JOURNALOFNEUROPHYSIOLOGY Vol. 63, No. 6, June 1990. Printed in U.S.A.
Muscle Afferent Responses to Isometric Contractions and Relaxations in Humans BEN~NI B. EDIN AND AKE B. VALLB~ Department of Physiology, University of Urned, SE-901 87 Umea”, Sweden SUMMARY
AND
CONCLUSIONS
gives rise to an internal shortening of the muscle, and, when the subject abruptly relaxes, the muscle spindles are stretched by the elastic pull from the tendon. In contrast, pull on tendon organs decreases with a decline of contractile force. With this view, tendon organ afferents should respond with a decelerated impulse rate, whereas muscle spindle afferents should increase their discharge. It has also been shown in animal experiments that spindles may produce a burst of impulses under these conditions (Larson et al. 198 1). Although this test has been used previously for identification purposes in human studies (Vallbo 1970, 1974a,b), its predictive power has yet to be analyzed. In the present study, it was found that an increased discharge on isometric relaxation occurred almost exclusively with muscle spindle afferents. In 1978, Burke et al. presented data on the recruitment of human muscle spindles during weak isometric contractions (Burke et al. 1978). If the task was performed under the same conditions, individual muscle spindle afferents were reproducibly recruited at particular levels of extrafusal contraction, the so-called ‘fusimotor threshold.’ A number of studies followed in which changes in the skeletomotor:fusimotor balance were assessed by relying on stable fusimotor thresholds (Burke et al. 1980a,b; Westerman et al. 198 1). However, the variability of recruitment levels for individual units was not reported, and it is therefore difficult to evaluate the main claim, viz., that fusimotor thresholds are fixed and reproducible. In the present experiments, we failed to demonstrate fixed recruitment levels for muscle spindle afferents. Rather, individual muscle spindles were recruited at sometimes highly variable levels of extrafusal contraction, and the variation in thresholds was not related to subject-specific factors.
contraction
1. One hundred and two single afferents from the finger extensor muscles of humans were studied with the microneurography technique. 2. The afferents were provisionally classified as primary muscle spindle afferents (62/102), secondary spindle afferents (22), and Golgi tendon organ afferents (18) on the basis of their responsesto four tests: 1) ramp-and-hold stretch, 2) 20- and 50-Hz small-amplitude sinusoidal stretch superimposed on ramp-andhold stretch, 3) maximal isometric twitch contraction, and 4) stretch sensitization. 3. The response profiles of the three unit types were analyzed during slowly rising isometric contraction terminating with an abrupt relaxation. About 75% (6 l/84) of all muscle spindle afferents increased their discharge during isometric contraction, whereas the discharge was reduced for the remaining afferents. All Golgi tendon organs increased their discharge during the contraction. 4. The level of extrafusal contraction at which a spindle afferent increased its discharge rate often varied from trial to trial, speaking against a fixed fusimotor recruitment level of the individual spindle ending. 5. In 70% of the spindle afferents, a distinct burst of impulses appeared when the subject rapidly relaxed after the isometric contraction. The burst was more common and usually much more prominent with primary than secondary afferents, often reaching instantaneous discharge rates well above 100 Hz. 6. Whereas all Golgi tendon organ afferents displayed an increaseddischarge during the contraction phase, only one of them exhibited a rate acceleration close to the relaxation phase. However, this response could clearly be identified as being of different nature than the spindle bursts. 7. It seemed justified to conclude that a test involving an isometric contraction ending with an abrupt relaxation is a valuable test to classify muscle stretch receptor afferents, because this test discriminated ~70% of the spindle afferents in the present material from tendon organ afferents.
METHODS INTRODUCTION
Subjects
The identification of muscle stretch receptors on the basis of the impulse discharge in single afferents is an important step in many studies of proprioceptive mechanisms. Although a number of identification tests have been employed with reduced preparations (anesthetized or decerebrated animals), these tests are not always useful in experiments with human subjects (for references see Prochazka and Hulliger 1983). On the other hand, the ability of human subjects to follow fairly complex instructions invites the development of new identification tests. One such test uses isometric relaxation. An isometric
The experiments were performed with 49 subjects, 19 males and 27 females, 18-66 yr old (median 26.5). Before the experiments, all subjects gave their informed consent in accordance with the Declaration of Helsinki.
0022-3077/90
$1 SO Copyright
Neurophysiological
technique and classiJication procedures
The neurophysiological technique as well as methods of data storage and off-line analysis were described in a companion paper (Edin and Vallbo 1990a). The units were provisionally classified as primary or secondary muscle spindle afferents or Golgi tendon organ afferents on the
0 1990 The
American
Physiological
Society
1307
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1308
B. B. EDIN AND A. B. VALLBO
basis of their responses to 1) maximal twitch contraction test Auditory and visual signals that indicated the target level were (Edin and Vallbo 1987, 1990a), 2) ramp-and-hold stretch (Edin provided, but an exactly uniform time course of torque change and Vallbo 1990a), 3) stretch sensitization (Edin and Vallbo was not required. The task performances shown in Figs. 1 and 2 1988), and 4) sinusoidal stretch (Edin and Vallbo 1990b). By use are representative. At least three contractions were recorded with of criteria and strategiessimilar to those described in a companion each unit. Surface electrodes were positioned to obtain optimal paper (Edin and Vallbo 1990a), individual units were assigned to recordings of the electromyographic activity of the extensor inthe most likely classeswhen the responses to the four tests had dicis muscle, individual portions of the extensor digitorum musbeen intuitively weighed together. cle, and the finger flexor muscles. The latter were used to assess the presence of significant antagonist activity during relaxation. In 39 consecutively recorded muscle spindle afferents that inExperimental procedure creased their discharge rate during contraction, the threshold for Subjects’ position and connections of finger to the servo-actua- recruitment was assessedby measuring the force level at which the tor were described in a companion paper (Edin and Vallbo discharge rate began to accelerate. 1990a). The resting position of the finger was carefully determined before unit recording. The subject was asked to remain completely Statistics and data-processing relaxed while a seriesof torque pulses were applied to the metaThe x2 test was used to test for differences between primary and carpophalangeal joint. The pulses alternated in direction between flexion and extension and were of successivelydecreasing ampli- secondary muscle spindle afferents (Siegel 1956). Data sampling, tudes. The position at which the finger came to rest when the statistical analyses, and illustrations were done with the use of the amplitude of the torque pulses approached zero was reproducible software package FYSTAT developed by Lars Backstrom, Dewithin a few degrees and was defined as the resting position. The partment of Physiology, University of Umea. isometric contractions were performed at this resting position, and the maximal voluntary torque (MVT) of the finger was meaRESULTS sured at this position or close to it. The subjects were instructed to perform slowly increasing exSubjects’ performance tension at the metacarpophalangeal joint of the test finger under isometric conditions at a rate of ~2% MVT/s up to 10% MVT. Subjects were asked to perform a slowly increasing isoWhen reaching this level, they were asked to maintain the conmetric contraction, then to maintain a steady contraction traction stable for 2 s and then rapidly relax. Extension of neighfor a few seconds, and finally to relax as quickly as possible. boring fingers was allowed to avoid undesirable contractions of the flexor muscles required to compensate for extensor force. Although exact performance was not stressed in the inBefore the microneurography experiment, subjects were familiar- structions, most subjects produced uniform contractions as ized with this task sequence and performed at least nine isometric in Fig. 1, whereas some deviated from the regular course as contractions. illustrated in Figs. 2 and 4.
