Download Muscle damage is not a function of muscle force but active muscle

Muscle damage is not a function
but active muscle strain
of muscle force
RICHARD
L. LIEBER
AND JAN FRIDfiN
of Orthopedics and Biomedical Sciences Graduate Group, University of California
Department
and Veterans Affairs Medical Centers, San Diego, California 92161; and Departments of Anatomy
Hand Surgery, University of Urned, S901 87 Ume&, Sweden
LIEBER,RICHARD L., ANDJANFRID~N.
Muscledamageis
not
a function of muscle force but active muscle strain. J. Appl. Physiol. 74(Z): 520-526, 1993.-Contractile
properties
of rabbit tibialis anterior muscles were measured after eccentric contraction to investigate the mechanism of muscle injury. In the first
experiment,
two groups of muscles were strained 25% of the
muscle fiber length at identical
rates. However, because the
timing of the imposed length change relative to muscle activation was different, the groups experienced
dramatically
different muscle forces. Because muscle maximum tetanic tension
and other contractile parameters measured after 30 min of cyclic activity with either strain timing pattern were identical
(P > 0.4), we concluded that muscle damage was equivalent
despite very different imposed forces. This result was supported by a second experiment
in which the same protocol was
performed at one-half the strain (12.5% muscle fiber length).
Again, there was no difference in maximum
tetanic tension
after cyclic 12.5% strain with either strain timing. Data from
both experiments
were analyzed by two-way analysis of variance, which revealed a highly significant effect of strain magnitude (P < 0.001) but no significant
effect of stretch timing
(P > 0.7). We interpret these data to signify that it is not high
force per se that causes muscle damage after eccentric contraction but the magnitude of the active strain (i.e., strain during
active lengthening).
This conclusion was supported by morphometric analysis showing equivalent area fractions of damaged
muscle fibers that were observed throughout
the muscle cross
section. The active strain hypothesis is described in terms of
the interaction
between the myofibrillar
cytoskeleton,
the sarcomere, and the sarcolemma.
eccentric contraction;
muscle injury; stress; cytoskeleton;
grins; intermediate
filaments; desmin
inte-
KNOWN that eccentric contractions
(EC) result in high forces and are a normal part of the gait cycle,
especially in the extensor muscles (8, 11). Despite their
common occurrence, the underlying basis for the behavior of muscle during EC is not as well understood as muscle behavior during concentric (shortening) contractions.
For example, Harry et al. (10) recently performed frog
sartorius muscle EC at high velocities and concluded that
the shape of the lengthening portion of the force-velocity
curve could not be explained on the basis of simple modification of the cross-bridge theory proposed by Huxley
(14). The experimental data of Harry et al. were accompanied by an intriguing explanation of muscle lengthening
behavior based not on cross-bridge properties but on in-
IT IS DELL
520
and
teraction between sarcomeres along the fiber length (21).
These two reports outlined the difficulty in understanding EC of skeletal muscle.
Chronic studies have also demonstrated that EC is selectively associated with muscle injury and muscle soreness in humans and in various animal models (see Ref. 4
for review). That injury occurs after EC is intriguing in
light of the observation that EC is very common during
normal gait. Because high muscle tension and muscle
injury can both be associated with EC, it has been proposed that high muscle force is responsible for the muscle injury observed. However, this proposal has not been
explicitly tested. The purpose of this study, therefore,
was to determine the effects of force and strain on the
tension generated by rabbit tibialis anterior (TA) muscles. These experiments were performed to test the hypothesis that muscle damage is due to high force. A brief
account of this work has been presented elsewhere (20).
METHODS
The muscle chosen for study was the TA of the New
Zealand White rabbit. This muscle was chosen primarily
because the muscle fibers are oriented with a pennation
angle of only 3O (16) and thus demonstrate negligible
angular rotation during lengthening.
Pilot experiments
with more highly pennated muscles, such as the gastrocnemius, revealed significant shear stresses that were
nonlinear and highly dependent on strain magnitude.
