Full text of DOI:10.1093/gerona/56.6.B240

Journal of Gerontology: BIOLOGICAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2001, Vol. 56A, No. 6, B240–B247
Skeletal Muscle Satellite Cell Characteristics in
Young and Older Men and Women After Heavy
Resistance Strength Training
S.M. Roth,1 G.F. Martel,1,2 F.M. Ivey,1 J.T. Lemmer,1 B.L. Tracy,1 E.J. Metter,3
B.F. Hurley,1 and M.A. Rogers1
1Department of Kinesiology, University of Maryland, College Park.
2Department of Physical Therapy, University of Maryland Eastern Shore, Princess Anne.
3Gerontology Research Center, National Institute on Aging, Baltimore, Maryland.
Skeletal muscle satellite cell proportions and morphology were assessed in healthy, sedentary
young and older men and women in response to heavy resistance strength training (HRST).
Fourteen young (20–30 years) men (n ꢀ 7) and women (n ꢀ 7) and 15 older (65–75 years) men
(n ꢀ 8) and women (n ꢀ 7) completed 9 weeks of unilateral knee extension exercise training 3
days per week. Muscle biopsies were obtained from each vastus lateralis before and after train-
ing, with the nondominant leg serving as an untrained control. All four groups demonstrated a
significant increase in satellite cell proportion in response to HRST (2.3 ꢁ 0.4% vs 3.1 ꢁ 0.4%
for all subjects combined, before and after training, respectively; p ꢂ .05), with older women
demonstrating the greatest increase (p ꢂ .05). Morphology data indicated a significant increase
in the proportion of active satellite cells in after-training muscle samples compared with before-
training samples and with control leg samples (31% vs 6% and 7%, respectively; p ꢂ .05). The
present results indicate that the proportion of satellite cells is increased after HRST in young and
older men and women, with an exaggerated response in older women. Furthermore, the propor-
tion of satellite cells that appear morphologically active is increased as a result of HRST.
HE loss of muscle mass and strength with age, termed
sarcopenia, can result in a loss of functional capacity in
ported no change in satellite cell proportions as a result of
training, but they did observe increases in myonuclear num-
ber proportional to the increases in muscle fiber area in both
the young and the older subjects (23). Satellite cell mor-
phology was not examined in either investigation (23,24).
Kadi and coworkers (25) recently reported an increased sat-
ellite cell number in young women following 10 weeks of
strength training. To our knowledge, no data exist on the
satellite cell response of older women to heavy resistance
strength training (HRST), nor are data available about the
morphology of human satellite cells in response to HRST.
Thus, the purpose of the present investigation was to assess
satellite cell proportions and morphology, quantified by
electron microscopy, in young and older men and women in
response to nine weeks of HRST. We hypothesized that
satellite cell proportions and activity would be increased
similarly in all groups as a result of HRST. A complete
analysis of the satellite cell characteristics of these healthy,
sedentary subjects before training has been previously pub-
lished (26).
T
older individuals (1,2). Although strength training is an im-
portant intervention for increasing muscle mass and strength
in older individuals (3,4), some evidence suggests that the
hypertrophic response of older animals and humans is re-
duced compared with that of their young counterparts (5–9).
Skeletal muscle satellite cells are important for muscle fiber
regeneration and hypertrophy (10,11). Given that skeletal
muscle regenerative capacity declines with age (12), age-
related changes in satellite cell activation in response to
muscle injury or overload could have important conse-
quences for older individuals (13,14). Lowe and Alway (6)
have suggested that age-related changes in satellite cell acti-
vation could contribute to age-related differences in the
muscle response to overload.
Although several investigations have demonstrated satel-
lite cell activation following muscle overload stimuli in
young and older animals (15–20), only limited research is
available regarding the role of strength training in satellite
cell activation in humans. Kadi and colleagues (21,22) re-
ported higher satellite cell proportions in male power lifters
compared with those in young untrained men. Hikida and
colleagues (23,24) have shown maintenance of myonuclear
density in young and older men following strength training,
an indication of satellite cell activation. The researchers re-
METHODS
Subject Selection
Twenty-nine healthy, young and older men and women
who volunteered for this study had adequate biopsy material
B240

AGING, STRENGTH TRAINING, AND SATELLITE CELLS
B241
for subsequent analysis. The subjects consisted of seven
young (20–30 years) and eight older (65–75 years) men and
seven young and seven older women. All subjects were
medically screened by a physician who performed a health
history, physical examination, and graded maximal tread-
mill exercise test (GXT). All subjects were nonsmokers and
were free of significant cardiovascular, metabolic, or mus-
culoskeletal disorders. Individuals enrolled in the study had
not participated in a regular exercise program for at least 6
months before their recruitment. After all methods and pro-
cedures were explained to them, the subjects read and
signed a written consent form that had been approved by the
Institutional Review Boards at the Baltimore Veterans Af-
fairs Medical Center and the University of Maryland, Col-
lege Park. All subjects were reminded throughout the study
not to alter their regular physical activity levels or dietary
habits for the duration of the investigation.