A
B
0.161
C
0.16
Torque a1 MCPjoint mm)
0.24
1
1
Discharge rate (imp/s)
-2 401
0
2
4
6
8
10
I
Discharge 20 rate ( imp/s 1
-2
0
2
4
6
8
-2
0
2
4
6
8
-2
0
2
4
6
Time (set 1 FIG. 1. Response of muscle spindle primary afferents to isometric contractions. Records from 3 different subjects illustrate the 3 different response patterns encountered (A-C). Top: records show from above torque, impulse discharge rate, and sequence of single unit impulses. Three vertical dotted lines indicate when the subject was instructed to start increase the torque, to keep it stable, and to rapidly relax, respectively. Bottom: responses of the units to a ramp-and-hold stretch of 20” and with 5O”/s stretch velocity.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
8
MUSCLE
AFFERENTS
AND ISOMETRIC
CONTRACTIONS
1309
Torque at MCPjoint mm)
-2
0
2
4
6
8
-2
0
2
4
6
8
-2
0
2
4
6
8
Time (set) FIG. 2. Response of muscle spindle secondary afferents to isometric contractions. Records from 3 different subjects illustrate the 3 different response patterns encountered (A-C). Top: records show from above torque, impulse discharge rate, and sequence of single unit impulses. Dotted vertical lines have the same significance as in Fig. 1. Bottom: responses of the units to a rampand-hold stretch of 20” and with 5O”/s stretch velocity.
Sample properties and unit identijication
I
when the rate change during the main part of the contraction was considered as well as the response to relaxation The unit sample consists of 102 muscle stretch receptor (Figs. 1 and 2, Table 2). In a majority of spindle afferents afferents from the finger extensor muscles. It represents a the discharge increased during the main part of the consubgroup of 124 units recorded from 53 subjects (Edin and traction (6 l/84), whereas the impulse rate clearly decreased Vallbo 1990a). Sixty-two units were classified as primary for the rest, presumably because of unloading (23/84). and 22 as secondary muscle spindle afferents, and 18 were A significantly higher proportion of primary afferents classified as Golgi tendon organ afferents. As many as exhibited an accelerated rate during the contraction com9 I/ 102 units exhibited at least three out of the four repared with the secondaries (49/62 vs. 12/22, P < 0.05). sponse features expected of the unit type in response to Moreover, the primaries displayed much more variable inidentification tests (Table 1, see METHODS). However, the terspike intervals during the voluntary contractions when stretch sensitization test was lacking with 12 units. the discharge exceeded the precontraction level, as illustrated in Figs. 1 and 2. A fair proportion of units responded Response to contraction with a delayed acceleration (Fig. 2C). Even an initial deThree types of response to the isometric voluntary con- crease preceding the acceleration was seen in a few spindle traction could be discerned among muscle spindle afferents afferents. TABLE
1.
Unit sample
TABLE 2. Number of Tests
Ia II CT0 Total
3
2
1
31 9 13
25 11 2
3 2 3
3
3
38
responsepatterns
Decrease
4
53
Contraction/rehxation
8
Entries give the number of units that exhibited 1-4 out of 4 response features considered typical of the unit type. The features considered were I) response to twitch test, 2) response to ramp-and-hold stretch (deceleration response and/or silencing during muscle shortening), 3) stretch sensitization, and 4) response to small sinusoids. GTO, Golgi tendon organ.
No burst
Increase Burst
No burst
Burst
Total
Ia II GTO
13 10
18 7 17
31 5
62 22
Total
23
42
37
18
1’
102
Decrease and increase refer to unit response during the main part of the contraction, whereas burst and no burst refer to discharge on relaxation. *During the rapid relaxation a single Golgi tendon organ afferent displayed an atypical increase that was clearly different from spindle bursts (cf. Fig. 4C.) GTO, Golgi tendon organ.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1310
B. B. EDIN AND A. B. VALLBO
The impulse rate of spindle afferents often reached a plateau early in the contraction while the torque continued to increase (Figs. 1 and 2).
A
I
Response to relaxation Units that showed increased discharge during the main part of the isometric contraction produced either of two opposite responses to the abrupt relaxation, i.e., a shortlasting burst of accelerated discharge or a prompt cessation of the discharge when the tension began to fall (Figs. 1, B and C, and 2, B and C). On the other hand, all units that displayed a decreased discharge during the main part of the contraction (23/84), produced a burst of impulses on isometric relaxation. As pointed out above, the latter group consisted of spindle afferents exclusively. The proportion of spindle afferents displaying the combination of accelerated rate and relaxation burst was larger for primary than for secondary afferents (3 l/49 vs. 5/ 12, P < 0.10). Moreover, the size of the relaxation burst was much larger for the primaries where instantaneous discharge rates well above 100 imp/s were often reached.
It seemed reasonable to conclude that the various response profiles to contraction and relaxation of both primary and secondary muscle spindle afferents were due to different types and/or amount of fusimotor drive, although other mechanisms are plausible as well. An analysis of intrasubject variations was pursued to explore the hypothesis that the varying response patterns in the sample were due to intersubject differences, i.e., that individual subjects employed a uniform fusimotor strategy. Recordings from three to five muscle spindle primary afferents in six subjects were available, constituting one-third of the sample of spindle primaries. The analysis showed that primary afferents from an individual subject displayed the same variation in response patterns as were noted in the whole sample (Table 3). Hence, the hypothesis was refuted that the varying response profiles were accounted for by subject-specific fusimotor strategies. In this context, it should also be pointed out that, whereas a considerable variability was observed within the groups of primary and secondary spindle afferents, individual afferents showed stereotypical response profiles to repeated isometric contractions. TABLE
3.
Responseprofiles to isometric contraction Decrease
Experiment Number 85 87 90 91 94 99 Total
units
0
B
No burst
Increase
Burst
No burst
Burst
1
1 1 2 2 2 1
1 1 1
1
3 1
1
2
Total 3 3 3 5 4 3 21
Entries give the number of units that exhibited 1 of 3 different response profiles recorded in individual experiments.
.. .. .
1
recruited
2
4
Normalized
isometric
0
10
6
8
contraction
l-0 level
12 (% MVT)
12 10
8
Level of
Intrasubject variations
“-““-““i.,..“.““““““““““...
, ;f---;; ......I; , I.---r ,:__ ii
1001
recruitment
6
(%MvT)
4 2 0 Senal order
of units accordjng
20
30
40
to minimal level of recruitment
FIG. 3. Recruitment of muscle spindle afferents. A: curves show the percentage of spindle afferents recruited at contraction levels indicated by the abscissa. Units with decreased firing during isometric contractions are not included. Each subject made 3 consecutive contractions corresponding to the curves labeled ‘First,’ ‘ Second,’ and ‘Third,’ respectively. Note that the curves level off at MVT >5%, suggesting a saturation of unit recruitment. B: range of recruitment level of individual muscle spindle afferents. Ordinate represents the recruitment level in percent of maximal voluntary torque (MVT). Each vertical line represents 1 unit. With 2 units a reliable measurement could only be done in 1 of 3 tests; these units were included in A but not in B.
Recruitment
threshold of muscle spindle aflerents
It has been claimed that individual spindle afferents are recruited at characteristic torque levels (Burke et al. 1978a). In relation to this claim, it was relevant to assess the amount of torque at the moment when the discharge rate of individual units accelerated. It is evident from the cumulative frequency histograms of Fig. 3A that the majority of muscle spindle afferents were recruited already at 5% MVT. Moreover, beyond 5% MVT the curves seem to show a saturation of the recruitment. However, the threshold for the individual units varied considerably between tests (Fig. 3B), as found in many subjects. For instance, the 14 units with the largest differences between minimal and maximal thresholds were recorded from 10 different subjects. On the average, there was a small but statistically significant decrease in the median recruitment level from the first to the third test (P < 0.02, paired Student’s t test), but the effect of repeated tests was highly irregular.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
MUSCLE
.-.--,w,,,-,,, -2 0 2 20I
-2
0
4
2
6
4
8
6
AFFERENTS
10
8
AND ISOMETRIC
CONTRACTIONS
b ~~-~~~~-2 0 2 4
6
-2
4
0
2
1 - -. 4 8 10
6
1311
b.--.---.----.--.,.,, 7 7.5 8 Time kc)
8.5
8
Time (set 1 FIG. 4. Response of Golgi tendon organ afferents to isometric contractions. Records from 2 different subjects illustrate the 2 different response patterns encountered, A and B. Top: records show from above torque, impulse discharge rate, and sequence of single unit impulses. Bottom: responses of the units to a ramp-and-hold stretch of 20” and with 5O”/s stretch velocity. Dotted time span in B is shown on an expanded time scale in C where the response of the tendon organ afferent to the rapid relaxation is contrasted to the response of a primary muscle spindle afferent recorded from the same subject; the torque profiles labeled ‘Ia’ and ‘GTO’ in C belong to the spike trains with the same labels.
