Clinical Science (1919) 56,41-52
Relaxation rate of constituent muscle-fibre types in human
quadriceps
C. M. W I L E S , A. Y O U N G , D. A. J O N E S A N D R. H. T . E D W A R D S
Department of Human Metabolism, University College Hospital Medical School, London
(Received 30 January 1978; accepted 14 August 1978)
time taken for electrically stimulated contraction
force to fall to that percentage of plateau value
shown by the subscript numerals.
Summary
1. Muscle fibres may be subdivided into type I
(with slow-twitch contractile properties) and type I1
(fast-twitch) depending on their myosin adenosine
triphosphatase activity. In voluntary isometric contractions type I fibres are utilized at low forces
((20% of maximum) whereas type I1 fibres are
recruited in addition at high forces. This physiological recruitment order has enabled us to measure
the relaxation rate of type I and I1 fibres in uiuo in
normal human subjects.
2. Relaxation rate was measured in 16 subjects
from low (10% of maximum) and maximum isometric quadriceps contractions and the musclefibre type composition determined from needlebiopsy specimens in 10 subjects. The relaxation
rate of type I1 fibres was calculated to be twice as
fast as that of type I.
3. It was not possible to estimate, from studies
in 33 quadriceps muscles (25 normal subjects), the
contribution of type I1 fibres to overall fibre area
from the relaxation rate as determined from
electrically stimulated isometric contractions.
Introduction
Human muscle fibres may be subdivided histochemically on the basis of their staining activity for
myosin adenosine triphosphatase (ATPase) after
preincubation at pH 9-4 into type I (low activity)
and type I1 (high activity) (Engel, 1962). Myosin
ATPase activity assayed biochemically in human
fibre fragments is about 2.5 times as great in type
I1 as in type I fibres (Essh, Jansson, Henriksson,
Taylor & Saltin, 1975). Bhrany (1967) showed that
speed of muscle contraction was related to myosin
ATPase activity in several animal species and
studies of single motor units in the cat (Burke,
Levine, Tsairis & Zajac, 1973) demonstrated that
units with fast contraction and relaxation times had
higher enzyme activities than slow units. Furthermore there is evidence of a h e a r relationship
between the relaxation rate of human muscle
studied in uitro and the percentage of type I1 fibres
in the biopsy (Moulds, Young, Jones & Edwards,
1977).
Motor.units comprise either type I or type I1
muscle fibres (Edstrom & Kugelberg, 1968).
During voluntary muscle contractions there is an
orderly recruitment of motor units with increasing
force. Units recruited at low-force thresholds tend
to have slower relaxation rates than those recruited
at high-force thresholds (Milner-Brown, Stein &
Yemm, 1973). In human quadriceps the patterns of
glycogen depletion reflecting fibre activity during
Key words: muscle-fibre types, human quadriceps,
relaxation rate.
Abbreviations: ATPase, adenosine triphosphatase;
MRR, maximum relaxation rate; MVC, maximum
voluntary contraction; SF,,, SF,,, expressions of
Correspondence:Professor R. H. T. Edwards, Department of Human Metabolism, University College Hospital
Medical School, The Rayne Institute, University Street,
London, WClE 655.
47
48
C. M . Wiles et al.
isometric contractions sustained at low and high
forces support the view that type I fibres are
utilized at low forces (e.g. (20% of maximum)
whereas type I1 fibres are recruited at high forces
(Gollnick, Karlsson, Piehl & Saltin, 1974). Relaxation rate might therefore be expected to increase
with force of voluntary contraction, reflecting the
increasing recruitment of ‘fast-twitch’ type I1 fibres.
We have measured the relaxation rate of human
muscle from low-force and maximum voluntary
contractions. According to the recruitment pattern
outlined above, the former should reflect type I
fibre relaxation whereas the latter should reflect the
relaxation of all fibres, both type I and 11. These
measurements have enabled us, in conjunction with
knowledge of the cross-sectional areas of the two
fibre types obtained from needle biopsy, to estimate
the relative relaxation rates of type I and I1 fibres in
uiuo. A preliminary communication of part of this
work has been made (Wiles, Jones, Young &
Edwards, 1978).