ing sessions, the untrained leg was placed in front of the
contralateral leg extension pad to prevent voluntary contrac-
tion of that muscle. The training program consisted of five
sets of high volume, heavy resistance leg extension exercise
performed 3 days per week. The resistance for each set was
based on the subject’s 5RM. After performing a first set of
five repetitions at 50% of the original 1RM for a warm-up,
the subject performed a second set of five repetitions at the
5RM resistance following a 30-second rest period. The third
set consisted of 10 repetitions, with the first four or five rep-
etitions at the current 5RM value. The resistance was then
lowered just enough for the subject to complete one or two
additional repetitions before reaching fatigue. The process
was repeated until a total of 10, 15, and 20 repetitions were
completed for the third, fourth, and fifth sets, respectively.
The third, fourth, and fifth sets were preceded by rest peri-
ods lasting a minimum of 90, 150, and 180 seconds, respec-
tively. An exercise specialist directly supervised the exer-
cise sessions of every subject at every training session to
verify compliance with the training protocol, record resis-
tance values and provide vocal encouragement. As in-
creases in muscle strength occurred throughout the 9-week
training program, the 5RM resistance was increased 1–2 kg
during the next training session.
Body Composition
Body composition was assessed before and after HRST
by using a Lunar DPXL dual-energy x-ray absorptiometer
(Lunar software version 1.22, GE Medical Systems, New
York, NY) as described previously (27). Subjects were in-
structed not to eat or drink for at least 8 hours prior to their
morning scan. The coefficient of variation for repeated mea-
surements of a calibration standard was less than 1.0%.
Body weight was measured weekly during an exercise ses-
sion by using a medical beam scale, and subjects were en-
couraged to maintain their before-training body weight.
Muscle Tissue Sampling, Fixation, and Analysis
Detailed procedures for sampling and fixation procedures
are outlined elsewhere (27). Muscle biopsies were taken
from each m. vastus lateralis of a subject before and after
the HRST program by the same investigator, using the per-
cutaneous needle biopsy technique. The initial sampling site
determined for each subject was 14 cm (women) or 16 cm
(men) from the proximal border of the patella at the midline
of the quadriceps. The biopsy sample taken after training
was obtained at a new site 2.5 mm proximal and lateral to
the original incision, with the biopsy needle directed into
approximately the same location and depth in the muscle as
the first biopsy. The before-training biopsy occurred 1 week
prior to the familiarization sessions; the after-training bi-
opsy occurred 24–48 hours after the last training session.
Muscle samples were minced for fixation into eight to ten
0.5- to 1.0-mm cubes that were fixed in a 2% solution of
glutaraldehyde in 0.12M phosphate buffer (Millonig’s
buffer) at room temperature for 1 hour and then refrigerated
(5–10ꢃC) until postfixation. The before- and after-training
tissue samples were postfixed at the same time, using 1%
osmium tetroxide, after which samples were stained with
2% uranyl acetate, dehydrated, and embedded longitudi-
nally in epoxy resin (Spurr’s). Five sample blocks were ob-
tained from each muscle biopsy sample. Thin sections (60–
70 nm) were obtained from each sample block and placed
on 75 ꢄ 300 copper grids. The sections on each grid were
stained with 2% uranyl acetate and 0.1% lead citrate. Pre-
pared grids were viewed on a Zeiss EM 10 CA electron mi-
croscope (Zeiss, Thornwood, NY) operated at 80 kV.
Strength Testing
Muscle strength was assessed in both quadriceps unilater-
ally by means of a one-repetition maximum (1RM) test us-
ing the Keiser K-300 pneumatic variable-resistance leg
extension machine (Keiser, Fresno, CA) as described previ-
ously (27). Briefly, three training sessions were conducted
by using light resistance prior to the strength testing in order
to familiarize the subjects with the equipment, training pro-
tocol, and proper exercise technique. For the 1RM test, a re-
sistance was chosen that was thought to be slightly below
the 1RM, and the subject performed one repetition. In-
creases in resistance between 1RM trials were adjusted to
minimize the total number of trials (6–8) required before the
true 1RM was obtained, and to keep the number of trials
similar before and after training. Rest periods of ꢀ60 sec-
onds were allowed between trials. The same investigator ad-
ministered the 1RM strength tests before and after training,
and all testing procedures were standardized based on spe-
cific seat and body positions. Following the 1RM test, each
subject’s five-repetition maximum (5RM) was also deter-
mined by using the same procedures to establish the initial
resistance for the first regular HRST session.
HRST Protocol
Each training session was preceded by a 3-minute warm-
up on a stationary cycle, followed by 5–10 minutes of static
stretching of both quadriceps. The training protocol con-
sisted of nine weeks of unilateral HRST of the knee exten-
sors of the dominant leg with the nondominant leg serving
as a control, as described previously (27). During the train-
Assessment of Satellite Cell Proportions
and Morphology
A detailed description of the analysis of satellite cells is
presented elsewhere (26). Briefly, each viable muscle fiber

B242
ROTH ET AL.