Golgi tendon organs All Golgi tendon organ afferents (n = 18) produced a sustained increase of impulse rate in response to the isometric contraction (Fig. 4, A and B). Moreover, their discharge rate was closely related to the active torque, not always during the whole contraction but during large periods of it. On the other hand, none of the Golgi tendon organ afferents showed any signs of unloading, not even a transient slowing at the beginning of the contraction. Only five tendon organ afferents displayed two or more stepwise accelerations during the slowly rising isometric contraction
(Crag0 et al. 1982; Vallbo 1973). Often each step commenced with a transient increase of impulse rate (Fig. 5). In one single Golgi tendon organ afferent, the isometric relaxation was associated with a burst of impulses (Fig. 4B). However, this burst was clearly of a different nature than the typical muscle spindle response, because the impulse acceleration was preceded by a short pause and its onset was delayed (Fig. 4C). Moreover, it was evident from the torque records that this subject produced a rapid sequence of flexion-extension-flexion contractions, probably with the purpose to reach quickly the desired low level of force. DISCUSSION
EMG Contraction 5. level (% MVT) OA
Disharge ~5~ rate ( imp/s 1 5 set
-
FIG. 5. Response of a Golgi tendon organ afferent to isometric contraction. Stepwise increases in the discharge rate of an afferent that was spontaneously active. Three steps are discernible, each one presumably reflecting the recruitment of one or several motor units.
Mechanisms accounting for spindle aferent burst on relaxation The burst of impulses that appeared on isometric relaxation was most likely due to an elastic recoil of the tendon and consequent stretch of the spindle. Alternative but rather unlikely mechanisms are spurious mechanical effects or a prompt increase of y-discharge when the muscle relaxed. Accepting the elastic recoil hypothesis, the presence or absence of a burst as well as its size would then depend on two factors: the stretch sensitivity of the end organ and the speed and amplitude of stretch of the muscle spindle. It is likely that the characteristics of the stretch seen by a spindle varied appreciably on its location in the muscle (MeyerLohman et al. 1974) and between tests as well; it was actually noticed that the size of the burst from an individual afferent was dependent on the rate of torque decline as recorded by the actuator system.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
B. B. EDIN
1312
AND
It is remarkable that all units that were unloaded during the main part of the contraction also presented a burst on relaxation. This finding of a short-lasting discharge with a rate exceeding the precontraction level indicates that the elastic recoil of the tendon was regularly sufficient to bring out a stretch response from primaries as well as secondaries when the fusimotor drive was small or absent. The situation is more complicated in the presence of fusimotor drive. At first glance, a burst on relaxation after a sustained rate increase indicative of fusimotor drive may be interpreted as an expression of a high dynamic sensitivity of the ending, considering that the response was similar to the typical response of a primary afferent to an imposed ramp-and-hold stretch. However, because the time course of the fusimotor drive was not known, it is feasible that a decline of fusimotor activity accounted for the fall in afferent impulse rate after the burst, and hence that the burst, in fact, reflects a high static sensitivity. Anyway, it seems clear that a burst on relaxation indicates a high stretch sensitivity but not necessarily a high dynamic sensitivity, at least when the burst is preceded by afferent rate acceleration during the main part of the contraction. It is therefore not difficult to accept the fact that a fair proportion of the secondaries indeed displayed an accelerated discharge on relaxation of the extrafusal fibers. Fusimotor drive during isometric contraction The present data indicate that weak isometric contractions were associated with a considerable fusimotor drive to -75% of the spindles, whereas -25% (23/84) of them were unloaded and therefore probably lacked a substantial fusimotor activation during the contraction. Whether y- or P-fibers or a combination of the two accounted for spindle acceleration is impossible to state. In one-third of the primaries (18/49), a burst on relaxation was lacking despite a rapid decline of the torque. This response pattern may be due to a high static fusimotor drive of the type that produces a decrease of stretch sensitivity (Boyd 1985; Crowe and Matthews 1964a,b; Hulliger et al. 1977a,b; Matthews 1962). An alternative explanation is that the effect of stretch was balanced by a concomitant decrease of fusimotor activity. It seems that the intrasubject variability between muscle spindle afferents as found in the present study fits with the finding of nonuniform connections to fusimotor neurons described by Appelberg et al. (1983a-c). If the neural input profile differs greatly between individual fusimotor neurons, it is reasonable that the fusimotor drive during voluntary contractions would vary between spindles as well. Studies in cat, with simultaneous recordings from several muscle spindle afferents, have also demonstrated differences in type of fusimotor drive during reflex activation (Johansson et al. 1987). Two groups of muscle spindles The present sample of spindle afferents fell into two main categories with regard to their response to isometric contractions, i.e., most of them accelerated, whereas - 25% decelerated. A first interpretation, consistent with the hypothesis of Burke et al. (1978a), is that the decelerating spindle afferents had higher thresholds in terms of isomet-
A. B. VALLBO
ric contraction force than was reached in our tests. Our findings would then be consistent with a continuum of fixed thresholds in the population of spindle afferents. However, further considerations suggested a different interpretation, i.e., a lack of fixed thresholds and a discontinuity with two subgroups of spindle afferents characterized by different fusimotor drive, one group that was recruited at low contraction levels and one that was not. Three points supported this interpretation. First, the shapes of the cumulative frequency histograms of Fig. 3A suggest that the recruitment of the first group approached saturation at quite low contraction forces because the curves level off beyond 5% MVT. Second, it was found that the threshold of the individual afferent was not constant but varied considerably between tests, as illustrated in Fig. 3B. Third, because of this threshold variation, it would be expected that some of the afferents having thresholds close to the plateau force, i.e., 10% MVT, would in some tests show acceleration and in others deceleration if the recruitment thresholds of the population formed a smooth continuum. However, all spindle afferents produced a uniform response in this respect, either decelerating or accelerating. These findings suggest that a subgroup of spindle afferents with low but unstable thresholds were fully recruited at - 10% MVT. This is reasonably consistent with previous findings that fusimotor neurons are largely recruited close to skeletomotor threshold, as has been reported for jaw muscles of the monkey (Lund et al. 1979) and triceps surae muscles of the cat (Post et al. 1980). Related to the saturation of spindle recruitment might be the finding that the afferent rate often reached a plateau quite early during the individual isometric contraction in spite of the continuously increasing force (Figs. 1, B and C, and 2, B and C). This is the expected response if the spindle was subjected to a fusimotor drive that first increased along with the contraction force up to 5- 10% of the maximal voluntary force and then leveled off (cf. Vallbo 1974b). Similar observations have been made in monkey (Larson et al. 198 1). Another subgroup was constituted by the spindle afferents that failed to accelerate in our test contractions. It remains to be clarified under which conditions their discharge may accelerate. Recruitment
thresholds
Because Burke et al. (1978a) did not report the variability of recruitment thresholds of individual spindle afferents, it is difficult to judge if there is a real difference between the data obtained from spindles in the lower limb and the present findings. It should be pointed out, however, that a high variability of thresholds as found in the present study was not a characteristic of particular subjects, because units with very small as well as units with very large differences in thresholds were recorded from the same individual subject. Moreover, although there was, on average, a slight decrease in threshold from the first to the third contraction, there was no indication that the variability decreased with moderate training ( lo-40 contractions). One of the main claims of Burke et al. (1978a) was that the muscle spindles are orderly in their recruitments when the muscle contraction increases. Because we recorded one
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
MUSCLE
AFFERENTS
AND ISOMETRIC