Methods
Subjects
In the first part of the study there were 16
subjects (1 1 male, five female) with a mean age of
30.8 years (range 22-47 years). In the second part
of the study there were 25 subjects (21 male and
four female) with a mean age of 30.6 years (range
20-51 years). Eight of these latter subjects also
took part in the first section of the study. All
subjects were healthy and had normal muscle
strength (Edwards, Young, Hosking & Jones,
1977).
Subjects were normal healthy volunteer subjects
who had given their informed consent to all
procedures performed. Approval for needle biopsy
had been obtained from the Committee on the
Ethics of Clinical Investigations at University
College Hospital.
Procedure
The relaxation rate of the quadricep.s muscle
from isometric voluntary and electrically
stimulated contractions of known force was
measured. The subject was seated in a muscle
testing chair (Edwards et al., 1977). A strap
around the subject’s ankle transmitted the force of
each contraction to a strain gauge, the output of
which was amplified and recorded on a rapidresponse oscillograph. All contractions were iso-
Stimulus
marker
7
---ns..
Maximum relaxation rate IMRRI
-7
a x 17.92
tan0 =
450
SF,,-
Farce
SF,.
(%of plateau force losdl0 ms)
Y
lm ms
Time
FIG. 1. Composite Figure to show technique of measuring the time course of relaxation after a brief tetanus at
30 Hz. SF,, and SF,, indicate the time from the last
electrical stimulus to 95% and 50% loss of force
respectively. Maximum relaxation rate (MRR) may be
determined from the differential force record or from the
slope of the initial phase of relaxation (=tan 0).
metric. Relaxation of muscle from electrically
stimulated contractions has previously been
described (Edwards et al., 1977) in terms of the
time taken for force to fall to 95% and 50% of the
plateau value from the last electrical impulse (SF,,
and SF,,). However, relaxation from voluntary
contractions cannot be measured in this way
because of inaccuracies in determining the end of
active contraction. To overcome this problem the
force signal was electronically differentiated with
respect to time and displayed on a second channel
of the oscillograph (Fig. 1).
The differentiator was calibrated against ramps
of known slope produced by a waveform generator.
A differential deflection of 1 cm was equivalent to a
ramp slope of 17.92 cm/s (r = 0.999, P < 0.001)
and the error was less than 2% for any given
differential within the range found in subjects.
Maximum relaxation rate (MRR) from a given
force was calculated as 17-92 x [force differential
deflection (a)/force deflection (b)] (Fig. l), and
gave the maximum percentage of plateau force lost
per 10 ms.
Experimental protocol
The force of a maximum voluntary isometric
contraction (MVC) of the quadriceps was established for each subject. With the help of a visual
force target on the oscillograph the subject could
then make voluntary contractions of 10% and 30%
of MVC. Each contraction was sustained at the
Human musclefunction
target force for about 2 s and on command the
subject then relaxed as completely and instantaneously as possible. A total of ten contractions at
10% of MVC, five at 30% of MVC and five maximum contractions were made by each subject at 1
min intervals (see below for order of contractions)
and the maximum percentage force loss/lO ms was
calculated for each relaxation as above. A reduced
number of high force contractions was used to
avoid fatigue.
Forces equivalent to 10% and 30% of MVC
were also generated by percutaneous electrical
stimulation to exclude the possibility that differences in relaxation rate seen with increasing
voluntary force were due to the elastic properties of
tendons, lower leg or strain-gauge strap. During
electrically stimulated contractions relaxation rate
was expected to be independent of the force of contraction unless the elastic properties of the system
were influencing it.