Contractile properties were measured and experimental
treatments were performed essentially as previously described (17).
Preparation and contractile measurements. Briefly,
rabbits were anesthetized with a subcutaneous injection
of a ketamine-xylazine-acepromazine
cocktail (50,5, and
1 mg/kg body mass, respectively) and mai ntained on halothane anesthesia. Heart and respiratory rate were monitored during muscle isolation and testing (model 78O7C,
Hewlett-Packard,
Palo Alto, CA), and anesthesia level
was adjusted as needed. All experimental
procedures
were performed in accordance with the guidelines set
forth by the National Institutes of Health “Guide for the
Care and Use of Animals.”
Great care was taken to minimize system compliance,
ensuring that the imposed deformation was taken up by
the muscle itself and not the apparatus or the external
tendon. The distal TA tendon was secured to a dualmode servomotor (model 310, Cambridge Technology,
0161-7567193 $2.00 Copyright 0 1993 the American Physiological
Society
MUSCLE
DAMAGE
IS NOT
Early stretch
Late stretch
Moderate
Very high
25
25
125
125
Early stretch
Late stretch
Low
Moderate
12.5
12.5
63
63
L,, fiber length.
DUE
TO
HIGH
521
FORCE
nominally
55 mm, absolute strain magnitude and strain
rate were -13 mm and 65 mm/s, respectively.
In the LS group (late stretch), muscle stretch was delayed for 200 ms while the muscle developed tension.
Then the muscle was stretched with the identical pattern
as the ES group. Thus stretch rate and magnitude were
identical
between groups. H owever, because of the
stretch timing, the peak force reached in the LS group
was significantly greater than that of the ES group(Fig.
1). In this way, both groups received similar deformation
patterns but experienced dramatically
different forces. It
should be noted that the 200-ms delay is probably longer
than the normal 60- to SO-ms delay usually observed between muscle activation and lengthening
(9). However,
we increased the delay to produce dramatically
different
forces in the two experimental
groups.
Variation of muscle strain. The experi ments described
above were repeated at one- ,half the total strain (12.5%
L,) in two separate groups of animals (n = Wgroup).
Thus, combining the results of both experiments, high
and low strains were imposed at high and low forces
to test the effect of force and strain independently
(Table 1).
It was imperative that the two groups have nearly the
same metabolic energy requirements so that the experiment tested only mechanical difference s between groups
and not differen .ces based on cellular metabolism
(2).
Therefore, all experimental activation cycles consisted of
400-ms trains of 40-Hz pulses. A separate experimental
group that only experienced cyclic isometric activation
(n = 8) was also studied for comparison.
Treatment was performed on-line while the computer
synchronized muscle activation and length change and
stored contractile data in real time at specified intervals.
In addition to data acquisition and storage, the computer
also performed real-time integration
of force with respect to time for each contraction according to
Cambridge, MA) and aligned with the motor’s measuring
and translation axis. System compliance, including the
transducer, was 1.3 pm/g.
The peroneal nerve was isolated for direct muscle activation. Muscle temperature
was then maintained
at
37°C with radiant heat, mineral oil, and a servo-temperature controller (model 73A, Yellow Springs Instruments,
Ye1.low Springs, OH). Under computer control (19), muscle length was adjusted to the length at which twitch
tension was maximum
(L,), and contractile
properties
were determined before experimental treatment .(see below). Contractile properties measured included time to
peak twitch tension, the rate of rise of twitch and tetanic
tension (dP/dt), twitch half-relaxation
time, maximum
twitch tension, passive force in response to 12.5 or 25%
strain at 63% muscle fiber length (L,) /s or 125% L,ls (see
below), and contractile tension at stimulation
frequencies of 5, 25, 50, 75, and 100 Hz. Maximum
tetanic tension (P,) was defined as the tension measured while stimulated at 100 Hz, the peak of the force-frequency relationship.
Half-fusion
frequency was defined as the
frequency at which the tension was 50% P,. Measurements of L, were made on each muscle from th .e tibia1
tubercl .e to the most distal muscle fiber insertion. On the
basis of previous architectural studies (16), L, was calculated for each muscle as 0.67L,.