spindle afferent at a time, we cannot exclude mechanisms that simultaneously changed the thresholds of all or many spindle endings. If this was the case, the present findings might still be consistent with an orderly recruitment. On the other hand, orderly recruitment seems hard to reconcile with our finding that some muscle spindles recorded from individual subjects showed small differences in thresholds, whereas others showed very large differences. Moreover, some spindles recorded from individual subjects exhibited a successive decrease, whereas others exhibited a successive increase of the thresholds when very similar contractions were repeated. Furthermore, a lack of orderly recruitment has been emphasized on the basis of animal data (Murphy 198 1). Classijkation of muscle aRerents The present analysis indicates that a burst on isometric relaxation strongly supports the interpretation that the afferent originates from a muscle spindle rather than a tendon organ. The findings suggest that -70% of the muscle spindle afferents would have been identified as such with the use of this criterion alone, whereas, at the most, 5% of the tendon organ afferents would have been misclassified as muscle spindle afferents (see Fig. 4, B and C), assuming that our provisional classification was reasonably correct. It seems, therefore, recommendable to include isometric contractions with rapid relaxation in the test battery for classification of single muscle afferents in human microneurography studies along with ramp-and-hold stretch, twitch contraction, and stretch sensitization. The authors are greatly indebted to L. Backstrom and S.-O. Johansson for valuable technical assistance. The study was supported by the Swedish Medical Research Council (Grant 14X-3548) Forskningsradsnamnden, Gunvor och Josef An&-s Stiftelse, Torsten och Ragnar Soderbergs Stiftelse, Magn. Bergvalls Stiftelse, and Umea University (Fonden fdr Medicinsk forskning, ograduerade forskare). Present address of A. B. Vallbo: Dept. of Physiology, University of Goteborg, Box 330 31, SE-400 33 Goteborg, Sweden. Address for reprint requests: B. Edin, Dept. of Physiology, University of Umea, SE-901 87 Umea, Sweden. Received 5 July 1988; accepted in final form 24 January 1990. REFERENCES B., HULLIGER, M., JOHANSSON, H., AND SOJKA, P. Actions on gamma-motoneurones elicited by electrical stimulation of group II muscle afferent fibers in the hind limb of the cat. J. phvsiol. Lond. 335: 255-273, 1983a. APPELBERG, B., HULLIGER, M., JOHANSSON, H., AND SOJKA, P. Actions on gamma-motoneurones elicited by electrical stimulation of group III muscle afferent fibers in the hind limb of the cat. J. Physiol. Lond. 335: 275-292, 1983b. APPELBERG, B., HULLIGER, M., JOHANSSON, H., AND SOJKA, P. Actions on gamma-motoneurones elicited by electrical stimulation of group III muscle afferent fibers in the hind limb of the cat. J. Physiol. Lond. 335: 293-305, 1983~. BOYD, I. A. Muscle spindles and stretch reflexes. In: Scientific Basis of Clinical Neurology, edited by M. Swash and C. Kennard. London: Churchill, 1985, p. 74-97. BURKE, D., HAGBARTH, K-E., AND SKUSE, N. F. Recruitment order of human spindle endings in isometric voluntary contractions. J. Physiol. Lond. 285: 101-l 12, 1978. BURKE, D., MCI&ON, B., SKUSE, N. F., AND WESTERMAN, R. A. Anticipation and fusimotor activity in preparation for a voluntary contraction. J. Physiol. Lond. 306: 337-348, 1980a. APPELBERG,
CONTRACTIONS
1313
D., MCI&ON, B., AND WESTERMAN, R. A. Induced changes in the thresholds for voluntary activation of human spindle endings. J. Physiol. Land. 302: 17 l- 18 1, 1980b. CRAGO, P. E., HOUK, J. C., AND RYMER, W. 2. Sampling of total muscle force by tendon organs. J. Neurophysiol. 47: 1069- 1083, 1982. CROWE, A. AND MATTHEWS, P. B. C. The effect of stimulation of static and dynamic fusimotor fibres on the response to stretching of the primary endings of muscle spindles. J. Physiol. Land. 174: 109- 13 1, 1964a. CROWE, A. AND MATTHEWS, P. B. C. Further studies of static and dynamic fusimotor fibers. J. Physiol. Land. 174: 132- 15 1, 1964b. EDIN, B. B. AND VALLBO, A. B. Twitch contraction for identification of human muscle afferents. Acta Physiol. Stand. 13 1: 129- 138, 1987. EDIN, B. B. AND VALLBO A. B. Stretch sensitization in human muscle spindles. J. Physiol. Lond. 400: 10 1- 111, 1988. EDIN, B. B. AND VALLBO, A. B. Dynamic response of human muscle spindle afferents to stretch. J. Neurophysiol. 63: 1297- 1306, 1990a. EDIN, B. B. AND VALLBO, A. B. Classification of human muscle stretch receptor afferents: A Bayesian approach J. Neurophysiol. 63: 1314-1322, 1990b. HULLIGER, M., MATTHEWS, P. B. C., AND NOTH, J. Static and dynamic fusimotor action on the response of Ia fibers to low frequency sinusoidal stretching of widely ranging amplitudes. J. Physiol. Lond. 267: 81 l-836, 1977a. HULLIGER, M., MATTHEWS, P. B. C., AND NOTH, J. Effects of combining static and dynamic fusimotor stimulation on the response of the muscle spindle primary ending to sinusoidal stretching. J. Physiol. Lond. 267: 839-856, 1977b. JOHANSSON, H., SJ~LANDER, P., SOJKA, P., AND WADELL, I. Fusimotor reflexes to antagonistic muscles simultaneously assessedby multi-afferent recordings from muscle spindle afferents. Brain Res. 435: 337-342, 1987. LARSON, C. R., SMITH, A., AND LUSCHEI, E. S. Discharge characteristics and stretch sensitivity of jaw muscle afferents in the monkey during controlled isometric bites. J. Neurophysiol. 46: 130- 142, 198 1. LUND, J. P., SMITH, A. M., SESSLE, B. J., AND MURAKAMI, T. Activity of trigeminal alpha- and gamma-motoneurones and muscle afferents during performance of a biting task. J. Neurophysiol. 42: 7 10-725, 1979. MATTHEWS, P. B. C. The differentiation of two types of fusimotor fibre by their effects on the dynamic response of muscle spindle primary endings. Q. J. Exp. Physiol. 47: 324-333, 1962. MEYER-L• HMAN, J., RIEBOLD, W., AND ROBRECHT, D. Mechanical influence of the extrafusal muscle on the static behavior of deefferented primary muscle spindle endings in cat. Pfluegers Arch. 352: 267-278, 1974. MURPHY, P. R. The recruitment order of gamma-motoneurones in the decerebrate rabbit. J. Physiol. Land. 3 15: 59-67, 198 1. POST, E. M., RYMER, W. Z., AND HASAN, 2. Relation between intrafusal and extrafusal activity in triceps surae muscles of the decerebrate cat: evidence for beta action. J. Neurophysiol. 44: 383-404, 1980. PROCHAZKA, A. AND HULLIGER, M. Muscle afferent function and its significance for motor control mechanisms during voluntary movements in cat, monkey and man. In: Motor Control Mechanisms in Health and Disease, edited by J. E. Desmedt. New York: Raven, 1983, p. 93-132. SIEGEL, S. Nonparametric Statistics for the Behavioral Sciences. Tokyo: McGraw-Hill, 1956. VALLBO, A. B. Slowly adapting muscle receptors in man. Acta Physiol. Stand. 78: 315-333, 1970. VALLBO, A. B. Muscle spindle afferent discharge from resting and contracting muscles in normal human subjects. In: New Developments in Electromyographic and Clinical Neurophysiology, edited by J. E. Desmedt. Basel: Karger, vol. 3, 1973, p. 25 1-26 1. VALLBO, A. B. Afferent discharge from human muscle spindles in noncontracting muscles. Steady state impulse frequency as function of joint angle. Acta Physiol. Stand. 90: 303-3 18, 1974a. VALLBO, A. B. Human muscle spindle discharge during isometric voluntary contractions. Amplitude relations between spindle frequency and torque. Acta Physiol. Stand. 90: 3 19-336, 1974b. WESTERMAN, R. A., BURKE, D., MCI&ON, B. B., AND SKUSE, N. F. Muscle spindle activity in man. Flexibility of skeletomotor:fusimotor balance during various movements. In: Brain Mechanisms and Perceptual Awareness, edited by 0. Pompeiano and C. Ajmone Marsan. New York: Raven, 198 1. BURKE,
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
Muscle Afferent Responses to Isometric Contractions and Relaxations in Humans BEN~NI B. EDIN AND AKE B. VALLB~ Department of Physiology, University of Urned, SE-901 87 Umea”, Sweden SUMMARY
AND
CONCLUSIONS
gives rise to an internal shortening of the muscle, and, when the subject abruptly relaxes, the muscle spindles are stretched by the elastic pull from the tendon. In contrast, pull on tendon organs decreases with a decline of contractile force. With this view, tendon organ afferents should respond with a decelerated impulse rate, whereas muscle spindle afferents should increase their discharge. It has also been shown in animal experiments that spindles may produce a burst of impulses under these conditions (Larson et al. 198 1). Although this test has been used previously for identification purposes in human studies (Vallbo 1970, 1974a,b), its predictive power has yet to be analyzed. In the present study, it was found that an increased discharge on isometric relaxation occurred almost exclusively with muscle spindle afferents. In 1978, Burke et al. presented data on the recruitment of human muscle spindles during weak isometric contractions (Burke et al. 1978). If the task was performed under the same conditions, individual muscle spindle afferents were reproducibly recruited at particular levels of extrafusal contraction, the so-called ‘fusimotor threshold.’ A number of studies followed in which changes in the skeletomotor:fusimotor balance were assessed by relying on stable fusimotor thresholds (Burke et al. 1980a,b; Westerman et al. 198 1). However, the variability of recruitment levels for individual units was not reported, and it is therefore difficult to evaluate the main claim, viz., that fusimotor thresholds are fixed and reproducible. In the present experiments, we failed to demonstrate fixed recruitment levels for muscle spindle afferents. Rather, individual muscle spindles were recruited at sometimes highly variable levels of extrafusal contraction, and the variation in thresholds was not related to subject-specific factors.
