| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Physiology (2002), 538.1, pp. 209-218
© Copyright 2002 The Physiological Society
DOI: 10.1113/jphysiol.2001.012785
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
To investigate the possibility of a peripheral contribution to the perturbations of force sensation reported to occur after eccentric exercise, responses to passive and active tension were recorded from Golgi tendon organs in the medial gastrocnemius muscle of the anaesthetised cat, before and after a series of eccentric contractions. After the eccentric contractions, nearly all tendon organs commenced firing at a shorter muscle length during slow passive stretch than before, probably because of a rise in whole muscle passive tension. There was a small drop in the sensitivity to incremental tension, but no mean change in tension threshold. Following the eccentric contractions, there was a small, but not significant, increase in tendon organ sensitivity to active tension, which was graded using a method of optimised, distributed stimulation of divided ventral roots. Sensitivity was estimated as the mean response over a range of tensions and as the change in discharge rate in response to incremental tension. The experiments provided the opportunity of comparing tendon organ sensitivities to graded passive and active whole muscle tension. In agreement with previous work in which whole muscle nerve stimulation was employed, little difference was found. It was concluded that the peripheral contribution to perturbations of force perception after eccentric exercise is likely to be small and that the centrally derived sense of effort plays the dominant role. Tendon organs appear to be remarkably reliable in signalling whole muscle tension, whether passive or active, and even after the muscle's force production has been disturbed by fatigue or eccentric exercise.
(Received 28 May 2001; accepted after revision 4 October 2001)
Corresponding author J. E. Gregory: Department of Physiology, PO Box 13F, Monash University, Victoria 3800, Australia. Email: ed.gregory{at}med.monash.edu.au
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
It is generally accepted that there are peripheral and central components contributing to force perception, or the sensation of the force produced by the muscles (McCloskey et al. 1974; McCloskey, 1978; Gandevia, 1996). It is thought that Golgi tendon organs provide the principal afferent input for a peripherally derived sense of muscle tension, while a centrally derived sense of effort originates from a corollary discharge of the motor command to the muscles (McCloskey, 1981).
A variety of evidence suggests that the centrally derived sense of effort is a major contributor to force perception. This evidence comes mainly from observations of errors in force estimation in certain pathological conditions and in fatigued or experimentally paralysed muscle. The error is in the direction of an overestimation of the force actually produced, in circumstances where it is assumed, but perhaps not always correctly, that there has been no change in the sensitivity of the tension sensors in the muscle. It is therefore presumed that the error is central in origin.
The contribution of the sense of tension to force perception is less clear. It is known that information from tendon organs reaches the cerebral cortex, this being a prerequisite for conscious sensation (McIntyre et al. 1984). It has been shown, in carefully designed experiments, that the sense of tension can be used for force estimation (Roland & Ladegaard-Pedersen, 1977), but the contribution it makes to force perception under various conditions remains a matter of speculation.
Muscle force may decline after concentric or isometric exercise because of fatigue, and this is accompanied by perturbations of force perception. In eccentric exercise, where the active muscle is forcibly lengthened, components of the force drop are thought to result from the disruption of sarcomeres and damage to some muscle fibres, which may eventually develop localised contractures and lose their ability to generate active tension. The experimentally measured consequences of eccentric exercise are a decline in force-generating capacity that persists long after the metabolic effects of fatigue have gone, an increase in the muscle length at which maximal isometric tension is generated (Wood et al. 1994; Jones et al. 1997; Whitehead et al. 1998), and an increase in the resting tension in the passive muscle (Whitehead et al. 2001).
Two recent studies have shown that eccentric exercise also results in proprioceptive perturbations (Saxton et al. 1995; Brockett et al. 1997). In one of these (Brockett et al. 1997), there was a suggestion that the observed errors in force estimation were at least partly peripheral in origin. To investigate whether eccentric exercise results in any change in the signalling properties of the peripheral tension sensors, we have recorded the responses of tendon organs in the cat medial gastrocnemius muscle to passive and active tension before and after periods of eccentric contractions. The motivation for these experiments came originally from work in which we showed that passive tension rises after eccentric exercise (Whitehead et al. 2001), and it was of interest to know whether this additional tension is signalled by tendon organs.
While this was the main purpose of the experiments, the results provided the opportunity to compare directly the sensitivity of tendon organs to passive tension with that to the active tension produced by graded contraction of the whole muscle, and thus add a comment to an old debate, now largely settled, about the primary function of tendon organs. There was found to be little difference in the sensitivities measured in the way we have done here.
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The experiments were carried out on 11 cats, of both sexes, weighing between 3.0 and 5.4 kg. The experiments were undertaken with approval from the Monash University Committee for Ethics in Animal Experimentation. Anaesthesia was induced with 40 mg kg-1 sodium pentobarbitone (Nembutal, Rhone Merieux), administered I.P., and maintained with 12 mg ml-1 sodium pentobarbitone, given I.V. as needed to maintain a deep level of anaesthesia, via a cannula in a cephalic vein. The trachea was cannulated and expired CO2 concentration was monitored. Artificial ventilation was employed occasionally as necessary. Rectal temperature was monitored and the animal heated with a feedback-regulated electric blanket if the temperature fell below 38 °C.
