USA), aliquoted andstored at 80C until further analysis. The RT-PCRreactions for peroxisome proliferator-activated receptor coactivator 1 (PGC-1) mRNA expression were runusing forward (5-CAT CAA AGA AGC CCA GGT ACA-3) and reverse primers (5-GGA CTT GCT GAG TTGTGC ATA-3) in combination with SYBR green/ROXfluorescence chemistry (PerfeCTa; Quanta Biosciences,Gaithersburg, MA, USA). Reactions were run on a thermalcycler (Applied Biosystems), and expression levels werenormalized to the housekeeper gene 2-microglobulin(forward, 5-GGC TAT CCA GCG TAC TCC AA-3; andreverse, 5-GAT GAA ACC CAG ACA CAT AGC A-3),which was verified to be unchanged in response to ourexercise interventions (data not shown).

Table 2. Performance characteristics for the acute INT andCONT high-intensity exercise sessions

Characteristics INT CONT

Total work (kJ) 66.8 6.8 67.0 6.8Peak power output (W) 824 126 510 101Mean power output (W) 557 90 281 46Work duration (min:s) 2:00 0:00 4:02 0:26Ratings of perceived exertion 18.1 1.2 18 1.8

Values are means SD; n = 8 subjects. Abbreviations: CONT,continuous trial; and INT, intermittent trial. P " 0.05 versusINT, unpaired students t-test.

Maximal activity of citrate synthase. Approximately20 mg of wet muscle was homogenized using glasstissue pestles in 10 volumes of buffer containing 70 mMsucrose, 220 mM mannitol and 10 mM Hepes (pH 7.4),supplemented with protease inhibitors (Complete Mini R;Roche Applied Science). The maximal activity of citratesynthase (CS) was then quantified as we have describedpreviously (Gibala et al. 2006; Little et al. 2010). Theprotein content of the homogenate was determined bythe BCA method using a commercial assay (Pierce), andenzyme activity was expressed as millimoles per kilogramof protein per hour wet weight.

Statistical analyses

Exercise data from study 1 were analysed using Studentspaired t tests, while all muscle data from study 1 wereanalysed using a two-factor repeated-measured ANOVA,followed where appropriate by Tukeys HSD post hoc test.All data from study 2 were analysed using Students pairedt tests. The level of significance was set at P " 0.05 for allanalyses, and all analyses were conducted using SigmaStat3.1 software (Systat Software, Chicago, IL, USA). All dataare presented as means SD.

Results

Acute investigation

Performance data are presented in Table 2. Total work andratings of perceived exertion were not different betweentrials (P = 0.71, total work and P = 0.81, ratings ofperceived exertion, respectively). Peak power output andmean power output, averaged over the four Wingate testsin the INT trial, was higher than the respective valuescalculated for the CONT trial. Conversely, total exerciseduration in the CONT trial (!4 min) was approximatelydouble that of the INT trial (2 min, i.e. four bursts of30 s), although the latter session required a total of 14 minincluding recovery between intervals.

Muscle glycogen content was reduced by !25%, andmuscle lactate concentration was elevated !10-fold afterexercise, with no difference between protocols (main



effects for time, P < 0.01; Fig. 1A and B). Phosphorylationof p38 MAPK and ACC serine-79 increased immediatelyafter exercise by !3-fold and !2.5-fold, respectively, withno difference between treatments (P < 0.05, main effectfor time; Fig. 2A and B). The mRNA expression of PGC-1was increased !4-fold from rest after 3 h of recovery, withno difference between conditions (P < 0.05, main effectfor time; Fig. 3).

Chronic investigation

Subjects completed 99% of the assigned training sessions.Mean total work was 77 12 kJ, completed in an averagetime of 391 75 s (6:31 00:47 min:s) at a mean poweroutput of !212 49 W. Mean heart rate during trainingsessions was 182 9 beats min1, which was equivalentto 95 2% of maximal heart rate. The average value forratings of perceived exertion was 18 1. The time requiredto complete designated training work quotas was increasedin association with workload progression, but there wereno other significant changes in time to complete training(data not shown).