Experimental treatment. Deformation
patterns were
rt=400
ms
imposed on muscles at two strain magnitudes and using
MI (gas) = I
P( t)dt
two timing patterns relative to muscle activation (Table
1). The experimental
design was that of a two-way analywhere P(t) is the muscle force during activation
sis of variance (ANOVA), with two levels of strain (12.5 the mechanical impulse calculated.
and 25%) and two levels of timing (“early” and “late”
A
stretch).
::
Variation of muscle force using different stimulation
timing. To test the hypothesis that muscle damage is di-
rectly related to fiber force, we imposed cyclic linear
length changes of identical magnitude and velocity on
the TA muscle. Cyclic length changes were completed in
400 ms and repeated, along with muscle activation, every
2 s for 30 min for a total of 900 stretches. The only difference between groups (n = ll/group)
was the timing of
the stretch. In the ES group (early stretch), the muscle
was stretched coincident with the onset of muscle activation (Fig. 1). The magnitude of the stretch was 25% of the
TA L, (determined individually
for each muscle), and the
strain rate was 125% LJs. Both the magnitude and rate
of stretch were within the physiological range based on
cat kinematic data, which demonstrate TA strains of 10
and 50% during the E, phase of the gait cycle that are
complete within 0.1 and 0.3 s during walking and galloping, respectively (cf. Fig. 12 of Ref. 8). Because L, was
and MI is
::.
..:
!
i
;
500 ms
1. Sample
contractile
data from early stretch
(A) and late
stretch (B) experimental
groups. Note that both groups receive identical deformation
patterns
(bottom).
However,
due to timing of applied
deformation,
late stretch group experiences
much higher forces (top).
Stippled
area beneath
force record represents
stimulation
duration.
L,!
muscle fiber length.
FIG.
522
MUSCLE
DAMAGE
IS NOT
DUE
TO
HIGH
FORCE
2. Contractile properties of muscles tested at 25% strain
TABLE
Early Stretch
Parameter
Maximum
tetanic tension,
Tetanic
dPldt, g/ms
Time to peak tension,
ms
Twitch
force, g
Half-relaxation
time, ms
Values
measured.
Post
Pre
g
1,325&103
47.3t4.0
25.4+0.62
227g22.2
44.8k2.6
Pre
514t56
7.47t1.08
22.1k2.32
31.9k4.6
35.7t4.0
are means + SE; n = Wgroup.
No significant
difference
dP/dt, rate of rise of twitch and tetanic tension.
was observed
After the treatment
period (30 min), animals were
maintained under anesthesia for 1 h to permit the early
properties
inflammatory
response. Then contractile
were again measured. Animals were killed, and the TA
was excised and submitted for light-microscopic
investigation.
Light microscopy. The TA was frozen in isopentane
cooled by liquid nitrogen (-159°C) and stored at -8OOC
for histochemical
processing. Muscle cross sections (8
,urn thick) taken from the TA midbelly were stained with
hematoxylin
and eosin to observe overall fiber appearance, location of nuclei, and appearance of connective
tissue. Enzyme assays were performed on selected muscles to demonstrate oxidative enzyme activity (succinate
dehydrogenase) (ZZ), glycolytic activity (a-glycerophosphate dehydrogenase) (ZZ), and myofibrillar
adenosinetriphosphatase activity (1). Muscle fibers were classified
as fast oxidative glycolytic, fast glycolytic, or slow oxidative, according to the classification scheme of Peter et al.
(23). The relative area (area fraction) of damaged fibers
(see below) was determined by the stereometric pointcounting technique of Weibel (26). Sampling protocol
consisted of fiber damage measurements
from 2 tissue
blocks/muscle, 3 sections/block, and 5-10 fields/section.
