JOURNALOF NEUROPHYSIOLOGY Vol. 65, No. 2, February 199 1. Prinrcd

in U.S.A.

Effects of Muscle Fatigue on Mechanically Sensitive Merents of Slow Conduction Velocity in the Cat Triceps Surae L. HAYWARD, U. WESSELMANN, AND W. 2. RYMER Departments of Physiology and Rehabilitation Medicine, Northwestern University Medical and Veterans Administration Lakeside Medical Center, Chicago, Illinois 60611

SUMMARY

AND

CONCLUSIONS

1. Group III and IV muscle afferents have been shown to be sensitive to both mechanical stimuli and metabolic and thermal changes in muscle. To establish the potential role of slowly conducting muscle afferents in regulating motor output during fatigue, we recorded from mechanically sensitive group III and nonspindle group II afferents originating in the triceps surae in barbiturate-anesthetized cats. We evaluated the response of these afferents to tetanic muscle contraction, stretch, and surface pressure, before, during, and after fatigue. 2. Our results show that muscle fatigue both increases spontaneous discharge in these mechanically sensitive afferents and sensitizes their response to muscle stretch, surface pressure, and, in a few instances, muscle contraction. These fatigue-induced changes typically occurred after 5-- 10 min of submaximal fatiguing stimulation. 3. During recovery from muscle fatigue, several contractionsensitive free nerve endings, which had become sensitized to contractions during fatigue, remained sensitized after 20-30 min of rest. 4. The results of this study provide support for the hypothesis that fatigue-induced excitation of slowly conducting afferents is significant in mediating fatigue-induced inhibition of motoneuron output. However, our finding that the discharge of many slowly conducting mechanoreceptor afferents declinesduring the initial phase of fatigue argues against a primary role for these afferents in mediating the initial decline in motoneuron rate that is so prominent in fatiguing maximum voluntary muscular contraction.

INTRODUCTION

Over the last ten years, there has been an increasing interest in the functional role(s) of small-diameter muscle afferents (Mense 1986; Mitchell and Schmidt 1983). Historically, group III and IV muscle afferents have been thought to mediate the sensation of muscle pain (Edwards 1988; Paintal 1960) and to evoke a flexion withdrawal response when noxious stimuli are applied to muscle (Eccles and Lundberg 1959). However, several investigators (Kaufman et al. 1983, 1984; Mense and Meyer 1985) have shown that many mechanically and thermally sensitive group III and IV muscle afferents are also responsive to nonnoxious stimuli. The mechanically sensitive afferents are of particular interest because they have been shown to respond quite vigorously to normal (and presumably innocuous) levels of muscle contraction and stretch. Several non-pain-related functions have also been attributed to mechanically sensitive group III and IV afferents, including activation of a 360

0022-3077/9

1 $1 SO Copyright

School

cardiovascular reflex at the onset of exercise (McCloskey and Mitchell 1972) and, more recently, a fatigue-induced inhibitory motor reflex (Bigland-Ritchie et al. 1986). The objective of the present study was to examine the effects of muscle fatigue on the discharge of slowly conducting muscle mechanoreceptor afferents, with a view toward establishing any reflexive role such afferents might play in mediating fatigue-related changes in motor output. A fatigue-induced inhibitory reflex has been proposed by Bigland-Ritchie and co-investigators to explain the decline in motor-unit firing rate observed during fatiguing maximum voluntary contractions in human subjects (BiglandRitchie et al. 1986). In their experiments, mean motor-unit firing rate has been shown to decline steadily during a maximum voluntary contraction, broadly paralleling the loss of force output of the fatiguing muscle. This rate reduction is hypothesized to be important for maintaining a match between motor-unit firing rate and the fatigue-induced changes in the muscle’s contractile properties (i.e., declining muscle fusion frequency) (Bigland-Ritchie et al. 1983a,b). Bigland-Ritchie and co-investigators have proposed that this fatigue-induced change in motoneuron discharge may arise as a result of feedback from the fatiguing muscle and have presented indirect evidence from human studies that this reflex exists (Bigland-Ritchie et al. 1986; Woods et al. 1987). On the basis of their known sensitivities to mechanical, thermal, and chemical stimuli, group III and IV muscle afferents appear to be the most likely sources for transmitting fatigue-related changes in muscle to the CNS. However, whether the primary stimulus associated with activating these afferents during prolonged contractions is mechanical or metabolic/thermal remains unclear. The effect of muscle fatigue on mechanically sensitive muscle free nerve endings has not been tested previously. Previous studies on the response characteristics of group III and IV muscle afferents suggest that the majority of the group IV afferents are mainly nociceptors, whereas the group III afferents mainly respond to nonnoxious mechanical stimuli (Mense and Meyer 1985). Several characteristics of the slow conducting mechanoreceptors indicate that they may be involved in relaying this fatigue-induced reflex. First, a subgroup of the mechanically sensitive afferents has been shown to be very responsive to muscle contraction (Kaufman et al. 1983, 1984; Mense and Meyer, 1985). Second, group III and IV mechanoreceptors have been shown to be activated both by increased muscle temperatures, which are known to occur during prolonged muscle contrac-

0 199 1 The American

Physiological

Society

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

EFFECTS

OF

FATIGUE

ON

MUSCLE

tion (Hertel et al. 1976), and by chemical substances such as bradykinin, lactic acid, prostaglandins and K+ (Kaufman et al. 1982, 1983; Kniflki et al. 1978; Mense and Meyer 1988; Mense and Stahnke 1983; Rotto and Kaufman 1988; Rotto et al. 1990), which may be released during fatiguing contractions (Rotto et al. 1989; Thimm and Baum 1987; Vollestad and Sejersted 1988; Westra et al. 1988). However, the influence of any of these substances on the mechanical sensitivity of slowly conducting mechanoreceptors remains unclear (Mense and Meyer 1988; Rotto et al. 1990). The goal of this study was to determine the effects of muscle fatigue on the response characteristics of slowly conducting muscle mechanoreceptor afferents. We also compared the fatigue-induced changes of these muscle afferents to those recorded from several larger diameter muscle spindle and Golgi tendon organ afferents. Portions of these findings have been reported in abstract form (Hayward et al. 1988b). METHODS

