SUMMARY

1. Skeletal muscle is a complex and heterogenous tissue capable of remarkable adaptation in response to exercise train-ing. The role of gene transcription, as an initial target to control protein synthesis, is poorly understood.

2. Mature myofibres contain several hundred nuclei, all ofwhich maintain transcriptional competency, although the local-ized responsiveness of nuclei is not well known. Myofibres arecapable of hypertrophy. These processes require the activationand myogenic differentiation of mononuclear satellite cells thatfuse with the enlarging or repairing myofibre.

3. A single bout of exercise in human subjects is capable ofactivating the expression of many diverse groups of genes.

4. The impact of repeated exercise bouts, typical of exercisetraining, on gene expression has yet to receive systematic investigation.

5. The molecular programme elicited by resistance exerciseand endurance exercise differs markedly. Muscular hypertrophyfollowing resistance exercise is dependent on the activation ofsatellite cells and their subsequent myogenic maturation.Endurance exercise requires the simultaneous activation of mitochondrial and nuclear genes to enable mitochondrial biogenesis.

6. Future analysis of the regulation of genes by exercise maycombine high-throughput technologies, such as gene-chips,enabling the rapid detection and analysis of changes in theexpression of many thousands of genes.

Key words: mRNA, myofibre, satellite cells, training, tran-scription.

INTRODUCTION

Mature skeletal muscle is a remarkably adaptive tissue, able todemonstrate significant regeneration, hypertrophy and metabolicadaptation.1 One of the most powerful stimuli for inducing skeletal

muscle cellular re-organization is exercise training. In adults,depending on the type and duration of training, there can be appre-ciable changes in the size of the muscle mass and fibre-type com-position, in addition to increased contractile activity and metaboliccharacteristics of the myofibre population.2 Considerable researchhas focused on the phenotypic nature of the adaptations, measuredas alterations in protein abundance and activity, metabolic pathwayflux and anatomical alterations. Yet, surprisingly few studies havesystematically addressed the role and importance of gene expres-sion, despite the sometimes marked increase in protein levels. Inthe present review, the role of gene expression in skeletal muscleadaptability will be examined. Attention will be given to the dif-ferential gene programmes elicited by varied exercise interventions,the hypothetical adaptations in gene expression following repeatedexercise (training) and the emerging technologies that hold promisein analysing the responsiveness of many thousands of genes to exer-cise interventions.

REGULATION OF GENE TRANSCRIPTION

The draft sequence of the human genome, published February 2001,had as a major finding that the human genome contains approxi-mately 30 000 genes,3 one-quarter of the number predicted only 12 months earlier.4 The smaller than anticipated gene numberencodes approximately 85 000 different mRNAs, due to alternativemRNA splicing and variable polyadenylation.5 Messenger RNAsynthesis or gene transcription is catalysed by multisubunit RNApolymerase II, which is under the interacting and complex regula-tion of the chromatin, histone acetylation and many nuclear proteins,including ‘transcription factors’.6 Transcription factors comprise adiverse population of DNA-binding proteins, which, in manyinstances, are the terminal protein linking a complex signallingsequence with the activation of gene transcription. The activationof gene transcription results in the transient synthesis of mRNA,which may undergo modification to generate the mature transcript.The mature transcript exits the nucleus via the nuclear pores and,when coupled to the translational machinery of the ribsome, is trans-lated into the encoded peptide.

UNIQUE REGULATION OF GENETRANSCRIPTION IN SKELETAL MUSCLE

Skeletal muscle is a complex tissue containing predominantly multi-nucleated mature myofibres. Each myofibre arises from the fusionof many hundred mononucleated progenitor cells, known as satel-lite cells, which are capable of undergoing myogenic programming

Proceedings of the Australian Physiological and Pharmacological SocietySymposium: Integrative Physiology of Exercise

EXERCISE AND SKELETAL MUSCLE GENE EXPRESSION

David Cameron-Smith

School of Health Sciences, Deakin University, Melbourne, Victoria, Australia

Correspondence: Dr David Cameron-Smith, School of Health Sciences,Deakin University, 221 Burwood Highway, Burwood, Victoria 3125,Australia. Email: davidcs@deakin.edu.au

Presented at the Australian Physiological and Pharmacological SocietySymposium Integrative Physiology of Exercise, November 2000. The papersin these proceedings have been peer reviewed.

Received 21 July 2001; revision 10 September 2001; accepted 11 September 2001.

