Presented at NACOB 98: North American Congress on Biomechanics Canadian Society for Biomechanics - American Society of Biomechanics University of Waterloo Waterloo, Ontario, Canada August 14-18, 1998

---

CONTRIBUTIONS FROM SLOW- AND FAST-TYPE MUSCLE FIBERS
TO STRETCH RESPONSES IN THE HUMAN TRICEPS SURÆ

J.A. Hoffer¹, A.G. Cresswell², A. Thorstensson^2
1 School of Kinesiology, Simon Fraser University, Burnaby, B.C., V5A 1S6 Canada
² Departments of Neuroscience, Karolinska Institute and Human Biology,
University College of Physical Education and Sports, Stockholm, S-114 86 Sweden

INTRODUCTION

When an active muscle is rapidly stretched, the elastic properties of passive elements and of engaged cross-bridges cause an immediate, large rise in tension. The short-range stiffness and force yielding properties of isolated slow- and fast-type muscle fibers have been shown to differ. We have investigated the expression of fiber-type-dependent stretch properties in the voluntarily activated human triceps suræ.

REVIEW AND THEORY

When the active cat soleus muscle is rapidly stretched it exhibits high short-range stiffness, but its force yields abruptly if the stretch exceeds 1% of muscle length (Rack and Westbury, 1974). Malamud et al. (1996) found the same stretch behavior in skinned type I muscle fibers; however, they also found that type II fibers have lower short-range stiffness and usually do not yield.

Inconsistent conclusions have emerged from previous studies on human ankle musculature. Allum and Mauritz (1982) showed considerable yield when stretching the ankle plantar flexors at low levels of voluntary effort (6-14 Nm), whereas Sinkjær (1997) reported that these muscles did not yield, except during the first stretch in a potentiated muscle. In both studies, compensation for muscle yield was attributed to stretch reflex activation, as was previously shown for the soleus muscle in the decerebrate cat (Nichols and Houk, 1976; Hoffer and Andreassen, 1981).

The inconsistent information available for human muscles may be attributed to several factors. Force yielding, if indeed present, might be masked by passive elastic components, inertial torques caused by the large moving mass and/or asynchronous stretching of fibers in the various muscles involved.

To investigate fiber-type-dependent stretch properties in the human triceps suræ, it is necessary to i) provide very secure coupling to the foot, ii) study the full range of muscle activation levels, and iii) produce rapid joint rotations. This requires a strong motor and footplate with a low moment of inertia, as rapid rotations generate inertial torques that can obscure the tension produced by the stretched muscle fibers. With such a system, and using simultaneous accelerometer recordings, the inertial component of the total system can be estimated and subsequently subtracted from the measured ankle torque.

PROCEDURES

Nine adults (4F, 5M; 48-91 kg) lay prone on a bench with legs extended and one foot tightly secured to a servo-controlled motor aligned with the ankle axis. Subjects maintained a series of 3-5 s long, isometric plantar flexions at torques ranging from 5 to 65-237 Nm. Once during each contraction, the motor rapidly dorsiflexed the ankle either 10° in 100 ms or 20° in 150 ms (peak velocity [approximately equal to] 170°/s), and after 500 ms returned to the initial ankle angle. Soleus and medial gastrocnemius EMG, ankle torque, angular position, angular velocity and foot acceleration signals were sampled at 2 kHz and analyzed using Spike2 and MATLAB. A least-squares approach gave a scalar factor that minimized the difference between the measured torque and foot acceleration for the 5 Nm torque trials. This factor was used to compute the inertial torque component that needed to be subtracted from each trial.

RESULTS

Removal of the inertial torque component resulted in a family of net muscle torque curves (Fig. 1A). These curves, after subtraction shifting to a common initial torque value (Fig. 1B), revealed that:
1. For all subjects at all levels of maintained effort, muscle torque rose steeply upon stretch and reached a peak by [approximately equal to] 25 ms, when the ankle had been rotated < 2°.
2. The initial peak increased nonlinearly with increasing level of effort.
3. For low levels of maintained effort, muscle torque yielded after reaching the first peak.
4. At increasingly high levels of maintained effort, there was progressively less yield and the slope of renewed force development increased as a function of the level of background muscle activation.
5. Reflex EMG activity started 40-60 ms after stretch onset.
6. A second, more variable torque peak occurred 70-130 ms after onset of stretch (earliest for highest initial torque levels).

Figure 1. A: family of net muscle torque curves (Subject 8). Three trials were made at each level of effort (0-100 Nm). Time 0 indicates start of ankle rotation. B: the active muscle torque curves (5-100 Nm) shown after alignment of their initial values to compare incremental stretch responses. C: angular displacement of the ankle joint.

DISCUSSION

It was found that the human triceps suræ exhibited a short-range stiffness region followed by yielding behavior. This was most clearly evident for low levels of maintained voluntary effort. The progressive reduction in yield seen for increasing levels of effort was interpreted to result from intrinsic muscle properties, as it preceded any additional force recruitment by reflex action.

In light of Malamud et al.’s (1996) findings for single fibers, it is suggested that the early rise in muscle torque mainly reflects passive muscle stiffness and the high short-range stiffness of engaged cross-bridges in slow-type muscle fibers. The peak and subsequent yield at low levels of effort indicate the time when the cross- bridges are stretched beyond their elastic limit. At higher levels of effort, the increasing participation of fast-type fibers provides considerable yield compensation before and until any reflex action takes place.

REFERENCES

Allum, J.H.J. and Mauritz, K.H. J. Neurophysiol. 52:797-818, 1982.

Hoffer, J.A. and Andreassen, S. J. Neurophysiol. 45:267-285, 1981.

Malamud, J.G., Godt, R.E. and Nichols, T.R. J. Neurophysiol. 76:2280-2289, 1996.

Nichols, T.R. and Houk, J.C. J. Neurophysiol. 39:119-142, 1976.

Rack, P.M.H. and Westbury, D.R. J. Physiol. Lond. 24:331-350, 1974.

Sinkjær, T. Acta Physiol. Scand. Suppl. 170:1-28, 1997.

ACKNOWLEDGMENTS

JAH was a visiting professor funded by the Swedish Central Association for the Promotion of Sports. Y. Chen and K. Daggfeldt provided helpful suggestions.