General Experiments were performed in 45 barbiturate-anesthetized cats of either sex, 1.5-3 kg. Cats were initially anesthetized with a gaseous mixture of halothane-nitrous oxide-oxygen and later switched to pentobarbital sodium (40 mg/kg iv). The trachea, carotid artery, and external jugular vein were cannulated, and blood pressure was monitored from the right carotid artery. The left triceps surae muscles, medial and lateral gastrocnemius (MG and LG), and soleus (SOL), were isolated from adjacent muscles and severed from their calcaneal attachments. This included cutting the tendon of plantaris from its calcaneal attachment and sectioning the tendinous connection between MG and LG. The left hind limb was immobilized by knee and ankle clamps attached to a muscle puller. Skin flaps from the leg were pulled up to form an oil pool over the exposed muscles and nerves. To isolate afferent input from the triceps surae muscles, we sectioned the nerves innervating all the hip muscles and all muscle and cutaneous nerves in the popliteal fossa, except those innervating the triceps surae muscles. Afferent activity was recorded from sectioned lumbar dorsal rootlets exposed after a dorsal laminectomy. An oil pool was formed over the exposed spinal cord with the use of skin flaps from the back. Peripheral muscle and spinal cord oil pool temperatures were maintained by radiant heat, and core temperature was monitored by an esophageal probe. Body temperature was maintained between 36 and 38” by a ventral heating pad positioned under the cat and by radiant heat.

Single-unit

isolation and ident&ation

Single dorsal root afferents, originating from one of the tricep surae muscles, were isolated by progressive teasing of sectioned L,, or S, dorsal rootlets. Afferent discharge was recorded with monopolar platinum-iridium wire electrodes. Afferent activity was amplified 1OO- 10,000 times and bandpass filtered between 0.1 and 10 kHz. The action potential of isolated units was passed through a window discriminator, which issued a 5 V pulse when the action potential satisfied specific time and voltage conditions. These pulses were used to reset a computer clock, and interstimulus interval was recorded on-line from the window discriminator TTL output to a resolution of 0.1 ms. Afferents were classified initially by conduction velocity and further by the characteristics of their responses to light surface pressure. muscle stretch. vibration. and contraction. Conduction

MECHANORECEPTORS

361

velocity was calculated from the latency of the dorsal root discharge elicited in response to electrical stimulation of the peripheral nerve (group I, 120-72 m/s; group II, 72-30 m/s; group III, 30-2.5 m/s; group IV, ~2.5 m/s) (Eccles and Lundberg, 1959). Group I fibers were classified as either muscle spindle (Ia), or Golgi tendon organ (Ib) afferents. Muscle spindle afferents respond dynamically to muscle stretch and are unloaded during contraction. Golgi tendon organ afferents respond vigorously to muscle contraction but respond more modestly to passive stretch (Matthews 1972). Group II afferents were classified as either muscle spindle or nonspindle afferents by their response to muscle stretch and light surface pressure. Muscle spindle group II afferents respond well to stretch of muscle and are also unloaded during muscle contraction (Matthews, 1972). In contrast, the nonspindle group II afferents may respond quite vigorously to muscle contraction, stretch, and light or moderate surface pressure (Paintal, 1960). On the basis of the similarities between their response characteristics, all group III and nonspindle group II afferents were classified as “mechanically sensitive free nerve endings.” This broad classification included afferents that responded quite vigorously to muscle contraction, stretch, and surface pressure and afferents that were relatively insensitive to muscle contraction but responded well to innocuous levels of muscle stretch and surface pressure.

Data collection and experimental

protocol

After isolated afferents were classified, afferent discharge rate was recorded in response to muscle stretch, vibration, contraction, and surface pressure. Muscle length changes were generated by a muscle stretcher, configured as a position servo under the control of a computer. Ramp stretches (0.5 s in duration) varied in velocity between 2 and 20 mm/s between trials. Vibration was induced by small-amplitude, high-frequency, longitudinal stretches (20100 pm, 50-300 Hz). Surface pressure was applied over the receptive field with the use of different stimulus intensities, including brushing the muscle surface with a soft paintbrush, von Frey hairs, and tendon squeezing with smooth forceps at both moderate and strong (presumably noxious) intensities. Isometric contractions (0.5-l s in duration) were elicited by electrical stimulation (pulse duration, 0.2 ms), delivered by bipolar platinum-iridium wire electrodes placed on the isolated muscle nerve. Stimulus intensity was set at 1.3 times motor threshold, and muscle force levels were changed by varying the stimulation frequency between 15 and 50 Hz. Muscle force was recorded from a load cell placed in series with the attachment of the muscle tendon to the muscle puller. We recorded afferent interspike interval, muscle force, and length during muscle stretch, muscle contraction, or application of surface pressure. Force and length signals were low-pass filtered at 50 Hz and sampled at 100 Hz by A/D converters. Before fatigue, afferent discharge was recorded during a sequence of 2- 15 trials of each type of stimulus: repeated stretches of two different amplitudes, varying tetanic contraction forces, and different intensities of surface pressures. Spontaneous activity was recorded if present. After collection of control values, the muscle was fatigued by electrical stimulation of the muscle nerve at 25 Hz, with stimulus intensity set to 1.3 times motor-axon threshold. Stimulation was applied for 800 ms and repeated every second. Muscle force was continuously monitored on the oscilloscope, and fatigue was defined to be present when force output fell to 530% of its original value. Typically, force output for both the MG and LG muscles dropped to 30% of the original force after 8- 10 min of fatiguing stimulation. Once fatigue was present, afferent discharge was recorded during repeated trials of stretch, contraction, and surface manipulation. Between groups of trials. muscle fatigue was main-

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

362

HAYWARD,

WESSELMANN,

tained by continuing electrical stimulation of the muscle nerve (with the use of the same stimulation parameters used to fatigue the muscle initially). The time of fatigue was calculated by summing the total number of minutes of repeated fatiguing stimulation. In three experiments, afferent firing rate was also recorded during ischemia combined with fatigue. Ischemia was produced by reversibly blocking blood flow in the popliteal artery just before the offset of a period of fatiguing stimulation. Postfatigue responses or recovery responses were typically recorded for 20-30 min after the last fatiguing stimulation period. Typically, force output had not returned to the prefatigue levels after 20 min of rest; however, it might be expected that the metabolic and thermal state of the muscle would have recovered within this time period (Miller et al., 1987).