Clinical and Experimental Pharmacology and Physiology (2002) 29, 209–213

210 D Cameron-Smith

and differentiation into mature myocytes.7 Satellite cells remain present within skeletal muscle, even into old age, retaining their proliferative potential.8 Satellite cell proliferation and subsequentdifferentiation into myocytes enables repair and hypertrophy of exist-ing myofibres or the generation of new myofibres.

Within mature myofibres, the many nuclei remain transcription-ally competent and theoretically capable of upregulated gene syn-thesis.9 However, significant compartmentalization of geneexpression is evident, with genes for the �-subunit of the acetyl-choline receptor and �-dystrobrevin (involved in synapse matura-tion) predominantly expressed by nuclei in close proximity to theneuromuscular junction.10 This is further supported by the limiteddiffusion of mRNA within structurally complex cells.11 Yet, theexpression of both endogenous muscle-specific genes and house-keeping genes is not uniform along the length of a myofibre, despiteshared cytoplasm and equal protein requirement along the myofi-bre.9 The differences in individual nuclei action may be explainedin terms of instability of the transcriptional complexes, such thatsmall fluctuations in low-abundance enhancers or transcription factors increase stability and regulate gene expression stochastically,in an off/on mode, rather than regulating the rate of transcription.12

To date, it has yet to be determined whether exercise elicits thewidespread activation of nuclei or, alternatively, localized nucleihave sustained periods of transcriptional activity. However, the largenumber of nuclei and the involvement of the entire length of themyofibre in contractile events suggest marked synthesis of newmRNA species is possible in myofibres.

EXERCISE REGULATES GENETRANSCRIPTION

Numerous studies have now demonstrated that, following a singleexercise bout, significant elevations in the concentration of mRNAspecies, including metabolic,13–15 coordinatory16,17 and immuno-modulatory18 genes, are observed in muscle samples from healthyhuman subjects. In response to exercise, gene transcription may beactivated within seconds of contraction initiation through to hoursafter the cessation of exercise or following the restoration of nutri-ent stores.19 For example, the mRNA abundance of the fos proto-oncogene family (c-fos and fosB) has been reported to be more than20-fold higher 4 min after the commencement of treadmill runningin adult subjects.16 Yet, the majority of studies have demonstratedthat gene expression is most significantly enhanced in the recoveryperiod, following the completion of the exercise bout.19 Using anovel run-on protocol to measure nuclear mRNA abundance,Pilegaard et al.20 have demonstrated widespread activation of genesin the hours following varying modes of exercise. Thus, the pre-dominant transcriptional response to exercise may be present in therecovery phase of exercise. However, during exercise, the expres-sion of many genes may be suppressed. In preliminary data fromour laboratory, the expression of 184 genes was analysed simulta-neously using gene-array technology (Research Genetics, HumanGenefilter, Huntsville, AL, USA) in human subjects from whombiopsies were collected prior and immediately at the end of 40 mincycling exercise (70% VO2max). Of the genes analysed, more than85% demonstrated a greater than 1.5-fold reduction in abundance(D Cameron-Smith, unpubl. data, 2001), demonstrating the pre-dominant action may be the inhibition of transcription or degrada-

tion of mRNA during exercise prior to selective transcriptional activity once exercise has ceased.

EXERCISE TRAINING AND GENETRANSCRIPTION

The impact on gene expression of exercise performed without priorfamiliarization or training is likely to differ markedly from theresponse to repeated exercise bouts or the trained response. Few studies have systematically examined the changes in mRNA abun-dance following repeated bouts of the same exercise protocol, eithermaintained at the same absolute workload or matched relative tothe improvements in exercise performance observed with training.Furthermore, the contribution of transcriptional (mRNA synthesis)versus translational (mRNA stability or translation efficiency)adaptations to a training-induced increase in protein levels are poorlyunderstood. It is difficult to correlate mRNA with protein levels dueto both the transient nature of mRNA following exercise and theprotracted synthesis and longer half-lives of proteins. Therefore, caution must be applied to the analysis of adaptive changes in bothmRNA responses to exercise and the impact of transcriptional com-pared with translational events.

