Available online at www.sciencedirect.com

Journal of Exercise Science & Fitness 10 (2012) 16e22www.elsevier.com/locate/jesf

Original article

GLUT4 is not essential for exercise-induced exaggerated muscle glycogendegradation in AMPKa2 knockout mice

Haojie Gong a,b, Ying Zhang b,*

aHenan Scientific and Technical University, Luoyang, ChinabBeijing Sport University, Beijing, China

Received 8 March 2010; revised 16 August 2010; accepted 19 January 2012

Available online 8 June 2012

Abstract

The purpose of this study was to explore the effects of AMPKa2 knockout on expression of GLUT4 after exercise with different intensities.This was investigated in the quadriceps femoris muscle from control, lower intensity (12 m/minute) and higher intensity (20 m/minute) runninggroups of wild-type (WT) and AMPKa2 knockout (KO) mice. The skeletal muscle glycogen content was also measured to investigate the role ofAMPK in regulation of glycogen metabolism. GLUT4 mRNAwas increased immediately at the end of 60 minutes of running, but there were nodifferences between WT and KO mice, and between lower intensity and higher intensity groups. Pre-exercise muscle glycogen levels were notdifferent between WT and KO; however, the lack of the a2-isoform was associated with a generally lower level of muscle glycogen after60 minutes of running either with lower intensity or higher intensity. The finding that KO of the AMPKa2 isoform had no effect on expressionsof GLUT4 mRNA and protein expression in mouse skeletal muscle and that muscle glycogen degradation was greater in AMPKa2 KO mice thanthat in WT mice after different intensity exercises, provides further evidence that AMPK is dispensable in exercise-induced expression ofGLUT4, and suggests alternative pathways other than GLUT4 being involved in the regulation of muscle glycogen degradation by AMPK. Also,the present investigation demonstrated that higher intensity (20 m/minute) treadmill running, lasting 1 h, increased GLUT4 mRNA in miceskeletal muscle to a level similar to that attained after 1 h of lower intensity (12 m/minute) treadmill running.Copyright � 2012, The Society of Chinese Scholars on Exercise Physiology and Fitness. Published by Elsevier (Singapore) Pte Ltd. All rightsreserved.

Keywords: AMPK; Blood glucose; Exercising muscles; Glucose transport; GLUT4; Intracellular signaling

Introduction

Because skeletal muscle is an important tissue in maintainingglucose homeostasis, and the content of GLUT4 is a primaryfactor in determining the maximal rate of glucose transport intoskeletal muscle,1 increasing GLUT4 protein is a potentialmethod to treat the hyperglycemia associated with diabetes.Previous studies have shown that exercise can induce a rapidadaptive increase in GLUT4 expression.2e4 Therefore, it isimportant to understand how exercise regulates GLUT4

* Corresponding author. Beijing Sport University, Beijing 100084, China.

Tel.: þ86 10 62989584; fax: þ86 10 62989444.

E-mail address: zhyi9256@126.com (Y. Zhang).

1728-869X/$ - see front matter Copyright � 2012, The Society of Chinese Scholars on Exerci

doi:10.1016/j.jesf.2012.04.001

expression so that therapeutic strategies might be designed tohelp increasemuscle glucose disposal as a treatment for diabetes.

One possible mediator that might be involved is 50AMP-activated protein kinase (AMPK). Exercise increases AMPKactivity in skeletal muscle,5e7 and activation of AMPKby 5-aminoimidazole-4-carboxamide-1-b-4-ribofuranoside(AICAR) is associated with increases in GLUT4 mRNA andprotein content.8e10 Further evidence that AMPK plays a rolein regulating GLUT4 expression comes from observations thatthe time course for GLUT4 mRNA upregulation is the samefor both AICAR and treadmill exercise. Likewise, it appearsthat the same DNA promoter sequences are required for theGLUT4 response to AICAR and exercise. When the micecarrying only e730 base pairs (bp) of the human GLUT4promoter are exercised, there is no GLUT4 mRNA increase.11

se Physiology and Fitness. Published by Elsevier (Singapore) Pte Ltd. All rights reserved.