Data analysis Afferent discharge rate, muscle length, and force were quantified by averaging the responses over specified time windows. For example, afferent firing rate and muscle force were averaged during 700 ms of tetanic contraction to provide a profile of average afferent firing rate as a function of mean force output. For muscle spindle and Golgi tendon organ afferents, responses to stretch were averaged during both the peak of stretch (peak dynamic response, 50-ms window) and during the hold phase (static stretch, IOO-ms window during static stretch). Because the discharge rates of the group III and nonspindle group II afferents were generally lower, dynamic stretch responses were averaged over the entire stretch (0.5 s), and static stretch responses were averaged over the entire hold period (3.5 s). If an afferent was discharging spontaneously, the spontaneous discharge rate was subtracted from the average firing rate during ramp stretch and hold. All values are expressed as means t SD. The percent change in afferent discharge during fatigue compared to prefatigue was calculated by subtracting the average prefatigue rate from the mean fatigue-induced rate and dividing this mean change in firing rate by the mean prefatigue rate. Paired Student’s t tests (P < 0.05) were used to compare the significance of the changes in afferent firing rate during fatigue to prefatigue rates. Averaged afferent firing rates were paired by stimulation conditions (i.e., stretch amplitude) and trial number. Pairing by trial number was important to accommodate for adaptation, which occurs over repeated trials of the same stimulus for the small-diameter mechanically sensitive afferents (Cleland et al. 1990; Paintal 1960). When multiple comparisons were made, Bonferroni adjustments were made to find acceptable probability levels. RESULTS

We successfully recorded the discharge of 56 mechanically sensitive muscle afferents before and during muscle fatigue. Twenty-nine were classified as mechanically sensitive free nerve endings, including 23 group III afferents (conduction velocity range, 5.5-28 m/s) and 6 nonspindle group II afferents (conduction velocity range, 3 l-67.7 m/s). The remaining afferents (27/56) were classified as either muscle spindle ( 14 group Ia and 9 group II) or Golgi tendon organ afferents (8 group Ib). The majority of the group II (nonspindle) and group III afferents was isolated from the MG muscle (22/29) and had receptive fields that were located superficially, either in the distal muscle-tendon aponeurosis or in the tendon itself (see insets in Fig. 1). The remaining smaller diameter muscle afferents were found in either SOL (5/29) or LG muscles (2/29). Most of the larger

AND

RYMER

diameter (Ia and Ib) afferents were also recorded from the MG muscle (17/27). The remainder were located in either LG (9/27) or SOL muscles ( l/27). This distribution of afferents reflects our preference to record small-diameter muscle afferents originating in the MG muscle, a fatigable muscle that could be selectively stimulated through its isolated muscle nerve. Response characteristics of free nerve endings TO CONTRACTION. Of the 29 mechanically sensitive free nerve endings recorded, more than one-half responded vigorously to tetanic muscle contractions (59%, including 11 group III and 6 group II afferents). The response to contraction typically began immediately after the onset of contraction and maintained discharge throughout the contraction (see Fig. 1, A and B, left panels). Overall, there was a tendency for the discharge rate of the contraction-sensitive afferents to increase with increasing force of contraction, as illustrated on the lefi side of Fig. 1A (mean contraction sensitivity, 4.86 t 2.77 imp s-l N-l; n = 12). Several of the free nerve endings that were sensitive to contraction (7/ 17) appeared to respond more to the fluctuations in force induced by unfused contractions than to the mean force level (exemplified in the left panel of Fig. 1B). For example, at the lower forces these free nerve endings discharged several times with each “ripple” of the unfused contraction. At higher stimulation frequencies, as fusion frequency was approached and muscle force production increased, the afferents would discharge fewer spikes, in association with each smaller “ripple.” As a result, the response of these afferents to a stronger, fused contraction was not much greater compared with their response to the lower force, unfused contractions. A minority of the 29 mechanically sensitive free nerve endings ( 12/29) were relatively insensitive to tetanic muscle contraction (as shown in the left panel of Fig. 1C) but responded well to other mechanical stimuli (i.e., stretch and surface pressure). RESPONSE

l

l

TO STRETCH. Twenty-seven of the 29 mechanically sensitive free nerve endings demonstrated sensitivity to muscle stretch within the physiological range (including the 17 afferents that were also sensitive to contraction and 10 free nerve endings that were relatively insensitive to muscle contraction). The stretch responses of those free nerve endings insensitive to contraction were indistinguishable from those group III and nonspindle afferents sensitive to contraction (see Fig. 1, B and C). All 27 free nerve endings discharged in association with the dynamic aspects of stretch and showed an increase in discharge rate with increasing stretch amplitude. The sensitivity to dynamic stretch of 6 mm ranged from 1 to 18 spikes discharged during the 0.5-s ramp stretch period (mean dynamic stretch sensitivity, 1.02 t 0.95 imp s-l mm-‘; n = 15). Eighteen of the 27 afferents sensitive to dynamic stretch also discharged during the static stretch. The mean static rate during the 3.5-s hold period was 1.22 t 0.93 imp/s (for a mean final length of 2.22 t 1.5 mm less than physiological maximum, n = 12). Two afferents also discharged during the release of stretch. RESPONSE

l

l

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

EFFECTS OF FATIGUE

A

contraction-sensitive

Force

Rate

ON MUSCLE

receptive region with a soft paintbrush as shown in Fig. 1A. Typically, afferent discharge rate increased in response to increasing surface pressure (light to moderate to noxious squeezes). The mean response to moderate pressure was 8 9 -+ 7.1 imp/s (n = 18). Before fatigue, 8 of the 29 free nerve endings showed spontaneous activity (mean rate, 0.39 t 0.36 imp/s).

free nerve ending

101 30 -

(imp / s> 0

paintbrush

B

contraction-sensitive

Rate

(imp/s)

free nerve ending

m

stretch

C

contraction-insensitive Force

N

moderate tendon squeeze

40 0 II.