(27,28) was analyzed for total myonuclei and satellite cell
proportions by using an electron microscope, with all poten-
tial satellite cells photographed (4000-12,500ꢄ). Each re-
sulting micrograph was then reassessed later to ensure that
the mononucleate satellite cell was within the basement
membrane yet outside the sarcolemma of the muscle fiber
(29). All myonuclei and satellite cells were quantified, and
the satellite cell proportion [satellite cells/(myonuclei ꢅ sat-
ellite cells)] was calculated (29–31). In addition, the number
of central myonuclei, those completely surrounded by myo-
filaments and centered in a muscle fiber, (32) was assessed
for each of the four groups.
Satellite cell morphology was assessed to indirectly esti-
mate changes in satellite cell activity as a result of HRST, as
well as to determine possible age or gender differences in the
response. Each micrograph was developed into a photo-
graphic print that was scanned and converted to a graphic
computer image file. The following morphological character-
istics were examined in each satellite cell (26): satellite cell
area, satellite cell nucleus area (nuclear area), and the nuclear
area/cell area ratio, nuclear chromatin structure (percent het-
erochromatin), endoplasmic reticulum (ER) and golgi appa-
ratus, mitochondrial number, ribosomes, pinocytotic vesicles,
and the existence of disorganized myofilaments.
Cell and nuclear areas (26,33) were measured by using
UTHSCSA ImageTool imaging software (version 1.27,
University of Texas Health Science Center at San Antonio,
TX) and the nuclear area/cell area ratio was calculated. The
percentage of heterochromatin (26,34) was assessed by
using histographic–density analysis tools available in the
ImageTool software. All area measurements were obtained
in duplicate with the investigator blind to age, gender, leg,
and training time point. Reliability was assessed by the re-
peated measuring of multiple cells on separate days (Pear-
son r ꢀ .99).
The organelle morphological characteristics (e.g., ER,
golgi, and mitochondria) were assessed by using the photo-
graphic prints based on semiquantitative procedures out-
lined in previous research (26,33,34). Organelles were
ranked according to the frequency and extent of their ap-
pearance, such that each organelle was described as follows:
not observed (ꢆ), and rarely (ꢅ), seldom (ꢅꢅ), occasion-
ally (ꢅꢅꢅ), or frequently (ꢅꢅꢅꢅ) observed. Changes in
cell morphology as a result of HRST were assessed within
and among groups. An increased frequency or abundance of
these organelle structures has been used previously to indi-
cate satellite cell activation indirectly (33). As such, a rating
of “Active” or “Inactive” was assigned to each cell follow-
ing the morphological analysis. For this categorization, the
investigator was similarly blind to condition and the rating
assignment was based on the general pattern of scores for
each of the morphological characteristics. All satellite cells
assigned as Active exhibited at least four organelle catego-
ries with scores of ꢅꢅꢅ or greater. All cells were mea-
sured in duplicate, and a reliability analysis was performed
as described above (Pearson r ꢀ .96).
chromatin percentage as a result of HRST were analyzed by
using a three-factor (Experimental Condition ꢄ Age ꢄ
Gender) repeated-measures analysis of variance (ANOVA).
Each subject’s mean value for each variable was used in
those analyses. The morphological characteristics of satel-
lite cells were analyzed by chi-square analysis, with
Fisher’s Exact Test used when the number of cases fell be-
low five. Statistical significance for all analyses was ac-
cepted at p ꢇ .05. All data are reported as means ꢁ SE.
RESULTS
Physical Characteristics
Subject characteristics are presented in Table 1. All sub-
jects completed at least 27 HRST sessions over approxi-
mately 9 weeks. Both before and after HRST, the 1RM
strength values of the older men (73.7 ꢁ 2.9 kg and 94.6 ꢁ
3.6 kg, respectively) and young men (79.4 ꢁ 8.0 kg and
99.8 ꢁ 8.6 kg, respectively) were significantly higher than
those of the older women (42.7 ꢁ 2.1 kg and 52.1 ꢁ 3.0 kg,
respectively; p ꢂ .05). The strength of the young women
(61.9 ꢁ 6.6 kg and 85.1 ꢁ 7.1 kg, respectively) was signifi-
cantly greater than that of the older women before and after
HRST (p ꢂ .05). HRST resulted in a significant increase in
percent muscle strength in all groups (p ꢂ .05) with no sig-
nificant differences among groups.
