Muscle metabolism during exercise in young

and older untrained and endurance-trained men

ANDREW
R. COGGAN,
AMIR M. ABDULJALIL,
SCOTT C. SWANSON,
MARGARET
S. EARLE,
JAMES W. FARRIS,
LISA A. MENDENHALL,
AND PIERRE-MARIE
ROBITAILLE
Exercise Physiology Laboratory, School of Health, Physical Education, and Recreation, and Departments
Radiology and Medical Biochemistry, The Ohio State University, Columbus, Ohio 43210
COGGAN,ANDREW R., AMIR M. ABDULJALIL, SCOTT C.
SWANSON,MARGARETS.EARLE,JAMES W. FARRIS,LISA A.
MENDENHALL,AND~IERRE-MARIEROBITAILLE.MUX~~
metabolism during exercise in young and older untrained and endurance-trained men. J. Appl. Physiol. 75(5): 2125-2133, 1993.To examine effects of aging and endurance training on human
muscle metabolism
during exercise, lP magnetic resonance
spectroscopy was used to study the metabolic response to exercise in young (21-33 yr) and older (58-68 yr) untrained
and
endurance-trained
men (n = 6/group).
Subjects performed
graded plantar flexion exercise with the right leg, with metabolic responses measured using a lP surface coil placed over
the lateral head of the gastrocnemius
muscle. Muscle biopsy
samples were also obtained for determination
of citrate synthase activity. Rate of increase in Pi-to-phosphocreatine
ratio
with increasing power output was greater (P < 0.01) in older
untrained
[0.058 k 0.022 (SD) W-l] and trained men (0.042 t
0.010 W-) than in young untrained
(0.038 t 0.017 W-l) and
trained men (0.024 k 0.010 W-l). Plantar flexor muscle crosssectional area and volume (determined using H magnetic resonance imaging) were ll-12%
(P < 0.05) and 16-18% (P < 0.01)
smaller, respectively, in older men. When corrected for this
difference in muscle mass, age-related differences in metabolic
response to exercise were reduced by -50% but remained significant (P < 0.05). Citrate synthase activity was -20% lower
(P < 0.001) in older untrained
and trained men than in corresponding young groups and was inversely related to Pi-phosphocreatine
slope (r = -0.63, P < 0.001). Age-related
reductions in exercise capacity were associated with an altered muscle metabolic response to exercise, which appeared to be due to
smaller muscle mass and lower muscle respiratory capacity of
older subjects. Endurance
training
was associated with 60100% higher muscle citrate synthase activity and improved metabolic responses in young and older men but apparently could
not prevent an age-related decrement in these variables or an
age-related decrease in muscle mass.
aging; master athletes; muscle atrophy;
netic resonance spectroscopy; magnetic

citrate synthase; magresonance imaging

MUSCULARENDURANCE
in rats( 11,15)
and humans (10). In rats, this increase in muscle fatigability with age appears to be due, at least in part, to an
altered metabolic response to exercise (2, 11, 13, 15, 25).
Thus the muscles of older rats undergo greater depletion
of ATP, phosphocreatine (PCr), and glycogen and accumulate more lactate during electrical stimulation of the
motor nerve than the muscles of young rats (11,15). SimiAGINGREDUCES

0161-7567/93

$2.00

of

larly, the rate of muscle glycogen utilization during

treadmill running is greater in older rats than in young
rats (2). These differences between young and older rats
in the metabolic response to exercise appear to be largely
the result of an age-related decrease in skeletal muscle
mitochondrial respiratory capacity (1, 2).
Little is known about the effects of aging in humans on
the metabolic response of skeletal muscle to exercise. We
have found, however, that mitochondrial marker enzyme
activities are 15-30s lower in the muscles of healthy but
sedentary older (60-70 yr) subjects than in those of
young (20-30 yr) subjects (9). Similarly, others have reported that the capacity of human muscle to utilize O,,
measured in vitro using either whole muscle homogenates (26, 34) or isolated mitochondria (34), is 25-50%
lower in older subjects. It therefore seemspossible that,
as in rats, an age-related reduction in human muscle respiratory capacity could alter the metabolic response to
exercise and thus contribute to the reduction in muscular
endurance observed in older subjects (10). In keeping
with this hypothesis, McCully et al. (24) reported that
the rate of muscle PCr resynthesis after exercise is significantly reduced in older subjects. Others, however, found
no effect of aging on human muscle metabolism during or
after exercise (33). Thus, whether aging alters muscle
metabolic responses during exercise in humans remains
uncertain.
Reduced skeletal muscle respiratory capacity and altered metabolic responses during exercise in older rats or
humans may be due to a decrease in habitual physical
activity rather than the aging process per se. In support
of this hypothesis, young and older rats that have been
trained using an identical exercise program have similar
muscle mitochondrial respiratory capacities and use similar amounts of muscle glycogen during exercise (1,2). In
older men and women, we found that endurance exercise
training increases muscle mitochondrial marker enzyme
activities by ~50% (8), and Meredith et al. (26) demonstrated an even larger training-induced increase in in vitro muscle respiratory capacity. It is not known, however, whether endurance training of older humans can
prevent or ameliorate possible age-related alterations in
muscle metabolism during exercise.
The purpose of the present study, therefore, was to
determine the effects of aging and endurance training on