Although few studies have examined the transient changes inmRNA levels following an initial exercise bout and again after anadaptation or training period, significant adaptation is likely to occur.Of the available data, it has been shown that 8 weeks of leg–kneeextensor training markedly attenuated (three-fold) the expression ofvascular endothelial growth factor (VEGF) mRNA at a workloadadjusted to increase as exercise capacity improved with training.21

An alternative adaptive response to repeated exercise may be theselective activation of mRNA species that require repeated bouts ofexercise prior to any measurable increase in mRNA abundance. In our laboratory, the expression of several genes involved in skeletal muscle fat transport (fatty acid translocase) and oxidation(carnitine palmitoyltransferase 1) were not increased following a single exercise bout, but demonstrated increased basal and post-exercise expression following 9 days of 1 h cycle training (RJTunstall, unpubl. data, 2001). Similarly, adaptations, such as inhi-bitions in gene expression, requiring repeated bouts of exercise havealso been shown. After a single exercise bout, no measurable impacthas been observed on the expression of the myosin heavy chain(MHC) IIa isoform, yet 7 days of training led to reduced mRNAlevels after the last exercise bout, demonstrating that suppressionwas evident only after repeated exercise bouts.22 Further studies areclearly required to elucidate the differential impact exercise train-ing has on the expression of genes and the contribution this exertsin adaptive alterations in the cellular protein concentrations.

POST-TRANSLATIONAL EVENTS ANDEXERCISE TRAINING

In experimental models of increased muscle loading, an increasedprotein synthesis rate proceeds changes in mRNA abundance, imply-ing that the efficiency of protein synthesis is enhanced.23 Severalstudies have now examined the activation of the rate-limiting stepof protein synthesis, the initiation step in which the mRNA tran-script is coupled to the ribosomal machinery. Several kinasesinvolved in the initiation step of protein synthesis, including; the

Exercise and gene expression 211

70 kDa AS6 protein kinase (p70S6K), protein kinase B (PKB/Akt)and terminal members of the mitogen-activated protein kinase kinasepathway (extracellular signal-regulated kinase and p38), have allrecently been identified to be upregulated by exercise.24

The extensive remodelling of muscle with exercise training alsorequires the activation of protein degradation pathways for theremoval of proteins no longer required. The ubiquitin–proteasomepathway is the major mechanism of selective protein degradation.25

Chronic motor nerve stimulation for 28 days of the rabbit tibialisanterior muscle markedly increased total proteasome activity of muscle extracts, with upregulated protein abundance of the 20S proteasome subunit and two regulatory proteins, namely PA700 andPA28.26

TRANSCRIPTIONAL REGULATION INEXERCISE

Resistance exercise

A pronounced adaptive response to high-intensity or weight-bearing exercise interventions is muscle hypertrophy. The increasedmass of active muscle groups is achieved by an increase in the volume of individual myofibres.27 The enlarged myofibre can onlyexpand with the insertion of new nuclei, because a constant ratio ofnuclei to cytoplasmic volume is maintained throughout all hyper-trophic responses.28 Thus, hypertrophy is dependent on the prolif-erative activation of satellite cells and their myogenic differentiation7

prior to fusion with the existing myofibre.29

Progression of satellite cells into myoblasts involves the regula-tion of muscle-specific proteins belonging to the basic-helix-loop-helix family of transcription factors. Members include MyoD,myogenin, myf-5, myogenic regulatory factor (MRF)-4 and myocyteenhancer factor (MEF)-2, which collectively function as dominantactivators of skeletal muscle differentiation.30 In the quiescent state,these myogenic factors are expressed at very low levels in satellitecells but, once activated, their abundance increases markedly.31

Considerable progress has been made in elucidating the signallingpathways responsible for the activation of the myogenic differenti-ation pathway, highlighting the pivotal role of endogenouslyderived factors regulated by shear or stress forces, including integrin-linked kinases,32 and increased intracellular calcium flux,activating the calcineurin intracellular signalling pathway.33

Exogenous hormone activation of insulin-like growth factor-1, whichmay participate in the activation of calcineurin,34 in addition toangiotensin II35 and fibroblast growth factors,36 may also coordinatemyogenic differentiation of satellite cells.

Endurance exercise

In contrast, endurance exercise typically results in a shift in myo-fibrillar components towards an increased abundance of slow isoform proteins, together with upregulated mitochondrial and oxida-tive metabolism enzyme levels.37 Interestingly, induction of slowmyofibrillar isoform genes may also be dependent on calcium andthe calcineurin pathway.38 Thus, an apparent anomalous situationarises in which calcium activation of calcineurin is pivotal in twovastly differing transcriptional programmes, either hypertrophy or the activation of the slow myofibre cellular pattern and mito-chondrial biogenesis. A partial explanation is evident in the patternof intracellular calcium released by tonic low-frequency motor nerve

activity typical of endurance exercise, as compared with the high-amplitude and short-duration response of intense hypertophic phys-ical activity.39 The resultant differing patterns of activation may resultin the recruitment of alternative calcium-responsive kinases40 ordepend on kinases being activated simultaneously to stimulate anintegrative downstream regulator.41 Such a downstream integrationof calcium signalling may be controlled by the myogenic transcription factor MEF-2, the activity of which is regulated by multiple signal cascades, including calcineurin and calmodulin-dependent protein kinase pathways.41 However, much is to be dis-covered about how these pathways may intersect with thosenecessary for mitochondrial proliferation, which requires the coordination of both nuclear genes and the genes contained withinthe mitochondria’s own DNA.42