contraction
1. One hundred and two single afferents from the finger extensor muscles of humans were studied with the microneurography technique. 2. The afferents were provisionally classified as primary muscle spindle afferents (62/102), secondary spindle afferents (22), and Golgi tendon organ afferents (18) on the basis of their responsesto four tests: 1) ramp-and-hold stretch, 2) 20- and 50-Hz small-amplitude sinusoidal stretch superimposed on ramp-andhold stretch, 3) maximal isometric twitch contraction, and 4) stretch sensitization. 3. The response profiles of the three unit types were analyzed during slowly rising isometric contraction terminating with an abrupt relaxation. About 75% (6 l/84) of all muscle spindle afferents increased their discharge during isometric contraction, whereas the discharge was reduced for the remaining afferents. All Golgi tendon organs increased their discharge during the contraction. 4. The level of extrafusal contraction at which a spindle afferent increased its discharge rate often varied from trial to trial, speaking against a fixed fusimotor recruitment level of the individual spindle ending. 5. In 70% of the spindle afferents, a distinct burst of impulses appeared when the subject rapidly relaxed after the isometric contraction. The burst was more common and usually much more prominent with primary than secondary afferents, often reaching instantaneous discharge rates well above 100 Hz. 6. Whereas all Golgi tendon organ afferents displayed an increaseddischarge during the contraction phase, only one of them exhibited a rate acceleration close to the relaxation phase. However, this response could clearly be identified as being of different nature than the spindle bursts. 7. It seemed justified to conclude that a test involving an isometric contraction ending with an abrupt relaxation is a valuable test to classify muscle stretch receptor afferents, because this test discriminated ~70% of the spindle afferents in the present material from tendon organ afferents.
METHODS INTRODUCTION
Subjects
The identification of muscle stretch receptors on the basis of the impulse discharge in single afferents is an important step in many studies of proprioceptive mechanisms. Although a number of identification tests have been employed with reduced preparations (anesthetized or decerebrated animals), these tests are not always useful in experiments with human subjects (for references see Prochazka and Hulliger 1983). On the other hand, the ability of human subjects to follow fairly complex instructions invites the development of new identification tests. One such test uses isometric relaxation. An isometric
The experiments were performed with 49 subjects, 19 males and 27 females, 18-66 yr old (median 26.5). Before the experiments, all subjects gave their informed consent in accordance with the Declaration of Helsinki.
0022-3077/90
$1 SO Copyright
Neurophysiological
technique and classiJication procedures
The neurophysiological technique as well as methods of data storage and off-line analysis were described in a companion paper (Edin and Vallbo 1990a). The units were provisionally classified as primary or secondary muscle spindle afferents or Golgi tendon organ afferents on the
0 1990 The
American
Physiological
Society
1307
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1308
B. B. EDIN AND A. B. VALLBO
basis of their responses to 1) maximal twitch contraction test Auditory and visual signals that indicated the target level were (Edin and Vallbo 1987, 1990a), 2) ramp-and-hold stretch (Edin provided, but an exactly uniform time course of torque change and Vallbo 1990a), 3) stretch sensitization (Edin and Vallbo was not required. The task performances shown in Figs. 1 and 2 1988), and 4) sinusoidal stretch (Edin and Vallbo 1990b). By use are representative. At least three contractions were recorded with of criteria and strategiessimilar to those described in a companion each unit. Surface electrodes were positioned to obtain optimal paper (Edin and Vallbo 1990a), individual units were assigned to recordings of the electromyographic activity of the extensor inthe most likely classeswhen the responses to the four tests had dicis muscle, individual portions of the extensor digitorum musbeen intuitively weighed together. cle, and the finger flexor muscles. The latter were used to assess the presence of significant antagonist activity during relaxation. In 39 consecutively recorded muscle spindle afferents that inExperimental procedure creased their discharge rate during contraction, the threshold for Subjects’ position and connections of finger to the servo-actua- recruitment was assessedby measuring the force level at which the tor were described in a companion paper (Edin and Vallbo discharge rate began to accelerate. 1990a). The resting position of the finger was carefully determined before unit recording. The subject was asked to remain completely Statistics and data-processing relaxed while a seriesof torque pulses were applied to the metaThe x2 test was used to test for differences between primary and carpophalangeal joint. The pulses alternated in direction between flexion and extension and were of successivelydecreasing ampli- secondary muscle spindle afferents (Siegel 1956). Data sampling, tudes. The position at which the finger came to rest when the statistical analyses, and illustrations were done with the use of the amplitude of the torque pulses approached zero was reproducible software package FYSTAT developed by Lars Backstrom, Dewithin a few degrees and was defined as the resting position. The partment of Physiology, University of Umea. isometric contractions were performed at this resting position, and the maximal voluntary torque (MVT) of the finger was meaRESULTS sured at this position or close to it. The subjects were instructed to perform slowly increasing exSubjects’ performance tension at the metacarpophalangeal joint of the test finger under isometric conditions at a rate of ~2% MVT/s up to 10% MVT. Subjects were asked to perform a slowly increasing isoWhen reaching this level, they were asked to maintain the conmetric contraction, then to maintain a steady contraction traction stable for 2 s and then rapidly relax. Extension of neighfor a few seconds, and finally to relax as quickly as possible. boring fingers was allowed to avoid undesirable contractions of the flexor muscles required to compensate for extensor force. Although exact performance was not stressed in the inBefore the microneurography experiment, subjects were familiar- structions, most subjects produced uniform contractions as ized with this task sequence and performed at least nine isometric in Fig. 1, whereas some deviated from the regular course as contractions. illustrated in Figs. 2 and 4.
A
B
0.161
C
0.16
Torque a1 MCPjoint mm)
0.24
1
1
Discharge rate (imp/s)
-2 401
0
2
4
6
8
10
I
Discharge 20 rate ( imp/s 1
-2
0
2
4
6
8
-2
0
2
4
6
8
-2
0
2
4
6
Time (set 1 FIG. 1. Response of muscle spindle primary afferents to isometric contractions. Records from 3 different subjects illustrate the 3 different response patterns encountered (A-C). Top: records show from above torque, impulse discharge rate, and sequence of single unit impulses. Three vertical dotted lines indicate when the subject was instructed to start increase the torque, to keep it stable, and to rapidly relax, respectively. Bottom: responses of the units to a ramp-and-hold stretch of 20” and with 5O”/s stretch velocity.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
8
MUSCLE
AFFERENTS
AND ISOMETRIC
CONTRACTIONS
1309
Torque at MCPjoint mm)
-2
0
2
4
6
8
-2
0
2
4
6
8
-2
0
2
4
6
8
Time (set) FIG. 2. Response of muscle spindle secondary afferents to isometric contractions. Records from 3 different subjects illustrate the 3 different response patterns encountered (A-C). Top: records show from above torque, impulse discharge rate, and sequence of single unit impulses. Dotted vertical lines have the same significance as in Fig. 1. Bottom: responses of the units to a rampand-hold stretch of 20” and with 5O”/s stretch velocity.