Copyright 0 1993 the American Physiological Society

2125

2126

AGING

AND

HUMAN

MUSCLE

1. Training duration, volume, and pace in young

and older trained men

METABOLISM

DURING

EXERCISE

was approved by The Ohio State University Human Subjects Review Committee, and all subjects provided informed written consent.
Young
Older
All subjects were normotensive nonsmokers who were
n
Trained
Trained
free of detectable cardiovascular, metabolic, or musculoDuration
of training,
yr
6
12+6
12+5
skeletal disease. None was taking prescription medicaTotal training
volume,
h/wk
6
7+2
B&4
tions. Preliminary testing included a medical history,
Running
distance, km/wk
5
55+19
56+24
physical examination, and a 75-g oral glucose tolerance
Running
pace, km/h
5
1421
12+1*
test. The older subjects also underwent two graded
Cycling distance, km/wk
4
137+77
190+114
Cycling pace, km/h
4
33s
29+2*
treadmill tests to volitional fatigue while blood pressure
and a 1%lead electrocardiogram were monitored. The
Values are means -t SD. * Significantly
(P < 0.001) different
from
first test, using a Bruce protocol, was used to screen for
young trained
men.
cardiovascular abnormalities, whereas the second test,
human skeletal muscle metabolism during exercise. To using an individually adjusted walking or running protodo so, we used 31Pmagnetic resonance spectroscopy (31P- col
. (ZO), was used to determine maximal 0, uptake
MRS) to examine the metabolic response of the plantar
(VO 2,,,). The young subjects performed only an individuflexor muscles during exercise in young and older un- ally adjusted running test to fatigue to determine VO, m8x.
trained and endurance-trained men. Muscle biopsy sam- In the young and older subjects, 0, uptake (vo2)
was
ples were also obtained from the lateral gastrocnemius measured every 30 s during exercise by use of an automuscle to examine the possible relationship between mated open-circuit system that incorporated a dry gas
age-related changes in mitochondrial respiratory capac- meter (Rayfield RAM-9200), mixing chamber, and elecity and metabolic responses during exercise.
tronic 0, and CO, analyzers (Applied Electrochemistry
S3-A and Beckman LB-Z, respectively). The mean of the
two highest consecutive 30-s values for VO, was defined
METHODS
as V0, max. To ensure that 7j02max
had &indeed been
Subjects and preliminary testing. Subjects were rereached, at least two of the following criteria had to be
cruited through advertisements placed with campus and
met: a plateau in VO, despite an increase in treadmill
community publications and media. Endurance-trained
speed and/or grade, a respiratory exchange ratio >l.lO,
subjects were also recruited using the published results of
or a heart rate within 10 beats/min of age-predicted
local running and/or triathlon competitions. After premaximal heart rate.
liminary screening of health status (seebelow), six young
Body fat and fat-free mass were estimated from skin[25 t 2 (SD) yr] untrained men, six older (63 t 3 yr)
untrained men, six young (27 t 4 yr) trained men, and six fold measurements (19).
Magnetic resonance experiments. Subjects first perolder (62 t 2 yr) trained men were enrolled in the study.
formed
the entire plantar flexion exercise protocol (see
Most of the young men were university students,
below)
at
the Exercise Physiology Laboratory. This iniwhereas most of the older men were employed in or retired from professional careers. All untrained subjects tial study served to minimize possible learning effects
and also enabled quantification of the subjects heart
performed normal daily activies (e.g., shopping, driving,
walking short distances), and some participated in recre- rate, blood pressure, and VO, during plantar flexion exational sports or games (e.g., squash, tennis). None, how- ercise. These latter measurements could not be performed in conjunction with the subsequent 31P-MRS exever, held a job requiring strenuous physical labor or properiment, because the powerful magnetic field precluded
longed walking, and none participated in endurance
the use of ferrous metal equipment.
sports such as distance running or cycling. On the other
Approximately 1 wk later, subjects were studied at the
hand, both the young and older trained men had been
training for and/or competing in endurance sports al- Magnetic Resonance Facility. Experiments were conmost continuously for Z-20 yr, with both groups consist- ducted 26 h after the subjects last meal and, for the
ing of two runners, three duathletes/triathletes,
and one trained men, 224 h after their last training session. To
cyclist. Whereas these men were competitive in local and correct for the possible effects of age-related muscle atroregional endurance competitions, most had not achieved phy on the metabolic response to exercise, the cross-secsuccesson a national level and would not be considered tional area and total volume of the plantar flexors were
first determined using H magnetic resonance imaging.
to be elite athletes. One of the older athletes, however,
had won the World Triathlon Championship (1.5 km Imaging was performed with a GE Signa 1.5-T instruswimming, 40 km cycling, 10 km running) in his age ment and a quadrature detection body coil with use of a
group -1 yr before being studied, whereas another older standard multislice multiecho sequence. Parameters
athlete had finished fourth in the same event and placed were as follows: TE (echo time) = 20 ms, TR (repetition
third in the Hawaii Ironman Triathlon (4 km swimming, time) = 600 ms, 128 X 256 matrix, 1.6 X 0.8 X lo-mm
180 km cycling, 42 km running) after this study. Total
resolution, 2 NEX (no. of excitations). The subject was
years of training and average weekly training volume did placed in the magnet in the supine posture with the right
not differ significantly between the young trained men leg positioned in the center of the coil. The distance beand the older trained men (Table 1). Running and/or
tween the femoral epicondyles and the calcaneus was
cycling pace, however, was greater in the young athletes first determined by acquiring several sagittal images.
than in the older athletes (Table 1). The study protocol
Without moving the subject, 35-40 sequential transaxial
TABLE

AGING

AND

HUMAN

MUSCLE

METABOLISM

images (10 mm thick, 1 mm interspacing)

