Professional Documents
Culture Documents
Short, Kevin R., Janet L. Vittone, Maureen L. Bigelow, David not appear to change significantly with age in human leg
N. Proctor, Jill M. Coenen-Schimke, Paul Rys, and K. Sreekuma- muscles, but that the size of type II fibers was selectively
ran Nair. Changes in myosin heavy chain mRNA and protein ex- reduced in the vastus lateralis and gastrocnemius of older
pression in human skeletal muscle with age and endurance exercise people (6, 11, 15, 27).
training. J Appl Physiol 99: 95–102, 2005. First published March 3,
The use of myosin ATPase staining to determine muscle
2005; doi:10.1152/japplphysiol.00129.2005.—Aging is associated
with reduced muscle strength and atrophy of type II muscle fibers. fiber type is based on the relative expression of isoforms of
Muscle fiber type and contractile function are primarily determined by myosin heavy chain (MHC), the major contractile protein in
myosin heavy chain (MHC) isoforms. There are few data available on skeletal muscle. Contractile phenotype in adult muscle fibers is
the effects of aging on MHC isoform expression in humans. In the primarily determined by the relative expression of the MHC
present study, we tested the hypothesis that MHC isoform protein isoforms I, IIa, and IIx (5, 9), which correspond to the histo-
composition and mRNA abundance would favor a fast-to-slow iso- chemically determined fiber types I, IIa, and IIb in humans
form shift with aging and in response to endurance exercise training. (29). Han et al. (9) demonstrated that ATP consumption rate
Muscle biopsies were obtained from previously sedentary, healthy and tension cost of single human muscle fibers vary with the
men and women, aged 21– 87 yr before (n ⫽ 77) and after (n ⫽ 65) expression of MHC isoforms. Whereas most muscle fibers
16 wk of bicycle training (up to 45 min at 80% peak heart rate, 3– 4 appear to express a single MHC isoform, isoform coexpression
days/wk). At baseline, MHC I mRNA was unchanged with age,
whereas IIa and IIx declined by 14 and 10% per decade, respectively
has been demonstrated in up to ⬃30% of fibers and may vary
(P ⬍ 0.001). MHC IIa and IIx protein declined by 3 and 1% per with age and exercise training (1, 16, 38). Thus measurement
decade with a reciprocal increase in MHC I (P ⬍ 0.05). After training, of fractional MHC isoform composition of muscle provides an
MHC I and IIa mRNA increased by 61 and 99%, respectively, and IIx important index of the regulation of contractile phenotype.
decreased by 50% (all P ⬍ 0.001). The increase in MHC I mRNA was The mechanisms that regulate the expression of MHC iso-
positively associated with age, whereas the changes in MHC IIa and forms in aging muscle are not yet clear. It has been reported
IIx mRNA were similar across age. MHC I protein increased by 6% that the synthesis rate of MHC protein is reduced with age in
and was positively related to age, whereas IIx decreased by 5% and humans (2, 10), but it is not known whether this is a global
was inversely related to age. These results suggest that the altered change that affects all MHC isoforms or whether there is
expression of MHC isoforms with aging is transcriptionally regulated. isoform-specific regulation. Isoform-specific regulation may
In response to endurance exercise, regulation of MHC isoform tran-
scripts remains robust in older muscle, but this did not result in
occur at the pretranslational level. Previous work from our
corresponding changes in MHC protein expression. laboratory demonstrated that the mRNA abundance of MHC
IIa and IIx was reduced in vastus lateralis biopsies from
muscle contractile proteins; gene expression; isokinetic strength; mus- healthy older vs. younger people, whereas MHC I mRNA
cle size content was unchanged (3). This finding is consistent with a
report by Klitgaard et al. (15) in which older men were found
AGING IS ASSOCIATED WITH REDUCED size and strength of human to have smaller type II muscle fibers than young men and a
skeletal muscles (13, 19, 23). With aging, there are also corresponding decrease in the fractional amount of MHC IIx
changes in the size, number, and contractile phenotype of protein. In contrast, in studies by Welle and colleagues (36, 37)
individual muscle fibers that may impact muscle function. and Marx et al. (24), there was no difference detected between
Classic studies by Larsson et al. (18) and Lexell (22) showed groups of younger and older people (n ⫽ 8 –16 people per
that the decline in size of the vastus lateralis with normal aging group) in mRNA abundance of MHC isoforms. The study by
could be attributed to fewer total muscle fibers and preferential Marx et al. also reported that the fractional protein content of
atrophy of fibers classified as type II (fast twitch) by histo- MHC isoforms was also unchanged with age.