ANALYSIS STRATEGIES

Current individual mRNA species are analysed using eitherhybridization techniques (northern blotting or RNAse protection)or polymerase chain reaction (PCR) amplification of individualgenes. Normalization of gene abundance is based on comparisonswith genes that have a constant gene expression, namely the house-keeping genes. These techniques enable the semiquantitative analysis of single or small numbers of genes and are labour inten-sive. Given that any phenotypic adaptation may require the activa-tion and inhibition of many thousands of genes simultaneously, thesetechniques will provide limited new data on the scope or complex-ity of the adaptation.

New technologies, particular high-throughput gene-scanningtools (including gene-chips), will provide dramatic insights into thecomplex patterns of gene expression necessary for the phenotypeadaptation in muscle. The use of gene-chip technology to probe forpatterns of gene expression have been applied to the analysis of fibre-type differences in gene expression,43 skeletal muscle ageingand species differences44 and the molecular pathophysiology of muscular dystrophies.45 Gene-chip technology, not withstanding thepotential pitfalls,46 will enable the mass screening of genetic responseto a wide variety of stimuli, including different exercise inter-ventions. This technique holds particular promise with respect tothe sensitivity and reproducibility of gene analysis, enabling quanti-tative detection of many regulatory genes that are often expressedat relatively low abundance.47 Furthermore, the inclusion on gene-chips of expressed sequence tags, which represent partial genesequences of genes of no described function, may aid in the identi-fication of new and novel genes involved in the adaptive responseof exercise to physical activity.

FUTURE ISSUES

There is much to be discovered in the pathways of gene regulationin response to exercise. Currently, little is known of the complexintegration of competing signalling cascades prior to the activationof exercise-sensitive genes. Further confounding these investigationsare the large differences in individual responsiveness to exercise.The population differences in exercise capacity are unlikely to bedue to polymorphisms of single genes, although the recent exami-nation of the angiotensin-converting enzyme gene has yielding inter-esting associations with performance in several,48,49 but not all,50

212 D Cameron-Smith

studies. The recently published genomic scan of the HERITAGEfamily training study identified a number of potential genetic markers that correlate with changes in body composition.51 Suchgene-scanning studies will increasingly define areas of the humangenome that are linked to individual differences in exercise respon-siveness. Beyond these issues will be the difficult task of convert-ing genetic findings into outcomes that can benefit the health ofdiseased individuals, including the obese and diabetic.

ACKNOWLEDGEMENTS

The author acknowledges the collaborations and support of ProfessorMark Hargreaves, Professor Greg Collier and Dr Rodney Snow(School of Health Sciences, Deakin University).

REFERENCES

1. Goldspink G. Selective gene expression during adaptation of muscle inresponse to different physiological demands. Comp. Biochem. Physiol.B 1998; 120: 5–15.

2. Booth F, Tseng B, Fluck M, Carson J. Molecular and cellular adapta-tion of muscle in response to physical training. Acta Physiol. Scand.1998; 162: 343–50.

3. Venter JC, Adams MD, Myers EW, et al. The sequence of the humangenome. Science 2001; 291: 1304–51.

4. Liang F, Holt I, Pertea G, Karamycheva S, Salzberg SL, QuackenbushJ. Gene index analysis of the human genome estimates approximately120 000 genes. Nat. Gen. 2000; 25: 239–40.

5. Beaudoing E, Freier S, Wyatt JR, Claverie JM, Gautheret D. Patternsof variant polyadenylation signal usage in human genes. Genome Res.2000; 10: 1001–10.

6. Macfarlane WM. Demystified … transcription. Mol. Pathol. 2000; 53:1–7.

7. Seale P, Rudnicki MA. A new look at the origin, function, and ‘stem-cell’ status of muscle satellite cells. Dev. Biol. 2000; 218: 115–24.

8. Marsh DR, Criswell DS, Carson JA, Booth FW. Myogenic regulatoryfactors during regeneration of skeletal muscle in young, adult, and oldrats. J. Appl. Physiol. 1997; 83: 1270–5.