were then obtained covering this linear distance (total scan time -10 min).
The metabolic response to exercise was then studied
using 31P-MRS. A 4cm-diam
31P-tuned surface coil was
positioned directly under the lateral head of the subjects
right gastrocnemius
muscle and secured to the leg with
adhesive tape. The subject was then repositioned within
the magnet, and the coil was tuned to 31P. Once the magnetic field had been shimmed, 31P spectra were acquired
at 25.86 MHz by use of a l-pulse sequence, 20 averages/
spectrum, a 2-kHz sweep width, and a l-kilobyte
block
size. To correct for saturation
effects, both fully relaxed
(10-s TR) and partially saturated (2-s TR) 31P spectra
were acquired before exercise. The subject then began
performing repeated plantar flexions at a rate of 3O/min
(paced by a metronome)
using a custom-built
exercise
device. This device consisted of a wooden foot pedal and
a cable-and-pulley
system arranged so that plantar flexion from 0 (vertical) to 24 raised a weight 10 cm. This
weight also served to return the pedal to O after each
movement. The axis of rotation of the foot pedal was
positioned at the ankle joint, thereby minimizing movement of the leg within the magnet. The foot was secured
to the pedal with a strap, and a wide strap was placed
tightly over the waist and upper thighs to minimize horizontal or vertical movement.
This second strap also
served to keep the knee fully extended, thereby maximizing involvement
of the gastrocnemius
muscle (31). The
initial power output was 0.75 W, which was maintained
for 3 min. Thereafter, the power output was increased by
0.75 W every 3 min until, despite strong verbal encouragement, the subject was I) unable to maintain the cadence
or 2) unable to move the foot pedal over the required
range of motion. Total exercise duration ranged from 12
to 30 min (4-10 stages). Partially saturated 31P spectra
were acquired during the last -60 s of each exercise
stage with use of a 2-s TR. Representative
spectra acquired from a young trained subject at rest and during
exercise are shown in Fig. 1.
Muscle biopsy. On a subsequent day, diet and exercise
controls were again imposed as described above, and a
muscle biopsy sample was obtained from the lateral head
of the right gastrocnemius
muscle by use of a 5mm
Bergstrom
needle (Stille-Werner,
Ronkonkoma,
NY).
The- muscle specimen was frozen in liquid N, and then
stored at -8OOC until subsequently
analyzed for citrate
synthase activity (4) and protein content (2l), as previously described (5-9). Briefly, a 5- to lo-mg portion of
each sample was homogenized in 50 vol of 20 mM sodium
phosphate buffer (pH 7.4) containing 0.5 mM EDTA, 5
mM ,8-mercaptoethanol,
0.02% citrate-free
bovine serum
albumin (BSA), and 50% glycerol. An aliquot of this homogenate was then diluted further (20-fold) with 20 mM
imidazole buffer (pH 7.0) containing 0.02% BSA. Citrate
synthase activity was measured by adding 2 ~1 of the
diluted homogenate to 50 ~1 of assay reagent [50 mM
tris(hydroxymethyl)aminomethane
buffer, pH 8.1, 400
PM acetyl-CoA,
500 PM oxaloacetate, 0.25% BSA] and
incubating at room temperature
(24 t O.IC) for 1 h. The
reaction was stopped, and excess oxaloacetate was destroyed by adding 5 ~1 of 0.5 N NaOH and boiling for 5
min. The citrate formed was then measured fluorometri-

DURING

2127

EXERCISE

ATP

CP

0 min

__h__ihc?s*_ij_l
9 min

12 min

@--++fi

-A-

15 min

27 min

10

-6

-14

-22

-30

PPm
FIG. 1. lP magnetic
gastrocnemius
muscle
min) and during plantar
phate.

resonance
spectra acquired
of a young endurance-trained
flexion exercise (3-27 min).

from right1 lateral

man at rest (0
CP, creatine phos-

tally using citrate lyase and malate dehydrogenase, as

previously described in detail (4, 5).
Protein concentration of the homogenate was measured calorimetrically
(21) using BSA as the protein
standard.
Data analyses. H magnetic resonance imaging scans
were printed on film and digitized using a video analysis
system. This system consisted of a light box and a video
camera interfaced with a laboratory computer operating
commercial software (JAVA, Jandel Scientific, Corta
Madera, CA). For each image, the cross-sectional areas
of three separate regions were determined by manually
tracing the desired muscle area with a mouse-driven cursor, carefully excluding all visible noncontractile tissue
(i.e., blood vessels, nerves, tendons, connective tissue,
fat). The three areas quantified were I) the lateral gastrocnemius, 2) the medial gastrocnemius, and 3) the remaining plantar flexor muscles of the posterior compartment. The latter area included the soleus, tibialis posterior, flexor digitorum long-us, flexor hallucis longus, and
(where present) plantaris but excluded the popliteus and
peroneus longus and brevis. Each cross-sectional area

2128

AGING

HUMAN

MUSCLE
m

0.75

1 0.50
0

-.

Y=0.099+0.021

R=0.96

AND

*x

0.25

m
I

0.00

0.0

1.5

3.0
Power

4.5

METABOLISM

DURING

EXERCISE

SULTS), no significant interaction effects were observed,

indicating that the effects of aging were generally similar
in the untrained and the trained men. The P values reported here therefore refer to these significant main effects. For comparison purposes, however, data for the
young and older untrained and trained men are presented separately as means t SD.

6.0

(W)

RESULTS

2. Relationship
of Pi/PCr
to power output
for subject whose
data are shown in Fig. 1. Initial
linear rate of increase in P;/PCr
with
increasing
power output
(i.e., Pi/PCr
slope) was calculated
by regression analysis (solid line) and was used as an index of muscle metabolic
stress. These analyses were performed
with power output expressed
in
absolute
terms (W) and relative
to muscle cross-sectional
area (W/
cm2), muscle volume
(W/l),
and peak plantar
flexor
power (%peak
power).
Because P;/PCr is unitless,
corresponding
slopes have units of
W-l, cm2/W, l/W, and %peak power-,
respectively.
FIG.