Satellite Cell Proportions
In the trained leg, a total of 2500 myonuclei (33 central
myonuclei) were observed in 564 fibers in before-HRST
samples compared with a total of 2661 total myonuclei (36
central myonuclei) observed in 618 fibers in after-HRST
samples, with an average of ꢀ90 myonuclei per subject per
time point. Including samples obtained from the control
limb, 7922 total myonuclei were observed in 1863 fibers, in-
cluding 183 satellite cells; these numbers are similar to those
found in previous work (29,35,36). The number of muscle fi-
bers analyzed before and after HRST did not differ signifi-
cantly. Within the trained leg, a total of 58 satellite cells
were evident in before-HRST samples compared with a total
of 86 satellite cells in after-HRST samples (p ꢂ .05). The in-
crease in satellite cell number after HRST resulted from a
significant increase observed in the older women (22 cells
observed before HRST, and 44 cells after HRST; p ꢂ .05).
Figure 1 is a representative micrograph of a skeletal mus-
cle satellite cell in relation to a myonucleus. The satellite
Table 1. Subject Characteristics
Parameter
Older Men Older Women Young Men Young Women
n
8
7
7
7
Age (y)
69 ꢁ 3
67 ꢁ 3
25 ꢁ 3
26 ꢁ 1
Height (cm)
Weight (kg)
% Body fat
Before HRST
After HRST
173.3 ꢁ 4.3
83.6 ꢁ 10.5
157.0 ꢁ 6.1 180.4 ꢁ 10.7 168.5 ꢁ 5.8
71.1 ꢁ 8.8
93.0 ꢁ 18.3
69.2 ꢁ 6.4
30 ꢁ 6
30 ꢁ 5
41 ꢁ 6*
39 ꢁ 7*
26 ꢁ 8
26 ꢁ 8
31 ꢁ 5
32 ꢁ 3
Statistical Analysis
Changes in satellite cell proportions as well as cell and
nuclear areas, the nuclear area/cell area ratio, and hetero-
Notes: Data are means ꢁ SE. HRST ꢀ heavy resistance strength training.
*Significantly greater than all other groups (p ꢂ .05).

AGING, STRENGTH TRAINING, AND SATELLITE CELLS
B243
Figure 2. Satellite cell proportions [satellite cells/(myonuclei ꢅ sat-
ellite cells)] from the vastus lateralis muscle for healthy young and
older men and women before and after heavy resistance strength
training (HRST). Before HRST ꢀ data from dominant leg before
HRST; after HRST ꢀ data from dominant leg after HRST. All un-
trained (UT) samples ꢀ pooled data for the untrained leg only. *Sig-
nificantly different from before HRST, omnibus F(1,25) ꢀ 5.60; p ꢂ
.05. †Significant Age ꢄ Sex interaction, F(3,25) ꢀ 3.05; p ꢂ .05. Data
are means (%) ꢁ SE.
Satellite Cell Morphology
Young women had significantly greater satellite cell ar-
eas compared with those of young and older men, both be-
fore and after HRST (p ꢂ .05), although no significant dif-
ferences were noted for nuclear area or nuclear area/cell area
ratio among the groups either before or after HRST. Be-
cause both control leg samples (before and after HRST) did
not receive the HRST stimulus and no significant changes
in satellite cell proportion or morphology were seen be-
tween the samples, the data from all untrained samples were
combined (before-training sample of the trained leg ꢅ both
control leg samples) and compared with data from the after-
HRST sample. In this analysis, a significant increase in sat-
ellite cell area and nuclear area was noted in the after-HRST
samples (p ꢂ .05; Table 2). No differences were noted for
nuclear area/cell area ratio or heterochromatin percentage
between the pooled untrained samples and the after-training
samples (Table 2).
Figure 1. Electron micrograph representative of a human skeletal
muscle satellite cell (S) compared with a myonucleus (M; bar ꢀ 1 ꢈm).
Note the presence of the basement membrane surrounding both nu-
clei (arrowhead), while the satellite cell is located outside of the mus-
cle fiber’s sarcolemma (separation indicated by arrow).
cell proportion [satellite cells/(myonuclei ꢅ satellite cells)]
increased significantly as a result of HRST in the trained leg
[omnibus F(1,25) ꢀ 5.60; p ꢂ .05; Figure 2], with a signifi-
cant Age ꢄ Gender interaction resulting from a signifi-
cantly greater increase in proportion in the older women
compared with the other groups [F(3,25) ꢀ 3.05; p ꢂ .05;
Figure 2]. When the data for all subjects were pooled,
HRST elicited a significant increase in satellite cell propor-
tion in the trained leg (2.3 ꢁ 0.4% vs 3.1 ꢁ 0.4%, respec-
tively; p ꢂ .05). No changes in satellite cell proportion oc-
curred in the untrained (control) limb following HRST in
any of the four groups. A histochemical analysis of fiber
type and area indicated no significant differences among the
groups for fiber-type before or after HRST, no significant
change in fiber type proportions after HRST, and significant
increases in fiber area for all fiber types in all groups (un-
published observations).