17H. Gong, Y. Zhang / Journal of Exercise Science & Fitness 10 (2012) 16e22

In vivo treatment of transgenic mice with AICAR also resultedin a significant increase in relative GLUT4 mRNA levels in thee895 bp, but not in the e730 bp, promoter.12

Glycogen is an important energy source for the workingmuscle, especially at moderate to high intensities. Studieshave shown that AMPK phosphorylates glycogen synthase(GS) on site 2 with a concomitant reduction in GS activity,13,14

and the deactivation of GS induced by activation of AMPKusing AICAR is abolished in AMPKa2 knockout mice.15 Asomewhat surprising finding was that while AICAR stimula-tion decreased GS activity, the glycogen content in skeletalmuscle was increased rather than decreased.16,17 Given thatthe content of GLUT4 is the primary factor in determining therate of glucose transport into skeletal muscle,1 and latterfinding also proved that the rate of glycogen synthesis isdepending on the rate of glucose transport rather than theactivity of GS,18,19 we proposed that it could be caused by thestimulatory effect of AMPK on the expression of GLUT4,which might override the inhibitory action of AMPK on GSactivity, resulting in enhanced glycogen synthesis in skeletalmuscle.

Because AMPK activity is exercise intensity depend-ent,20e22 the purpose of this study was to explore the effects ofAMPKa2 knockout on expression of GLUT4 after exercisewith different intensities. Also the skeletal muscle glycogencontent was measured to detect the role of AMPK in regula-tion of glycogen metabolism.

Methods

Animals

The study protocol was reviewed and approved by theAnimal Care and Use Community of Beijing Sport University.Two-month-old C57/BL mice were housed with controlledroom temperature and lighting (20e25 �C and 12:12 hourlightedark cycle) and free access to food and water. AMPKa2knockout (KO) mice specimens were kindly provided byBenoit Viollet (Department of Endocrinology, Metabolism andCancer, Institute Cochin, University Paris Descartes, France),and bred by the institute of Laboratory Animal Science,Chinese Academy of Medical Sciences (CAMS) & PekingUnion Medical College (PUMC). Wild-type (WT) and KOmice were littermates produced by intercross breading ofheterozygote parents and the genotype of the offspring wasdetermined by Southern blotting on DNA extracted froma tail-piece.

Treadmill exercise protocol

Prior to the experiment, all mice were acclimatized totreadmill running at days e3 and e2, by running for 5 minutesat 10 m/minute, 4 minutes at 15 m/minute, and 2 minutes at20 m/minute. During acclimatization and the experiment, anair-gun was used to encourage running if necessary. On theexperimental day, wild-type (WT; n¼ 30) and AMPKa2knockout (KO; n¼ 30) mice were randomly subdivided into

three groups: control group; lower intensity (0% incline, 12 m/minute) group; and higher intensity (0% incline, 20 m/minute)group.23 All mice in the exercised groups completed60 minutes of continuous running. Immediately after the60 minutes running, control and exercised mice were anes-thetized with an intraperitoneal injection of pentobarbital,5 mg/100 g of body weight, and eye blood sample wascollected for the measurement of blood glucose. Quadricepsfemoris muscles were excised. Immediately after removal, thewet weight of muscle tissue for the muscle glycogen wasmeasured, whereas muscles for the remaining analyses werequick-frozen with aluminum tongs, precooled in liquidnitrogen, and stored at e80 �C.