Sample properties and unit identijication
I
when the rate change during the main part of the contraction was considered as well as the response to relaxation The unit sample consists of 102 muscle stretch receptor (Figs. 1 and 2, Table 2). In a majority of spindle afferents afferents from the finger extensor muscles. It represents a the discharge increased during the main part of the consubgroup of 124 units recorded from 53 subjects (Edin and traction (6 l/84), whereas the impulse rate clearly decreased Vallbo 1990a). Sixty-two units were classified as primary for the rest, presumably because of unloading (23/84). and 22 as secondary muscle spindle afferents, and 18 were A significantly higher proportion of primary afferents classified as Golgi tendon organ afferents. As many as exhibited an accelerated rate during the contraction com9 I/ 102 units exhibited at least three out of the four repared with the secondaries (49/62 vs. 12/22, P < 0.05). sponse features expected of the unit type in response to Moreover, the primaries displayed much more variable inidentification tests (Table 1, see METHODS). However, the terspike intervals during the voluntary contractions when stretch sensitization test was lacking with 12 units. the discharge exceeded the precontraction level, as illustrated in Figs. 1 and 2. A fair proportion of units responded Response to contraction with a delayed acceleration (Fig. 2C). Even an initial deThree types of response to the isometric voluntary con- crease preceding the acceleration was seen in a few spindle traction could be discerned among muscle spindle afferents afferents. TABLE
1.
Unit sample
TABLE 2. Number of Tests
Ia II CT0 Total
3
2
1
31 9 13
25 11 2
3 2 3
3
3
38
responsepatterns
Decrease
4
53
Contraction/rehxation
8
Entries give the number of units that exhibited 1-4 out of 4 response features considered typical of the unit type. The features considered were I) response to twitch test, 2) response to ramp-and-hold stretch (deceleration response and/or silencing during muscle shortening), 3) stretch sensitization, and 4) response to small sinusoids. GTO, Golgi tendon organ.
No burst
Increase Burst
No burst
Burst
Total
Ia II GTO
13 10
18 7 17
31 5
62 22
Total
23
42
37
18
1’
102
Decrease and increase refer to unit response during the main part of the contraction, whereas burst and no burst refer to discharge on relaxation. *During the rapid relaxation a single Golgi tendon organ afferent displayed an atypical increase that was clearly different from spindle bursts (cf. Fig. 4C.) GTO, Golgi tendon organ.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
1310
B. B. EDIN AND A. B. VALLBO
The impulse rate of spindle afferents often reached a plateau early in the contraction while the torque continued to increase (Figs. 1 and 2).
A
I
Response to relaxation Units that showed increased discharge during the main part of the isometric contraction produced either of two opposite responses to the abrupt relaxation, i.e., a shortlasting burst of accelerated discharge or a prompt cessation of the discharge when the tension began to fall (Figs. 1, B and C, and 2, B and C). On the other hand, all units that displayed a decreased discharge during the main part of the contraction (23/84), produced a burst of impulses on isometric relaxation. As pointed out above, the latter group consisted of spindle afferents exclusively. The proportion of spindle afferents displaying the combination of accelerated rate and relaxation burst was larger for primary than for secondary afferents (3 l/49 vs. 5/ 12, P < 0.10). Moreover, the size of the relaxation burst was much larger for the primaries where instantaneous discharge rates well above 100 imp/s were often reached.
It seemed reasonable to conclude that the various response profiles to contraction and relaxation of both primary and secondary muscle spindle afferents were due to different types and/or amount of fusimotor drive, although other mechanisms are plausible as well. An analysis of intrasubject variations was pursued to explore the hypothesis that the varying response patterns in the sample were due to intersubject differences, i.e., that individual subjects employed a uniform fusimotor strategy. Recordings from three to five muscle spindle primary afferents in six subjects were available, constituting one-third of the sample of spindle primaries. The analysis showed that primary afferents from an individual subject displayed the same variation in response patterns as were noted in the whole sample (Table 3). Hence, the hypothesis was refuted that the varying response profiles were accounted for by subject-specific fusimotor strategies. In this context, it should also be pointed out that, whereas a considerable variability was observed within the groups of primary and secondary spindle afferents, individual afferents showed stereotypical response profiles to repeated isometric contractions. TABLE
3.
Responseprofiles to isometric contraction Decrease
Experiment Number 85 87 90 91 94 99 Total
units
0
B
No burst
Increase
Burst
No burst
Burst
1
1 1 2 2 2 1
1 1 1
1
3 1
1
2
Total 3 3 3 5 4 3 21
Entries give the number of units that exhibited 1 of 3 different response profiles recorded in individual experiments.
.. .. .
1
recruited
2
4
Normalized
isometric
0
10
6
8
contraction
l-0 level
12 (% MVT)
12 10
8
Level of
Intrasubject variations
“-““-““i.,..“.““““““““““...
, ;f---;; ......I; , I.---r ,:__ ii
1001
recruitment
6
(%MvT)
4 2 0 Senal order
of units accordjng
20
30
40
to minimal level of recruitment
FIG. 3. Recruitment of muscle spindle afferents. A: curves show the percentage of spindle afferents recruited at contraction levels indicated by the abscissa. Units with decreased firing during isometric contractions are not included. Each subject made 3 consecutive contractions corresponding to the curves labeled ‘First,’ ‘ Second,’ and ‘Third,’ respectively. Note that the curves level off at MVT >5%, suggesting a saturation of unit recruitment. B: range of recruitment level of individual muscle spindle afferents. Ordinate represents the recruitment level in percent of maximal voluntary torque (MVT). Each vertical line represents 1 unit. With 2 units a reliable measurement could only be done in 1 of 3 tests; these units were included in A but not in B.
Recruitment
threshold of muscle spindle aflerents
It has been claimed that individual spindle afferents are recruited at characteristic torque levels (Burke et al. 1978a). In relation to this claim, it was relevant to assess the amount of torque at the moment when the discharge rate of individual units accelerated. It is evident from the cumulative frequency histograms of Fig. 3A that the majority of muscle spindle afferents were recruited already at 5% MVT. Moreover, beyond 5% MVT the curves seem to show a saturation of the recruitment. However, the threshold for the individual units varied considerably between tests (Fig. 3B), as found in many subjects. For instance, the 14 units with the largest differences between minimal and maximal thresholds were recorded from 10 different subjects. On the average, there was a small but statistically significant decrease in the median recruitment level from the first to the third test (P < 0.02, paired Student’s t test), but the effect of repeated tests was highly irregular.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
MUSCLE
.-.--,w,,,-,,, -2 0 2 20I
-2
0
4
2
6
4
8
6
AFFERENTS
10
8
AND ISOMETRIC
CONTRACTIONS
b ~~-~~~~-2 0 2 4
6
-2
4
0
2
1 - -. 4 8 10
6
1311
b.--.---.----.--.,.,, 7 7.5 8 Time kc)
8.5
8
Time (set 1 FIG. 4. Response of Golgi tendon organ afferents to isometric contractions. Records from 2 different subjects illustrate the 2 different response patterns encountered, A and B. Top: records show from above torque, impulse discharge rate, and sequence of single unit impulses. Bottom: responses of the units to a ramp-and-hold stretch of 20” and with 5O”/s stretch velocity. Dotted time span in B is shown on an expanded time scale in C where the response of the tendon organ afferent to the rapid relaxation is contrasted to the response of a primary muscle spindle afferent recorded from the same subject; the torque profiles labeled ‘Ia’ and ‘GTO’ in C belong to the spike trains with the same labels.