9. Newlands S, Levitt L, Robinson C et al. Transcription occurs in pulsesin muscle fibers. Genes Dev. 1998; 12: 2748–58.

10. Fontaine B, Sassoon D, Buckingham M, Changeux JP. Detection of thenicotinic acetylcholine receptor alpha-subunit mRNA by in situhybridization at neuromuscular junctions of 15-day-old chick striatedmuscles. EMBO J. 1988; 7: 603–9.

11. Rossi SG, Vazquez AE, Rotundo RL. Local control of acetyl-cholinestrerase gene expression in multinucleated skeletal muscle fibers:Individual nuclei respond to signals from the overlying plasma mem-brane. J. Neurosci. 2000; 20: 919–28.

12. McAdams HH, Arkin A. Stochastic mechanisms in gene expression.Proc. Natl Acad. Sci. USA 1997; 94: 814–19.

13. Koval JA, DeFronzo RA, O’Doherty RM et al. Regulation of hexokinaseII activity and expression in human muscle by moderate exercise. Am. J. Physiol. 1998; 274: E304–8.

14. Kraniou Y, Cameron-Smith D, Misso M, Collier G, Hargreaves M.Effects of exercise on GLUT-4 and glycogenin gene expression inhuman skeletal muscle. J. Appl. Physiol. 2000; 88: 794–6.

15. Wadley GD, Tunstall RJ, Sanigorski A, Collier GR, Hargreaves M,Cameron-Smith D. Differential effects of exercise on insulin-signalinggene expression in human skeletal muscle. J. Appl. Physiol. 2001; 90:436–40.

16. Puntschart A, Wey E, Jostarndt K et al. Expression of fos and jun genesin human skeletal muscle after exercise. Am. J. Physiol. 1998; 274:C129–37.

17. Gustafsson T, Puntschart A, Kaijser L, Jansson E, Sundberg J. Exercise-induced expression of angiogenesis-related transcription and growth factors in human skeletal muscle. Am. J. Physiol. 1999; 276:H679–85.

18. Febbraio M, Koukoulas I. HSP72 gene expression progessivelyincreases in human skeletal muscle during prolonged, exhaustive exer-cise. J. Appl. Physiol. 2000; 89: 1055–60.

19. Neufer P, Ordway G, Williams R. Transient regulation of c-fos, �B-crystallin, and hsp70 in muscle during recovery from contractileactivity. Am. J. Physiol. 1998; 274: C341–6.

20. Pilegaard H, Ordway G, Saltin B, Neufer P. Transcriptional regulationof gene expression in human skeletal muscle during recovery from exer-cise. Am. J. Physiol. 2000; 279: E806–14.

21. Richardson R, Wagner H, Mudaliar S, Saucedo E, Henry R, Wagner P.Exercise adaptation attenuates VEGF gene expression in human skeletal muscle. Am. J. Physiol. 2000; 279: H772–8.

22. O’Neill D, Zheng D, Anderson W, Dohm G, Houmard J. Effect ofendurance exercise on myosin heavy chain gene regulation in humanskeletal muscle. Am. J. Physiol. 1999; 276: R414–19.

23. Welle S, Bhatt K, Thornton C. Stimulation of myfibrillar synthesis byexercise is mediated by more efficient translation of mRNA. J. Appl.Physiol. 1999; 86: 1220–5.

24. Nader GA, Esser KA. Intracellular signaling specificity in skeletal mus-cle in response to different modes of exercise. J. Appl. Physiol. 2001;90: 1936–42.

25. Lecker SH, Solomon V, Mitch WE, Goldberg AL. Muscle protein breakdown and the critical role of the ubiquitin–proteasome pathwayin normal and disease states. J. Nutr. 1999; 129 (Suppl. 1S): S227–37.

26. Ordway G, Neufer P, Chin E, DeMartino G. Chronic contractile activ-ity upregulates the proteasome system in rabbit skeletal muscle. J. Appl.Physiol. 2000; 88: 1134–41.

27. Green H, Goreham C, Ouyang J, Ball Burnett M, Ranney D. Regulationof fiber size, oxidative potential, and capillarization in human muscleby resistance exercise. Am. J. Physiol. 1999; 276: R591–6.

28. McCall G, Allen D, Linderman J et al. Maintenance of myonucleardomain size in rat soleus after overload and growth hormone/IGF-I treat-ment. J. Appl. Physiol. 1998; 84: 1407–12.