RNA isolation and RT

Total RNAwas isolated from approximately 20 mg crushedmuscle tissue using the guanidinium thiocyanate-phenol-chloroform extraction technique.24 Reverse transcription oftotal RNA to cDNA was performed using the AMV ReverseTranscriptase Kit (Promega A3500; Promega, Madison, WI,USA) in a Perkin-Elmer DNA Thermal Cycler (Perkin-ElmerApplied Biosystems, Foster City, CA, USA). First-strandcDNA was synthesized from 1 mg total RNA in a 20 mLreaction using random hexamers as primers. The reaction wasrun at 42 �C for 90 minutes and thereafter at 72 �C for10 minutes.

Real time PCR (RT-PCR)

RT-PCR was performed in a GeneAmp 7300 Real TimePCR System using the GeneAmp SYBR� Green kit (Perkin-Elmer, Applied Biosystems, Foster City, CA, USA) with thepreviously synthesized cDNA as template in a 20 mL reaction.A no-template control was included in all experiments. TheGeneAmp 7300 system monitors the amplification of DNA inreal time using an optical imaging system, via the binding ofa fluorescent dye to double-stranded DNA. Forward andreverse primers complementary to the mouse GLUT4 gene(GenBank NM_009204) were designed using Primer Expresssoftware (Perkin-Elmer, Applied Biosystems, Foster City, CA,USA). The GLUT4 forward primer sequence (50 to 30) wasCTT GGC TCC CTT CAG TTT GG, whereas the reverseprimer sequence was CTA CCC AGC CAC GTT GCATT. Theexpression of GLUT4 was compared with expression of b-actin, which was continuously expressed at constant amountsin cells. The primers for b-actin were forward: AGG CAAACC GTG AAA AGATG and reverse: CAC AGC CTG GATGGC TAC GT. Initial conditions for PCR were (in centigrade)55 �C for 2 minutes, 95 �C 10 minutes and then cycling for 40cycles between 95 �C for 30 seconds and 60 �C for 1 minute.This was followed by dissociation, 1 cycle of 95 �C for15 seconds, 60 �C for 30 seconds and 95 �C for 15 seconds.The cycle threshold (CT) of GLUT4 was subtracted from theCT of b-actin giving the dCT. The value of the control dCT(wild type) was set to one, and the control (wild type) dCTwassubtracted from the experimental dCT yielding the ddCT. The

18 H. Gong, Y. Zhang / Journal of Exercise Science & Fitness 10 (2012) 16e22

difference in expression between control and experimentalwas calculated as 2�ddCT, as described.25

Immunoblot analysis

For the measurement of concentrations of GLUT4 andAMPKa2, electrophoresis and Western blotting were per-formed on homogenates extracted from the quadriceps fem-oris muscles. Protein content in lysates was measured by thebicinchoninic acid method (Pierce, Rockford, IL, USA). Aquantity of 20 mg of protein was resolved on sodium dodecylsulfate-polyacrylamide gel electrophoresis, followed byelectrophoretic transfer to nitrocellulose filter (NC)membranes. After blocking with a 5% skim milk suspensionfor 1 hour at 22 �C, the membranes were incubated overnightat 4 �C with goat anti-GLUT4/AMPKa2 polyclonal (1:1000)obtained form Santa Cruz Biotechnology (Santa Cruz, CA,USA), and then in incubated with anti-goat horseradishperoxidase (Santa Cruz Biotechnology, Santa Cruz, CA,USA) for 1 hour at 22 �C. To confirm equal protein loadingand transfer, membranes were stripped and reprobed withmonoclonal anti-b-actin (1:4000, Santa Cruz Biotechnology,Santa Cruz, CA, USA) and then incubated with anti-goathorseradish peroxidase (1:20,000). All antibodies werediluted in 1% polyvinylpyrrolidone in Tris buffered saline(TBS), and membranes were washed between antibodyincubations three times with Tris buffer saline with Tween-20(TBST) for 10 minutes. Membranes were exposed toElectrochemiluminescence (ECL) solution for 1 minute andplaced in an autoradiography cassette. Immunoreactive bandswere highlighted by ECL, exposed to light-sensitive film for15e60 seconds and quantified by densitometry using imageanalysis software (Kodak Digital Science, New York, NY,USA). The individual values were originally expressed asa percentage of a standard (b-actin content) and thenexpressed as a percentage of the control group (WT) value.The control group value of WT was set at 100%.