Golgi tendon organs All Golgi tendon organ afferents (n = 18) produced a sustained increase of impulse rate in response to the isometric contraction (Fig. 4, A and B). Moreover, their discharge rate was closely related to the active torque, not always during the whole contraction but during large periods of it. On the other hand, none of the Golgi tendon organ afferents showed any signs of unloading, not even a transient slowing at the beginning of the contraction. Only five tendon organ afferents displayed two or more stepwise accelerations during the slowly rising isometric contraction
(Crag0 et al. 1982; Vallbo 1973). Often each step commenced with a transient increase of impulse rate (Fig. 5). In one single Golgi tendon organ afferent, the isometric relaxation was associated with a burst of impulses (Fig. 4B). However, this burst was clearly of a different nature than the typical muscle spindle response, because the impulse acceleration was preceded by a short pause and its onset was delayed (Fig. 4C). Moreover, it was evident from the torque records that this subject produced a rapid sequence of flexion-extension-flexion contractions, probably with the purpose to reach quickly the desired low level of force. DISCUSSION
EMG Contraction 5. level (% MVT) OA
Disharge ~5~ rate ( imp/s 1 5 set
-
FIG. 5. Response of a Golgi tendon organ afferent to isometric contraction. Stepwise increases in the discharge rate of an afferent that was spontaneously active. Three steps are discernible, each one presumably reflecting the recruitment of one or several motor units.
Mechanisms accounting for spindle aferent burst on relaxation The burst of impulses that appeared on isometric relaxation was most likely due to an elastic recoil of the tendon and consequent stretch of the spindle. Alternative but rather unlikely mechanisms are spurious mechanical effects or a prompt increase of y-discharge when the muscle relaxed. Accepting the elastic recoil hypothesis, the presence or absence of a burst as well as its size would then depend on two factors: the stretch sensitivity of the end organ and the speed and amplitude of stretch of the muscle spindle. It is likely that the characteristics of the stretch seen by a spindle varied appreciably on its location in the muscle (MeyerLohman et al. 1974) and between tests as well; it was actually noticed that the size of the burst from an individual afferent was dependent on the rate of torque decline as recorded by the actuator system.
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
B. B. EDIN
1312
AND
It is remarkable that all units that were unloaded during the main part of the contraction also presented a burst on relaxation. This finding of a short-lasting discharge with a rate exceeding the precontraction level indicates that the elastic recoil of the tendon was regularly sufficient to bring out a stretch response from primaries as well as secondaries when the fusimotor drive was small or absent. The situation is more complicated in the presence of fusimotor drive. At first glance, a burst on relaxation after a sustained rate increase indicative of fusimotor drive may be interpreted as an expression of a high dynamic sensitivity of the ending, considering that the response was similar to the typical response of a primary afferent to an imposed ramp-and-hold stretch. However, because the time course of the fusimotor drive was not known, it is feasible that a decline of fusimotor activity accounted for the fall in afferent impulse rate after the burst, and hence that the burst, in fact, reflects a high static sensitivity. Anyway, it seems clear that a burst on relaxation indicates a high stretch sensitivity but not necessarily a high dynamic sensitivity, at least when the burst is preceded by afferent rate acceleration during the main part of the contraction. It is therefore not difficult to accept the fact that a fair proportion of the secondaries indeed displayed an accelerated discharge on relaxation of the extrafusal fibers. Fusimotor drive during isometric contraction The present data indicate that weak isometric contractions were associated with a considerable fusimotor drive to -75% of the spindles, whereas -25% (23/84) of them were unloaded and therefore probably lacked a substantial fusimotor activation during the contraction. Whether y- or P-fibers or a combination of the two accounted for spindle acceleration is impossible to state. In one-third of the primaries (18/49), a burst on relaxation was lacking despite a rapid decline of the torque. This response pattern may be due to a high static fusimotor drive of the type that produces a decrease of stretch sensitivity (Boyd 1985; Crowe and Matthews 1964a,b; Hulliger et al. 1977a,b; Matthews 1962). An alternative explanation is that the effect of stretch was balanced by a concomitant decrease of fusimotor activity. It seems that the intrasubject variability between muscle spindle afferents as found in the present study fits with the finding of nonuniform connections to fusimotor neurons described by Appelberg et al. (1983a-c). If the neural input profile differs greatly between individual fusimotor neurons, it is reasonable that the fusimotor drive during voluntary contractions would vary between spindles as well. Studies in cat, with simultaneous recordings from several muscle spindle afferents, have also demonstrated differences in type of fusimotor drive during reflex activation (Johansson et al. 1987). Two groups of muscle spindles The present sample of spindle afferents fell into two main categories with regard to their response to isometric contractions, i.e., most of them accelerated, whereas - 25% decelerated. A first interpretation, consistent with the hypothesis of Burke et al. (1978a), is that the decelerating spindle afferents had higher thresholds in terms of isomet-
A. B. VALLBO
ric contraction force than was reached in our tests. Our findings would then be consistent with a continuum of fixed thresholds in the population of spindle afferents. However, further considerations suggested a different interpretation, i.e., a lack of fixed thresholds and a discontinuity with two subgroups of spindle afferents characterized by different fusimotor drive, one group that was recruited at low contraction levels and one that was not. Three points supported this interpretation. First, the shapes of the cumulative frequency histograms of Fig. 3A suggest that the recruitment of the first group approached saturation at quite low contraction forces because the curves level off beyond 5% MVT. Second, it was found that the threshold of the individual afferent was not constant but varied considerably between tests, as illustrated in Fig. 3B. Third, because of this threshold variation, it would be expected that some of the afferents having thresholds close to the plateau force, i.e., 10% MVT, would in some tests show acceleration and in others deceleration if the recruitment thresholds of the population formed a smooth continuum. However, all spindle afferents produced a uniform response in this respect, either decelerating or accelerating. These findings suggest that a subgroup of spindle afferents with low but unstable thresholds were fully recruited at - 10% MVT. This is reasonably consistent with previous findings that fusimotor neurons are largely recruited close to skeletomotor threshold, as has been reported for jaw muscles of the monkey (Lund et al. 1979) and triceps surae muscles of the cat (Post et al. 1980). Related to the saturation of spindle recruitment might be the finding that the afferent rate often reached a plateau quite early during the individual isometric contraction in spite of the continuously increasing force (Figs. 1, B and C, and 2, B and C). This is the expected response if the spindle was subjected to a fusimotor drive that first increased along with the contraction force up to 5- 10% of the maximal voluntary force and then leveled off (cf. Vallbo 1974b). Similar observations have been made in monkey (Larson et al. 198 1). Another subgroup was constituted by the spindle afferents that failed to accelerate in our test contractions. It remains to be clarified under which conditions their discharge may accelerate. Recruitment
thresholds