Satellite cell morphology was assessed for each time
point and a global assignment of Active or Inactive was
assigned based on this analysis. The number and proportion
of satellite cells scored as Active increased as a result of
HRST (Figure 3). Chi-square analysis revealed a significant
difference between the proportion of active cells in after-
Table 2. Satellite Cell Morphological Characteristics in Trained
Leg Samples vs All Untrained Samples
Parameters
All Untrained Samples
After HRST
n
94
72
Cell area (ꢈ2)
12.4 ꢁ 0.7
7.4 ꢁ 0.4
59.1 ꢁ 1.3
39.5 ꢁ 1.6
15.3 ꢁ 1.2*
9.1 ꢁ 0.7*
60.4 ꢁ 1.4
39.0 ꢁ 1.0
Nuclear area (ꢈ2)
Nucleus/cell area ratio (%)
Heterochromatin (%)
The proportion of myonuclei that were central (i.e., located
within the myofibrils) was significantly higher in the older
men compared with that in all other groups before HRST
(3.3% compared with 0.5–1.0%, respectively; p ꢂ .05). This
significant difference remained after HRST, as the proportion
of central myonuclei did not change as a result of HRST.
Notes: Data are means ꢁ SE. HRST ꢀ heavy resistance strength training.
“All untrained samples” are defined as any muscle sample not subjected to the
HRST stimulus (before-training samples pooled with both control leg samples).
*Significantly greater than all untrained samples (p ꢂ .05). No other signif-
icant differences.

B244
ROTH ET AL.
some reports have found no change in satellite cell activa-
tion with an overload stimulus (6,37). Umnova and Seene
(16) reported an increase in satellite cell proportion from
3.4% to 9.6% in rats following 6 weeks of treadmill exer-
cise training, and Hanzlikova and colleagues (15) reported
an increase from 5.8% to 16.6% in rats following synergist
ablation. Although the increase in satellite cell proportion
noted in the present investigation is modest (from 2.3% to
3.1%), a direct comparison with animal studies is difficult
as a result of species differences and the variable nature of
the various overload stimuli used. The HRST program used
in the present study resulted in three relatively short periods
of stimulation each week, whereas many animal studies al-
low for a constant 24-hour stimulus over days or weeks
(15,17,38). Furthermore, increases in satellite cell propor-
tion are seen most dramatically early in an intervention, pre-
sumably during the height of cell proliferation (17,30,38).
Given that the present study investigated satellite cell char-
acteristics at the end of a 9-week HRST program, we specu-
late that peak satellite cell activation likely occurred earlier
in the training period and that substantial differentiation of
cells had occurred by 9 weeks. This observation would cor-
respond well with the data of Hikida and colleagues, (23)
who demonstrated no change in satellite cell proportion af-
ter 8–16 weeks of strength training in young and older men,
and Kadi and colleagues, (25) who reported an increase in
satellite cell proportion from 3.7% to 5.4% after 10 weeks
of strength training in young women. In addition to the sig-
nificant increase in satellite cell proportion, the present data
indicated no change in the satellite cell proportions of the
untrained leg and provided morphological evidence for in-
creased cellular activity. Thus, these results provide substan-
tial evidence for satellite cell activation and proliferation in
the muscle response to HRST in young and older men and
women.
Additional evidence for satellite cell activation in humans
in response to strength training exists, although data are lim-
ited. Kadi and coworkers (21,22) recently reported satellite
cell proportions of ꢀ7% in the trapezius muscle of elite
male power lifters compared with values of ꢀ4% in un-
trained controls. Although the data were cross sectional,
satellite cell activation resulting from long-term strength
training was indicated indirectly by the maintenance of my-
onuclear density despite substantial increases in muscle fi-
ber size in the power lifters (21,22). Recently, Hikida and
coworkers (23) assessed satellite cell proportions in muscle
fiber cross sections in young and older (65 years) men fol-
lowing lower body strength training. The researchers re-
ported no differences in satellite cell proportions with age or
with strength training (ꢀ2.4% in both groups before and af-
ter training), although differences in the duration of the
strength training stimulus for young versus older men (8 vs
16 weeks) limits comparison between the groups. Hikida
and colleagues (23,24) have also observed that myonuclear
number increased proportionately to muscle fiber area in re-
sponse to strength training in both young and older men, in-
directly indicating satellite cell activation in both age
groups (39). Using an electron microscopic analysis of lon-
gitudinally oriented muscle fibers, the current investigation
indicated both an increased proportion of satellite cells and
Figure 3. Number of satellite cells labeled “Active” and “Inactive”
in before T, after T, and both UT leg samples (combined) for all sub-
jects combined. Before T ꢀ before-training samples; After T ꢀ after-
training samples; UT ꢀ untrained pooled control leg samples. *Signif-
icantly different from Before T and UT samples. Data are presented
as total cell number, with the percentage of active cells presented for
the corresponding column.
training muscle samples compared with all untrained mus-
cle samples (21 of 67 satellite cells in after-HRST samples
compared with 6 of 88 satellite cells from before-HRST ꢅ
control leg samples; p ꢂ .05). Although not significantly
different (p ꢀ .14), a higher proportion of active satellite
cells was found in older women than in other groups after
HRST (36% compared with 17–25%, respectively).