Physical characteristics. Physical characteristics of the

subjects are shown in Table 2. The older men were significantly shorter and had a lower fat-free mass than the
younger men, regardless of training status. Percent body
fat, on the other hand, differed with age and with training
status, being higher in the older than in the young men
but lower in the trained than in the untrained subjects.
As a result of this lower body fat content, the trained
was measured three to eight times, with an average coeffisubjects weighed significantly less than the untrained
cient of variation of ~2%. The volume of each imaged subjects, regardless of age.
muscle segment was calculated by multiplying the crossvo 2max was -45% lower in the older untrained men
sectional area by the image thickness (i.e., 10 mm). The than in the young untrained men, whether expressed in
volume of muscle between the imaged segments was cal- liters per minute or in milliliters per minute per kiloculated assuming that each such nonimaged segment had gram. Differences in VO 2maxbetween the young and older
the shape of a truncated cone (27). Total plantar flexor
trained men were somewhat smaller (32 and 22% for l/
muscle volume (in liters) was calculated by summation of min and ml min- kg-, respectively), but this apparent
the volume of the imaged and nonimaged muscle seg- interaction between the effects of aging and the effects of
ments.
endurance training on VO 2maxwas not statistically signifiA ~-HZ line broadening was applied to the lP-MRS
cant (P = 0.14).
signal before Fourier transformation. Peak areas were
Maximal heart rate was ~40 beats/min lower (P <
integrated by a blinded technician using commercial soft0.001) in the older untrained men than in the young unware (GE) to give relative concentrations of ATP (cu,,& trained men, consistent with their roughly four-decade
and y peaks), PCr, and Pi. Intracellular pH was calcu- difference in age. Despite a similar age differential, howlated by measuring the PCr-P; chemical shift difference
ever, maximal heart rate of the older trained men was
(30). The rate of increase in the Pi/PCr ratio with in- only 20 beats/min less (P < 0.01) than that of the young
creasing power output was used as an indicator of muscle trained men and, consequently, was significantly higher
metabolic stress (3, 22) (Fig. 2). Although Pi/ATP and (P < 0.05) than that in the older untrained men. AlPCr/ATP are more directly relevant from a thermodythough based on a relatively small number of subjects,
namic perspective, differences between groups in these these cross-sectional observations are consistent with
ratios did not achieve statistical significance when ana- recent longitudinal data, suggesting that endurance
lyzed as a function of power ouput. This is most likely
training may slow the fall in maximal heart rate with
due to an inadequate signal-to-noise ratio in the 31P age (28).
spectra. Thus, whereas P;/PCr may be less thermodyMuscle citrate synthase activity. Citrate synthase activnamically appropriate, we chose it as a more sensitive ity (in mol h-l kg muscle protein-) averaged 3.68 t
measure because both the numerator and denominator
0.52 in the young untrained men, 3.20 t 0.62 in the older
of this ratio reflect changes in bioenergetic status. untrained men, 7.50 t 0.99 in the young trained men, and
Changes in Pi/PCr under equilibrium conditions have 5.62 t 0.44 in the older trained men. The effect of age and
been theoretically linked to changes in free ADP, pro- the effect of training were statistically significant (P <
vided that ATP and H+ are constant (3). It is uncertain,
0.05 and P < 0.001, respectively).
however, whether equilibrium conditions were achieved
Plantar flexor muscle cross-sectional area and volume.
in the present studies; furthermore, pH decreased during
Maximal plantar flexor cross-sectional areas in the older
exercise, albeit only near fatigue (see RESULTS). Neveruntrained men (49.5 t 6.0 cm2) and older trained men
theless, although changes in Pi/PCr cannot be easily in- (48.3 t 6.9 cm2) were ll-12%
smaller (P < 0.05) than
terpreted in terms of thermodynamically significant reac- those in the young untrained men (56.5 t 8.5 cm2) and
tions, this ratio remains an accurate gross measure of young trained men (54.5 t 4.5 cm2). Age-related differbioenergetic status, and as such we have used this ratio
ences in muscle cross-sectional area were greater near
while recognizing its shortcomings.
the calcaneus and especially the popliteal space. ConseStatistical analyses. Data were analyzed using two-way
quently, total plantar flexor muscle volume in the older
(age X training) analyses of variance. Significance was untrained men (0.89 t 0.13 liter) was 18% lower (P <
defined as P 5 0.05. Significant main effects for age and 0.01) than that in the young untrained men (1.09 t 0.20
training were obtained for a number of variables. Howliters). These values are almost identical to those obever, with the exception of maximal heart rate (see RE- tained previously by Rice et al. (27) using computed tol

AGING

TABLE

AND

HUMAN

MUSCLE

METABOLISM

2129

EXERCISE

2. Subject characteristics
Untrained

Trained

Young

Height, cm
Weight,
kg
Body fat, %
Fat-free
mass, kg
VO

DURING

Age
Effect

Older

Young

Older

177-t9
77.0* 14.0
15.1t6.9
64.4k9.3

170s
74.4k9.1
22.2k4.0
57.6k5.5

179k5
70.4k5.9
7.5-tl.7
65.1k5.9

16825
61.3k4.7
11.5-tl.4
54.2k4.0

P < 0.01

3.62kO.49
47.6k4.8
199+9

2.14-t0.26
29.2k5.7
155t9

4.6620.38
66.324.4
189&6

3.19kO.33
52.Ok2.7
169&6*

P < 0.001
P < 0.001

P < 0.05
P < 0.001

2 max

l/min
ml min-
kg-
HR max, beatslmin
l

Val ues are means -t SD for 6 subjs/group.

2-way (age X training)
analysis of variance.

ii0 2 max, maximal

0, uptake;
HR,,,,
m aximal heart rate.
* Significantly
higher (P < 0.01) than in older untrained

mography. A similar age-related difference in muscle volume was observed in the trained subjects, with plantar
flexor volume averaging 0.86 t 0.13 liter in the older
trained men compared with 1.02 t 0.10 liters in the
young trained men. Differences in muscle volume between young and older subjects were only slightly reduced (to 12-13%, P < 0.05) when the data were expressed relative to height or tibia1 length (data not
shown), indicating that the reduced muscle volume of the
older men was not simply due to their smaller stature.
Responsesduringplantar flexion exercise. VO, was measured at rest and during plantar flexion exercise during
the initial experiment performed at the Exercise Physiology Laboratory. Resting VO, was -20% lower (P <
0.001) in the older untrained and older trained men than
in the corresponding young groups (Table 3). This difference is probably partially due to the lower fat-free mass
of the older subjects. At any given power outp.ut during
exercise, however, the increase in whole body VO, above
resting was not significantly different among the four
groups (Table 3).
The subjects were subsequently studied at the Magnetic Resonance Facility. At rest, Pi/PCr averaged 0.10 t
0.02 in the young untrained men, 0.12 t 0.02 in the older
untrained men, 0.10 t 0.04 in the young trained men, and
0.15 t 0.03 in the older trained men. Although Pi/PCr
tended to be higher in the older subjects, especially in the
trained men, this difference did not achieve statistical
significance (P = 0.064). During exercise, however, the
P;/PCr ratio rose more rapidly with increasing power
3. Resting %2 and increase in L%I~above resting
during one-leggedplantar flexion exercise
TABLE

Ai02

Stage 1

Stage

2701~25
205+21*

44+15
30+18

79+17
63+24

113221
96+35

147+27
128k46

277219
223+20*

34+20
34+18

75222
71+22

115k25
108k30

176~~51
143+37

Resting

Untrained
Young
Older
Trained
Young
Older

VO,

Stage

Stage

Values-are
m$ans -t- SD for 6 subjs/group
in ml/min.
Data for increase in VO, (AVO,) are presented
for the first 4 exercise stages, which
were completed
by all subjs. * Significantly
lower (P < 0.001) in older
untrained
and trained
men than in young untrained
and trained
men,
respectively.