Five electrically stimulated contractions equivalent to 10% of MVC and 30% of MVC were made
in each subject. The quadriceps was made to
contract by a pulsed electrical stimulus through
flexible pad electrodes applied proximally and
distally to the anterolateral aspect of one thigh
(Edwards el al., 1977). The stimulus was unidirectional 50 ps square-wave pulse administered
for 2 s at 80 Hz. The maximum relaxation rate was
calculated as described above. The voluntary and
electrically stimulated contractions were performed
in an identical order in each subject:
49
sectional area was calculated for each type. Since
the numerical proportion of type I and type I1
fibres was known, the relative contribution of each
type to the overall cross-sectional area could be
calculated. The contribution of type I1 fibres to
overall cross-sectional area has been measured in
duplicate needle biopsies of quadriceps from 17
limbs. The coefficient of variation within each
biopsy pair was 17%.
Results
Relaxation rate from voluntary isometric quadriceps contractions increased with force of contraction (Fig. 2, filled-in circles). The highly
significant increase (P < 0.01, paired t-test) in
relaxation rate from 9.6% of MVC to 33.5% of
MVC is in”contrast to the results from electrically
stimulated contractions of the same magnitude
when there was no significant change (P > 0.1)
from 9.7% of MVC to 32.9% of MVC (Fig. 2,
open circles). Relaxation rates from electrically
stimulated contractions were not significantly
different from those from maximum voluntary contractions (P > 0.1). It is clear therefore that the
increases in relaxation rate with increasing forces
of voluntary contraction are not a function of the
elastic properties of the system but depend upon
the mode (volitional or electrical stimulation) of
muscle ,activation.
(1) Maximum voluntary contraction.
(2) Alternate 30% of MVC voluntary and
stimulated contractions (x 5).
(3) Alternate 10% of MVC voluntary and stimulated contractions (x5).
(4) Alternate maximum and 10% of MVC voluntary contractions (x 5).
Needle-biopsy specimens of muscle were obtained from the lateral part of the same quadriceps
at approximately the junction of the middle and
distal third of the thigh (for technique and initial
preparations of sample see: Bergstrom, 1962;
Edwards & Maunder, 1977). Muscle fibres were
classified as type I or type I1 on the basis of their
histochemical reaction to myosin ATPase after preincubation at pH 9.4 (Dubowitz & Brooke, 1973).
The ‘lesser fibre diameter’ (Dubowitz & Brooke,
1973) of 100 fibres of each type was measured by
using an eyepiece micrometer. Assuming each fibre
to have a circular cross-section of diameter equal to
the ‘lesser fibre diameter’ the mean fibre cross-
0
20
40
60
80
100
Force (% of MVC)
FIG. 2. Relationship between force of isometric quadriceps contractions and relaxation rate. 0 , Voluntary contractions; o , electrically stimulated contractions. Mean
values with bars indicating +SEM are shown; n = 16
subjects.
C. M . Wiles et al.
50
TABLE1. Individual values of percentage type II fibre cross-sectional area and relaxation rates from maximum and
10% of maximum voluntary contractions
The two right-hand columns show the calculated rates for type I1 fibres in each subject and the ratio of relaxation
rates (type II/type I).
Subject
Type I1
Relaxation rate (% of force loss/lO ms)
(% cross-sectional
1 (D.T.)
2 (S.G.)
3 (C.M.)
4 (H.S.)
5 (A.Y.)
6 (M.J.)
7 (C.W.)
8 (G.G.)
9 (J.R.)
10 (T.S.)
Mean
SD
area)
10% of MVC
MVC
45.4
57.9
68.1
56.5
45.3
49.5
32.4
58.6
54.2
51.8
52.0
9.7
6.65
7.17
6.11
8.15
7.71
9.30
8.06
6.20
7.19
9.43
7.60
1.17
10.88
9.44
10.61
12.37
13.49
12.11
13.14
8.98
11.56
12.99
11.56
1.55
1001
: I
80
e
cn
J
.*
**
n
40
*c
.t:*
**:
**
''I
20
0
I
2
4 '
6
8
10
12
14
Relaxation rate (% of plateau force loss/lO ms)
FIG. 3. Relationship between relaxation rate and %
cross-sectional area of type I1 fibres in 25 normal
subjects (eight of whom had measurements made on
both quadriceps). Each point represents the mean value
of three measurements of relaxation rate. r = 0.13; P >
0.4; n = 33.