Active satellite cells were generally scored higher for all
morphological traits assessed, especially for ER, pinocy-
totic vesicles, and mitochondrial number. Figures 4A–4F
depict cells representative of active satellite cells or mor-
phological traits representative of active satellite cells.
Older women had more satellite cells scored with high val-
ues for lipofuscin granules, but no other significant group,
age, or gender differences were noted in cell morphology.
Lipofuscin granule content did not change as a result of
HRST. A complete analysis of the satellite cell populations
in the before-training and untrained (control leg) muscle
samples is presented elsewhere in the context of aging skel-
etal muscle (26). Cilia were noted in several satellite cells in
muscle samples from all time points, most of which were
categorized as Active (Figures 4E and 4F).
DISCUSSION
In the present investigation, satellite cell proportion in
young and older men and women was significantly in-
creased as a result of 9 weeks of HRST. Additionally, an as-
sessment of satellite cell morphology after HRST indicated
a significant increase in the proportion of active satellite
cells, cells with higher levels of mitochondria, ER, and pi-
nocytotic vesicles. Interestingly, older women demonstrated
a significantly greater increase in satellite cell proportion
(significant interaction, p ꢂ .05) and the largest increase in
the number of active satellite cells in response to HRST (p ꢀ
.14) compared to the other groups.
Substantial previous work in young and older animals has
demonstrated increases in satellite cell proportion in re-
sponse to various muscle overload stimuli (15–20), although

AGING, STRENGTH TRAINING, AND SATELLITE CELLS
B245
Figure 4. Representative micrographs of satellite cells obtained from after-training muscle samples labeled as “Active” based on morpholog-
ical criteria. For reference, Fig. 1 is representative of an inactive satellite cell. A, Satellite cell exhibiting expanded cytoplasm and endoplasmic
reticulum (ER), and numerous mitochondria (arrow; bar ꢀ 1 ꢈm). B, Active satellite cell showing numerous mitochondria (arrowheads) and pi-
nocytotic vesicles (bar ꢀ 1 ꢈm). C, Satellite cell (S) demonstrating expanded cytoplasm in the form of a lamellapodia (arrowheads), as well as
euchromatic chromatin appearance (bar ꢀ 2 ꢈm). This cell appears to be traveling along the muscle fiber, over the top of myonuclei (M). D, Sat-
ellite cell exhibiting numerous pinocytotic vesicles (arrows; bar ꢀ 1 ꢈm). Note also the wrapping of the muscle cell’s cytoplasm (arrowheads)
around the periphery of the satellite cell. E, Satellite cell exhibiting expanded cytoplasm and ER, and numerous mitochondria (arrow), along
with euchromatic chromatin appearance (bar ꢀ 1 ꢈm). A cilium (arrowhead) also appears to be present in the cytoplasm of the cell. F, Active
satellite cell exhibiting a cilium (arrow; bar ꢀ 1 ꢈm). Note also the substantial festooning of the nucleus in this cell, also demonstrated in A and
B. This nuclear structure was seen regularly throughout active and inactive satellite cells.
an increased number of active satellite cells following 9
weeks of HRST, providing evidence for satellite cell prolif-
eration in both young and older men and women. In the
present investigation, increases in satellite cell proportions
and activity accompanied increases in both muscle strength
and mass (outlined in detail elsewhere; see Ref. 40) follow-
ing 9 weeks of HRST. Substantial evidence indicates that
the number of myonuclei increases in proportion to in-
creases in muscle fiber volume such that myonuclear den-
sity is maintained (24,39), and that the increase in myonu-

B246
ROTH ET AL.
3. Fiatarone MA, Marks EC, Ryan ND, Meredith CN, Lipsitz LA, Evans
WJ. High-intensity strength training in nonagenarians. JAMA. 1990;
263:3029–3034.
4. Hurley BF, Hagberg JM. Optimizing health in older persons: aerobic
or strength training? In: Holloszy JO, ed. Exercise and Sport Sciences
Reviews. Baltimore: Williams and Wilkins; 1998:61–90.
5. Blough ER, Linderman JK. Lack of skeletal muscle hypertrophy in
very aged male Fischer 344 X Brown Norway rats. J Appl Physiol.
2000;88:1265–1270.
clear number is a result of satellite cell proliferation and
differentiation (41). The present results provide further evi-
dence for satellite cell proliferation in response to HRST,
and the first evidence for a possible gender difference in the
satellite cell proliferative response. These data provide indi-
rect evidence for a role for satellite cells in the muscle mass
response to HRST.