chemical staining of myosin ATPase. Those authors also re- Therefore, it is not yet resolved whether expression of MHC
ported that the percentage of each of the three major fiber types mRNA and protein is altered with age in human skeletal
in adult muscle, type I (slow twitch), IIa, and IIb, was largely muscle, and to our knowledge only one study has examined the
unchanged with age. Subsequent investigations by other age effect on both MHC mRNA and protein (24). Some of the
groups also showed that the percentage of each fiber type did variation among studies may be due to the use of a small
number of subjects, different methods of sample analysis, or
http://www. jap.org 8750-7587/05 $8.00 Copyright © 2005 the American Physiological Society 95
Downloaded from journals.physiology.org/journal/jappl (103.084.005.071) on November 15, 2021.
96 AGING AND EXERCISE EFFECT ON MHC
variable control for physical activity status. Thus the first Table 1. Subject characteristics
purpose of the present study was to test the hypothesis that
Younger Middle-aged Older
expression of MHC II mRNA and protein is reduced with age Variable Gender (21–37 yr) (40 –58 yr) (60 – 87 yr)
in human skeletal muscle. To address this question, muscle
biopsies of the vastus lateralis were obtained from 77 healthy, No. of participants W 10 12 17
M 12 11 15
non-exercise-trained men and women between the ages of 21 Weight, kg W 68.7⫾1.8 69.1⫾2.2 69.5⫾2.0
and 80 yr. We also examined the strength of the association M 85.7⫾2.7 90.4⫾2.8 86.0⫾2.8
between MHC expression and muscle strength in these people. BMI, kg/m2 W 24.2⫾0.5 25.1⫾0.6 26.3⫾0.6
Larsson et al. (18) reported that the total cross-sectional area of M 25.8⫾0.6 28.1⫾0.6 27.6⫾0.7
type II fibers was significantly related to muscle strength in Body fat, % W 34.1⫾1.2 37.2⫾1.1 39.3⫾1.1
M 23.9⫾1.0 25.8⫾1.7 27.1⫾1.3
younger and older people. Klitgaard et al. (15), however, did Fat-free mass, kg W 41.4⫾0.8 39.3⫾0.9 38.7⫾0.9
not find a significant relationship between muscle strength and M 60.4⫾1.7 60.9⫾2.6 56.6⫾1.4
fiber area or MHC isoform content in young and older un- V̇O2peak, l/min W 1.95⫾0.05 1.66⫾0.06 1.26⫾0.05
trained men, but this negative finding may be due to the use of M 3.15⫾0.11 2.73⫾0.19 1.96⫾0.12
a small number of subjects (7– 8 per group). Values are means ⫾ SE. W, women; M, men; BMI, body mass index;
The second purpose of the study was to determine whether V̇O2peak, peak oxygen uptake. Statistically significant differences between men
an endurance exercise training program would alter MHC and women were present for each variable. There was not a statistically
isoform expression and whether the effect would vary with significant change in BMI with age. Body fat increased (r ⫽ 0.40), whereas
fat-free mass (r ⫽ ⫺0.46), and V̇O2peak (r ⫽ ⫺0.81) decreased with age, P ⬍
age. We are aware of only one study that prospectively exam- 0.01.
ined the effect of endurance exercise training on MHC mRNA
or protein expression in human skeletal muscle. O’Neill and
colleagues (25) found that the mRNA abundance for MHC IIx
training or control phases. During each study period, a weight-
was reduced in vastus lateralis muscle in young men after
maintaining diet (55% carbohydrate, 30% fat, and 15% protein) was
performing 7 consecutive days of bicycle training. This ap- provided for the first 4 days. On the morning of day 4, subjects were
pears to be in agreement with previous studies that have admitted to the General Clinical Research Center (GCRC). The
reported that endurance training results in a decrease in the following morning (day 5), muscle biopsy samples from the vastus
percentage of fibers classified as type IIb by myosin ATPase lateralis (33) were obtained.
histochemistry (7, 12). These changes in fiber type percents The exercise program was performed on a stationary bicycle.