A laminectomy was performed to expose dorsal and ventral roots L7 and S1, which were cut where they entered the spinal cord. The left leg was dissected to expose the calf muscles. Markers were placed on the tibia and the distal tendon of medial gastrocnemius, and with the leg in the position it would occupy during the experiment, the distance between these markers was measured with the ankle fully flexed. This enabled the length of medial gastrocnemius, the muscle to be used, to be referred to its maximal physiological length (Lmax).
The muscle was dissected free and an extensive denervation of the hind limb carried out, leaving only the medial gastrocnemius, with its nerve and tendons intact, attached to the calcaneum. The calcaneum was severed, leaving a piece attached to the tendon. A hole 2 mm in diameter was drilled into this piece, through which was passed a length of threaded rod, and to which it was secured between a pair of nuts and washers. The other end of the rod was attached, with a strain gauge interposed, to an electromagnetic muscle stretcher, which was regulated by position feedback from a linear variable differential transformer displacement transducer.
The leg was fixed in position by means of opposed clamps or metal pins in the hips, knee and ankle. Exposed tissues were covered with mineral paraffin oil retained in baths fashioned from skin flaps.
Dorsal roots were divided into filaments containing single functional Golgi tendon organ afferents, which were identified by their conduction velocity and responses to muscle twitches and passive stretch. Filaments were mounted on a multiple monopolar electrode array. Nerve impulses, together with muscle length and tension, were recorded, stored and processed using an analog-to-digital converter (PCI-MIO-16E-4, National Instruments, Austin, TX, USA) and custom-designed software written in the program IgorPro (WaveMetrics, Lake Oswego, OR, USA).
Muscle active tension was produced by electrical stimulation, through bipolar platinum electrodes, of the muscle nerve or, as occasion demanded, of large ventral root divisions or of ventral root filaments containing a few or single motor units.
In some experiments, muscle contractions were produced by distributed, sequential stimulation of five or six ventral root divisions, which generated approximately equal tensions and together contained the entire motor supply to the muscle. This allowed tension to be graded by varying the stimulus rate over a wider range, for a similar amount of tension oscillation at unfused frequencies, than is possible with synchronous stimulation of the whole motor supply.
A refinement of the technique originally described by Rack & Westbury (1969) was used here (Wise et al. 2001). With this refinement, it is not necessary that equal tensions are generated by the ventral root divisions being stimulated. Tension inequalities are compensated for by suitable adjustment of the intervals between stimuli delivered to each ventral root division. The intervals between stimulus channels are thus not equal, but each channel delivers stimuli at a constant rate. The intervals were optimised by computer to minimise ripple in the tension recorded at the tendon (Brown et al. 1999). Stimulus rates between 7 and 50 pulses s-1 (pps) per channel were used, allowing tension to be graded over a 5- to 10-fold range in what was considered to be a nearly physiological manner.
This method of stimulation was particularly appropriate in the present circumstances (i.e. where a muscle composed of different fibre types is to undergo repeated eccentric contractions), and the tension decline owing to this may not have been the same in each division. Repeated re-optimisation of the interstimulus intervals allowed muscle tension to remain optimally smooth during distributed stimulation without physical re-arrangement of ventral root filaments.
In all experiments, recordings from tendon organs were made before and after a series of eccentric contractions of the whole muscle, spaced 20 s apart. Each contraction consisted of a 400 ms tetanic contraction at a stimulation rate of 80 pps with a 6 mm muscle stretch at 50 mm s-1 starting 150 ms after the start of the tetanus. The stretch started at a length usually 3, but up to 6 mm shorter than Lmax. Between 50 and 150 eccentric contractions were performed, until there was a significant drop in maximal tension and an increase in Lmax.
In some experiments, where the whole muscle force was greater than the equipment could withstand, the motor supply was divided into two approximately equal parts, which were stimulated alternately. The length-tension relationship was measured before and after the series of eccentric contractions, using stimulation at 80 pps for 250 ms.
At the end of each experiment, the animal was killed with 300 mg sodium pentobarbitone, given I.V.
Statistical analyses were performed using the program DataDesk (Data Description, Ithaca, NY, USA). Means are quoted with S.E.M.
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In all experiments, the series of eccentric contractions was effective in reducing the maximal tetanic tension produced by the muscle, in increasing the length at which maximal tension was produced, and in increasing the passive tension in the unstimulated muscle at all lengths above that at which it was completely slack (Whitehead et al. 2001). The active tension drop was between 40 and 67 % (mean 52 ± 3 %) in different experiments, the shift in the peak of the length-tension relationship was from 3.3 to 6.3 mm (mean 4.4 ± 0.3 mm), and the increase in passive tension at Lmax was 6-61 % (mean 33 ± 7 %). The maximal increase in passive tension was actually at a shorter length than Lmax, as can be seen in Fig. 1.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. Illustration of the procedure used to test the sensitivity of tendon organs to passive stretch of the muscle before and after a series of eccentric contractions The bottom trace shows the slow triangular stretch and release from a muscle length of maximum physiological length (Lmax) - 20 mm to Lmax. The pair of traces in the middle show the tension recorded at the tendon: the tension is higher at all muscle lengths after the eccentric contractions (continuous trace) than before (dashed trace). The upper pair of traces show the response of one tendon organ as instantaneous impulse frequency before (open symbols) and after (filled symbols) the eccentric contractions. Downward-pointing arrows indicate where the threshold tensions were measured from the respective tension traces, before and after the eccentric contractions. The upward-pointing arrow shows where the maximal difference in tension occurred, before and after the eccentric contractions. | ||
Responses to passive stretch
The response of 43 tendon organs to passive tension in the unstimulated muscle was tested with a triangular stretch and release at 1 mm s-1 from a length of Lmax - 20 mm to Lmax. Responses were recorded before and after the series of eccentric contractions (Fig. 1). At Lmax - 20 mm, the muscle lay completely slack, both before and after the eccentric contractions. In nearly all cases, firing commenced at a shorter muscle length after the eccentric contractions than before, and there was an overall increase in the response to the stretch.