Although the percentage is not significantly different (p ꢀ
.14), older women in the present study tended to have a
higher proportion of active satellite cells than other groups
(36% compared with 17–25%, respectively), and they dem-
onstrated the largest increase in satellite cell proportion at
the end of 9 weeks of HRST (p ꢂ .05). Alternatively, given
that satellite cell proliferation peaks early following an
overload stimulus (17,30,38), this difference might indicate
delayed satellite cell activation in the older women, if we
assume that similar increases in cell proliferation had oc-
curred earlier in the other groups. Except that the increases
in muscle fiber area as a result of HRST were not signifi-
cantly different among the groups, we have no additional
data to support or refute this alternative explanation. In pre-
vious work from our laboratory (27,42), these same older
female subjects demonstrated a significantly greater number
of muscle fibers exhibiting ultrastructural muscle damage
(17%) as a result of HRST compared with the other groups
(3–7% of fibers exhibiting damage). Muscle activity and
other possible mediators of satellite cell proliferation and
activation independent of muscle damage cannot be ruled
out as stimuli in the present study.
6. Lowe DA, Alway SE. Stretch-induced myogenin, MyoD, and MRF4
expression and acute hypertrophy in quail slow-tonic muscle are not
dependent upon satellite cell proliferation. Cell Tissue Res. 1999;296:
531–539.
7. Carson JA, Alway SE. Stretch overload-induced satellite cell activa-
tion in slow tonic muscle from adult and aged Japanese quail. Am J
Physiol. 1996;270:C578–C584.
8. Häkkinen K, Kallinen M, Izquierdo M, et al. Changes in agonist-
antagonist EMG, muscle CSA, and force during strength training in
middle-aged and older people. J Appl Physiol. 1998;84:1341–1349.
9. Welle S, Totterman S, Thornton C. Effect of age on muscle hypertro-
phy induced by resistance training. J Gerontol Med Sci. 1996;51A:
M270–M275.
10. Rosenblatt JD, Yong D, Parry DJ. Satellite cell activity is required for
hypertrophy of overloaded adult rat muscle. Muscle Nerve. 1994;17:
608–613.
11. Phelan JN, Gonyea WJ. Effect of radiation on satellite cell activity and
protein expression in overloaded mammalian skeletal muscle. Anat
Rec. 1997;247:179–188.
12. Sadeh M. Effects of aging on skeletal muscle regeneration. J Neurol
Sci. 1988;87:67–74.
13. Carlson BM. Factors influencing the repair and adaptation of muscles
in aged individuals: satellite cells and innervation. J Gerontol Biol Sci.
1995;50A:B96–B100.
14. Chakravarthy MV, Davis BS, Booth FW. IGF-I restores satellite cell
proliferative potential in immobilized old skeletal muscle. J Appl
Physiol. 2000;89:1365–1379.
In summary, the data from the present investigation are
the first to indicate that HRST elicits increases in satellite
cell proportion in young and older men and women, with a
possible Age ꢄ Gender interaction in the response. Further-
more, the present data provide the only known evidence for
an increased proportion of morphologically active satellite
cells following HRST in these groups.
15. Hanzlikova V, Mackova EV, Hnik P. Satellite cells of the rat soleus
muscle in the process of compensatory hypertrophy combined with
denervation. Cell Tissue Res. 1975;160:411–421.
16. Umnova MM, Seene TP. The effect of increased functional load on the
activation of satellite cells in the skeletal muscle of adult rats. Int J
Sports Med. 1991;12:501–504.
17. Snow MH. Satellite cell response in rat soleus muscle undergoing hy-
pertrophy due to surgical ablation of synergists. Anat Rec. 1990;227:
437–446.
18. Darr KC, Schultz E. Exercise-induced satellite cell activation in grow-
ing and mature skeletal muscle. J Appl Physiol. 1987;63:1816–1821.
19. Jacobs SCJM, Wokke JHJ, Bar PR, Bootsma AL. Satellite cell activa-
tion after muscle damage in young and adult rats. Anat Rec. 1995;242:
329–336.
Acknowledgments
This work was supported by NIH Research Contract AG-42148 (to B.F.
Hurley and M.A. Rogers). S.M. Roth was supported by NIA Grants AG-
00268 and AG-05893.
20. McCormick KM, Thomas DP. Exercise-induced satellite cell activa-
tion in senescent soleus muscle. J Appl Physiol. 1992;72:888–893.
21. Kadi F, Eriksson A, Holmner S, Butler-Browne GS, Thornell LE. Cel-
lular adaptation of the trapezius muscle in strength-trained athletes.
Histochem Cell Biol. 1999;111:189–195.
22. Kadi F, Eriksson A, Holmner S, Thornell LE. Effects of anabolic ste-
roids on the muscle cells of strength-trained athletes. Med Sci Sports
Exerc. 1999;31:1528–1534.
23. Hikida RS, Walsh S, Barylski N, Campos G, Hagerman FC, Staron
RS. Is hypertrophy limited in elderly muscle fibers? A comparison of
elderly and young strength-trained men. Basic Appl Myol. 1998;8:
419–427.