Statistical analysis. For each muscle, pre- and postexercise contractile parameters were obtained. Only postexercise morphological
parameters were obtained from
these specimens, although contralateral
muscles (untested and untreated) were used as controls. For each
muscle, the difference between the pre and post value of
a parameter was expressed as a percentage change in
that parameter. First, it was determined
whether the
percent change was significantly different from zero using a one-sample t test, i.e., whether treatment significantly changed the parameter. For the overall experiment, a two-way ANOVA was used with strain (12.5 vs.
25%) and stretch timing (early vs. late) as the grouping
factors. Assumptions of the ANOVA and t tests (normalTABLE
Late Stretch
1,474?64
54.8t2.60
25.8t0.66
276.7529.6
43.8k2.6
between
any pre-
(Pre)
59Ok52
9.01+1.01
22.9t2.26
44.3t7.4
37.9t4.0
or postcontractile
(Post)
parameters
ity and equality of variances between groups) were explicitly tested using the diagnostic software in SuperAnova
(Abacus Concepts, Berkeley, CA) by plotting
group
means vs. variance and by visual inspection of the ANOVA residuals. To quantify the relative effects of strain
and timing, data were submitted to a stepwise regression
model where the dependent variable was PO after treatment and the independent variables were strain and peak
force achieved during treatment. Data are means t SE.
Significance level was chosen as cy= 0.05. Power analysis
revealed that the statistical power (1 - p) for these experiments exceeded 80% for all parameters except area
fraction of damaged fibers (see RESULTS).
RESULTS
Contractile properties after high vs. low force. Two-way
ANOVA with repeated measures revealed no significant
difference between experimental
groups for any contractile properties measured before and after treatment for
either the 25% strain group (P > 0.4; Table 2) or the
12.5% strain group (P > 0.6; Table 3). The 25 and 12.5%
experiments yielded qualitatively
similar results in terms
of the tension time course during treatment. Results for
the 25% strain group revealed that the initial peak force
of the LS group was 40% greater than that of the ES
group and remained significantly higher throughout the
treatment period (Fig. ZA). Interestingly,
the mechanical
impulse (i.e., the integrated force-time record) experienced by the two groups was also significantly different
(P < 0.01) but in a way that was not expected. Whereas
peak force was always greater in the LS group, the initial
mechanical
impulse of the ES group was 520 go s,
whereas that of the LS group was only 460 g s (Fig. ZB).
The relative difference between groups also changed significantly over time (Fig. 3). Peak force in the LS group
was significantly
greater than in the ES group, but the
magnitude of this difference decreased rapidly during the
l
3. Contractile properties of muscles tested at 12.5% strain
Early Stretch
Parameter
Maximum
tetanic tension,
Tetanic
dPldt, g/ms
Time to peak tension,
ms
Twitch
force, g
Half-relaxation
time, ms
Values
Post
are means
Late Stretch
Pre
g
& SE; n = 8/group.
1,274+54
52.0t3.90
25.020.46
233t42.9
37.9tl.O
No significant
difference
Post
Pre
749t36
15.2k1.75
24.1k1.12
47.5t8.2
37.1k1.9
was observed
between
Post
1,244t34
49.3A3.11
23.9t0.61
253k35.3
38.1k1.3
any pre-
or postcontractile
741k38
13.8k2.30
26.3k0.94
49.2k9.4
38.8k1.4
parameters
measured.
MUSCLE
DAMAGE
IS NOT
Early Stretch
Late Stretch
0
0
a
600
'0"
0
o!
0
0
1
I
10
20
Treatment
0
1
30
Time (min)
2. Mechanical
environment
of muscles
during
experimental
treatment.
A: peak force during experimental
treatment.
Note that late
stretch group experiences
higher forces throughout
treatment
period.
B: mechanical
impulse
during experimental
treatment.
Note that despite the fact that late stretch group experiences
higher peak forces,
early stretch
group experiences
a greater
mechanical
impulse.
Data
shown for 25% strain.