Blood glucose

The glucose concentration was measured by a glucoseoxidase method as described previously.26,27

Muscle glycogen

Muscle glycogen concentrations were determined byenzymatic methods according to Lowry and Passonneau afteracid hydrolysis.28 Results are expressed as millimoles ofglucose per kilogram of wet weight.

Statistical analysis

Fig. 1. (A) Representative immunoblots of AMPKa2 protein; (B) AMPKa2

protein content response to exercise in wild-type (WT) and AMPKa2

knockout (KO) mice. KO, nondetectable. n¼ 9e10.

All values are reported as means� standard deviation (SD).Mean differences from each experiment were analyzed bytwo-way ANOVA, and a Tukey post hoc test was used whensignificance was found. Statistical significance was set atp� 0.05.

Results

AMPKa2 protein content

Results from immunoblot analysis showed (Fig. 1) nodifferences in muscle AMPKa2 protein content betweencontrol and exercised groups in WT mice.

GLUT4 protein content

Results from immunoblot analysis showed (Fig. 2) nodifferences in muscle GLUT4 protein content between controland exercised groups in either the WTor KO mice and betweenWT and KO mice in either the control or exercised groups.

GLUT4 mRNA content

Results from real time PCR showed (Fig. 3) no differencesin muscle GLUT4 mRNA between WT and KO mice in eitherthe control or exercised groups. The increase in GLUT4mRNA in response to exercise was statistically significant( p< 0.05), and no differences between lower intensity andhigher intensity groups.

Muscle glycogen

Muscle glycogen content of WT and a2-KO quadricepsfemoris (Fig. 4) was reduced by ( p< 0.01) 28% and 49% after1 hour of treadmill running with lower intensity, respectively,and reduced by ( p< 0.01) 44% and 59% after 1 hour of

Fig. 2. (A) Representative immunoblots of GLUT4 protein; (B) GLUT4

protein content response to exercise in wild-type (WT) and AMPKa2

knockout (KO) mice; n¼ 9e10.

Fig. 4. Muscle glycogen content response to exercise in wild-type (WT) and

AMPKa2 knockout (KO) mice; n¼ 9e10. *Significantly different from

control group ( p< 0.05). &Significantly different from low intensity group

( p< 0.05). #Significantly different from WT values ( p< 0.05).

19H. Gong, Y. Zhang / Journal of Exercise Science & Fitness 10 (2012) 16e22

treadmill running with higher intensity, respectively, con-firming that WT and a2-KO animals had performeda substantial amount of running. After 1 hour of treadmillrunning, the muscle glycogen content in higher intensity groupwas less ( p< 0.05) than that in lower intensity group. Pre-exercise muscle glycogen levels showed no differencebetween WT and KO. However, the lack of the a2-isoformwas associated with a generally lower ( p< 0.05) level ofmuscle glycogen after 1 hour of treadmill running.

Fig. 3. GLUT4 mRNA content response to exercise in wild-type (WT) and

AMPKa2 knockout (KO) mice; n¼ 9e10. *Significantly different from

control group ( p< 0.05).

Blood glucose

Treadmill running with lower intensity reduced ( p< 0.05)blood glucose concentration (Fig. 5) by 41% and 56% in WTand a2-KO mice, respectively, compared with control groups,whereas the blood glucose concentration was unaffected bythe higher intensity treadmill running and there was nodifference between WT and a2-KO mice.