Ten of the 16 subjects had a needle biopsy of the
same quadriceps muscle as was tested in the above
procedure. Assuming that the relaxation rate from
voluntary contractions of 10% of MVC reflects the
relaxation rate of type I fibres and that that from a
MVC reflects the relaxation rate of type I1 and type
I in proportion to their cross-sectional areas, the
relaxation rate for type I1 fibres can be calculated
for each subject. These data and the ratio of
relaxation rates type II/type I are shown in Table
1; they suggest that, on average, type I1 fibres relax
twice as fast as type I in vivo.
Relaxation rate
for type I1
(calculated: % of
force loss/lO ms)
Relaxation rate ratio
(type II/type I)
15.96
11.10
12.72
15.61
20.40
14.98
23.72
10.94
15.25
16.31
15.70
3.95
2.40
1.55
2.08
1.92
2.63
1.61
2.94
1.77
2.12
1.73
2.08
0.46
We have also measured the relaxation rate from
electrically stimulated quadriceps contractions in
25 normal subjects (of whom eight had bilateral
studies) to see whether this index of contractility
could be used to predict the cross-sectional area of
type I1 fibres in needle-biopsy specimens of the
same muscle. In these studies the mean relaxation
rate from three isometric contractions each induced
by a 1 s, 30 Hz tetanus was calculated as
450/SFS,-,, (Fig. l), the values for SF,, and SF,,
falling within the normal range as defined by
Edwards et al. (1977). The relaxation rate for the
33 quadriceps muscles was 10.3 t 1.14% (mean
value f SD) plateau force loss/lO ms. The coefficient of variation for relaxation rate measurements was 3.6% in 17 tests on six subjects.
In the muscle samples (from the same 33
quadriceps) the frequency of type I1 fibres was
41.7 k 14.00% (mean ? SD) and the percentage
type I1 fibre cross-sectional area was 42.2 f 13.98
(mean f SD). Fig. 3 shows that there was no
significant correlation between relaxation rate and
the percentage cross-sectional area of type I1 fibres
(r = 0.13, P > 0.4). Since both the dependent and
independent variable were subject to error, Bartlett's three-group method for model I1 regression
was used in calculating this correlation coefficient
(Sokal & Rohlf, 1969).
Discussion
In an individual subject the relaxation rate
increases with the force of voluntary contraction.
This increase did not have a simple mechanical
Human musclefunction
TABLE2. Relative characteristics of type I and type 11
muscle fibres (expressed as ratio type IIltype I ) derived
from published work
Superscript numerals indicate the following references:
* Burke, Levine, Tsairis & Zajac (1973), * Biscoe &
Taylor (1967), Andersen & Sears (1964), Buchthal
& Schmalbruch (1970), Thorstensson, Grimby &
Karlsson (1976), 6Moulds, Young, Jones & Edwards
(1977), Esstn, Jansson, Henriksson, Taylor & Saltin
(1975) and * Keul, Doll & Keppler (1972).
Ratio
type Wtype I
Contractile properties
(a) Contraction rate
Cat gastrocnemius’
(twitches,single motor units)
Cat intercostals2V3
(admixture of fast and slow muscle)
Human biceps4
(small fibre groups)
Human quadricepss
(maximum angular velocity)
(b) Relaxation rate
Cat intercostals2*3
(admixture of fast and slow muscle)
Human muscle6
(in uitro)
Present study
Enzyme activities
(a) Myosin ATPase’
(human quadriceps, fibre fragments)
(b) Glycolytic potential’
(phosphofructokinase/aglycerophosphate dehydrogenase)
(c) Lactate dehydrogenase
(human muscle)*
(d) Oxidative potential
(succinate dehydrogenase)’
(e) Creatine kinasee
2.0
1.8
1.9
1.6
1.7
4.3
2.1
2.5
1.9
1.9
0.7
1.6
explanation, e.g. non-linearity of elastic elements in
series with the muscle or recording system, since
relaxation rate was the same when forces of
approximately 10% of MVC and 30% of MVC
were generated by electrical stimulation and were
not significantly different from that from a maximum voluntary contraction. During electrical
stimulation all muscle fibres, of both types, in a
volume of muscle are made to contract, the volume
depending on the stimulus voltage. Since the two
fibre types probably had a homogeneous dispersion
through the quadriceps (Edgerton, Smith &
Simpson, 1975), relaxation from electrically
stimulated contractions should be independent of
the force of contraction, and, in the limit, prove to
be the same as that from a maximum voluntary
contraction.