FIG.
first few minutes (Fig. 3A). Thereafter, the ratio between
the groups slowly increased for the next 15 min. The
mechanical impulse of the ES group was greater than
that of the LS group, and the magnitude of this difference increased as a function of time (Fig. 3B). This appeared to be due to the rapidly decreasing peak force in
the LS group. Again, most of the change occurred in the
first few minutes of treatment, rapidly rising to 1.5
within 3 min and increasing in a relatively linear fashion
to 4.0 over the remaining 27 min. This was because the
LS group impulse decreased more rapidly during the first
few minutes than did the ES group (Fig. 2B).
Thus the peak and integrated tensions experienced by
the two experimental groups were different, yet their
contractile properties after 30 min of such treatment
were identical. As an example, P, and tetanic tension
dPldt after cyclic 25% strain decreased by the same
amount in both groups, suggesting an identical decrease
in performance and speed, independent of treatment
tension (Fig. 4).
Contractile properties after high vs. low strain. On the
basis of the observation that varied force did not cause a
change in contractile properties, we tested the effect of
altered strain on contractile properties. Ideally, we would
have applied several different strains to the muscle at
identical forces. However, to accomplish this, the strain
rates required were well out of the physiological range
(200-400%/s). We thus tested the effect of strain per se
by performing the identical experiments described above
DUE
TO
HIGH
523
FORCE
but at 12.5% L, strain. The results were qualitatively similar to the 25% strain experiments; there were no significant differences between strain timing groups for all contractile parameters measured (P > 0.6; Fig. 5). In addition, maximum tetanic tension from all EC groups was
significantly lower (P < 0.05) than that measured after
treatment with only isometric contractions (Fig. 5).
Differential effects of force and strain. Two-way ANOVA of all experimental data grouped by strain (12.5 vs.
25%) and timing (early vs. late stretch) revealed no significant effect of timing (P > 0.7), a significant effect of
strain (P < O.OOl), and no significant strain X timing
interaction (P > 0.9) on P, measured after EC treatment.
To determine the dependence of posttreatment P, on
strain and force, a multiple regression model was developed that demonstrated that strain accounted for ~50%
of the experimental variability in P, measured after EC
treatment, whereas peak tension achieved during treatment accounted for only an additional 8% of the variability. This analysis reinforces the idea that strain rather
than force is more important in determining the magnitude of the tension decrease after EC.
Muscle morphology after eccentric exercise. To investigate the structural basis for the contractile properties
measured after 25% strain, multiple serial sections were
obtained from the midbelly of the treated muscles. On
the basis of previous studies that documented significantly enlarged, pale staining, and rounded fibers of the
fast glycolytic fiber type that were selectively associated
with muscle injury (6,17), we examined the area fraction
of these enlarged fibers.
Comparisons between the ES and LS groups of the
A
1.4
5
1.3
0
t 0
1
1.2
l o.e
l
0.
I
1.1
i
1.0
0
2
2
r
1
20
30
40
I
30
1
40
1
6
.-0
5
III
I
10
0
1.8
l
l
1.6
-E
1.0 !
0
1
10
Treatment
I
20
Time (min)
3. Peak force (A) and mechanical
impulse
(B) ratio between
early and late stretch groups of 25% strain. Note that ratio continually
changes throughout
time course of treatment.
FIG.
524
MUSCLE
2000
1A
0
q
DAMAGE
IS NOT
Pm-Exercise
Post-Exercise
Early
Stretch
Early
Stretch
Late
Stretch
Late Stretch
Group
FIG. 4. Contractile
properties
after experimental
treatment
at 25%
strain. A: maximum
tetanic tension. B: maximum
rate of rise of tetanic
tension (dPldt).
Despite widely different
mechanical
treatments,
there
was no significant
difference
between groups for any contractile
properties measured.
first experiment revealed no significant difference in the
area fraction of damaged fibers (P > 0.6). Unfortunately,
there was a great deal of variability in the morphometric
data. Whereas for the contractile parameters, coefficients of variation ranged from 10 to 35%, morphometric
values for area fraction had a coefficient of variation of
-80%. This of course made it extremely difficult to detect significant differences between groups. Power analysis revealed that to have an 80% chance of detecting a 3%
difference between groups in area fraction of damaged
fibers (given the actual variability of the data), a sample
size of ~60 would be required. Statistical power for the
given data set was thus only 40%.