Because Burke et al. (1978a) did not report the variability of recruitment thresholds of individual spindle afferents, it is difficult to judge if there is a real difference between the data obtained from spindles in the lower limb and the present findings. It should be pointed out, however, that a high variability of thresholds as found in the present study was not a characteristic of particular subjects, because units with very small as well as units with very large differences in thresholds were recorded from the same individual subject. Moreover, although there was, on average, a slight decrease in threshold from the first to the third contraction, there was no indication that the variability decreased with moderate training ( lo-40 contractions). One of the main claims of Burke et al. (1978a) was that the muscle spindles are orderly in their recruitments when the muscle contraction increases. Because we recorded one
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
MUSCLE
AFFERENTS
AND ISOMETRIC
spindle afferent at a time, we cannot exclude mechanisms that simultaneously changed the thresholds of all or many spindle endings. If this was the case, the present findings might still be consistent with an orderly recruitment. On the other hand, orderly recruitment seems hard to reconcile with our finding that some muscle spindles recorded from individual subjects showed small differences in thresholds, whereas others showed very large differences. Moreover, some spindles recorded from individual subjects exhibited a successive decrease, whereas others exhibited a successive increase of the thresholds when very similar contractions were repeated. Furthermore, a lack of orderly recruitment has been emphasized on the basis of animal data (Murphy 198 1). Classijkation of muscle aRerents The present analysis indicates that a burst on isometric relaxation strongly supports the interpretation that the afferent originates from a muscle spindle rather than a tendon organ. The findings suggest that -70% of the muscle spindle afferents would have been identified as such with the use of this criterion alone, whereas, at the most, 5% of the tendon organ afferents would have been misclassified as muscle spindle afferents (see Fig. 4, B and C), assuming that our provisional classification was reasonably correct. It seems, therefore, recommendable to include isometric contractions with rapid relaxation in the test battery for classification of single muscle afferents in human microneurography studies along with ramp-and-hold stretch, twitch contraction, and stretch sensitization. The authors are greatly indebted to L. Backstrom and S.-O. Johansson for valuable technical assistance. The study was supported by the Swedish Medical Research Council (Grant 14X-3548) Forskningsradsnamnden, Gunvor och Josef An&-s Stiftelse, Torsten och Ragnar Soderbergs Stiftelse, Magn. Bergvalls Stiftelse, and Umea University (Fonden fdr Medicinsk forskning, ograduerade forskare). Present address of A. B. Vallbo: Dept. of Physiology, University of Goteborg, Box 330 31, SE-400 33 Goteborg, Sweden. Address for reprint requests: B. Edin, Dept. of Physiology, University of Umea, SE-901 87 Umea, Sweden. Received 5 July 1988; accepted in final form 24 January 1990. REFERENCES B., HULLIGER, M., JOHANSSON, H., AND SOJKA, P. Actions on gamma-motoneurones elicited by electrical stimulation of group II muscle afferent fibers in the hind limb of the cat. J. phvsiol. Lond. 335: 255-273, 1983a. APPELBERG, B., HULLIGER, M., JOHANSSON, H., AND SOJKA, P. Actions on gamma-motoneurones elicited by electrical stimulation of group III muscle afferent fibers in the hind limb of the cat. J. Physiol. Lond. 335: 275-292, 1983b. APPELBERG, B., HULLIGER, M., JOHANSSON, H., AND SOJKA, P. Actions on gamma-motoneurones elicited by electrical stimulation of group III muscle afferent fibers in the hind limb of the cat. J. Physiol. Lond. 335: 293-305, 1983~. BOYD, I. A. Muscle spindles and stretch reflexes. In: Scientific Basis of Clinical Neurology, edited by M. Swash and C. Kennard. London: Churchill, 1985, p. 74-97. BURKE, D., HAGBARTH, K-E., AND SKUSE, N. F. Recruitment order of human spindle endings in isometric voluntary contractions. J. Physiol. Lond. 285: 101-l 12, 1978. BURKE, D., MCI&ON, B., SKUSE, N. F., AND WESTERMAN, R. A. Anticipation and fusimotor activity in preparation for a voluntary contraction. J. Physiol. Lond. 306: 337-348, 1980a. APPELBERG,
CONTRACTIONS
1313
D., MCI&ON, B., AND WESTERMAN, R. A. Induced changes in the thresholds for voluntary activation of human spindle endings. J. Physiol. Land. 302: 17 l- 18 1, 1980b. CRAGO, P. E., HOUK, J. C., AND RYMER, W. 2. Sampling of total muscle force by tendon organs. J. Neurophysiol. 47: 1069- 1083, 1982. CROWE, A. AND MATTHEWS, P. B. C. The effect of stimulation of static and dynamic fusimotor fibres on the response to stretching of the primary endings of muscle spindles. J. Physiol. Land. 174: 109- 13 1, 1964a. CROWE, A. AND MATTHEWS, P. B. C. Further studies of static and dynamic fusimotor fibers. J. Physiol. Land. 174: 132- 15 1, 1964b. EDIN, B. B. AND VALLBO, A. B. Twitch contraction for identification of human muscle afferents. Acta Physiol. Stand. 13 1: 129- 138, 1987. EDIN, B. B. AND VALLBO A. B. Stretch sensitization in human muscle spindles. J. Physiol. Lond. 400: 10 1- 111, 1988. EDIN, B. B. AND VALLBO, A. B. Dynamic response of human muscle spindle afferents to stretch. J. Neurophysiol. 63: 1297- 1306, 1990a. EDIN, B. B. AND VALLBO, A. B. Classification of human muscle stretch receptor afferents: A Bayesian approach J. Neurophysiol. 63: 1314-1322, 1990b. HULLIGER, M., MATTHEWS, P. B. C., AND NOTH, J. Static and dynamic fusimotor action on the response of Ia fibers to low frequency sinusoidal stretching of widely ranging amplitudes. J. Physiol. Lond. 267: 81 l-836, 1977a. HULLIGER, M., MATTHEWS, P. B. C., AND NOTH, J. Effects of combining static and dynamic fusimotor stimulation on the response of the muscle spindle primary ending to sinusoidal stretching. J. Physiol. Lond. 267: 839-856, 1977b. JOHANSSON, H., SJ~LANDER, P., SOJKA, P., AND WADELL, I. Fusimotor reflexes to antagonistic muscles simultaneously assessedby multi-afferent recordings from muscle spindle afferents. Brain Res. 435: 337-342, 1987. LARSON, C. R., SMITH, A., AND LUSCHEI, E. S. Discharge characteristics and stretch sensitivity of jaw muscle afferents in the monkey during controlled isometric bites. J. Neurophysiol. 46: 130- 142, 198 1. LUND, J. P., SMITH, A. M., SESSLE, B. J., AND MURAKAMI, T. Activity of trigeminal alpha- and gamma-motoneurones and muscle afferents during performance of a biting task. J. Neurophysiol. 42: 7 10-725, 1979. MATTHEWS, P. B. C. The differentiation of two types of fusimotor fibre by their effects on the dynamic response of muscle spindle primary endings. Q. J. Exp. Physiol. 47: 324-333, 1962. MEYER-L• HMAN, J., RIEBOLD, W., AND ROBRECHT, D. Mechanical influence of the extrafusal muscle on the static behavior of deefferented primary muscle spindle endings in cat. Pfluegers Arch. 352: 267-278, 1974. MURPHY, P. R. The recruitment order of gamma-motoneurones in the decerebrate rabbit. J. Physiol. Land. 3 15: 59-67, 198 1. POST, E. M., RYMER, W. Z., AND HASAN, 2. Relation between intrafusal and extrafusal activity in triceps surae muscles of the decerebrate cat: evidence for beta action. J. Neurophysiol. 44: 383-404, 1980. PROCHAZKA, A. AND HULLIGER, M. Muscle afferent function and its significance for motor control mechanisms during voluntary movements in cat, monkey and man. In: Motor Control Mechanisms in Health and Disease, edited by J. E. Desmedt. New York: Raven, 1983, p. 93-132. SIEGEL, S. Nonparametric Statistics for the Behavioral Sciences. Tokyo: McGraw-Hill, 1956. VALLBO, A. B. Slowly adapting muscle receptors in man. Acta Physiol. Stand. 78: 315-333, 1970. VALLBO, A. B. Muscle spindle afferent discharge from resting and contracting muscles in normal human subjects. In: New Developments in Electromyographic and Clinical Neurophysiology, edited by J. E. Desmedt. Basel: Karger, vol. 3, 1973, p. 25 1-26 1. VALLBO, A. B. Afferent discharge from human muscle spindles in noncontracting muscles. Steady state impulse frequency as function of joint angle. Acta Physiol. Stand. 90: 303-3 18, 1974a. VALLBO, A. B. Human muscle spindle discharge during isometric voluntary contractions. Amplitude relations between spindle frequency and torque. Acta Physiol. Stand. 90: 3 19-336, 1974b. WESTERMAN, R. A., BURKE, D., MCI&ON, B. B., AND SKUSE, N. F. Muscle spindle activity in man. Flexibility of skeletomotor:fusimotor balance during various movements. In: Brain Mechanisms and Perceptual Awareness, edited by 0. Pompeiano and C. Ajmone Marsan. New York: Raven, 198 1. BURKE,
Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (139.184.014.150) on August 9, 2018. Copyright © 1990 American Physiological Society. All rights reserved.