Discussion

Given that the content of GLUT4 is a primary factor indetermining the maximal rate of glucose transport into skeletalmuscle1 and over-expression of GLUT4 in skeletal muscleameliorates many of the symptoms associated with type 2diabetes,29e31 there has been considerable interest in under-standing the regulation of GLUT4 expression. Several studiesover the last decade have contributed to the current under-standing that exercise induces a rapid adaptive increase inGLUT4 expression.2e4

However, only little is yet known about the intracellularsignaling pathways involved in eliciting these acute responses.One possiblemediator thatmight be involved isAMPK, becauseAMPK is activated in skeletal muscle during exercise5e7 andpharmacological activation of AMPK is associated withincreases in GLUT4 mRNA and protein content.8e10 Althoughpharmacological activation of AMPK in resting muscle8e10 canactivate expression of GLUT4, the present finding that the lackof the AMPKa2 isoform did not have any effect on expressionsof GLUT4 mRNA and protein expression in mouse skeletalmuscle after different intensity exercises provides, however,further evidence that AMPKa2 is not essential for the acute,adaptive responses of GLUT4 to exercise. This finding is sup-ported by the fact that overexpression of a kinase dead AMPKconstruct did not reduce the exercise-induced activation of theGLUT4 gene in mouse muscle.32

Fig. 5. Blood glucose concentrations response to exercise in wild-type (WT)

and AMPKa2 knockout (KO) mice; n¼ 9e10. *Significantly different from

control group ( p< 0.05).

20 H. Gong, Y. Zhang / Journal of Exercise Science & Fitness 10 (2012) 16e22

The reason for the different results in resting and exercisingmuscles can probably be explained by alternative signalinginduced by exercise, e.g., the remaining a1 isoform orcalcium-calmodulin-activated kinase-dependent (CaMK)8,33,34

signaling. Because the AMPKa2 knock-out mice had a 2e3-fold increase in a1 expression and a 2-fold increase incontraction-stimulated a1 activity,14,15 it is possible that theup-regulation of a1 compensated for the loss of a2 function.Also, Studies have shown that incubated C2C12 myotubeswith caffeine is associated with increases in GLUT4 expres-sion,34,35 because CaMKII activity is also increased in skeletalmuscle during exercise,36,37 it seems reasonable to speculatethat the observed increase in GLUT4 expression in AMPKa2KO mice might be mediated by CaMKII. Further studies areneeded to clarify this.

Another interesting finding was that exercise bouts at12 m/minutes and 20 m/minutes increased GLUT4 mRNA inboth KO and WT mouse skeletal muscle to a similar extent.Contrary to our original hypothesis, the higher intensityexercise was no more effective than lower intensity exercisein eliciting increases in GLUT4 mRNA levels. The reason forthis is not clear but it could be that during relatively low-intensity exercise, there may have been a progressiveincrease in AMPK20e22 and CaMKII activity36,37 withincreasing exercise duration, and reached to the level atwhich the maximal amount of GLUT4 mRNA expressionwas induced. Alternatively, lower intensity exercise mayhave been sufficient to initiate and activate GLUT4 tran-scription, with no further increases at the higher intensityexercise.

In WT mice, protein of GLUT4 failed to increase inresponse to exercise, that is in contrast to previous studies.4

It is possible that there was an inadequate time for changein GLUT4 protein to become evident. There are importantmultiple steps in the regulation of cellular protein,including gene transcription, mRNA stability, protein

translation rate, translation efficiency, and posttranslationalmodifications, which could result in the expression ofGLUT4 protein later than gene. A previous study38 hasshown that the exercise-induced increase in GLUT4 proteinis delayed until 1.5 hours of recovery, and the 0 hours ofrecovery in the present study was most likely caused by aninadequate time for change in GLUT4 protein to becomeevident.