DUE
TO
HIGH
FORCE
logical differences. If one accepts the first experiment as
showing that muscle damage is not a function of force
during EC, then the conclusion from the second experiment is that muscle damage is a strong function of the
muscle fiber strain that occurs during lengthening of an
activated muscle (i.e., active strain). There is conceptual
support for this idea in the literature based on known
sarcomere structure.
The myofibrillar array is embedded in a complex extrasarcomeric cytoskeletal framework (24). The cytoskeleta1 matrix is joined to the muscle cell basal lamina and
sarcolemma via adhesive connections made by talin, vinculin, and a-actinin, among other proteins, and the integrin superfamily of adhesion receptors (25). It is likely
that active strain that exceeds the limits of these connections may result in cytoskeletal damage. We previously
documented the longitudinal extensions that interrupt
the normal desmin periodic pattern in muscles subjected
to EC. We also recently described subtle changes in fiber
integrity accompanying EC-induced injury (7). It is plausible that cytoskeletal damage may be the first structure
to yield after EC, as has been previously suggested (5).
This could then result in the myofibrillar disruption at
the level of the A band and Z disk, which has been demonstrated in the literature (5, 15, 17). The mechanism for
the damage would thus initially be cytoskeletal disruption followed by myofibrillar derangement. This could, in
principle, be similar to the myofibrillar disruption reported by Horowits and Podolsky (l3), who selectively
irradiated the intermyofilamentous protein titin (which,
by itself, did not result in myofibrillar alterations) and,
on contraction, observed significant misalignment of the
A band and “smearing” of the Z disk. The similarity between their micrographs and those reported by other investigators after EC (e.g., see Refs. 5, 15) is intriguing.
The implications of these findings for exercise involving EC are not clear. It is not appropriate to conclude
2ooo
El
q
Early Stretch
Late Stretch
DISCUSSION
The purpose of this study was to investigate mechanical factors contributing to muscle damage after EC. Previous investigators have suggested that the high forces
associated with EC may be responsible for the damage
observed. Although this is certainly an attractive hypothesis, the current experiments do not support this
idea.
Our data demonstrate that large differences in force
and mechanical impulse during EC did not result in muscle damage differences, as evidenced by identical contractile and morphological parameters. However, large
strain differences did result in contractile and morpho-
0
Control
Isometric
Experimental
12.5%
25%
Group
FIG. 5. Summary
of maximum
tetanic tension generated
by various
treatment
groups. No significant
difference
in maximum
tension generated was observed
between
groups that experienced
early stretch
compared
with late stretch
at either 25 or 12.5% strain. However,
a
significant
difference
in tension
generation
was observed
between
groups strained 25 vs. 12.5% of muscle fiber length. Isometric
data from
Lieber et al. (17). Control,
mean + SE of normal
rabbit tibialis anterior
muscles.
MUSCLE
DAMAGE
IS NOT DUE TO HIGH
simply that low-amplitude joint excursions will result in
low-strain muscle movements.
This is because, for a
given joint rotation, length change varies considerably
among muscles. For example, the rabbit TA and extensor
digitorum longus (EDL) have approximately
the same
moment arm at the ankle joint. However, because TA
fibers are nearly twice as long as EDL fibers (16), EDL
fiber length changes twice as much as TA fiber length for
a given amount of joint rotation. Further studies of muscle-joint interaction are required before isolated muscle
strain measurements can be extrapolated to joint angle
rotations in the intact individual.
Finally, potential difficulties in interpretation
of the
present data should be detailed. First, it is extremely difficult to define the strain that is experienced by the
various experimental groups at the sarcomere level. This
is because the deformation patterns are imposed on muscle-tendon units that are operating at different forces.