The maximal activity of CS was unchanged aftertraining compared with pretraining (Fig. 4). Proteins

Figure 1. Muscle glycogen (A) and lactate concentrations (B)measured before (PRE) and after (POST) performing !67 kJ ofwork intermittently (INT) or continuously (CONT) at maximaleffortValues are shown as means + SEM for eight subjects. P < 0.05Two way mixed ANOVA, main effect for time.

representative of each of the complexes of the electrontransport chain were also unchanged after training(P # 0.10), the one exception being cytochrome c oxidasesubunit IV, which showed a 20% increase (P = 0.014;Fig. 5). Likewise, the protein contents of GLUT4, MCT1and MCT4 were unchanged after training (data notshown).

Peak oxygen uptake was increased by 6% after training(P < 0.05; Fig. 6), while time to complete 250 kJ ofimproved by !9% (P < 0.001; Fig. 7). There wereno differences in heart rate, respiratory exchange ratioor ventilation during steady-state cycling at 65% ofpretraining VO2peak before and after training (Table 3).

Figure 2. Changes in protein phosphorylation of p38mitogen-activated protein kinase (p38 MAPK; Thr180/Tyr182;A) and acetyl-CoA carboxylase (ACC; Ser79; B) before (PRE)and after (POST) !67 kJ of intermittent (INT) and continuous(CONT) exercise at maximal effortValues are shown as means + SEM for eight subjects. P < 0.05Two way mixed ANOVA, main effect for time.



Discussion

The overriding goal of the present study was to determinewhether the characteristic pulsed nature of high-intensityinterval exercise is critical to maximize adaptation to thistype of training. While training using brief intermittentbursts of all-out exercise is a potent stimulus to induceskeletal muscle remodelling towards a more oxidativephenotype (Gibala et al. 2006; Burgomaster et al. 2008),it is unclear whether the alternating pattern of hardeasyeffort is fundamental to the training response. The resultsof our acute investigation (study 1) showed that bothINT and CONT protocols elicited similar increases insignalling cascades linked to mitochondrial biogenesis,including the protein phosphorylation of ACC and p38MAPK and the mRNA expression of PGC-1. Despite this,

Figure 3. Peroxisome proliferator-activated receptor coactivator 1 (PGC-1) mRNA expression before (PRE) andafter 3 h of recovery (3 h POST) from !67 kJ of workperformed intermittently (INT) or continuously (CONT) atmaximal effortThe housekeeping gene 2-microglobulin was used fornormalization. Values are shown as means + SEM for eightsubjects. P < 0.05, main effect for time Two way mixedANOVA.

Figure 4. Maximal activity of citrate synthase (CS) measuredin resting muscle biopsy samples before (PRE-TR) and after6 weeks of low-volume, all-out CONT training (POST-TR)Values are shown as means + SD for nine subjects.

a range of mitochondrial enzyme markers were generallyunchanged after 6 weeks of training with the CONTprotocol (study 2). This finding is in contrast to the robustincreases in mitochondrial protein content and maximalenzyme activities that we have repeatedly observed after26 weeks of the INT training protocol (Gibala et al. 2006;Burgomaster et al. 2008).

Figure 5. Mitochondrial protein content before (PRE-TR) andafter 6 weeks of low-volume, CONT training at maximal effort(POST-TR)Values are shown as means + SD for nine subjects.Abbreviations: ATP Synthase , catalytic -subunit of ATPsynthase; Core protein 2, ubiquinolcytochrome c reductaseassembly protein; NDUFA9, NADH dehydrogenase 1subcomplex subunit 9; Subunit 70 kDA, succinatedehydrogenase 70 kDa flavoprotein subunit; and Subunit IV,cytochrome c oxidase subunit 4. P " 0.05 versus pretrainingPaired students t-test.

Figure 6. Peak oxygen uptake (VO2peak) relative to total bodyweight before (PRE-TR) and after 6 weeks of low-volume,all-out CONT training (POST-TR)Values are shown as means + SD for six subjects. P " 0.05versus pretraining Paired students t-test.



An obvious limitation of the present work was thelack of a direct comparison between the CONT andINT protocols. With respect to our measurements ofskeletal muscle adaptation, it has been proposed that CS isone of the most appropriate indicators of mitochondrialcontent in human skeletal muscle, because it is highlycorrelated with gold-standard measures of mitochondrialcontent made by electron microscopy (Larsen et al. 2012).Interestingly, a recent review by Bishop et al. (2013)suggested that training volume is more important forincreasing mitochondrial content than training intensity,which may in part explain the lack of change in CS activity.The CONT training also had no effect on other markers ofskeletal muscle adaptation, including the protein contentof GLUT4, MCT1 and MCT4, which we have previouslyshown to be increased by INT training (Burgomaster et al.2007). Overall, these data suggest that the intermittentnature of the HIIT stimulus may be important formaximizing skeletal muscle adaptations, at least when arelatively small total volume of high-intensity exercise isperformed in an all-out manner. Additional studies withlarger sample sizes and more comprehensive assessment ofphysiological adaptation are warranted in order to supportor refute this hypothesis.