51
The increase in relaxation rate with voluntary
force of contraction can be explained by current
theories of motor unit recruitment, which indicate
that whereas low-force (e.g. <20% of MVC) isometric contractions are sustained by type I fibres at
higher forces there is a progressive recruitment of
type I1 fibres (Gollnick et al., 1974). This physiological arrangement has enabled us to make
estimates for the relative rates of relaxation of the
two fibre types in human quadriceps in vivo. The
ratio of relaxation rates (type Wtype I) was 2.1/1.
This estimate is based on two main assumptions.
First it is assumed that the types I and I1 generate
the same force per unit cross-sectional area, and
secondly that the relaxation curves for the two
types are similar, i.e. the fastest portions, corresponding to the peak of the differential trace,
occur at the same time.
There are indications from animal studies that
type I1 fibres may generate more force per unit
cross-sectional area (Burke & Edgerton, 1975), but
whether this is also true for human muscle is not
known. If the human type I1 fibres do generate
more force this would have led us to over-estimate
the speed of the type I1 fibres. In contrast, if the
differentiated force peaks for the separate populations of type I or type I1 fibres do not coincide
this will reduce the height of the summated peak
and thus lead an underestimate of the speed of
the type I1 fibres.
Despite these reservations, the present findings
are in general agreement with other studies characterizing the different contractile properties of type I
and I1 fibres in man and the cat (Table 2). The
relative differences in contractile properties between type I and I1 fibres are matched by similar
differences in the activities of certain key enzymes
necessary for the contractile process: thus type I1
fibres, estimated to be about twice as fast as type I
fibres, have roughly twice the myosin ATPase
activity and glycolytic potential as type I fibres.
The ratio for the oxidative potential of the two fibre
types is perhaps higher than might be expected.
This results from a proportion of type I1 fibres
(type IIa: Dubowitz & Brooke, 1973) having an
oxidative capacity intermediate between type I and
the rest of type I1 (IIb) fibres.
Although our present results allow a confident
estimate to be made of the relative relaxation rates
of the type I and type I1 fibres, we were unable to
demonstrate a clear relation between relaxation
rate and fibre-type composition in a group of 33
normal quadriceps (25 subjects). This may be
because the sources of uncertainty mentioned
52
C. M . Wiles et al.
before become so great when comparisons are
made between individuals that no correlation is
obtained. There is, however, another possibility
that, although the relative speeds of the two fibre
types may be constant between individuals, the
absolute speed of a fibre, identified by its histochemical staining characteristics, may vary from
person to person.
Nevertheless, within individuals, changes in
relaxation rate of the quadriceps with force of contraction can be demonstrated and in conjunction
with needle biopsy provide a simple method of
estimating relative contractile properties of the two
main fibre types in vivo.
Acknowledgments
Support from the Wellcome Trust, the Muscular
Dystrophy Group of Great Britain and the Institute
of Sports Medicine is gratefully acknowledged.
References
ANDERSEN,
P. & SEARS,T.A. (1964). The mechanical properties and innervation of fast and slow motor units in the intercostal muscles of the cat. Journal of Physiology (London),
173,114-129.
BARANY,M. (1967) ATPase activity of myosin correlated with
speed of muscle shortening. Journal of General Physiology,
50,197-216.