Thus, in the case of the ES group where force and stiffness are lower, one might expect that the more compliant
muscle might be strained to a greater extent than the LS
group where force and stiffness are higher. However, this
effect is opposed by the fact that the tendon is also more
compliant in the ES group and thus tends to absorb more
of the deformation. Pilot experiments in which load-deformation curves were generated for rabbit TA tendons
(as in Ref. 18) revealed large tendon stiffness differences
for the two conditions. For example, at low forces similar
to the ES group, tendon stiffness was 300-400 MPa,
whereas at high forces similar to the LS group, stiffness
was l-2 GPa. It is therefore plausible that the two competing effects (increased muscle stiffness at high force
tending to decrease muscle strain in the LS group vs.
increased tendon stiffness at high force tending to increase muscle strain in the LS group) might converge to
make muscle strain in the two groups nearly identical.
Real-time
sarcomere length measurements
with laser
diffraction might specifically answer this question. Finally, because EC-induced injury is more commonly associated with the antigravity
muscles (e.g., quadriceps
and plantarflexors)
than the pretibial
flexor studied
here, it may be inappropriate
to generalize the current
findings to all skeletal muscles. Future studies on the
ankle extensors will specifically address this issue.
In summary, skeletal muscle injury after cyclic EC
with the use of various deformation paradigms suggests
that muscle damage is not simply a function of peak
muscle force but rather is due to the magnitude of the
strain experienced by the muscle during contraction. Further studies are underway to characterize the specific
cellular structures affected by the EC-induced damage.
The authors acknowledge Cindy Brown, Lena Carlsson, AnnaKarin Nordlund, Abbe Zaro, Chris Giangreco, Christy Trestik, and
Mary Schmitz for technical assistance.
This work was supported by the Veterans Affairs, the University of
California, San Diego Academic Senate, National Institute of Arthritis
and Musculoskeletal and Skin Diseases Grant AR-40050, the Research
Council of the Swedish Sports Federation, the Swedish Society of Medicine, and the University of Ume&.
Address for reprint requests: R. L. Lieber, Dept. of Orthopedics
(V-151), Univ. of California San Diego School of Medicine and Vet-
525
FORCE
erans Affairs Medical Center, 3350 La Jolla Village Dr., San Diego, CA
92161.
Received 11 May 1992; accepted in final form 3 August 1992.
REFERENCES
M. H., AND K. K. KAISER. Muscle fiber types: how many
and what kind? Arch. Neural. 23: 369-379, 1970.
2. CURTIN,
N. A., AND R. E. DAVIES. Chemical and mechanical
changes during stretching of activated frog skeletal muscle. Cold
1. BROOKE,
Spring Harbor
3. EISENBERG,
Symp.
Quant.
Biol.
37: 619-626,
1973.
B. R., AND R. MILTON. Muscle fiber termination at the
tendon in the frog’s sartorius: a stereological study. Am. J. Anat.
171: 273-284,1984.
4. EVANS, W. J., AND
J. G. CANNON. The metabolic effects of exerciseinduced muscle damage. Exercise Sport Sci. Reu. 19: W-125, 19%.
5. FRIDBN,
J., U. KJ~RELL,
AND L.-E. THORNELL.
Delayed muscle
soreness and cytoskeletal alterations: an immunocytological study
in man. Int. J. Sports Med. 5: 15-18, 1984.
6. FRIDBN, J., AND R. L. LIEBER. The structural and mechanical basis
of exercise-induced muscle injury. Med. Sci. Sport Exercise 24: 521530, 1992.
7. FRID~N, J.,
R. L. LIEBER, AND L.-E. THORNELL.
Subtle indications
of muscle damage following eccentric contractions. Acta Physiol.
&and.
142: 523-524,199l.
GOSLOW,
G., JR.,R. REINKING,
AND D. STUART. The cat step cycle:
hind limb joint angles and muscle lengths during unrestrained locomotion. J. Morphol.
141: l-42, 1973.
9. GREGOR, R. J., R. R. ROY,W. C. WHITING,
R. G. LOVELY,J. A.
HODGSON,
AND V. R. EDGERTON.