Despite the lack of change in most markers of skeletalmuscle oxidative or metabolite transport capacity, 6 weeksof CONT training improved the time to complete 250 kJ ofwork. While numerous factors are involved in determiningexercise performance one factor that may have contributedto the improved performance in the present study was anenhanced whole-body aerobic capacity (Bassett & Howley,2000), as reflected by the significant 6% increase in VO2peakafter training (despite being measured in only six ofour nine subjects). This observation supports the idea

Figure 7. Total time to complete 250 kJ of mechanical workbefore (PRE-TR) and after 6 weeks of low-volume, all-outCONT training (POST-TR)Values are shown as means + SD for nine subjects. P " 0.001versus pretraining Paired students t-test.

Table 3. Cardiorespiratory data during cycling exercise at65% of peak oxygen uptake before and after 6 weeks ofCONT-based training

Parameter Pretraining Post-training

Heart rate (beats min1) 158 14 157 16Respiratory exchange ratio 0.87 0.04 0.86 0.02Ventilation (l min1) 55.6 8.1 54.2 6.2Oxygen uptake (l min1) 2.14 0.41 2.07 0.39

Values are means SD; n = 9 subjects. Abbreviation: CONT,continuous trial.

that brief bouts of very intense exercise can improvecardiorespiratory fitness. Tjnna et al. (2013) recentlyreported a 10% improvement in VO2peak after 10 weeksof training, in which overweight but healthy subjectsperformed a single 4 min bout of continuous exercise atan intensity that elicited 90% of maximal heart rate, threetimes per week. Subjects in that study performed a 10 minwarm-up at 70% of maximal heart rate, followed by 4 minat 90% of maximal heart rate and 5 min cool-down at 70%of maximal heart rate, for a total time commitment of19 min (Tjnna et al. 2013). In the present study, subjectsperformed only 2 min of unloaded cycling as a warm-up,and thus our data show that VO2peak can be enhancedby a training protocol consisting of


adaptive response to exercise (Birk & Wojtaszewski, 2006;Pogozelski et al. 2009), and it is therefore possible thatsubtle differences in activation could not be resolved byour Western blotting techniques. Our data also highlightthe need for studies examining both acute responsesand training responses within the same individuals.Indeed, we are unaware of any evidence reporting thatsubject-to-subject variability in AMPK, p38 MAPK orPGC-1 is correlated with training adaptation in humanmuscle, and this has led some to question the purposeof focusing so much research upon upstream signallingevents (Timmons, 2011). Furthermore, our findingsunderscore that changes in mRNA expression do notnecessarily confer a similar change in functional proteinor enzyme activity, and that relatively little is knownat present regarding the effects that different types ofexercise may have on processes downstream from mRNAexpression in human skeletal muscle. These factors includemRNA stability and turnover, protein translation, proteinimport and assembly, mitochondrial fusion/fission andmitophagy. Any combination of these processes may beresponsible for the diversion between mRNA and proteinexpressions. More work must be done to examine theeffects that exercise intensity and duration and factors suchas intermittency may have on the intervening biologicalprocesses between mRNA content and functional proteinexpression.

In summary, we have shown that performing a givenamount of work using an all-out effort results in similaractivation of signalling cascades linked to mitochondrialbiogenesis, regardless of whether the exercise is performedin an intermittent or a continuous manner. Despite similaracute signalling responses to the CONT and INT protocols,a range of mitochondrial enzyme markers were generallyunchanged after 6 weeks of training with the CONTprotocol, which, although not measured in the presentstudy, is in contrast to the robust increases we havepreviously reported after 2 and 6 weeks of training with theINT protocol (Gibala et al. 2006; Burgomaster et al. 2008).Thus, the acute responses were not necessarily predictive oftraining-induced adaptations. Despite the lack of skeletalmuscle mitochondrial adaptations, our data show thata single session of exercise lasting


Gibala MJ, Little JP, van Essen M, Wilkin GP, Burgomaster KA,Safdar A, Raha S & Tarnopolsky MA (2006). Short-termsprint interval versus traditional endurance training: similarinitial adaptations in human skeletal muscle and exerciseperformance. J Physiol 575, 901911.