BERGSTROM, J. (1962) Muscle electrolytes in man.
Scandinavian Journal of Clinical Laboratory Investigation,
14 (Suppl. 68).
BISCOE,T.J. & TAYLOR,A. (1967) The effect of admixture of
fast and slow muscle in determining the form of the muscle
twitch. Medical and Biological Engineering, 5,473-479.
BUCHTHAL,
F. & SCHMALBRUCH,
H. (1970) Contraction times
and fibre types in intact human muscle. Acia Physiologica
Scandinavica, 79,435-452.
BURKE,R.E. & EDGERTON,V.R. (1975) Motor unit properties
and selective involvement in movement. In: Exercise and
Sports Sciences Reviews, vol. 3, pp. 31-81. Academic Press,
New York.
BURKE,R.E., LEVINE,D.N., TSAIRIS,P. & ZAIAC,F.E. (1973)
Physiological types and histochemical profiles in motor units
of the cat gastrocnemius. Journal of Physiology (London),
234,723-748.
DUBOWITZ,V. & BROOKE,M.H. (1973) Muscle Biopsy: A
Modern Approach, p. 99. Saunders, London.
EDGERTON,
V.R., SMITH,J.L. & SIMPSON,
D.R. (1975) Muscle
fibre type populations of human leg muscles. Histochemical
Journal, 7,259-266.
EDSTROM,L. & KUGELBERG,
E. (1968) Histochemical composition, distribution of fibres and fatiguability of single motor
units. Journal of Neurological and Neurosurgery and
Psychiatry, 3 1 , 4 2 4 4 3 3 .
EDWARDS,R.H.T. & MAUNDER,C.A. (1977) Muscle biopsy.
Hospital Update, 3,569-581.
EDWARDS,
R.H.T., YOUNG,A., HOSKING,G.P. & JONES,D.A.
(1977) Human skeletal muscle function: description of tests
and normal values. Clinical Science and Molecular Medicine,
52,283-290.
ENGEL, W.K. (1962) The essentiality of histo- and cytochemical studies of skeletal muscle in the investigation of
neuromuscular disease. Neurology, 12,778-784.
ESSEN, B., JANSSON,E., HENRIKSSON,
J., TAYLOR,A.W. &
SALTIN,B. (1975) Metabolic characteristics of fibre types in
human skeletal muscle. Acla Physiologica Scandinavica, 95,
153-165.
GOLLNICK,
P.D., KARLSSON,
J., PIEHL,K. & SALTIN,B. (1974)
Selective glycogen depletion in skeletal muscle fibres of man
following sustained contractions. Journal of Physiology
(London), 241,59-67.
KEUL, J., DOLL,E. & KEPPLER,D. (1972) Energy metabolism
of human muscle. In: Medicine and Sport, vol. 7, pp. 11-12.
S. Karger, Basel.
MILNER-BROWN,
H.S., STEIN,R.B. & YEMM,R. (1973) The
orderly recruitment of human motor units during voluntary
isometric contractions. Journal of Physiology (London), 230,
35 9-3 70.
MOULDS,R.F.W., YOUNG,A., JONES, D.A. & EDWARDS,
R.H.T. (1977) A study of the contractility, biochemistry and
morphology of an isolated preparation of human skeletal
muscle. Clinical Science and Molecular Medicine, 52, 291297.
SOKAL,R.R. & ROHLF, F.J. (1969) Biomefry, pp. 481-486.
W.H. Freeman, San Francisco.
THORSTENSSON,
A., GRIMBY,G. & KARLSSON,J. (1976)
Force-velocity relations and fiber composition in human knee
extensor muscles. Journal ofApplied Physiology, 40,12-16.
WILES,C.M., JONES,D.A., YOUNG,A. & EDWARDS,
R.H.T.
(1978) Relaxation rate of type I and type I1 fibres in human
skeletal muscle. Clinical Science and Molecular Medicine,
54,33P.