were observed after 6 wk of bicycle training in young men (12) Training started with three sessions per week, lasting 20 min each, at
or after a 9- to 12-mo walking and jogging program in older an intensity eliciting 70% of the peak heart rate. Peak heart rate during
men and women (7). To our knowledge, however, the effects cycling was measured during an aerobic capacity test as described
of endurance exercise training on MHC mRNA or protein below. Intensity, duration, and number of sessions were gradually
expression have not been compared in younger and older increased so that the final month of training consisted of four sessions
per week at 80% of maximal heart rate for 40 min. Exercise specialists
people. Therefore, subjects in the present study were tested supervised each session and recorded heart rates. Compliance with the
before and after completion of a 16-wk program of standard- target workloads and number of sessions was ⬎90%. The exercise
ized bicycle exercise training to determine whether the hypoth- protocol was completed by 41 of 47 participants originally assigned.
esized shift from fast to slow MHC isoform expression would The control group was taught a series of flexibility exercises and
vary with age. encouraged to perform them at home while maintaining their regular
lifestyle. Follow-up tests were available for 37 of the 43 people in the
METHODS control group. Subsequently, 24 control group members opted to
complete the exercise program, yielding a total of 65 people studied
Participants. Healthy men and women who exercised ⬍30 min on before and after exercise training.
two or fewer occasions per week during the previous 9 mo were Because the goal of the study was to examine effects of the exercise
recruited from the local community. Physical activity levels were program, participants were instructed to maintain body weight.
confirmed by questionnaire (33). Health status was assessed by Weight was recorded weekly, and the GCRC dietary staff provided
medical history, physical examination, blood chemistries (complete further guidance if weight changed ⬎2%. Only one person in the
blood count and comprehensive chemistry panel, including liver exercise group discontinued the study because of excessive weight
enzymes, creatinine, electrolytes, and glucose), urinalysis, and resting loss.
electrocardiogram. Exclusion criteria included body mass index ⬎32 Body composition and aerobic capacity. Fat and fat-free mass were
kg/m2, tobacco use, use of -blockers, diabetes or other metabolic or determined by dual-energy X-ray absorptiometry (DPX-L, Lunar,
endocrine disorders, history of alcohol or substance abuse, and debil- Madison, WI). Thigh muscle area was measured by using a single-
itating chronic illness. Thirty-nine women and 38 men between the slice (6-mm thickness) computed tomography scan (model C-150,
ages of 21 and 87 yr met these criteria and were enrolled after Imatron, San Francisco, CA) at the midpoint between the knee and
providing written and oral consent. Consent was obtained before any hip. Areas of muscle, fat, and bone were estimated by manual
tests were performed. The Mayo Foundation Institutional Review planimetry using custom software (21).
Board approved the study. Physical characteristics of the men and A standard treadmill stress test was performed initially to ensure
women in each decade have been published elsewhere (30, 31). A cardiovascular health and was followed on another day with measure-
summary version of those data is provided in Table 1, in which the ment of peak oxygen uptake (V̇O2 peak) on a bicycle ergometer (26).
participants were grouped as young (21–37 yr), middle aged (40 –56 The volume and composition of expired gases using breath-by-breath
yr), or older (60 – 87 yr). analysis, heart rate by 12-lead electrocardiogram, and blood pressure
Study protocol. Participants were randomized to either a 16-wk were continuously monitored throughout the tests (26). The posttrain-
endurance exercise or control program. A similar 5-day protocol was ing assessment was made within 3 days of the completion of the last
completed at baseline and again within 1 wk of completion of the training bout.
proportional content of each of the three MHC isoforms, but it was not Values for muscle MHC isoform mRNA and protein content
possible to determine whether the total MHC concentration varied at baseline are shown in Fig. 3. At the mRNA level, MHC I
with age or exercise training. abundance did not vary with age, whereas both MHC IIa and
Statistical analysis. Summary data are reported as means ⫾ SE. IIx transcripts declined by 14 and 10% per decade, respec-
Differences between men and women and between pre- and posttest-
tively. At the protein level, the relative content of MHC I
ing within groups were analyzed by using unpaired and paired t-tests,
respectively. Pearson correlation coefficients were used to measure increased with age by 3.5% per decade, whereas there was a
association among selected variables. Multiple regression analysis decline of 3% per decade for MHC IIa and 1% per decade for
was used to determine the contributions of leg muscle size and MHC MHC IIx. There were no differences between men and women
protein content to variance in knee extensor torque. Covariate adjust- in average MHC mRNA abundance or fractional protein con-
ment of knee extensor torque for muscle cross-sectional area was tent nor any significant differences in the changes in MHC
performed (34) so that the residual effect of age could be graphically expression with age.