(6 mm)

0u

light surface pressure

E&cts of muscle fatigue on free nerve ending

response characteristics During muscle fatigue, there was an increase in both the mechanical sensitivity and spontaneous discharge of the muscle free nerve endings recorded. The fatigue-induced increases typically developed during 1O-20 min of fatiguing contractions, although they were often evident much earlier. These fatigue-induced changes were maintained by repeated periods of fatiguing submaximal contractions (typically 5 min in duration).

A Force PO

(_MG

i

Pre-fatigue 81

free nerve ending

121 m

//

s

0

Rate

(imp/s)

30

0 1 stretch

(4 mm)

moderate pressure

Response characteristics of mechanically sensitive muscle free nerve endings. In this and subsequent figures single trials of afferent instantaneous discharge rate are plotted as a function of time, solid bars indicate the duration of stimulus application, and insets show the location of each afferent’s receptive field (r.f.). A and B: response of characteristics of contraction-sensitive free nerve endings [conduction velocity (c.v.) 3 1 and 19 m/s, respectively] to different intensities of muscle contraction (left panels), surface pressure applied to the afferent’s receptive field, and/or stretch (right panels). C: response of a contraction-insensitive free nerve ending (c.v. 12 m/s) to similar mechanical stimuli. FIG.

Fatigue

30

IS

Rate

(imp/s>

363

MECHANORECEPTORS

B IS

1.

Force

(N)

50 I

10 min. fatigue

I I

During repeated stretches the response of all free nerve C endings adapted to 50% of the original response within three to five trials (mean number of trials for dynamic adaptation, 4.7 t 2.5 trials; n = 17; mean number of trials for static adaptation, 5.4 t 2.6 trials; n = 17). All of the five free nerve endings tested were insensitive to longitudinal tendon vibration. All 29 mechanically RESPONSE TO SURFACE PRESSURE. sensitive free nerve endings responded to light or moderate FIG. 2. Changes in the contraction sensitivity of mechanically sensitive local surface pressure. Figure 1, right, illustrates the re- muscle free nerve endings during fatigue. In this and subsequent figures, the hatch-filled boxes reflect the time of fatiguing stimulation applied besponse to surface pressure of two free nerve endings that fore the afferent’s response. A: after 3 min of fatiguing stimulawere also sensitive to contraction and one afferent that was tion recording this group III afferent [conduction velocity (c.v.) 19 m/s] was unreinsensitive to contraction. The mean responses to light and sponsive to muscle contraction but recovered to near prefatigue levels of moderate surface pressure of the two types of afferents (con- contraction sensitivity after further fatiguing stimulation. B: after 10 min of fatiguing stimulation, the response of this contraction-sensitive free traction sensitive or contraction insensitive) were indistinending (c.v. 67.7 m/s) was sensitized above its prefatigue response. guishable. Fifteen of the 29 free nerve endings were very nerve C: this group III free nerve ending (c.v. 15.2 m/s) responded well to contracsensitive to mechanical pressure applied over their recep- tion after 10 min of fatiguing stimulation but was insensitive to contractive fields and discharged briskly during brushing of the tion before fatigue. Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

HAYWARD,

364

WESSELMANN,

TO CONTRACTION. Figure 2A illustrates the typical pattern of changing contraction sensitivity observed during fatigue in those free nerve endings sensitive to contraction. Before fatigue, this free nerve ending discharged quite vigorously during moderate, unfused contractions. Three minutes into the fatigue protocol, this afferent only discharged 3 or 4 spikes in response to a similar level of contractile force. Later, after an additional 10 min of repeated fatiguing stimulation, there was some recovery of the afferent’s sensitivity to contractions produced by the same frequency of stimulation. The afferent’s response to varying force levels (matched to prefatigue levels by increasing the frequency of stimulation) was comparable or slightly below the prefatigue rates (see Fig. 3A). This pattern of changing contraction sensitivity during fatigue was observed for all 17 free nerve endings sensitive to contraction before fatigue. Typically, there was a decline in contraction sensitivity during the first 5 min of fatiguing stimulation. After 1O-20 min of repeated fatiguing stimulation, the majority (1 l/l 7) of the responses of those free nerve endings sensitive to contraction before fatigue returned to prefatigue force-rate relationships or were slightly reduced (see MG unit in Fig. 3A). Two contraction-sensitive free nerve endings ( 1 located in SOL, 1 in MG) showed an increase in their response to contractile force after lo-20 min of fatiguing stimulation. Figure 2B illustrates one of these afferents. As shown in the right panels, after a total of 10 min of fatiguing stimulation, the response of this afferent to comparable levels of force was considerably enhanced compared to prefatigue responses. Figure 3B illustrates the fatigue-induced change in the force-rate relationship for this afferent. The increase in the afferent’s response to lower force contractions suggest that, in this instance, muscle fatigue may have enhanced the sensitivity of this afferent to the fluctuations in muscle force rather than to the absolute force of contraction. RESPONSE