29. Garry D, Meeson A, Elterman J, Zhao Y, Yang P, Bassel-Duby R.Myogenic stem cell function is impaired in mice lacking the fork-head/winged helix protein MNF. Proc. Natl Acad. Sci. USA 2000; 97:5416–21.

30. Perry R, Rudnick M. Molecular mechanisms regulating myogenic deter-mination and differentiation. Front. Biosci. 2000; 5: 750–67.

31. Cooper RN, Tajbakhsh S, Mouly V, Cossu G, Buckingham M, Butler-Browne GS. In vivo satellite cell activation via Myf5 and MyoD in regenerating mouse skeletal muscle. J. Cell Sci. 1999; 112: 2895–901.

32. Huang Y, Li J, Zhang Y, Wu C. The roles of integrin-linked kinase inthe regulation of myogenic differentiation. J. Cell Biol. 2000; 150:861–72.

33. Olson E, Williams R. Calcineurin signaling and muscle remodeling. Cell2000; 101: 689–92.

34. Musaro A, McCullagh K, Naya F, Olson E, Rosenthal N. IGF-1 inducesskeletal myocyte hypertrophy through calcineurin in association withGATA-2 and NF-ATc1. Nature 1999; 400: 581–5.

35. Gordon SE, Davis BS, Carlson CJ, Booth FW. ANG II is required foroptimal overload-induced skeletal muscle hypertrophy. Am. J. Physiol.2001; 280: 150–9.

36. Scata KA, Bernard DW, Fox J, Swain JL. FGF receptor availability regulates skeletal myogenesis. Exp. Cell Res. 1999; 250: 10–21.

37. Holloszy JO, Coyle EF. Adaptations of skeletal muscle to enduranceexercise and their metabolic consequences. J. Appl. Physiol. 1984; 56:831–8.

38. Delling U, Tureckova J, Lim H, De Windt L, Rotwein P, Molkentin J.A calcineurin–NFATc3-dependent pathway regulates skeletal muscle dif-ferentiation and slow myosin heavy-chain expression. Mol. Cell. Biol.2000; 20: 6600–11.

Exercise and gene expression 213

39. Chin ER, Olson EN, Richardson JA et al. A calcineurin-dependent tran-scriptional pathway controls skeletal muscle fiber type. Genes Dev. 1998;12: 2499–509.

40. Dolmetsch RE, Lewis RS, Goodnow CC, Healy JI. Differential acti-vation of transcription factors induced by Ca2+ response amplitude andduration. Nature 1997; 386: 855–8.

41. Wu J, Naya FJ, McKinsey TA et al. MEF2 responds to multiple calcium-regulated signals in the control of skeletal muscle fibre type.EMBO J. 2000; 19: 1963–73.

42. Hood DA. Contractile activity-induced mitochondrial biogenesis inskeletal muscle. J. Appl. Physiol. 2001; 90: 1137–57.

43. Campbell WG, Gordon SE, Carlson CJ, Pattison JS, Hamilton MT,Booth FW. Differential global gene expression in red and white skele-tal muscle. Am. J. Physiol. 2001; 280: C763–8.

44. Welle S, Brooks A, Thornton CA. Senescence-related changes in geneexpression in muscle. Similarities and differences between mice andmen. Physiol. Genom. 2001; 5: 67–73.

45. Chen Y-W, Zhao P, Borup R, Hoffman EP. Expression profiling in themuscular dystrophies. Identification of novel aspects of molecular patho-physiology. J. Cell. Biol. 2000; 161: 1321–36.

46. Knight J. When the chips are down. Nature 2001; 410: 860–1.47. Duggan DJ, Bittner M, Chen Y, Meltzer P, Trent JM. Expression

profiling using cDNA microarrays. Nat. Genet. 1999; 21: 10–14.48. Williams A, Rayson M, Jubb M et al. The ACE gene and performance.

Nature 2000; 403: 614.49. Folland J, Leach B, Little T et al. Angiotensin-converting enzyme geno-

type affects the response of human skeletal muscle to functional over-load. Exp. Physiol. 2000; 85: 575–9.

50. Rankinen T, Perusse L, Gagnon J et al. Angiotensin-converting enzymeID polymorphism and fitness phenotype in the HERITAGE family study.J. Appl. Physiol. 2000; 88: 1029–35.

51. Chagnon YC, Rice T, Perusse L et al. Genomic scan for genes affecting body composition before and after training in caucasians fromHERITAGE. J. Appl. Physiol. 2001; 90: 1777–87.