Miyamoto et al16 have shown that while AICAR stimula-tion decreased GS activity, the glycogen content in skeletalmuscle was unchanged, and proposed that it might be causedby the stimulatory effect of AMPK on the expression ofGLUT4. AMPK is activated in skeletal muscle during exerciseand exercise also induces a rapid adaptive increase in GLUT4expression, so we hypothesized that the lack of the AMPKa2isoform would result in a deceased GLUT4 expression, whichwould reduce the glycogen synthesis during exercise, thereforethe glycogen content in AMPKa2 KO mice would be lowerthan that in WT mice after exercise. As expected, in thepresent study muscle glycogen degradation was greater inAMPKa2 KO mice than that in WT mice, but the finding thatthe lack of a2-AMPK isoform had no effect on the exercise-induced GLUT4 expression suggests that pathways otherthan GLUT4 are involved in the regulation of the glycogensynthesis by AMPK. As Glycogen is an important energysource for the working muscle, and previous studies haveshown that a2-KO muscles were more metabolically stressedthan WT muscles during exercise,39 so the exaggeratedbreakdown of glycogen in KO mice could perhaps be causedby higher metabolic stress during exercise as reportedpreviously.

The blood glucose concentration was reduced by 41% and56% in WT and a2-KO mice after treadmill running withlower intensity, respectively, compared with control groups,whereas the blood glucose concentration was unaffected bythe higher intensity treadmill running. We also found that after1 hour treadmill running, the muscle glycogen content in thehigher intensity group was lower than that in the lowerintensity group. Therefore one scenario could be that duringhigher intensity exercise, the increasing consumption ofmuscle glycogen would have slowed down the consumption ofblood glucose. Furthermore, blood glucose during exercisereflects a balance between liver output and uptake of themainly working muscles, as liver glucose output increaseswith exercise intensity, this might also make sense. In addi-tion, while previous work40 in AMPKa2 KO mice had showna higher concentration of blood glucose than that in WT mice,in the present study, the blood glucose concentration seemedto be higher in AMPKa2 KO mice than WT mice, but thedifference was not significant. Further studies might reveal ifthis was caused by the sample size used.

In conclusion, the finding that the lack of AMPKa2 isoformhad no effect on exercise-induced expression of GLUT4, butmuscle glycogen degradation was greater in AMPKa2 KOmice than that in WT mice, suggesting that AMPK isdispensable in exercise-induced expression of GLUT4, thussuggests an alternative pathway other than GLUT4 might be

21H. Gong, Y. Zhang / Journal of Exercise Science & Fitness 10 (2012) 16e22

involved in the regulation of the glycogen metabolism byAMPK. Also, the present investigation demonstrated thathigher intensity treadmill running (20 m/minute), lasting1 hour, increased GLUT4 mRNA in mouse skeletal muscle toa level similar to that attained after 1 hour of lower intensitytreadmill running (12 m/minute).

Acknowledgments

This study was supported by the National Natural ScienceFoundation of China (30971412), the Industry Foundation ofMinistry of Health, China (200802036), and the NaturalScience Foundation of Beijing (5102024). We thank Mr.Benoit Viollet in Department of Endocrinology, Metabolismand Cancer, Institute Cochin, University Paris Descartes,France, for providing us two AMPKa2 knockout mice. Inaddition, the technical assistance of Mr. Zhang Lianfeng inInstitute of Laboratory Animal Sciences, Chinese Academy ofMedical Sciences & Peking Union Medical College is greatlyappreciated.

References

1. Dohm GL. Invited review: regulation of skeletal muscle GLUT4 expres-

sion by exercise. J Appl Physiol. 2002;93:782e787.2. Houmard JA, Hickey MS, Tyndall GL, et al. Seven days of exercise

increase GLUT4 protein content in human skeletal muscle. J Appl

Physiol. 1995;79:1936e1938.

3. Kraniou Y, Cameron-Smith D, Misso M, et al. Effects of exercise on

GLUT-4 and glycogenin gene expression in human skeletal muscle. J Appl

Physiol. 2000;88:794e796.

4. Terada S, Yokozeki T, Kawanaka K, et al. Effects of high-intensity

swimming training on GLUT-4 and glucose transport activity in rat

skeletal muscle. J Appl Physiol. 2001;90:2019e2024.