represented. A P value ⬍ 0.05 was considered statistically significant The association between the fractional MHC protein content
for all tests. and peak knee extensor torque is shown in Fig. 4. Peak torque
was negatively related to the fraction of MHC I protein and
RESULTS positively related to the fraction of MHC IIa protein. After
covariate adjustment of peak torque values for differences in
Subject characteristics at baseline are shown in Table 1. The thigh muscle cross-sectional area, however, the fraction of
data are summarized into three age groups to simplify the MHC protein isoforms was no longer significantly correlated
presentation, but analysis of age effects was performed by with peak torque (data not shown). Multivariate regression
linear regression. As expected, there were significant differ- confirmed that MHC protein did not make a significant con-
ences between men and women for most variables. With the tribution to the explanation of the variance in knee extensor
exception of body weight and body mass index, there were torque once thigh muscle area was entered into the model.
significant age-related changes in body composition, reflecting
increasing adiposity and decreasing fat-free mass. The peak
aerobic capacity (V̇O2 peak) declined with age by 8% per decade
in both men and women.
There were no statistically significant changes in the phys-
iological characteristics of the control group between measure-
ments made at baseline and 16 wk later. In exercisers, there
were small but statistically significant (P ⬍ 0.001) reductions
in body weight (0.6 ⫾ 0.2 kg) and body mass index (0.2 ⫾ 0.1
kg/m2) but no change in dual-energy X-ray absorptiometry-
determined body fat or fat-free mass. V̇O2 peak during cycling
improved by 9.5% on average (P ⬍ 0.001). The magnitude of
posttraining changes in body mass, body mass index, and
maximal oxygen uptake in the exercise participants was not
statistically different between men and women and did not vary
with age of the participant. These physiological characteristic
data have been presented previously (30, 31).
Peak isokinetic knee extensor torque was on average 48%
higher in men than women (P ⬍ 0.001) and declined with age
at a rate of 13% per decade in men and 8% per decade in
women (Fig. 2A). Cross-sectional area of the midthigh muscle
was on average 32% higher in men than women (P ⬍ 0.001)
and declined 5% per decade in men and 4% per decade in
women (Fig. 2B). Peak torque was strongly associated with
thigh muscle area in men and women combined (r ⫽ 0.88, P ⬍
0.001). After covariate adjustment of peak torque for differ-
ences in muscle area, differences in knee extensor torque
between men and women were eliminated, but there was still a
significant decline in peak torque with age of 5% per decade
(Fig. 2C). In the control group, there was no change from
baseline to follow-up in muscle size or strength (data not
shown). The exercise group had an average 6% increase in
peak isokinetic knee extensor torque (75.7 ⫾ 4.8 N䡠m pre- Fig. 3. Age effect on expression of MHC mRNA and protein isoforms. The
training, 78.5 ⫾ 5.2 N䡠m posttraining; P ⬍ 0.01) after training, relative abundance of mRNA transcripts for each isoform (left) are shown in
but the change was similar in young (7%), middle-aged, (4%), arbitrary units (AU) after normalization for 28S rRNA. Protein expression of
and older (6%) groups. Leg muscle area did not change each isoform (right) is shown as the fraction of each isoform relative to the
total MHC in each gel lane. There were no statistical differences in expression
significantly in response to training in the exercise group between women and men, so a single regression line is shown in each panel
(overall average, 129 ⫾ 5 cm2 pretraining, 133 ⫾ 4 cm2 except for MHC I mRNA, where there was not a statistically significant
posttraining; P ⫽ 0.17). association with age.