Training
Effect.

P < 0.05
P < 0.001

P < 0.001
P < 0.001

I < 0.001

P values

refer

to significant

main effects

bY

men.

output in the older untrained men than in the young

untrained men. A similar difference existed between the
young and older trained men, although Pi/PCr increased
more slowly in athletes than in nonathletes regardless of
age. Thus the mean Pi/PCr slope during exercise differed
significantly with age (P < 0.01) and with training status
(P < 0.05; Fig. 3A). When power output was expressed
relative to muscle cross-sectional area or muscle volume
to correct for the reduced muscle mass of the older men,
differences between the young and older subjects in the
metabolic response to exercise were reduced by -50%
but remained significant (P < 0.05; Fig. 3, B and C). Similar results were obtained when power output was expressed relative to peak plantar flexor power (Fig. 30).
Resting muscle pH did not differ with age or with
training status, averaging 7.07 t 0.06, 7.06 t 0.04, 7.05 t
0.03, and 7.03 t 0.03 in the four groups. Muscle pH did
not change during the initial stages of exercise but then
decreased to -6.9 at fatigue (Fig. 4). Although muscle
pH tended to decrease earlier but to a lesser extent in the
older subjects, these differences were not statistically significant because of large intersubject variability and/or
the small number of subjects in each group.
Peak plantar flexor power averaged 4.6 t 1.0 W in the
young untrained men, 3.9 t 0.7 W in the older untrained
men, 6.1 t 0.8 W in the young trained men, and 4.6 t 0.9
W in the older trained men and was inversely correlated
(r = -0.70, P < 0.001) to the Pi/PCr slope (expressed in
W-l). These results are in keeping with prior observations indicating that maximal exercise capacity is inversely related to muscle metabolic stress during submaximal exercise, as assessedusing 31P-MRS (22).
Relationship between metabolic responsesand skeletal
muscle characteristics. The Pi/PCr slope (in W-) was inversely related to muscle cross-sectional area (r = -0.48,
P < 0.05), muscle volume (r = -0.51, P < 0.05), and
muscle citrate synthase activity (r = -0.55, P < 0.01; Fig.
5A). When power output was expressed relative to muscle cross-sectional area or volume, the latter correlation
coefficient increased to r = -0.63 (P < 0.001; Fig. 5, B and
C). Again, similar results were obtained when power output was expressed relative to peak plantar flexor power
(Fig. 50).
DISCUSSION

Numerous studies have demonstrated that the reduction in exercise capacity that accompanies aging is asso-

2130

AGING

AND

HUMAN

MUSCLE

METABOLISM

DURING

EXERCISE

A
Age:

0.060

raining:

PcO.01

Age:
Training:

PcO.05

PcO.05
PcO.01

Young
Untrained

Old
Untrained

Young
Trained

Old
Trained

Young
Untrained

Old
Untrained

Young
Trained

Old
Trained

0.060
Age:
Training:

3
2

PcO.05
PcO.05

0.040

it
0

PcO.05
PcO.01

g 0.001
0

v)

Young
Untrained

Old
Untrained

Young
Trained

Old
Trained

FIG. X Mean
Pi/PCr
slopes during exercise. Significant
main effects
analysis of variance
regardless
of whether
power output
was expressed
cross-sectional
area (B), muscle volume
(C), or peak power (D).

ciated with a decrease in VO 2max resulting from a decline

in central cardiac function (cf. 30). Studies of rats, however, suggest that decreases in Vo2max and alterations in
skeletal muscle metabolism contribute to the deterioration in exercise capacity with aging (2,11,13,15,25).
The
purpose of the present study was to determine whether
aging also affects muscle metabolism during exercise in
humans and, if so, to ascertain whether this effect could
be prevented or ameliorated by endurance exercise training. The exercise model that we used, i.e., one-legged
plantar flexion, was chosen in part because it facilitated
noninvasive assessment of muscle metabolism using 31PMRS. This model was also chosen, however, because the
metabolic response to exercise with a small muscle mass
would be less likely to be influenced by the age-related
decline in maximal cardiac output. Indeed, Martin et al.
7.20

Training:

g 0.002
Q
sw

z
6 0.020
CL
2
0.000

Age:

:
Q
Y

Power

(W)

FIG. 4. Intramuscular
pH during
plantar
flexion
exercise in the 4
subject groups: l , young untrained
men; 0, older untrained
men; H,
young trained
men; 0, older trained
men. Standard
deviations
(typically -0.05-0.10
pH units) have been omitted
for clarity.

6
s 0.000
L--

Young
Untrained

for age and training

in absolute
terms

Old
Untrained

Young
Trained

Old
Trained

status were indicated

by
(A) or relati ve to muscle

(23) demonstrated
that maximal calf muscle blood flow is
unaffected by aging in healthy men.
Consistent with previous studies of aged rats (2,11,13,
l&25), we found that aging apparently altered the metabolic response to exercise in humans. Thus, although at
rest Pi/PCr did not differ significantly
between the young
and older men, during exercise Pi/PCr increased more
rapidly with increments in power output in the older subjects. Muscle pH also tended to decrease at a lower power
output in the older subjects, although this latter difference was not statistically
significant.
The physiological
importance
of this greater disturbance
to energetic homeostasis during exercise is demonstrated
by the significant correlation between the Pi/PCr slope and the exercise capacity of the plantar flexor muscles, as measured
by peak power output.
Changes in Pi and PCr concentrations
during exercise
reflect the balance between the rate of ATP hydrolysis
and the rate of ATP resynthesis.
Thus the altered metabolic response of the older subjects could theoretically
be
due to a faster rate of ATP utilization,
a slower rate of
ATP production,
or a combination
of these factors. It is
unlikely that the absolute rate of ATP hydrolysis
was
higher in the older men, because exercise efficiency (as
indicated by the relationship
between VO, and power
output during plantar flexion exercise) did not differ
with age. However, at any given absolute power output,
the rate of ATP hydrolysis per gram of muscle was probably higher in the older men because of their smaller muscle mass. Thus a greater demand for ATP per unit of
muscle tissue probably contributed
to the altered metabolic response of the older subjects during exercise. In

AGING
0.080
-&

0 . 060

*.