AND

RYMER

Four of the 17 free nerve endings sensitive to contraction before fatigue did not recover their contraction sensitivity during fatigue (2 were located in MG, 1 in LG, and 1 in SOL). Before fatigue, three of these free nerve endings appeared to be more sensitive to the movement associated with the unfused tetani than the force of contraction (see above). Despite a loss of contraction sensitivity during fatigue, these same afferents did show fatigue-induced increases in their responses to other types of mechanical stimulation (stretch, pressure) or increases in spontaneous activity. Figure 2C illustrates the changes in the contraction sensitivity of one afferent that first became sensitive to contraction during fatigue. After a total of 15 min of fatiguing stimulation, this afferent was extremely sensitive to muscle contraction and behaved similar to a free nerve ending sensitive to contraction before fatigue. RESPONSE TO STRETCH. Figure 4 illustrates the typical increase in the response of a mechanically sensitive free nerve ending to stretch (Fig. 4A) during fatigue. Seventeen of the 27 free nerve endings sensitive to stretch before fatigue showed similar increases in their response to dynamic stretch of 4,6, or 8 mm after 15 min of fatigue. The other 10 free nerve endings showed little or no change or a slight decrease in their response to dynamic stretch (see Table 1). During fatigue the mean dynamic sensitivity for a 6-mm ramp stretch was 1.12 t 0.85 imp s-l . mm-’ (n = 13). The fatigue-induced mean rate during static stretch was 2.09 t 1.92 imp/s (for a final stretch length 2.22 t 1.5 mm less than physiological maximum, n = 10). There was no significant change in the pattern of adaptation for dynamic stretch (mean number of trials for dynamic adaptation, 3.6 t 2 trials; n = 14) but there was a significant decrease in the mean number of trials for adaptation to 50% of the original response during static stretch (mean number of trials for static adaptation, 3.6 t 1.9 trials; n = 13). l

B

A 60

40

Rate

Rate (imp/s)

(imp/

lo+----0

MG force (N)

1

2

”

”

3 4 SOL force (N)

”

5

’

6

between mechanoreceptor discharge and force before and during muscle fatigue. The discharge rate FIG. 3. Relationship of contraction-sensitive free nerve endings during short-duration tetanic contractions (600 ms) of varying intensities are shown. Open squares represent the response before fatigue, and the filled triangles illustrate the response during fatigue. A: prefatigue vs. fatigue force-rate relationship of the same group III mechanoreceptor [conduction velocity (c.v.) 19 m/s] illustrated in Fig. 2A. B: force-rate relationship of the same nonspindle group II (c.v. 67.7 m/s) mechanoreceptor shown in Fig. 2B.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

EFFECTS

A

be-fatigue

OF

FATIGUE

ON

MUSCLE

10 /

GzGiE*

.f . I

40 2s

Rate

(imp/s)

. I moderate squeeze

paintbrush on surface

-paintbrush on surface

moderate squeeze

FIG. 4. Fatigue-induced changes in the stretch and pressure sensitivity of a mechanically sensitive free nerve ending. A: after 10 min of fatiguing stimulation the response of this free nerve ending (also shown in Fig. 2C) to 6-mm stretch was enhanced compared to prefatigue levels. B: response of the same afferent to light and moderate surface pressure was also enhanced after 15 min of fatiguing stimulation.

TO SURFACE PRESSURE. The responses of 2 1 of the 29 free nerve endings were tested both before and during fatigue to surface pressure. Overall, muscle fatigue induced a moderate increase in the pressure sensitivity of these afferents (see Fig. 4B). Fourteen of the 21 afferents showed increases in their response to surface manipulation. Seven free nerve endings showed no change or a decrease in sensitivity to surface pressure during fatigue (see Table 1). Seven of the eight free nerve endings that were discharging spontaneously before fatigue increased their spontaneous discharge during fatigue (mean percent increase, 17 1.1 t 135.1%; n = 7). The remaining afferent did not change its spontaneous discharge rate during fatigue. Also, six free nerve endings that had no spontaneous discharge before fatigue began discharging during fatigue (mean fatigue-induced spontaneous discharge rate, 0.95 $- 0.4 imp/ s; n = 6; see Table 1 for total mean percent change).

RESPONSE

TABLE

In three experiments we combined fatigue with ischemia, testing the responses of three mechanically sensitive free nerve endings. Typically, after a total of 35 min of fatiguing stimulation, the blood supply to the muscle was blocked for 5 min. During this period of ischemia combined with fatigue, all three afferents showed an increase in spontaneous discharge and/or stretch sensitivity, above the fatigue-induced changes. After release from ischemia and lo- 15 min rest, the afferents’ spontaneous discharge rate and responses to stretch declined to near prefatigue rates. RESPONSE TO STRETCH. Figure 5A illustrates the stretch responses of one of these free nerve endings before fatigue, during fatigue, and during fatigue combined with ischemia. During fatigue the mean response of this afferent to stretch did not change significantly (see 2nd panel on the left in Fig. 5A). After a total of 35 min of fatiguing stimulation, the blood supply to the muscle was blocked for 5 min. During this ischemic period, the afferent’s response to stretch increased significantly above the previous fatigue response (see 3rdpanel Fig. 5A; increase in stretch response was 64%, P < 0.05). After release from ischemia and 10 min of rest, the response of the afferent to stretch returned to prefatigue levels (see right panel). All three afferents that were tested during ischemia responded to stretch before fatigue. During fatigue the stretch sensitivity of these free nerve endings increased only slightly above prefatigue levels (mean percent increase, 14.0 t 33.8%; n = 3). During ischemia the sensitivity to stretch increased above the fatigue-induced levels (mean percent increase above fatigue level, 40.0 t 89.4%; n = 3). SPONTANEOUS DISCHARGE. Figure 5B illustrates the effect of fatigue and ischemia on spontaneous discharge of the same free nerve ending shown in Fig. 5A. During fatigue the spontaneous discharge of this afferent increased 128% above the prefatigue rate (P < 0.05). During the 5 min of ischemia, the afferent’s spontaneous discharge rose steadily, increasing 60% above previous fatigue rates. After release from ischemia and 10 min rest, the afferent’s spontaneous discharge rate declined to below the fatigue-induced rate, returning close to the prefatigue rate. Two of the three afferents recorded in response to fatigue and ischemia were spontaneously discharging before. fatigue (including the afferent described in Fig. 5B). Muscle fatigue induced an increase in the spontaneous discharge rates of these two afferents (mean percent increase, 118.0 t

Fatigue-relatedchangesin mechanoreceptor .free nerve endingresponses

1.