5. Fujii N, Hayashi T, Hirshman MF, et al. Exercise induces isoform-specific

increase in 50AMP-activated protein kinase activity in human skeletal

muscle. Biochem Biophys Res Commun. 2000;273:1150e1155.

6. Kahn BB, Alquier T, Carling D, et al. AMP-activated protein kinase:

ancient energy gauge provides clues to modern understanding of metab-

olism. Cell Metab. 2005;1:15e25.7. Wojtaszewski JF, Nielsen P, Hansen BF, et al. Isoform-specific and

exercise intensity-dependent activation of 50-AMP-activated protein

kinase in human skeletal muscle. J Physiol. 2000;528.1:221e226.

8. Li LG, Chen HQ. Coordinate regulation of contraction signal-induced

GLUT4 transcription by CaMK and AMPK pathways in cultured skel-

etal muscle cells. Prog Biochem Biophys. 2009;36:471e479.

9. Nakano M, Hamada T, Hayashi T, et al. Alpha2 isoform-specific activa-

tion of 5’-adenosine monophosphate activated protein kinase by 5-

aminoimidazole-4-carboxamide-1-beta-d-ribonucleoside at a physiolog-

ical level activates glucose transport and increases glucose transporter 4 in

mouse skeletal muscle. Metabolism. 2006;55:300e308.

10. Park SK, Sheffler TL, Spurlock ME, et al. Chronic activation of 50-AMP-

activated protein kinase changes myosin heavy chain expression in

growing pigs. J Anim Sci. 2009;87:3124e3133.

11. MacLean PS, Zheng D, Jones JP, et al. Exercise-induced transcription of

the muscle glucose transporter (GLUT4) gene. Biochem Biophys Res

Commun. 2002;292:409e414.

12. Zheng D, MacLean PS, Pohnert SC, et al. Regulation of muscle GLUT-4

transcription by AMP-activated protein kinase. J Appl Physiol.

2001;91:1073e1083.

13. Halse R, Fryer LG, McCormack JG, et al. Regulation of glycogen syn-

thase by glucose and glycogen: a possible role for AMP-activated protein

kinase. Diabetes. 2003;52:9e15.

14. Jorgensen SB, Nielsen JN, Birk JB, et al. The a2-50AMP-activated protein

kinase is a site 2 glycogen synthase kinase in skeletal muscle and is

responsive to glucose loading. Diabetes. 2004;53:3074e3081.

15. Jorgensen SB, Viollet B, Andreelli F, et al. Knockout of the a2 but not a1

50-AMP-activated protein kinase isoform abolishes 5-amino-imidazole-4-

carboxamide-1-b-4-ribofuranosidebut not con-traction-induced glucose

uptake in skeletal muscle. J Biol Chem. 2004;279:1070e1079.

16. Miyamoto L, Toyoda T, Hayashi T, et al. Effect of acute activation of 50-AMP-activated protein kinase on glycogen regulation in isolated rat

skeletal muscle. J Appl Physiol. 2007;102:1007e1013.

17. Winder WW, Holmes BF, Rubink DS, et al. Activation of AMP-activated

protein kinase increases mitochondrial enzymes in skeletal muscle. J Appl

Physiol. 2000;88:2219e2226.

18. McGee SL, Hargreaves M. Exercise and myocyte enhancer factor 2

regulation in human skeletal muscle. Diabetes. 2004;53:1208e1214.

19. Viollet B, Andreelli F, Jorgensen SB, et al. Physiological role of AMP-

activated protein kinase (AMPK): insights from knockout mouse

models. Biochem Soc Trans. 2003;31:216e219.

20. Chen ZP, Stephens TJ, Murthy S, et al. Effect of exercise intensity on

skeletal muscle AMPK signaling in humans. Diabetes. 2003;52:

2205e2212.

21. Stephens TJ, Chen ZP, Canny BJ, et al. Progressive increase in human

skeletal muscle AMPKa2 activity and ACC phosphorylation during

exercise. Am J Physiol Endocrinol Metab. 2002;282:E688e694.