Tendon organ response was plotted against both muscle length and passive tension during the slow stretch (Fig. 2). From these plots, tendon organ length and tension thresholds were determined; that is, the lengths and tensions at which the receptors discharged the first impulse during the stretch, before and after the eccentric contractions. Thresholds determined in this way were well defined and repeatable.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. Discharge of one tendon organ plotted against tension (A) and length (B) during stretch and release of the passive muscle Responses are shown as instantaneous impulse frequency before (open symbols) and after (filled symbols) eccentric contractions. These plots are from the same data used in Fig. 1. For this receptor, there was no change in the tension at which the discharge commenced, but there was a reduction in the sensitivity measured as the slope of the tension-firing rate relationship, after the eccentric contractions (A). After the eccentric contractions, the discharge started at a muscle length 3.4 mm shorter than before (B). | ||
For the 40 tendon organs where the whole muscle had undergone eccentric contractions, length thresholds for all except two decreased after the eccentric contractions by between 0.3 and 5.7 mm (Fig. 3). One of the remaining two showed no change and the other increased by 0.3 mm. The mean change in length threshold was a significant decrease of 2.4 ± 0.2 mm (Student's t test, P < 0.0001). A regression line fitted to the data in Fig. 3 had a slope significantly greater than zero (P = 0.003), indicating some association between the length and tension threshold changes.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. Change in tendon organ tension and length thresholds in response to passive stretch after eccentric muscle contractions The abscissa shows, for each of 40 receptors (filled circles), the difference in the length at which the discharge in response to the slow stretch commenced before and after eccentric contractions. Negative values indicate that the discharge started at a shorter length after the eccentric contractions. The ordinate shows the change in the tension at which firing commenced, negative values indicating that the discharge started at a lower tension after eccentric contractions. For each experiment, values on the ordinate have been normalised to the tension recorded at Lmax at the end of the slow stretch. The open square shows the mean values for these 40 receptors. The open circles show threshold changes for an additional three tendon organs, the motor units with muscle fibres inserting into which had not been subjected to eccentric contractions. | ||
The change in tendon organ length threshold was not strongly correlated with the shift in the length at which optimal whole muscle tension was produced (r2 = 7 %), and the slope of the relationship was not significantly greater than zero. This was possibly because of the large variation in the change in length threshold for receptors in any particular experiment, combined with the restricted range of values for the shift in tension optimum. However, on the assumption that if the optimal length did not change, this should be associated with a lack of change in the threshold length of tendon organ firing (that is, that the curve should pass through the origin), the slope became significantly different from zero (P < 0.0001).
Tension thresholds measured before the eccentric contractions were between 0.1 and 12.5 N (mean 4.0 ± 0.5 N). Changes in threshold showed no particular trend. Some tendon organs had a higher threshold after the eccentric contractions, while for others it was lower (Fig. 3). In order to allow for different-sized muscles, tension thresholds were normalised to the passive tension at Lmax measured before the eccentric contractions, during the triangular stretch and release. In all cases, the change was quite small, 10 % or less of the passive tension at Lmax, which was between 24 and 44 N in different animals. The mean change in tension threshold was an increase of 0.6 ± 0.6 % of the tension at Lmax, or 0.1 ± 0.2 N without normalisation. Neither was significant at the 0.05 level (t test).
For three tendon organs, the ventral root was repeatedly divided to isolate fine filaments that when stimulated produced a strong response from the receptor. These were probably single units or two to three units together. The object was to separate out the motor units having muscle fibres inserting into the receptor, together with as few as possible of those that did not. These filaments were then excluded from stimulation during the eccentric contractions. These three tendon organs are shown by the open circles in Fig. 3. Unlike the majority of the 40 other receptors, for none of them was there any decrease in length threshold. Our explanation for this result is that the decrease in length threshold is a consequence of the increase in passive tension following eccentric contractions. If the increased passive tension does arise in muscle fibres and not elsewhere, it will be present only in those fibres that have undergone eccentric contractions, not in fibres belonging to motor units spared during stimulation. A decrease in length threshold will be seen only for tendon organs with muscle fibres affected by the eccentric contractions inserting into them.