We thank Tim Maugel, Director of the Laboratory for Biological Ultra-
structure, and Dr. Stephen Wolniak for their expert insight and technical as-
sistance. This work is Contribution 88 from the Laboratory for Biological
Ultrastructure, University of Maryland, College Park (UMCP). We thank
Dr. Colleen Farmer and the staff of the Wellness Research Laboratory at
UMCP for the use of their facilities. We also extend our appreciation to
Dorothy O’Donnell, Diane Hurlbut, Dr. Mary Lott, and Nancy Roth for
their assistance with the project; Dr. Jerome Fleg for manuscript review;
and to the subjects who participated in this investigation. A preliminary re-
port of these results was presented at the 2000 American College of Sports
Medicine Annual Meeting (43).
Address correspondence to Stephen M. Roth, A300 Crabtree Hall, De-
partment of Human Genetics, University of Pittsburgh, Pittsburgh, PA
15261. E-mail: sroth@helix.hgen.pitt.edu
24. Hikida RS, Staron RS, Hagerman FC, et al. Effects of high-intensity
resistance training on untrained older men. II. Muscle fiber character-
istics and nucleo-cyoplasmic relationships. J Gerontol Biol Sci. 2000;
55A:B347–B354.
25. Kadi F, Thornell L-E. Concomitant increases in myonuclear and satel-
lite cell content in female trapezius muscle following strength training.
Histochem Cell Biol. 2000;113:99–103.
26. Roth SM, Martel GF, Ivey FM, et al. Skeletal muscle satellite cell pop-
ulations in young and older men and women. Anat Rec. 2000;260:
351–358.
References
1. Evans WJ, Campbell WW. Sarcopenia and age-related changes in
body composition and functional capacity. J Nutr. 1993;123:465–468.
2. Rantanen T, Guralnik JM, Ferrucci L, Leveille S, Fried LP. Coimpair-
ments: strength and balance as predictors of severe walking disability.
J Gerontol Med Sci. 1999;54A:M172–M176.
27. Roth SM, Martel GF, Ivey FM, et al. Ultrastructural muscle damage in

AGING, STRENGTH TRAINING, AND SATELLITE CELLS
B247
young vs. older men following heavy resistance strength training. J
Appl Physiol. 1999;86:1833–1840.
28. Roth SM, Martel GF, Rogers MA. Muscle biopsy and muscle fiber hy-
percontraction: a brief review. Eur J Appl Physiol. 2000;83:239–245.
29. Snow MH. The effects of aging on satellite cells in skeletal muscles of
mice and rats. Cell Tissue Res. 1977;185:399–408.
30. Winchester PK, Gonyea WJ. A quantitative study of satellite cells and
myonuclei in stretched avian slow tonic muscle. Anat Rec. 1992;232:
369–377.
31. Ishimoto S, Goto I, Ohta M, Kuroiwa Y. A quantitative study of the
muscle satellite cells in various neuromuscular disorders. J Neurol Sci.
1983;62:303–314.
32. Snow MH. A quantitative ultrastructural analysis of satellite cells in
denervated fast and slow muscles of the mouse. Anat Rec. 1983;207:
593–604.
33. Campion DR, McCusker RH, Richardson RL. Ultrastructure of muscle
satellite cells in hypersomatotropic rats. Acta Anat. 1987;128:67–70.
34. Wakayama Y, Schotland DL. Muscle satellite cell populations in
Duchenne dystrophy. In: Mauro A, et al., ed. Muscle Regeneration.
New York: Raven Press; 1979:121–129.
37. Lee J, Alway SE. Adaptations of myonuclei to hypertrophy in patagia-
lis muscle fibers from aged quail. Mech Ageing Devel. 1996;88:185–197.
38. Putman CT, Dusterhoft S, Pette D. Satellite cell proliferation in low
frequency-stimulated fast muscle of hypothyroid rat. Am J Physiol
Cell Physiol. 2000;279:C682–C690.
39. Roy RR, Monke SR, Allen DL, Edgerton VR. Modulation of myonu-
clear number in functionally overloaded and exercised rat plantaris
fibers. J Appl Physiol. 1999;87:634–642.
40. Ivey FM, Roth SM, Ferrell RE, et al. Effects of age, gender and myo-
statin genotype on muscle volume response to heavy resistance
strength training. J Gerontol Med Sci. 2000;55A:M641–M648.
41. Moss FP, Leblond CP. Satellite cells as the source of nuclei in muscles
of growing rats. Anat Rec. 1971;170:421–436.
42. Roth SM, Martel GF, Ivey FM, et al. High volume, heavy resistance
strength training and muscle damage in young and older women. J
Appl Physiol. 2000;88:1112–1118.
43. Roth SM, Martel GF, Ivey FM, et al. Satellite cell response to nine
weeks of strength training in young and older men and women. Med
Sci Sports Exerc. 2000;32:S294.
35. Schmalbruch H, Hellhammer U. The number of satellite cells in nor-
mal human muscle. Anat Rec. 1976;185:279–288.
Received September 21, 2000
36. Schultz E. Fine structure of satellite cells in growing skeletal muscle.
Am J Anat. 1976;147:49–70.
Accepted December 12, 2000
Decision Editor: John A. Faulkner, PhD