22. Wojtaszewski JF, Mourtzakis M, Hillig T, et al. Dissociation of AMPK

activity and ACC phosphorylation in human muscle during prolonged

exercise. Biochem Biophys Res Commun. 2002;298:309e316.23. Fernando P, Bonen A, Hoffman-Goeta L. Predicting submaximal oxygen

consumption during treadmill running in mice. Can J Physiol Pharmacol.

1993;71:854e857.24. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid

guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem.

1987;162:156e159.

25. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using

real-time quantitative PCR and the 2�ddCT method. Methods. 2001;25:

402e408.

26. Dahlqvist A. Determination of maltase and isomaltase activities with

a glucose-oxidase reagent. Biochem J. 1961;80:547e551.

27. Papadopoulos NM, Hess WC. Determination of neuraminic (sialic) acid,

glucose, and fructose in spinal fluid. Arch Biochem Biophys. 1960;88:

167e171.

28. Passonneau JV, Gatfield PD, Schulz DW, et al. An enzymic method for

measurement of glycogen. Anal Biochem. 1967;19:315e326.

29. Leturque A, Loizeau M, Vaulont S, et al. Improvement of insulin action in

diabetic transgenic mice selectively overexpressing GLUT4 in skeletal

muscle. Diabetes. 1996;45:23e27.

30. Ren JM, Marshall BA, Mueckler MM, et al. Overexpression of Glut4

protein in muscle increases basal and insulin-stimulated whole body

glucose disposal in conscious mice. J Clin Invest. 1995;95:429e432.31. Tsao TS, Burcelin R, Katz EB, et al. Enhanced insulin action due to

targeted GLUT4 overexpression exclusively in muscle. Diabetes.

1996;45:28e36.

32. Holmes BF, Lang DB, Birnbaum MJ, et al. AMP kinase is not required for

the GLUT4 response to exercise and denervation in skeletal muscle. Am J

Physiol Endocrinol Metab. 2004;287:E739e743.

33. Mugia M, Jensen TE, Cusinato M, et al. Multiple signaling pathways

redundantly control glucose transporter GLUT4 gene transcription in

skeletal muscle. J Physiol. 2009;587:4319e4327.

34. Ojuka EO, Jones TE, Nolte LA, et al. Regulation of glut4 biogenesis in

muscle: evidence for involvement of AMPK and Ca2þ. Am J Physiol

Endocrinol Metab. 2002;282:E1008eE1013.

35. Mukwevho E, Kohn TA, Lang D, et al. Caffeine induces hyperacetylation

of histones at the MEF2 site on the Glut4 promoter and increases MEF2A

binding to the site via a CaMK-dependent mechanism. Am J Physiol

Endocrinol Metab. 2008;294:E582eE588.

36. Fluck M, Waxham MN, Hamilton MT, et al. Skeletal muscle Ca(2þ)-

independent kinase activity increases during either hypertrophy or

running. J Appl Physiol. 2000;88:352e358.

22 H. Gong, Y. Zhang / Journal of Exercise Science & Fitness 10 (2012) 16e22

37. Rose AJ, Hargreaves M. Exercise increases Ca2þ-calmodulin-dependent

protein kinase II activity in human skeletal muscle. J Physiol Lond.

2003;553:303e309.

38. Kuo CH, Browning KS, Ivy JL. Regulation of GLUT4 protein expression

and glycogen storage after prolonged exercise. Acta Physiol Scand.

1999;165(2):193e201.

39. Jorgensen SB, Wojtaszewski JF, Viollet B, et al. Effects of a-AMPK

knockout on exercise-induced gene activation in mouse skeletal muscle.

FASEB J. 2005;19:1146e1148.

40. Viollet B, Andreelli F, Jorgensen SB, et al. The AMP-activated protein

kinase a2 catalytic subunit control whole-body insulin sensitivity. J Clin

Invest. 2003;111:91e98.