Threshold is one measure of receptor sensitivity. Another is the relationship between the change in tension, when tension is varied over a range, and the change in the rate of receptor discharge. This relationship was measured here as the slope of a straight line fitted to the plot of firing rate versus passive tension during triangular stretch (an example of which is shown in Fig. 2A), over the range of increasing tension from threshold to peak. The sensitivity after the eccentric contractions was measured up to the same tension value as before, stopping before the peak, which was of course higher after the eccentric contractions. Figure 4 plots the sensitivities (i.e. the increase in firing rate per unit tension increase) before and after eccentric contractions. There was a small reduction in sensitivity measured in this way after the eccentric contractions, and the mean sensitivity dropped from 2.93 ± 0.15 to 2.58 ± 0.11 impulses s-1 N-1. To assess whether the change was significant, the ratio of the sensitivity after the eccentric contractions to that before was calculated, and was found to be between 0.44 and 1.54 for individual afferents. The mean ratio (0.91 ± 0.03) was significantly less than unity (t test, P = 0.01). That is, there was a small but significant decrease in tendon organ sensitivity to passive tension after the eccentric contractions.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. Tendon organ sensitivity to passive muscle stretch Sensitivity was estimated as the relationship between impulse frequency and tension during slow stretch from Lmax - 20 mm to Lmax and measured as the slope of a straight line fitted to plots like those shown in Fig. 2A. Filled circles show values for 40 tendon organs before (abscissa) and after (ordinate) eccentric contractions. The open circles show values for the three receptors not subjected to eccentric contractions. | ||
It was thought possible that because the passive tension rose to a higher level after the eccentric contractions than before, the resulting difference in the rate of rise of tension might be partly responsible for the changed sensitivity because of some adaptive change in receptor response. To test this possibility, some receptors were re-tested using a smaller muscle stretch after the eccentric contractions, so that the peak passive tension reached the same level at the same time, before and after the contractions. This did not change the measured sensitivity.
Responses to active contraction
Responses of tendon organs to active tension were recorded while the tension was graded using three different methods. The first was to vary stimulus strength at constant, maximal frequency, thus engaging a variable proportion of the muscle. The second was to stimulate maximally at a range of muscle lengths, thus utilising the muscle's length-tension relationship to vary tension. The third was to use optimised distributed stimulation of the ventral root divided into five or six portions at a range of frequencies at maximal stimulus strength and a fixed length.
Each method produced a reasonably smooth gradation of tendon organ response. Distributed stimulation was found to be the most satisfactory method and was used to study 15 tendon organs, while the other methods provided a useful comparison in six more. It should be noted that although smooth tension profiles may be obtained at low stimulus rates using distributed stimulation, the resulting tendon organ rate profiles are not necessarily equally smooth, since the ventral root divisions prepared for distributed stimulation do not necessarily affect a particular tendon organ in proportion to their contribution to the tension seen at the muscle's tendon. Typical records of tendon organ responses to distributed stimulation at three frequencies are shown in Fig. 5. From these records, the mean impulse frequency was measured from the adapted response during the last 1 s of the 2 s period of stimulation and plotted against the mean tension measured over the same period.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. Response of a tendon organ to optimised, distributed stimulation of the muscle The response (upper record in each panel), shown as instantaneous impulse frequency during stimulation, at the rates shown (10, 20 and 30 pps), of each of five ventral root divisions that were not equal in terms of the tension they produced when stimulated separately. Adjustment of the stimulus intervals between divisions produced minimal ripple in the tension profile (lower trace in each panel), but less than optimal smoothness in the receptor's response. | ||
Figure 6 shows examples of the plots of mean afferent discharge rate against mean tension before and after eccentric contractions, when the tension was graded by each of the three methods. The extent of the range of common tension shared by the before and after points in each plot, or the degree of tension overlap, depended upon how much the tension was reduced following the eccentric contractions. Inspection of these plots for all of the receptors studied did not reveal any obvious, overall change in sensitivity to active muscle tension resulting from the eccentric contractions.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 6. Tendon organ responses to graded isometric muscle tension before and after eccentric contractions A, the mean frequency of discharge of one tendon organ during stimulation of the muscle at maximal rate and strength, where tension was graded by changing muscle length, before (open symbols) and after (filled symbols) eccentric contractions. B, responses of another tendon organ recorded at a single muscle length, where muscle tension was graded by changing stimulus strength at maximal frequency, thus engaging a variable proportion of the muscle. C, the responses of a third tendon organ to optimised, distributed stimulation over a range of frequencies at a single length and maximal strength. All methods produced a smooth gradation of receptor response. There was no obvious change in responsiveness resulting from the eccentric contractions. Tendon organ sensitivity measurements were made from straight lines fitted to the curves over the region of common tension before and after eccentric contractions, as illustrated in C. | ||
The responses to distributed stimulation were analysed further by fitting straight lines to the plots like those shown in Fig. 6C over the region of tension overlap before and after the eccentric contractions. Two measures of a change in sensitivity were derived from this. Firstly, the change in the sensitivity to varying tension was estimated as the difference in the slope of the fitted lines. Secondly, the overall change in response was estimated by measuring the average vertical displacement of the fitted lines, representing the change in average firing rate to the tension in the fitted range.
For the 15 receptors there was a small, but not significant, increase in the mean sensitivity estimated in both ways. The mean sensitivity to varying tension increased from 1.59 ± 0.22 impulses s-1 N-1 (range -0.26 to 2.86 impulses s-1 N-1) before the eccentric contractions to 1.74 ± 0.25 impulses s-1 N-1 (range -0.13 to 3.67 impulses s-1 N-1) after, and the mean of the individual average firing rates increased from 98.4 ± 9.5 impulses s-1 (range 23.5-178.6 impulses s-1 ) to 108.5 ± 13.4 impulses s-1 (range 3.4-208.7 impulses s-1). These values quoted for the average firing rates of individual receptors, as distinct from their differences before and after the eccentric contractions, are somewhat arbitrary and depend on the range over which tension overlapped, which was different in each experiment.
Active and passive sensitivity
For nine tendon organs in two experiments, data on both active and passive sensitivities were available, allowing a direct comparison to be made between the two.