The major findings from the present study were the age- and
exercise-related changes in expression of MHC isoform
mRNA and protein in human skeletal muscle. In a relatively
large group of healthy men and women, we found that both
MHC mRNA and protein expression of MHC II isoforms
declined with age, in support of our hypothesis. In response to
a 4-mo program of moderate-intensity endurance exercise
training, MHC I and IIa mRNA increased and IIx decreased,
shifting transcript expression from fast to slow isoforms in all
age groups. However, a training effect on MHC isoform
protein expression in this same direction was detected in
younger people, but there was no change in the composition of
MHC isoform protein in older people. This suggests that some
adaptations to exercise remain robust in older muscles, i.e.,
3. Balagopal P, Schimke JC, Ades P, Adey D, and Nair KS. Age effect on 21. Levine JA, Abboud L, Barry M, Reed JE, Sheedy PF, and Jensen MD.
transcript levels and synthesis rate of MHC and response to resistance Measuring leg muscle and fat mass in humans: comparison of CT and
exercise. Am J Physiol Endocrinol Metab 280: E203–E208, 2001. dual-energy X-ray absorptiometry. J Appl Physiol 88: 452– 456, 2000.
4. Blough ER, Rennie ER, Zhang F, and Reiser PJ. Enhanced electro- 22. Lexell J. Human aging, muscle mass, and fiber type composition. J
phoretic separation and resolution of myosin heavy chains in mammalian Gerontol A Biol Sci Med Sci 50A: 11–16, 1995.
and avian skeletal muscles. Anal Biochem 233: 31–35, 1996. 23. Lindle RS, Metter EJ, Lynch NA, Fleg JL, Fozard JL, Tobin J, Roy
5. Bottinelli R. Functional heterogeneity of mammalian single muscle fibres: TA, and Hurley BF. Age and gender comparisons of muscle strength in
do myosin isoforms tell the whole story? Pflügers Arch 443: 6 –17, 2001. 654 women and men aged 20 –93 yr. J Appl Physiol 83: 1581–1587, 1997.
6. Coggan AR, Spina RJ, King DS, Rogers MA, Brown M, Nemeth PM, 24. Marx JO, Kraemer WJ, Nindl BC, and Larsson L. Effects of aging on
and Holloszy JO. Histochemical and enzymatic comparison of the gas-
human skeletal muscle myosin heavy-chain mRNA content and protein
trocnemius muscle of young and elderly men and women. J Gerontol 47:
isoform expression. J Gerontol A Biol Sci Med Sci 57A: B232–B238,
71B–76B, 1992.
7. Coggan AR, Spina RJ, King DS, Rogers MA, Brown M, Nemeth PM, 2002.
and Holloszy JO. Skeletal muscle adaptations to endurance training in 60- 25. O’Neill DS, Zheng D, Andersen WK, Dohm GL, and Houmard JA.
to 70-yr-old men and women. J Appl Physiol 72: 1780 –1786, 1992. Effect of endurance exercise on myosin heavy chain gene regulation in
8. Frontera WR, Suh D, Krivickas LS, Hughes VA, Goldstein R, and human skeletal muscle. Am J Physiol Regul Integr Comp Physiol 276:
Roubenoff R. Skeletal muscle fiber quality in older men and women. Am J R414 –R419, 1999.
Physiol Cell Physiol 279: C611–C618, 2000. 26. Proctor DN and Beck KC. Delay time adjustments to minimize errors in
9. Han YS, Proctor DN, Geiger PC, and Sieck GC. Reserve capacity for breath-by-breath measurements of V̇O2 during exercise. J Appl Physiol 81:
ATP consumption during isometric contraction in human skeletal muscle 2495–2499, 1996.
fibers. J Appl Physiol 90: 657– 664, 2001. 27. Proctor DN, Sinning WE, Walro JM, Sieck GC, and Lemon PW.
10. Hasten DL, Pak-Loduca J, Obert KA, and Yarasheski KE. Resistance Oxidative capacity of human muscle fiber types: effects of age and training
exercise acutely increases MHC and mixed muscle protein synthesis rates status. J Appl Physiol 78: 2033–2038, 1995.
in 78 – 84 and 23–32 yr olds. Am J Physiol Endocrinol Metab 278: 28. Rooyackers OE, Balagopal P, and Nair KS. Measurement of synthesis
E620 –E626, 2000. rates of specific muscle proteins using needle biopsy samples. Muscle
11. Houmard JA, Weidner ML, Gavigan KE, Tyndall GL, Hickey MS, Nerve Suppl 5: S93–S96, 1997.
and Alshami A. Fiber type and citrate synthase activity in the human 29. Schiaffino S and Reggiani C. Molecular diversity of myofibrillar pro-
gastrocnemius and vastus lateralis with aging. J Appl Physiol 85: 1337– teins: gene regulation and functional significance. Physiol Rev 76: 371–
1341, 1998.