AND

H UMAN
-r
s
$
-0
g

I+-0.55
P<O.Ol

4.000
3.000
2.000

MUSCLE

METABOLISM

DURING

0
r
- P.
O. 0
, 0 . . *. . .
*.*
0
0. * .

EXERCISE

2131

R--0.63
P<O.OOl

a
0.000

3
L 1.000
0
a.
;o.oool

10

0 *

10

h
L

C
-2
1

0.060

g 0.003

W-0.63
P<O.OOl

0.080
0

D-

z
Y

0.002

0.0

8
* . *.

0
*. * .

g
-

V
g 0.001

R--0.55
PKO.01

0
0

. *.

n n

0 * . *. * *
0

0 0

0.020

* . * 0.

*.*.

n
n

-. * . . .

FIG. 5. Relationship
between
citrate
synthase
activity
of lateral
gastrocnemius muscle
and Pi/PCr
slope during
plantar
flexion
exercise
in young
untrained
men (o), older untrained
men
(0), young trained
men (m), and older
trained
men (0). A significant
inverse
relationship
was
observed
whether
power output
was expressed
in absolute
terms (A) or relative
to muscle crosssectional
area (B), muscle volume (C), or
peak power (U).

3;

cL 0.000 I
2
Citrate

4
synthase

6
activity

8
(mol/h/kg)

IO

6
a.

0.000

2
Citrate

4
synthase

keeping with this hypothesis, differences in the Pi/PCr

slope between the young and older subjects were significantly reduced when corrected for differences in plantar
flexor muscle cross-sectional area or volume. An age-related reduction in muscle mass has long been recognized
as an important factor in the decrease in muscular
strength that occurs with aging (cf. 16). Little attention
has been directed, though, toward the effect of this decline in muscle mass on metabolic responses and performance during dynamic exercise requiring submaximal
force production. The present results suggest, however,
that a decrease in muscle mass with aging also plays a
very important role in the reduced exercise capacity observed under these conditions.
Although expressing power ouput relative to muscle
cross-sectional area or volume reduced the effect of age
on the Pi/PCr slope, significant differences remained between the young and older subjects. This is unlikely to
have been due to a greater amount of noncontractile (i.e.,
fat or connective) tissue in the muscles of the older subjects, because maximal voluntary torque per unit of muscle cross-sectional area or volume did not differ with age
(12). Histochemical analysis of selected biopsy samples
also revealed no differences in noncontractile tissue between the young and older men (S. E. Alway and A. R.
Coggan, unpublished observations). Factors in addition
to a reduction in contractile tissue must therefore have
contributed to the altered metabolic response and reduced exercise capacity of the older men. One likely factor is muscle mitochondrial respiratory capacity. The
mitochondrial content of muscle is an important determinant of the metabolic response to exercise (6; cf. 18), and
several (9, 26, 34) [but not all (14)] prior investigations
have found that muscle respiratory capacity is reduced in
older subjects. This was also true in the present subjects,
inasmuch as citrate synthase activity was 13 and 25%
lower in the older untrained and older trained men than
in the corresponding young groups. Studies of rats have
demonstrated that a decline in muscle respiratory capac-

6
activity

10

(mol/h/kg)

ity with aging plays a major role in the altered muscle

metabolism and impaired exercise performance of older
animals (1, 2). This also appeared to be true in the present subjects, inasmuch as both the Pi/PCr slope and peak
power output were significantly correlated with muscle
citrate synthase activity (r = -0.63, P < 0.001 and r =
0.66, P < 0.001, respectively). Indeed, these correlations
were almost perfect (r = -0.98, P < 0.02 and r = 0.99, P <
0.001, respectively) when variability was reduced by using group mean data instead of individual data in the
calculations.
Alterations in muscle mass and muscle respiratory capacity (and therefore in exercise metabolism) with advancing age could be due to a reduction in habitual physical activity rather than aging per se. Therefore, in addition to studying young and older sedentary individuals,
we also studied young and older men who were training
vigorously for endurance competitions. As expected,
muscle respiratory capacity was roughly twofold higher
and the Pi/PCr slope was substantially lower in the
young trained men than in young untrained men. Peak
plantar flexor power was also higher in the young trained
men. Muscle respiratory capacity and peak plantar
flexor power were also higher and the Pi/PCr slope was
also lower in the older trained men than in the older
untrained men. The higher muscle respiratory capacity
in the older trained men is consistent with prior studies
demonstrating that older subjects can adapt to endurance exercise training with an increase in muscle respiratory capacity, provided that the training stimulus is adequate (8, 26). Furthermore the present results suggest
that this increase in muscle respiratory capacity reduces
muscle metabolic stress during submaximal exercise. Although expected on the basis of studies of young individuals, the latter adapation to endurance training has not
been previously demonstrated in older humans. The
benefits of endurance training are emphasized by the fact
that, in absolute terms, the Pi/PCr slope and peak plantar flexor power of the older trained men were similar to