Handschin C, Chin S, Li P, Liu F, Maratos-Flier E, LebrasseurNK, Yan Z & Spiegelman BM (2007). Skeletal musclefiber-type switching, exercise intolerance, and myopathy inPGC-1 muscle-specific knock-out animals. J Biol Chem282, 3001430021.

Jager S, Handschin C, St-Pierre J & Spiegelman BM (2007).AMP-activated protein kinase (AMPK) action in skeletalmuscle via direct phosphorylation of PGC-1. Proc NatlAcad Sci USA 104, 1201712022.

Larsen S, Nielsen J, Hansen CN, Nielsen LB, Wibrand F, StrideN, Shroder HD, Boushel R, Helge JW, Dela F &Hey-Mogensen M (2012). Biomarkers of mitochondrialcontent in skeletal muscle of healthy young human subjects.J Physiol 590, 33493360.

Little JP, Safdar A, Wilkin GP, Tarnopolsky MA & Gibala MJ(2010). A practical model of low-volume high-intensityinterval training induces mitochondrial biogenesis in humanskeletal muscle: potential mechanisms. J Physiol 588,10111022.

Perry CG, Heigenhauser GJ, Bonen A & Spriet LL (2008).High-intensity aerobic interval training increases fat andcarbohydrate metabolic capacities in human skeletal muscle.Appl Physiol Nutr Metab 33, 11121123.

Pogozelski AR, Geng T, Li P, Yin X, Lira VA, Zhang M, Chi JT& Yan Z (2009). p38 mitogen-activated protein kinase is akey regulator in skeletal muscle metabolic adaptation inmice. PLoS One 4, e7934.

Puigserver P, Rhee J, Lin J, Wu Z, Yoon JC, Zhang CY, Kraus S,Mootha VK, Lowell BB & Spiegelman BM (2001). Cytokinestimulation of energy expenditure through p38 MAP kinaseactivation of PPAR coactivator-1. Mol Cell 8, 971982.

Talanian JL, Galloway SD, Heigenhauser GJ, Bonen A & SprietLL (2007). Two weeks of high-intensity aerobic intervaltraining increases the capacity for fat oxidation duringexercise in women. J Appl Physiol 102, 14391447.

Timmons JA (2011). Variability in training-induced skeletalmuscle adaptation. J Appl Physiol 110, 846853.

Tjnna AE, Leinan IM, Bartnes AT, Jennsen BM, Gibala MJ,Winett RA & Wislff U (2013). Low- and high-volume ofintensive endurance training significantly improves maximaloxygen uptake after 10-weeks of training in healthy men.PLoS One 8, e65382.

Winder WW, Holmes BF, Rubink DS, Jensen EB, Chen M &Holloszy JO (2000). Activation of AMP-activated proteinkinase increases mitochondrial enzymes in skeletal muscle. JAppl Physiol 88, 22192226.

Additional Information

Competing interests

None declared.

Author contributions

Conception and design of the experiments: A.J.R.C., M.J.G.,J.P.L., J.B.G. and M.A.T. Collection, analysis and interpretationof the data: A.J.R.C., M.E.P., S.T., J.P.L., N.C., J.B.G., M.A.T. andM.J.G. Drafting the article or revising it critically for importantintellectual content: A.J.R.C., M.E.P., S.T., J.P.L., N.C., J.B.G.,M.AT. and M.J.G. All authors approved the final version forpublication.

Funding

This project was supported by operating grants from the NaturalSciences and Engineering Research Council of Canada (NSERC)to M.J.G. and M.A.T. A.J.R.C. was supported by a NSERC PGS-Dscholarship, and J.P.L. held a NSERC CGS-D scholarship. M.E.P.held a NSERC CGS-M, N.C. held a NSERC PGS-D, and J.B.G.held a NSERC CGS-M.

Acknowledgements

We would like to thank our subjects for their commitment andeffort, as well as Todd Prior, Adeel Safdar and Mahmood Akhtarfor their technical assistance.


2014 cochran, intermittent and continuous high-intensity exercise training induce similar acute but...

Documents