When the passive responses to slow stretch and the active responses to distributed stimulation were plotted together, little difference in sensitivity was apparent overall from inspection of the plots. There were, however, quite large differences between individual receptors, some showing a higher passive than active sensitivity and others vice versa.
Figure 7A shows a typical example, where the response to the lengthening phase only of the slow triangular stretch-and-release movement described earlier has been plotted along with the response to optimised, distributed stimulation. In a similar way to that described for Fig. 6, straight lines were fitted to the responses for each tendon organ over the range of common tension, or tension overlap, extending in this case from the tension produced by the lowest rate of distributed stimulation used, 7 pps, to the tension recorded at the peak of the passive stretch (i.e. at Lmax). Active tensions of similar magnitude to the passive tension measured at Lmax were produced by distributed stimulation at rates of about 10-20 pps. Stimulation at 7 pps produced tension of a similar magnitude to the passive tension recorded at about Lmax - 1 mm to Lmax - 5 mm.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 7. Comparison of tendon organ sensitivity to passive and active tension A, the responses of one tendon organ to passive stretch (filled circles) and distributed stimulation (open circles) are shown as instantaneous frequency and mean rate, respectively. Lines have been fitted to each curve over the range of common tension. The arrow shows where the average difference in response was measured as the vertical separation of the fitted lines, the values of which are plotted on the ordinate in B. B, the abscissa shows the difference in sensitivity to active and passive tension, measured as the difference in slope of the fitted lines illustrated in A. The ordinate shows the average difference in active and passive response over the same tension range. Positive values indicate that active sensitivity is higher than passive sensitivity. Filled circles show individual values for nine tendon organs. Mean sensitivity (open square) was higher for passive than for active tension. | ||
As before, two measures of receptor sensitivity were derived from the fitted lines. Firstly, the slope of the lines was calculated, as a measure of the sensitivity to varying tension. Values for the nine receptors ranged between 0.74 and 2.16 impulses s-1 N-1 (mean 1.49 ± 0.13 impulses s-1 N-1) for active tension and between 1.50 and 4.08 impulses s-1 N-1 (mean 2.27 ± 0.25 impulses s-1 N-1) for passive tension. Secondly, the average firing rate was calculated as a measure of the overall sensitivity of the receptor, or average response to the tension in the fitted range. As noted before, the absolute values of this measure depend upon the range of tension overlap between the active and passive curves. Since the overlap was different in the two experiments, only the difference between the active and passive average responses was calculated, not the absolute values.
When sensitivity was measured as the average response over similar ranges of tension, the sensitivity of this sample of tendon organs to passive tension was found to be as great as that to actively generated tension. When sensitivity was measured as the increment in firing rate per unit tension, the sensitivity to passive tension was significantly greater than to active tension (t test, P = 0.03), the mean difference being 0.78 ± 0.30 impulses s-1 N-1 (Fig. 7B). Figure 7A shows an example where the average response to active tension was higher, but the incremental sensitivity was lower.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The starting point for these experiments was the observation that immediately following a series of eccentric contractions, in both human and animal muscles, there is a rise in whole muscle passive tension (see Howell et al. 1993; Chleboun et al. 1998; Whitehead et al. 2001). Our explanation for this rise is that during the eccentric contractions some sarcomeres in muscle fibres are over-stretched to beyond myofilament overlap. Repeated eccentric contractions lead to disruptions and membrane damage (Morgan & Allen, 1999; Proske & Morgan, 2001). This leads to uncontrolled Ca2+ movements and the development of localised contractures (Ogilvie et al. 1988; Lieber & Friden, 1999). We had no idea how widespread such contraction clots were, although there were clearly measurable changes at the tendon. We therefore embarked on this study to look at the responses of tendon organs as indicators of the distribution of damaged muscle fibres. This investigation formed part of a wider study of the effects of eccentric exercise on tendon organ responses and the implications for proprioception.
No overall, large change in tendon organ sensitivity was seen after the eccentric contractions with the methods chosen to assess it. Thus, there was no detectable change in the threshold to passive tension, and a small, but significant decrease in the sensitivity measured as the slope of the relationship between firing rate and smoothly increasing tension. There were no detectable changes in the responses to active tension, whether these were measured as the average response over a range of tensions or the sensitivity to changing tension (i.e. the slope of the tension-firing rate relationship). While the results were suggestive of a small increase in sensitivity, this was found to be non-significant.
Quite large differences were noted between individual receptors, and this may have obscured some small changes produced by the eccentric contractions. A problem in the study of tendon organs is that the force actually applied to the receptor during a contraction cannot be measured directly; nor can it be assumed, for different motor units, that it is a constant and unchanging fraction of the tension measured at the muscle tendon. In fact, the converse is true during fatiguing contractions in muscles of mixed fibre type in both the normal and experimental situations, where fibres of different type fatigue at different rates. This is probably responsible for at least some of the scatter seen in the results, and it emphasises the importance of considering the responses of more than just one tendon organ, or what is usually called the ensemble response.
One clear result obtained here was the consistent change in length threshold after eccentric contractions. Nearly all of the receptors started firing at a shorter muscle length during passive ramp stretch after the eccentric contractions than before. Of itself, this could be taken to suggest an increase in sensitivity, but since at the same time there was no change in passive tension threshold and no other evidence of an appropriate change in sensitivity, the explanation must lie elsewhere. We propose that the observed rise in whole muscle passive tension following the eccentric contractions is responsible.