423, 1996.
12. Howald H, Hoppeler H, Claassen H, Mathieu O, and Straub R.
30. Short KR, Vittone JL, Bigelow ML, Procotor DN, and Nair KS. Age
Influences of endurance training on the ultrastructural composition of the
different muscle fiber types in humans. Pflügers Arch 403: 369 –376, 1985. and aerobic exercise training effects on whole body and muscle protein
13. Janssen I, Heymsfield SB, Wang Z, and Ross R. Skeletal muscle mass metabolism. Am J Physiol Endocrinol Metab 286: E92–E101, 2004.
and distribution in 468 men and women aged 18 – 88 yr. J Appl Physiol 89: 31. Short KR, Vittone JL, Bigelow ML, Proctor DN, Rizza RR, Coenen-
81– 88, 2000. Schimke JM, and Nair KS. Impact of aerobic training on age-related
14. Jubrias SA, Odderson IR, Esselman PC, and Conley KE. Decline in changes in insulin action and muscle oxidative capacity. Diabetes 52:
isokinetic force with age: muscle cross-sectional area and specific force. 1888 –1896, 2003.
Pflügers Arch 434: 246 –253, 1997. 32. Talmadge RJ and Roy RR. Electrophoretic separation of rat skeletal
15. Klitgaard H, Mantoni M, Schiaffino S, Ausoni S, Gorza L, Laurent- muscle myosin heavy-chain isoforms. J Appl Physiol 75: 2337–2340,
Winter C, Schnohr P, and Saltin B. Function, morphology and protein 1993.
expression of ageing skeletal muscle: a cross-sectional study of elderly 33. Taylor HL, Jacobs DR, Shucker B, Knudsen J, Leon AS, and De-
men with different training backgrounds. Acta Physiol Scand 140: 41–54, Backer G. A questionnaire for the assessment of leisure-time physical
1990. activities. J Chronic Dis 31: 741–755, 1978.
16. Klitgaard H, Zhou M, Schiaffino S, Betto R, Salviati G, and Saltin B. 34. Toth MJ, Goran MI, Ades PA, Howard DB, and Poehlman ET.
Ageing alters the myosin heavy chain composition of single fibers from Examination of data normalization procedures for expressing peak V̇O2
human skeletal muscle. Acta Physiol Scand 140: 55– 62, 1990. data. J Appl Physiol 75: 2288 –2292, 1993.
17. Krivickas LS, Suh D, Wilkins J, Hughes VA, Roubenoff R, and 35. Trappe S, Gallagher P, Harber M, Carrithers J, Fluckey J, and
Frontera WR. Age- and gender-related differences in maximal shortening
Trappe T. Single muscle fiber contractile properties in young and old men
velocity of skeletal muscle fibers. Am J Phys Med Rehabil 80: 447– 455,
and women. J Physiol 552: 47–58, 2003.
2001.
18. Larsson L, Grimby G, and Karlsson J. Muscle strength and speed of 36. Welle S, Bhatt K, and Thornton C. Polyadenylated RNA, actin mRNA,
movement in relation to age and muscle morphology. J Appl Physiol 46: and myosin heavy chain mRNA in young and old human skeletal muscle.
451– 456, 1979. Am J Physiol Endocrinol Metab 270: E224 –E229, 1996.
19. Larsson L and Karlsson J. Isometric and dynamic endurance as a 37. Welle S, Brooks AI, Delehanty JM, Needler N, and Thornton CA.
function of age and skeletal muscle characteristics. Acta Physiol Scand Gene expression profile of aging in human muscle. Physiol Genomics 14:
104: 129 –136, 1978. 149 –159, 2003.
20. Larsson L, Li X, and Frontera WR. Effects of aging on shortening 38. Williamson DL, Godard MP, Porter DA, Costill DL, and Trappe SW.
velocity and myosin isoform composition in single human skeletal muscle Progressive resistance training reduces myosin heavy chain coexpression
cells. Am J Physiol Cell Physiol 272: C638 –C649, 1997. in single muscle fibers from older men. J Appl Physiol 88: 627– 633, 2000.