Mechanical output of the cat
soleus during treadmill locomotion: in vivo vs. in situ characteristics. J. Biomech. 21: 721-732, 1988.
10. HARRY, J. D., A. W. WARD, N. C. HEGLUND,
D. L. MORGAN, AND
T. A. MCMAHON.
Cross-bridge cycling theories cannot explain
high-speed lengthening behavior in frog muscle. Biophys. J. 57:
8.
201-208,199O.
11. HOFFER, J.
A., A. A. CAPUTI, I. E. POSE, AND R. I. GRIFFITHS. Roles
of muscle activity and load on the relationship between muscle
spindle length and whole muscle length in the freely walking cat.
Prog. Brain Res. 80: 75-85, 1989.
12. HOROWITS,
R., E. S. KEMPNER,
M.
DOLSKY. A physiological role for titin
cle. Nature Land. 323: 160-164, 1986.
R., AND R. J. PODOLSKY.
13. HOROWITS,
E. BISHER, AND R. J. Poand nebulin in skeletal mus-
The positional stability of
thick filaments in activated skeletal muscle depends on sarcomere
length: evidence for the role of titin filaments. J. Cell Biol. 105:
2217-2223,
14.
1987.
HUXLEY,A. F. Muscle structure and theories of contraction.
Biophys.
15. JONES,
7: 255-318,
Prog.
1957.
NEWHAM,
D. A., D. J.
J. M. ROUND, AND S. E. J. TOLFREE.
Experimental human muscle damage: morphological changes in
relation to other indices of damage. J. Physiol. Land. 375: 435-448,
1986.
16. LIEBER,
R. L., AND F. T. BLEVINS. Skeletal muscle architecture of
the rabbit hindlimb: functional implications of muscle design. J.
Morphol.
17. LIEBER,
199: 93-101,
1989.
R. L., T. M. WOODBURN,
AND J. 0. FRIDBN. Muscle damage induced by eccentric contractions of 25% strain. J. Appl. Phys-
iol. 70: 2498-2507,
1991.
18. LIEBER, R. L., M. E. LEONARD,
C. G. BROWN, AND C. L. TRESTIK.
Frog semitendinosis tendon load-strain and stress-strain properties during passive loading. Am. J. Physiol. 261 (Cell Physiol. 30):
C86-C92,1991.
19. LIEBER, R. L., D. E. SMITH, AND A. R. HARGENS. Real-time acquisition and data analysis of skeletal muscle contraction in a multiPrograms
Biomed.
22: 259user environment. Comput. Methods
265,1986.
20. LIEBER,
R. L., C. L. TRESTIK, M. C. SCHMITZ, AND J. FRIDBN.
Damage following cyclic eccentric contraction is not a function of
muscle stress (Abstract). Trans. 38th Ortho. Res. Sot. 38: 259,1992.
21. MORGAN,
D. L. New insights into the behavior of muscle during
active lengthening. Biophys. J. 57: 209-221, 1990.
526
MUSCLE
DAMAGE
IS NOT DUE TO HIGH
22. PEARSE, A. G. E. Histochemistry;
Theoretical,
and Applied (4th ed.).
London: Churchill Livingston, 1961, vol. 2.
23. PETER, J.B.,R.J. BARNARD, V.R. EDGERTON,~. A. GILLESPIE,
AND K. E. STEMPEL. Metabolic profiles on three fiber types of skeletal muscle in guinea pigs and rabbits. Biochemistry
11: 2627-2733,
24.
1972.
THORNELL,
L.-E., AND M. G. PRICE. The cytoskeleton in muscle
cells
in relation
FORCE
to function.
Biochem.
Sot.
Trans.
19: 1116-1120,
1991.
25. TIDBALL, J. G. Myotendinous junction injury in relation to junction structure
and molecular composition. Exercise Sport Sci. Reu.
19: 419-446,
1991.
26. WEIBEL, E. R. Stereological Methods. Practical Methods
cal Morphometry.
New York: Academic, 1980, vol. 1.
for Biologi-