% Increase Spontaneous

discharge

n Stretch

response

n Surface

n Values prefatigue

365

Eflects of muscle fatigue and ischemia on free nerve ending response characteristics

Fatigue

min.

B

MECHANORECEPTORS

pressure

response

% Decrease

135.6 + 98.9 (10) 14 127.4 -t 88.9 (9) 16 70.5 AI 59.2 (8) 14

represent means + SD in fatigue-induced discharge rates (P < 0.05) are in parentheses; n is total number

rates vs. prefatigue of responses.

Total

0 (0) 1 25.9 t 22.8 (2) 10 6.5 AZ 11.3 (1) 6 rates;

numbers

of responses

% Change

126.5 + 101.5 (10) 15 91.1 x!I 155.4(11) 26 47.4 -t 61.2 (9) 20 that

were significantly

different

from

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

366

HAYWARD,

A

Pre-fatigue

Fatigue

WESSELMANN,

Ischemia

AND

RYMER

Recovery

.. ... ... ... : .lof

Len@(mm>

.min. *rest .. .. .. .. .. .. .. .. ....

Rate (imp/s

FIG. 5. Combined influence of muscle fatigue and ischemia on the stretch sensitivity and spontaneous discharge of a group III mechanically sensitive free nerve ending. A: during fatigue (2ndpanel) the response of this contraction-sensitive free nerve ending (c.v. 2 1.5 m/s) to stretch (6 mm) did not change significantly compared to prefatigue. During fatigue combined with ischemia, the response to stretch increased significantly (see text) above the fatigue responses. After ischemia the response of the afferent to stretch recovered to prefatigue levels. B: spontaneous discharge rate of the same afferent also increased during repeated fatiguing stimulation and fatigue combined with ischemia. Vertical bars reflect the average spontaneous discharge rate calculated over 14 s.

5 min. ischemia

B

Pre-fatigue

Recoverv

Fatigue Ischemia

Rate (imp/s)

1

10

20

30 time (min.)

intermittent fatiguing stimulation

40

50

ischemia

14.1%, n = 2), and fatigue combined with ischemia resulted in a further increase in spontaneous discharge (mean percent increase above fatigue-induced change, 59.0 t 1.4%, n = 2). The remaining afferent was not activated by either fatigue or fatigue combined with ischemia. Time course of recovery offree nerve endings after fatigue The responses of six free nerve endings were followed for 20-30 min after the last fatiguing stimulus. The stretch sensitivity of all six afferents returned to prefatigue levels after a minimum of 20 min rest. During recovery the contraction responses of all three afferents studied remained enhanced, TABLE 2.

60

becoming more sensitive to contraction tigue responses. Efect offatigue

compared to prefa-

on muscle spindle aferents

We recorded from 15 group Ia and 8 group II muscle spindle afferents, before and during muscle fatigue. Table 2 shows the fatigue-induced changes in spontaneous discharge and stretch sensitivity recorded from these larger diameter muscle afferents. The responses of both muscle spindles and Golgi tendon organ afferents to muscle contraction were not examined. The low electrical threshold of these larger diameter afferents made it impossible to electri-

Fatigue-relatedchangesin large-diameterafferent responses

Muscle spindle, Ia-afferents Spontaneous discharge n Stretch response n Muscle spindle, II afferents Spontaneous discharge n Stretch response n Golgi tendon organ, Ib-afferents Spontaneous discharge n Stretch response n

% Increase

% Decrease

Total % Change

42.0 IL 15.2 (3) 3 36.9 + 11.7 (8) 8

6.0 -t 4.3 (1) 12 2.0 Ik 2.0 (0) 2

3.6 -t 6.7 (4) 15 18.7 -t 8.0 (8) 15

0

(0)

0

(0)

0

(0)

0

0

70.7 t 28.9 (3) 3

23.5 AI 54.1 (3) 4

4 8.0 t- 8.0 (1) 5

4 13.8 -+ 10.0 (2) 6

4 21.5 k 17.9 (4) 8

4 21.5 IL 17.9 (5) 10

Legend, see Table 1.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

EFFECTS

OF

FATIGUE

ON

MUSCLE

MECHANORECEPTORS

367

tally evoke a muscle contraction without electrically driving the afferent fibers. RESPONSE TO STRETCH. Eight Ia-afferents showed a small increase in response to stretch after fatigue, and seven Iafibers showed either no change or a slight decrease in response to stretch during fatigue. Of the eight group II afferents recorded, three showed an increase in their response to stretch during fatigue, and five showed no change or a slight decrease during fatigue (see Table 2). SPONTANEOUS DISCHARGE. All 15 Ia-afferents had spontaneous discharge before fatigue. After 20 min of fatiguing stimulation, three Ia-afferents showed significant increases in their spontaneous discharge (P < 0.05) whereas the remaining afferents showed either no change or a slight decrease in their spontaneous activity (see Table 2). Four spindle group II afferents were spontaneously discharging before fatigue, but no change in afferent spontaneous discharge rate was recorded during fatigue.