Support for this is provided by the result of the experiment where the motor units acting on selected tendon organs were separated out and not subjected to eccentric contractions. These tendon organs did not show a change in length threshold. If, as is thought (Whitehead et al. 2001), the extra passive tension arises in the muscle fibres, it might be expected to be signalled only by those tendon organs connected in series with fibres affected by the eccentric contractions. If the rise in passive tension arose from other causes, such as muscle swelling or some change in tendon properties, it would be expected that all tendon organs would be affected.
This result also emphasises the specificity of tendon organ connections. The extra passive tension developed as a result of the eccentric contractions, although widespread throughout the muscle, was not signalled by the tendon organs whose motor units had not been stimulated. The first important conclusion is that the increase in passive tension has to be strictly in series with a tendon organ and not just a rise in adjacent muscle fibres for it to be signalled by the tendon organ. The second conclusion is that eccentric contractions of a muscle lead to damage, and a rise in passive tension, in sufficient numbers of muscle fibres for it to be signalled by nearly all of the muscle's tendon organs.
If it is assumed that about 10 muscle fibres insert into a tendon organ (Barker, 1974), and that eccentric damage to any one of these fibres is sufficient to influence the receptor's response, it may be inferred, since nearly all of the tendon organs were affected, that at least 10 % of the fibres in the muscle were damaged by the eccentric contractions in these experiments. There is evidence that fast-twitch muscle fibres are more susceptible to eccentric damage than are slow-twitch fibres (Brockett et al. 2002), and if this is so, the proportion of damaged fibres in susceptible motor units may be higher still. In any case, because motor units comprise many more than 10 muscle fibres, it is likely that all of the susceptible motor units were damaged by the eccentric contractions.
Another interesting result is that tendon organ sensitivities to active and passive tension were about the same. This result is in apparent conflict with the view (see, for example, Alnaes, 1967) that tendon organs are much more sensitive to active than to passive tension, as first shown by Houk & Henneman (1967). This conflict is resolved when it is remembered that Houk & Henneman (1967) compared whole muscle passive tension with the selected tension of motor units with muscle fibres inserting directly into the capsule of the tendon organ, and when this is the comparison being made, the sensitivity to active tension is indeed much higher. As recognised by Stuart et al. (1970) and Stephens et al. (1975), this does not apply if the active contractions are produced by whole muscle nerve stimulation: as did we, they found similar active and passive sensitivities in this situation.
In the present study, we actually found a slightly higher sensitivity to passive than to active tension. This may, however, be more the result of a difference in the tension profiles seen by the receptor than indicative of an inherent difference in sensitivity. The passive tension consisted of a smoothly increasing tension during ramp stretch of the muscle, while the active tension was a series of unfused contractions produced by distributed stimulation. Considering the known dynamic sensitivity of tendon organs, this might be expected to result in a different relationship between mean firing rate and mean muscle tension in the two situations. The dynamic sensitivity might be expected to contribute extra firing during unfused contractions compared to the discharge produced by an equivalent smooth tension. Since the tension profile was less smooth during low than high rates of stimulation, the effect would be to reduce the slope of the tension-firing rate relationship for active compared to passive tension by elevating the foot of the curve, as was actually observed. The contribution from dynamic responses, however, must have been quite small. Whatever the reason for the small sensitivity difference, the main point of interest is that passive tension is signalled about equally as well as active tension. This also adds further evidence that most passive muscle tension over this range of lengths is generated in structures in series with tendon organs, within the muscle fibres and not elsewhere (Magid & Law, 1985; Horowits, 1999).
A similar difference in tension profile might also explain the apparent contradiction in our findings of a small reduction in passive sensitivity after eccentric contractions and no change or a possible increase in active sensitivity. These sensitivities were also measured during smoothly rising passive tensions and unfused distributed stimulation. Moreover, the active tension profile after the eccentric contractions would differ from that before. Because of fatigue and eccentric damage, the production of the same tension after eccentric contractions would require a higher stimulation rate and so be smoother than before. There would be less dynamic contribution to receptor firing at low tensions and, by similar reasoning to that above, the foot of the curve relating tension and firing rate would be lowered after the eccentric contractions, thus increasing the slope. Any increase in smoothness would be even greater if, as seems likely, slow motor units made a larger contribution to the total tension after the eccentric contractions than before, because fast motor units are more susceptible to the effects of eccentric damage and fatigue (Brockett et al. 2002). It could be inferred that inherent receptor sensitivity is reduced by eccentric contractions and that this reduction is offset by the effects on the receptor of changes in the profile of the active tension. One cause of a short-term reduction in sensitivity could be the increased volume of afferent traffic generated by the eccentric contractions (Gregory & Proske, 1981).
The present study has a bearing on a subject of continuing debate on tendon organs, and that is whether the contraction of muscle fibres in the bulk of the muscle, lying in parallel with a tendon organ, could have an unloading effect on the in-series responses to the small bundle of muscle fibres inserting directly into the capsule. Significant unloading effects have been claimed in a number of previous studies (see Jami et al. 1992). On theoretical grounds, we considered that the effects of unloading would be, at most, transient unless the muscle was at very short lengths. We subsequently showed that, for the cat soleus muscle, this was indeed the case (Gregory et al. 1986).