2132

AGING

AND

HUMAN

MUSCLE

METABOLISM

those of untrained men -40 yr younger, even though

muscle size was smaller in the older trained men. Thus
the training-induced increase in muscle respiratory capacity was able to partially compensate for the reduction
in muscle mass with aging.
Despite this compensatory effect of endurance training, the Pi/PCr slope was higher and peak plantar flexor
power was lower in the older athletes than in the young
athletes. In part, this appears to have been due to the fact
that plantar flexor muscle cross-sectional area and volume were 11 and 16% smaller in the older trained men
than in the young trained men, similar to the 12 and 18%
differences observed between the young and older untrained men. Interestingly, using anthropometry, Grassi
et al. (17) found that leg muscle (plus bone) volume was
reduced by -10%
in older (60- to 69-yr-old) endurance
athletes compared with younger (17- to 26-yr-old) endurance athletes and that this decline paralleled that of nonathletes. The present results and those of Grassi et al.
therefore suggest that the decrease in muscle mass with
aging is due to factors other than or in addition to simple
inactivity and that endurance exercise training does not
entirely prevent this age-associated loss of muscle tissue.
Even when power output was expressed relative to
muscle cross-sectional area or volume, however, the Pi/
PCr slope still differed between the young and older
trained men. This appeared to be due to the 25% lower
muscle citrate synthase activity of the older athletes. The
latter was true even though total duration of training and
training volume were similar in the two groups of athletes. These findings seemingly conflict with our previous report that mitochondrial marker enzyme activities are actually higher in older athletes than in young
athletes with whom they are matched on the basis of
absolute training volume (7). However, in our prior
study, the young and older athletes were also matched on
the basis of absolute training intensity and absolute performance ability, whereas the young athletes of the present study trained and competed at higher absolute intensities than the older athletes. In this regard, the present
group of young trained men more closely resembles the
group of competitive young runners we studied previously who had muscle mitochondrial enzyme activities
14-23% higher than those of the older runners (7). Similarly, Saltin (29) reported that muscle citrate synthase
and ,&hydroxyacyl-CoA
dehydrogenase activies are
-15% higher in elite young orienteers than in elite older
orienteers. Thus a small but functionally significant reduction in muscle respiratory capacity may be an inevitable consequence of aging (34). Alternatively, it is possible
that the decrease in maximal cardiac output and therefore in VO 2maxwith age eventually limits absolute training intensity and therefore indirectly accounts for this
small reduction in muscle respiratory capacity. The latter possibility is consistent with animal studies demonstrating that mitochondrial respiratory capacity adapts
equally in young and older muscles exposed to the same
absolute training stimulus (1).
To summarize, in the present study we used 31P-MRS
to study muscle metabolic responses during exercise in
young and older untrained and endurance-trained men.
We found that, compared with young untrained men,

DURING

EXERCISE

older untrained men demonstrated greater muscle metabolic stress during exercise. This was apparently due to
both their smaller muscle mass and their lower muscle
respiratory capacity. Older trained men had higher muscle respiratory capacities that partially compensated for
their reduced muscle mass, such that their metabolic response and exercise performance were equal to or even
slightly better than those of untrained men -40 yr
younger. However, muscle mass, muscle respiratory capacity, metabolic responses, and exercise performance
still differed between young and older trained men, suggesting that endurance training cannot completely prevent age-related changes in these variables.
The authors
thank Drs. Steve Alway, Tim Kirby,
Rick Schaal, and
Richard
Strauss for assistance
with various
aspects of the study.
This study was supported
in part by a University
Seed Grant from
the Ohio State University
Research
Foundation.
Address
for reprint
requests:
A. R. Coggan,
Shriners
Burns Institute, Metabolism
Unit, 815 Market
St., Galveston,
TX 77550.
Received

19 February

1993; accepted

in final

form

22 June

1993.

REFERENCES
1. CARTEE, G. D., AND R. I? FARRAR. Muscle respiratory
capacity and
vo 2 max in identically
trained
young and old rats. J. Appl. Physiol.
63: 257-261,
1987.
2. CARTEE, G. D., AND R. P. FARRAR. Exercise training
induces glycogen sparing during exercise by old rats. J. Appl. Physiol. 64: 259265, 1988.
3. CHANCE, B., J. S. LEIGH, J. KENT, AND K. MCCULLY.
Metabolic
control
principles
and lP NMR.
Federation
Proc. 45: 2915-2920,
1986.
4. CHI, M. M.-Y., C. S. HINTZ, E. F. COYLE, W. H. MARTIN,
J. L. IVY,
P. M. NEMETH,
J. 0. HOLLOSZY,
AND 0. H. LOWRY. Effects
of
detraining
on enzymes
of energy metabolism
in individual
human
muscle fibers. Am. J. Physiol.
244 (Cell Physiol.
13): C276-C287,
1983.
5. COGGAN, A. R., R. J. SPINA, W. M. KOHRT, AND J. 0. HOLLOSZY.
Effect of prolonged
exercise on muscle citrate concentration
before
and after endurance
training
in men. Am. J. Physiol.
264 (Endocrinol. Metab. 27): E215-E220,
1993.
6. COGGAN, A. R., R. J. SPINA, W. M. KOHRT, J. P. KIRWAN,
D. M.
BIER, AND J. 0. HOLLOSZY.
Plasma glucose kinetics during exercise
in subjects with high and low lactate thresholds.
J. Appl. Physiol.
73: 1873-1880,
1992.
7. COGGAN, A. R., R. J. SPINA, M. A. ROGERS, D. S. KING, M. BROWN,
P. M. NEMETH,
AND J. 0. HOLLOSZY.
Characteristics
of skeletal
muscle in master athletes. J. Appl. Physiol. 68: 1896-1901,
1990.
8. COGGAN, A. R., R. J. SPINA, M. A. ROGERS, D. S. KING, M. BROWN,
P. M. NEMETH,
AND J. 0. HOLLOSZY.
Skeletal
muscle adaptations
to endurance
training
in 60-70 year old men and women. J. Appl.
Physiol. 72: 1780-1786,
1992.
9. COGGAN, A. R., R. J. SPINA, M. A. ROGERS, D. S. KING, M. BROWN,
P. M. NEMETH,
AND J. 0. HOLLOSZY.
Histochemical
and enzymatic
comparison
of the gastrocnemius
muscle of young and elderly men
and women. J. Gerontol.
47: B71-B76,
1992.
AND M. J. WHITE. Mechanical
10. DAVIES, C. T. M., D. 0. THOMAS,
properties
of young and elderly
muscle. Acta Med. Stand. Suppl.
711: 219-226,
1988.
11. DUDLEY, G. A., AND S. J. FLECK. Metabolite
changes in aged muscle during stimulation.
J. Gerontol.
39: 183-186,
1984.
12. EARLE, M. S., A. R. COGGAN, S. C. SWANSON, A. M. ABDULJALIL,
P.-M. ROBITAILLE,
AND S. E. ALWAY. Effect of aging and/or endurance training
on muscle strength
and volume
in men (Abstract).
Med. Sci. Sports. Exercise 25: S 150, 1993.
13. ERMINI,
M. Ageing changes in mammalian
skeletal muscle. Gerontology 22: 301-316,
1976.
14. ES&N-GUSTAVSSON,
B., AND 0. BARGES. Histochemical
and metabolic characteristics
of human skeletal
muscle in relation
to age.
Acta Physiol. Stand. 126: 107-114,
1986.
15. FITTS, R. H., J. P. TROUP, F. A. WITZMANN,
AND J. 0. HOLLOSZY.