tions (Herbaczynaska-Cedro et al. 1976; Rotto et al. 1989). Each of these substances has been shown to increase the spontaneous discharge of both group III and IV mechanoreceptors and nociceptors (Kniflki et al. 1978; Mense 198 1; Mense and Stahnke 1983; Mense and Meyer 1985; Rotto and Kaufman 1988). In addition, many of these substances have also been shown to increase the mechanical sensitivity of the nociceptors (Mense 1986). However, most studies have shown that the mechanical threshold of group III and IV mechanoreceptors is often greater in the presence of lactic acid, bradykinin, KCl, or local ischemia (each applied separately) (Kniflki et al. 1978; Mense 1986; Mense and Meyer 1988; Mense and Stahnke 1983). In this regard, it has been hypothesized that it may be the nociceptors and not the mechanically sensitive muscle free nerve endings that signal peripheral changes in the muscle as their mechanical threshold is lowered by metabolic by-products of muscle activation (Kaufman et al. 1983; Rotto et al. 1989). We did not record from any nociceptive free nerve endings in the present study. However, the results of our study Eflect of muscle fatigue on Golgi tendon organ aflerents demonstrate that the mechanical threshold of group III and nonspindle group II mechanoreceptors may also be lowered RESPONSE TO STRETCH. We recorded from eight Golgi by the metabolic and thermal changes that occur in muscle tendon organ (Ib) afferents, all of which were stretch reduring fatigue. These results suggest that free nerve ending sponsive before fatigue. After 1O-20 min of fatiguing stimuin addition to nociceptive free nerve lation, three Ib-fibers showed slight increases in their re- mechanoreceptors, endings, may be involved in relaying signals of peripheral, sponse to stretch (see Table 2), and five Ib-fibers showed no change or a slight decline in their stretch response. The fatigue-related changes in muscle. One possible explanation decline in stretch sensitivity of two of these Ib-fibers was for the increased mechanical sensitivity of the group III and significant (P < 0.05) and appeared to be related to a paral- nonspindle group II mechanoreceptors that we recorded is that, during muscle fatigue, a variety of metabolic sublel decline in passive force elicited during muscle stretch. stances released in parallel with thermal changes in the SPONTANEOUS DISCHARGE. Four of these aflYerents were muscle. are In other words, the change in mechanical sensitivspontaneously discharging before fatigue. There were no ity during muscle fatigue may reflect the combined eficts of fatigue-induced changes in the spontaneous discharge rate several dtrerent changes in the internal state of the muscle, of any of these Ib-fibers. all of which contribute to the fatigue-related changes in the contractile properties of muscle. The majority of studies DISCUSSION investigating the effect of metabolic stimuli on free nerve endings has examined the effects of injection of single subThe results of our study demonstrate that mechanically sensitive muscle free nerve endings, including group III and stances into the muscle’s blood supply on the afferent disnonspindle group II fibers, are excited after fatiguing mus- charge (Mense 1986). No study has attempted to examine cle contractions. The fatigue-induced excitation we ob- the combined and simultaneous effects of metabolic and served included both increases in spontaneous discharge thermal changes. and sensitization to mechanical stimuli, including muscle stretch, local pressure, and muscle contraction. In compariFatigue-induced changes in the contraction sensitivity of son, the response characteristics of the larger diameter, free nerve endings muscle spindle and Golgi tendon organ afferents were only Although we observed a general increase in stretch and slightly modified during muscle fatigue. Together, these results support the hypothesis that reflex modulation of mo- pressure sensitivity of slow conducting mechanoreceptors tor output during muscle fatigue may be induced by activa- during fatigue, the changes in contraction sensitivity that tion and sensitization of muscle free nerve endings (Big- these free nerve endings underwent were more complex. First, we reported a consistent decline in the contraction land-Ritchie et al. 1986). sensitivity of free nerve endings during the first 5 min of fatiguing stimulation. This was followed by a recovery of Fatigue-induced mechanisms for changes in free contraction sensitivity for the majority of the free nerve nerve endings responses endings. Mense and Stahnke ( 1983) recorded from group The concentrations of several metabolic by-products, in- III and IV mechanoreceptors during rhythmic muscle concluding lactic acid, H+, KCl, and phosphates (Miller et al. tractions (electrical stimulation was applied for 0.5 s every 1987; Westra et al. 1988) have been shown to increase dur- second) and demonstrated a moderate reduction in the afing prolonged, fatiguing contractions. In addition, other ferent response over 2 min. Moreover, Kaufman et al. substances, such as bradykinin and prostaglandins, have (1983) have demonstrated that when a contraction is susalso been shown to be released during prolonged contrac- tained without rest periods, the response of group III mechaDownloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

368

HAYWARD,

WESSELMANN,

noreceptors to the ongoing contraction declines very rapidly, within tens of seconds, despite little reduction in muscle force. Although the exact mechanism underlying this initial decline in afferent contraction sensitivity is unknown, a similar mechanism may have been responsible for the initial decline in sensitivity reported in the present study. It is possible that this initial decline in contraction sensitivity reflects the ongoing changes in the contractile properties of the muscle. For example, we have shown that many of the contraction-sensitive free nerve endings were very sensitive to the “ripple” or fluctuation in force associated with the unfused, electrically evoked contractions. As the muscle fatigues and contraction time slows, the contraction becomes more fused in response to a given stimulation frequency, and there is less fluctuation in force. In addition, free nerve endings have also been shown to adapt quickly to repeated stimuli (Cleland et al. 1990) and this initial decline in contraction sensitivity may simply reflect the receptor’s adaptation properties. Whether these same characteristics of changing responsiveness to continuous contractions occur during a naturally evoked contraction is unknown. Alternatively, the early decline in contraction sensitivity may be explained by the presence of ischemia or the early release of an endogenous substance in the muscle. For example, Mense and Stahnke (1983) have shown that the contraction sensitivity of group III and IV mechanoreceptors is diminished when the muscle is made ischemic. In addition, in some cases, the presence of bradykinin and prostaglandins has been shown to desensitize group III mechanoreceptors to muscle contraction (Mense and Meyer, 1988). After the initial decline in group III and nonspindle group II contraction sensitivity, we recorded a recovery in contraction sensitivity in more than one-half of the population or, in several cases, enhancement of the response to contraction. It is possible that the recovery of contraction sensitivity may have reflected the release of an endogenous substance, which in combination with the other metabolic and thermal changes produced a recovery in the contraction sensitivity of these afferents. In this regard, recently, Rotto and colleagues (1990) demonstrated that injection of arachidonic acid into the muscle’s blood supply produced an increase in the contraction sensitivity of group III mechanoreceptors. In addition, the presence of arachidonic acid was shown to elicit a similar triphasic response from the group III mechanoreceptors to the ongoing static contraction. The investigators hypothesized that the later increases in afferent discharge may have resulted from the release of metabolic by-products, because muscle force was unchanged or declining during that time. However, at this time the mechanism underlying the complex changes in contraction sensitivity recorded during muscle fatigue remains unclear. Recovery ofmechanically musclefatigue

sensitivefree nerve endings after

We also recorded from several mechanically sensitive free nerve endings during recovery from muscle fatigue. Several of these afferents were sensitized to muscle contrac-