If unloading is an important factor in tendon organ responses, it should have become apparent in both the passive and active responses we recorded here. The length threshold to passive stretch of those tendon organs whose motor units had not been subjected to eccentric contractions was unchanged after the eccentric contractions, although there had been a significant rise in passive tension in the rest of the muscle. This increased passive tension in muscle fibres not inserting directly into the receptor's capsule, as well as having no excitatory effect, had no unloading effect either. Active tension grading effected by the changing stimulus strength and during distributed stimulation would potentially include some degree of unloading, while grading by changing muscle length would be less likely to do so. All three methods resulted in a similar, relatively smooth gradation of tendon organ response, without the discontinuities or steps that might be expected with unloading.
A general conclusion that may be drawn from this study is that tendon organs, especially when their ensemble discharge is considered, are remarkably reliable in signalling whole muscle tension, whether this is produced by purely passive or a variety of active means, and they continue to signal tension accurately even after the muscle's force production has been disturbed by fatigue or eccentric exercise.
Caution should be exercised in using the results from acute animal experiments to explain psychophysical findings in the human. However, if human tendon organs are similar to those in the cat in showing little change in their responses after eccentric exercise, the implication of these results for proprioception is to reinforce the view that the sense of effort plays a dominant role in the perception of muscle force. If tendon organs continue to signal muscle force accurately, it seems unlikely that they would make a large contribution to force perception in circumstances where perturbations of force sense are seen (Cafarelli & Bigland-Ritchie, 1979; Jones & Hunter, 1983; Brockett et al. 1997). While it is known that the sense of tension can be used for force estimation in appropriately designed experiments (McCloskey et al. 1974; Roland & Ladegaard-Pedersen, 1977), its importance in a variety of normal situations remains an open question.
A speculation arising from this study is that the reduction in the tendon organ length threshold measured after eccentric contractions results in a conflict of information from length and tension sensors in the muscle (muscle spindles and tendon organs, respectively), in that the tendon organ discharge would be higher than expected at a particular length, as signalled by the spindles. It may be this conflict, rather than a change in the properties of the tension sensors, that contributes to disturbances of force perception. There is as yet no evidence that such a mechanism might operate, and this idea remains to be tested.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ALNAES, E. (1967). Static and dynamic properties of Golgi tendon organs in the anterior tibial and soleus muscles of the cat. Acta Physiologica Scandinavica 70, 176-187 | |
| BARKER, D. (1974). The morphology of muscle receptors. In Handbook of Sensory Physiology, vol. III, part 2, ed. HUNT, C. C., pp. 1-190. Springer, Berlin | |
| BROCKETT, C., MORGAN, D. L., GREGORY, J. E. & PROSKE, U. (2002). Damage from eccentric exercise in different types of motor units of the medial gastrocnemius muscle of the cat. Journal of Applied Physiology (in the Press) | |
| BROCKETT, C., WARREN, N., GREGORY, J. E., MORGAN, D. L. & PROSKE, U. (1997). A comparison of the effects of concentric versus eccentric exercise on force and position sense at the human elbow joint. Brain Research 771, 251-258 | [Medline] |
| BROWN, T. I., HUANG, Y., MORGAN, D. L., PROSKE, U. & WISE, A. (1999). A new strategy for controlling the level of activation in artificially stimulated muscle. IEEE Transactions on Rehabilitation Engineering 7, 167-173 | [Medline] |
| CAFARELLI, E. & BIGLAND-RITCHIE, B. (1979). Sensation of static force in muscles of different length. Experimental Neurology 65, 511-525 | [Medline] |
| CHLEBOUN, G. S., HOWELL, J. N., CONATSER, R. R. & GIESEY, J. J. (1998). Relationship between muscle swelling and stiffness after eccentric exercise. Medicine and Science in Sports and Exercise 30, 529-535 | [Medline] |
| GANDEVIA, S. C. (1996). Kinesthesia: role for afferent signals and motor commands. In Handbook on Integration of Motor, Circulatory, Respiratory and Metabolic Control During Exercise, ed. ROWELL, L. B. & SHEPHERD, J. T., pp. 128-172. American Physiological Society, Bethesda, MD, USA | |
| GREGORY, J. E., MORGAN, D. L. & PROSKE, U. (1986). The discharge of cat tendon organs during unloading contractions. Experimental Brain Research 61, 222-226 | |
| GREGORY, J. E. & PROSKE, U. (1981). Motor unit contractions initiating impulses in a tendon organ in the cat. Journal of Physiology 313, 251-262 | [Abstract] |
| HOROWITS, R. (1999). The physiological role of titin in striated muscle. Reviews of Physiology Biochemistry and Pharmacology 138, 57-96 | [Medline] |