AGING

16.

17.

18.

19.
20.

21.

22.

23.

24.

AND

HUMAN

MUSCLE

The effect of ageing and exercise

on skeletal
muscle
function.
Mech. Aging Dev. 27: 161-172,
1984.
FRONTERA,
W. R., V. A. HUGHES, K. J. LUTZ, AND W. J. EVANS. A
cross-sectional
study of muscle strength
in 45- to 7%yr-old
men
and women. J. Appl. Physiol. 71: 644-650,
1991.
GRASSI, B., P. CERETELLI,
M. V. NARICI, AND C. MARCONI.
Peak
anaerobic
power in master athletes.
Eur. J. Appl. Physiol.
Occup.
Physiol. 62: 394-399,
1991.
HOLLOSZY,
J. O., AND E. F. COYLE. Adaptations
of skeletal muscle
to endurance
exercise and their metabolic
consequences.
J. Appl.
Physiol. 56: 831-838,
1984.
JACKSON, A. S., AND M. L. POLLOCK.
Generalized
equations
for
predicting
body density of men. Br. J. Nutr. 40: 497-504,
1978.
KOHRT,
W. M., M. T. MALLEY,
A. R. COGGAN,
R. J. SPINA, T.
OGAWA, R. E. BOUREY, W. H. MARTIN,
AND J. 0. HOLLPSZY.
Effects of gender, age, and fitness level on response
of Vozmar to
training
in 60-71 yr olds. J. Appl. Physiol. 71: 2004-2011,
1991.
LOWRY, 0. H., N. J. ROSEBROUGH,
A. L. FARR, AND R. J. RANDALL.
Protein measurement
with the Folin phenol reagent. J. Biol. Chem.
193: 267-275,
1951.
MARSH,
G. D., D. H. PATERSON,
R. T. THOMPSON,
AND A. A.
DRIEDGER.
Coincident
thresholds
in intracellular
phosphorylation
potential
and pH during progressive
exercise. J. Appl. Physiol. 71:
1076-1081,
1991.
MARTIN,
W. H., T. OGAWA, W. M. KOHRT,
M. T. MALLEY,
E.
KORTE, P. S. KIEFFER, AND K. B. SCHECHTMAN.
Effects of aging,
gender, and physical training
on peripheral
vascular
function.
Circulation 84: 654-664,
1991.
MCCULLY,
K. K., M. A. FORCIEA, L. M. HICK, E. DONLON,
R. W.
WHEATLEY,
C. A. OATIS, T. GOLDBERG,
AND B. CHANCE. Muscle
metabolism
in older subjects
using 31P magnetic
resonance
spectroscopy.
Can. J. Physiol. Pharmacol.
69: 576-580,
1991.

METABOLISM

DURING

EXERCISE

2133

25. MCNAMARA,
M. C., A. T. MILLER,
A. L. SHEN, AND J. J. WOOD.
Restitution
of ATP and creatine
phosphate
after experimental
depletion in young, adult, and old rats. Gerontology
24: 95-103, 1978.
26. MEREDITH,
C. N., W. R. FRONTERA,
E. C. FISHER, V. A. HUGHES,
J. C. HERLAND,
J. EDWARDS, AND W. J. EVANS. Peripheral
effects
of endurance
training
in young and old subjects. J. Appl. Physiol.
66: 2844-2849,
1989.
27. RICE, C. L., D. A. CUNNINGHAM,
D. H. PATERSON, AND M. S. LEFCOE. Arm and leg composition
determined
by computed
tomography in young and elderly men. Clin. Physiol. Oxf. 9: 207-220,
1989.
28. ROGERS, M. A., J. M. HAGBERG, W. H. MARTIN,
A. A. EHSANI, AND
J. 0. HOLLOSZY.
Decline in VO 2 max with aging in master athletes
and sedentary
men. J. Appl. Physiol. 68: 2195-2199,
1990.
29. SALTIN, B. The aging endurance
athlete. In: Sports Medicine for the
Mature
Athlete, edited by J. R. Sutton and R. M. Brock. Indianapolis, IN: Benchmark,
1986, p. 59-80.
30. STAMFORD,
B. A. Exercise
and the elderly. In: Exercise and Sports
Sciences Reviews, edited by K. D. Pandolf.
New York: Macmillan,
1988, vol. 16, p. 341-379.
31. SIRLIN, A. V., AND A. E. PATLA. Myoelectric
changes in the triceps
surae muscles under sustained
contractions.
Eur. J. Appl. Physiol.
Occup. Physiol. 56: 238-244,
1987.
32. TAYLOR,
D. J., P. J. BORE, P. STYLES, D. G. GADIAN, AND G. K.
RADDA. Bioenergetics
of intact human muscle. A 31P nuclear magnetic resonance
study. Mol. Biol. Med. 1: 77-94, 1983.
33. TAYLOR, D. J., M. CROWE, P. J. BORE, P. STYLES, D. L. ARNOLD,
AND G. K. RADDA. Examination
of the energetics
of aging skeletal
muscle using magnetic
resonance
spectroscopy.
Gerontology
30:
2-7, 1984.
34. TROUNCE,
I., E. BYRNE, AND S. MARZUKI.
Decline in skeletal muscle mitochondrial
respiratory
chain function:
possible
factor in
ageing. Lancet 1: 637-689,
1989.