| HOUK, J. & HENNEMAN, E. (1967). Responses of Golgi tendon organs to active contractions of the soleus muscle of the cat. Journal of Neurophysiology 30, 466-481 | [Medline] |
| HOWELL, J. N., CHLEBOUN, G. & CONATSER, R. (1993). Muscle stiffness, strength loss, swelling and soreness following exercise-induced injury in humans. Journal of Physiology 464, 183-196 | [Abstract] |
| JAMI, L., PETIT, J., ZYTNICKI, D. & HORCHOLLE-BOSSAVIT, G. (1992). Unloading: a reason why individual Golgi tendon organs cannot measure muscle force. In Muscle Afferents and Spinal Control of Movement, ed. JAMI, L., PIERROT-DESEILLIGNY, E. & ZYTNICKI, D., pp. 71-76. Pergamon, Oxford | |
| JONES, C., ALLEN, T., TALBOT, J., MORGAN, D. L. & PROSKE, U. (1997). Changes in the mechanical properties of human and amphibian muscle after eccentric exercise. European Journal of Applied Physiology and Occupational Physiology 76, 21-31 | [Medline] |
| JONES, L. A. & HUNTER, I. W. (1983). Perceived force in fatiguing isometric contractions. Perception and Psychophysics 33, 369-374 | [Medline] |
| LIEBER, R. L. & FRIDEN, J. (1999). Mechanisms of muscle injury after eccentric contraction. Journal of Science and Medicine in Sport 2, 253-265 | [Medline] |
| MCCLOSKEY, D. I. (1978). Kinesthetic sensibility. Physiological Reviews 58, 763-820 | [Medline] |
| MCCLOSKEY, D. I. (1981). Corollary discharges: motor commands and perception. In Handbook of Physiology, section III, The Nervous System, vol. II, Motor Control, part 2, ed. BROOKHARD, J. M., MOUNTCASTLE, V. B. & BROOKS, V. B., pp. 1415-1448. American Physiological Society, Bethesda, MD, USA | |
| MCCLOSKEY, D. I., EBELING, P. & GOODWIN, G. M. (1974). Estimation of weights and tensions and apparent involvement of a 'sense of effort'. Experimental Neurology 42, 220-232 | [Medline] |
| MCINTYRE, A. K., PROSKE, U. & RAWSON, J. A. (1984). Cortical projection of afferent information from tendon organs in the cat. Journal of Physiology 354, 395-406 | [Abstract] |
| MAGID, A. & LAW, D. J. (1985). Myofibrils bear most of the resting tension in frog skeletal muscle. Science 230, 1280-1282 | |
| MORGAN, D. L. & ALLEN, D. G. (1999). Early events in stretch-induced muscle damage. Journal of Applied Physiology 87, 2007-2015 | [Abstract/Full Text] |
| OGILVIE, R. W., ARMSTRONG, R. B., BAIRD, K. E. & BOTTOMS, C. L. (1988). Lesions in the rat soleus muscle following eccentrically biased exercise. American Journal of Anatomy 182, 335-346 | [Medline] |
| PROSKE, U. & MORGAN, D. L. (2001). Muscle damage from eccentric exercise: mechanism, mechanical signs, adaptation and clinical applications. Journal of Physiology 537, 333-345 | [Abstract/Full Text] |
| RACK, P. M. & WESTBURY, D. R. (1969). The effects of length and stimulus rate on tension in the isometric cat soleus muscle. Journal of Physiology 204, 443-460 | [Medline] |
| ROLAND, P. E. & LADEGAARD-PEDERSEN, H. (1977). A quantitative analysis of sensations of tension and of kinaesthesia in man. Evidence for a peripherally originating muscular sense and for a sense of effort. Brain 100, 671-692 | [Medline] |
| SAXTON, J. M., CLARKSON, P. M., JAMES, R., MILES, M., WESTERFER, M., CLARK, S. & DONNELLY, A. E. (1995). Neuromuscular dysfunction following eccentric exercise. Medicine and Science in Sports and Exercise 27, 1185-1193 | [Medline] |
| STEPHENS, J. A., REINKING, R. M. & STUART, D. G. (1975). Tendon organs of cat medial gastrocnemius: responses to active and passive forces as a function of muscle length. Journal of Neurophysiology 38, 1217-1231 | [Medline] |
| STUART, D. G., GOSLOW, G. E., MOSHER, C. G. & REINKING, R. M. (1970). Stretch responsiveness of Golgi tendon organs. Experimental Brain Research 10, 463-476 | [Medline] |
| WHITEHEAD, N. P., ALLEN, T. J., MORGAN, D. L. & PROSKE, U. (1998). Damage to human muscle from eccentric exercise after training with concentric exercise. Journal of Physiology 512, 615-620 | [Abstract/Full Text] |
| WHITEHEAD, N. P., WEERAKKODY, N. S., GREGORY, J. E., MORGAN, D. L. & PROSKE, U. (2001). Changes in passive tension of muscle in humans and animals after eccentric exercise. Journal of Physiology 533, 593-604 | [Abstract/Full Text] |
| WISE, A. K., MORGAN, D. L., GREGORY, J. E. & PROSKE, U. (2001). Fatigue in mammalian skeletal muscle stimulated under computer control. Journal of Applied Physiology 90, 189-197 | [Abstract/Full Text] |
| WOOD, S. A., MORGAN, D. L., GREGORY, J. E. & PROSKE, U. (1994). Fusimotor activity and the tendon jerk in the anaesthetised cat. Experimental Brain Research 98, 101-109 | [Medline] |
Acknowledgements
This work was carried out with support from the National Health and Medical Research Council of Australia.
This article has been cited by other articles:
|
|
 |
|
  S. M. Marcora, A. Bosio, and H. M. de Morree Locomotor muscle fatigue increases cardiorespiratory responses and reduces performance during intense cycling exercise independently from metabolic stress Am J Physiol Regulatory Integrative Comp Physiol, March 1, 2008; 294(3): R874 - R883. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A Winter, T. J Allen, and U Proske Muscle spindle signals combine with the sense of effort to indicate limb position J. Physiol., November 1, 2005; 568(3): 1035 - 1046. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