AND

RYMER

tion during fatigue and maintained this enhanced sensitivity to contraction after 20 min of rest. The persistence of the altered mechanical sensitivity after recovery from fatigue suggests that these afferents could play an important role in relaying signals of muscle soreness or muscle damage resulting from fatigue (Edwards 1988). Fatigue-induced changes in the response of large myelinated aferents In contrast to the complex changes in the group III and nonspindle group II mechanoreceptors, we found relatively little change in either the stretch sensitivity and spontaneous activity of muscle spindle or Golgi tendon organ afferents during muscle fatigue, although it should be noted that our electrical stimuli were insufficient to drive y-motoneurons and may not have accurately replicated natural fatigue. Nelson and Hutton (1985) found short-term increases in spontaneous discharge and stretch sensitivity of both Ia and group II muscle spindle afferents when the whole muscle was fatigued with the use of a supramaximal stimulation intensity (relative to motor-axon threshold). In contrast, they reported little change in afferent sensitivity when the fatiguing stimulation was submaximal. Their results suggested that an increase muscle spindle stretch sensitivity during fatigue may not be due to metabolic or thermal changes in the muscle but may result from stimulation of the intrafusal fibers. Christakos and Windhorst (1986) reported an increase in the response of muscle spindles to contraction of isolated motor units during fatigue. These investigators hypothesized that changes in the muscle’s contractile properties and the associated changes in muscle length during contraction may have been responsible for changes in contraction sensitivity. Alternatively, Hertel et al. (1976) and Poppele and Bowman (1970) have shown that muscle spindle afferents (in particular Ia-afferents) are sensitive to muscle temperature increase. In a few instances, we recorded modest fatigue-induced increases in spontaneous activity and stretch sensitivity from muscle spindles. It is possible that these changes in spontaneous activity resulted from increases in muscle temperature during fatigue. However, these changes were modest compared to the changes the group III and non-spindle group II mechanoreceptors underwent. We also found modest changes in the stretch sensitivity of Golgi tendon organ afferents during fatigue. The changes that we reported with the Ib-afferents were always associated with changes in muscle force production. Hutton and Nelson (1986) have previously reported an overall decline in stretch sensitivity of Ib-afferents after supramaxima1 fatiguing stimulation. Thus it appears unlikely that these afferents are involved in relaying fatigue-related signals, unless it would be a decline in force production. Reflex actions of slow conducting muscle mechanoreceptors on spinal motoneurons It has been documented in several different muscles that motor-unit discharge rate generally declines in parallel with force during fatiguing maximum voluntary contractions (Bigland-Ritchie et al. 1983a). This modulation of motor-

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

EFFECTS

OF

FATIGUE

ON

MUSCLE

unit firing rate has been suggested to be important to maintain optimal stimulation of the muscle fibers as the contractile properties change during repeated activation (BiglandRitchie et al. 1983b). In particular, as a muscle fatigues, there is a slowing of twitch relaxation rate, which results in a decline in the fusion frequency of muscle. One potential mechanism hypothesized to underlie this rate decline is the summation of afterhyperpolarization currents of motoneurons activated during repetitive discharge. This hypothesis is supported by intracellular data (Kernel1 and Monster 1982) showing that the pattern of decline in discharge rate of a motoneuron activated by current injection is broadly similar to the decline in mean motor-unit discharge rate recorded from humans performing fatiguing maximum voluntary contractions (BiglandRitchie et al. 1983a). However, it remains to be clarified whether comparable afterhyperpolarization currents are activated during natural stimulation (Brownstone et al. 1987). In addition, if intrinsically activated currents are the only mechanism acting to modulate motoneuron discharge rate during prolonged contractions, there would be no way to compensate for any mismatch between motoneuron discharge history and unexpected peripheral changes in muscle mechanical properties. The other mechanism proposed to be involved in rate modulation during fatiguing contractions is fatigue-induced reflex inhibition of motoneuron rate. BiglandRitchie and co-workers (Bigland-Ritchie et al. 1986; Woods et al. 1987) demonstrated, in human subjects performing fatiguing maximum voluntary contractions, that the mean motor-unit discharge rate remained comparably low if the fatigued muscle was kept ischemic. Bigland-Ritchie and colleagues (Bigland-Ritchie et al. 1986) suggested that this fatigue-induced reflex must arise from the central actions of muscle receptors sensitive to the peripheral metabolic and thermal changes in muscle that accompany fatigue. In a previous study we confirmed the presence of such a fatigue-induced inhibitory reflex in an animal model (Hayward et al. 1988a). In those experiments, we examined the change in reflex inhibition induced by activation of MG’on SOL during selective fatigue of MG. We recorded significant increases in reflex inhibition of SOL after lo-20 min of intermittent fatiguing stimulation. Because the level of reflex inhibition onto SOL was dependent on MG force, this suggested that the reflex may arise from mechanically sensitive afferents. The time course of changes we reported in the present experiment for fatigue-induced changes in mechanically sensitive muscle free nerve ending sensitivity broadly parallels the increases in reflex inhibition we reported in our animal reflex experiment, further supporting the hypothesized role of these afferents in this reflex. The animal reflex experiment also suggested that the influence of the reflex pathway was widespread, acting presumably not only within a homonymous motoneuron pool, but also between close synergists. Central pathways for reflex action Several possible central spinal pathways have been suggested to mediate such a fatigue-induced inhibitory reflex (Hayward et al. 1988a), including the Ib inhibitory inter-

MECHANORECEPTORS

369

neuron reflex pathway or the recurrent inhibitory Renshaw cells. Both spinal pathways are known to receive excitatory inputs from high-threshold muscle afferents (Harrison and Jankowska 1985; Piercy and Goldfarb 1974) and both give rise to inhibition of both homonymous and heteronymous motoneuron pools (Baldissera et al. 198 1). Presumably, either of these reflex pathways would be activated during excitation of the motoneuron pool, either directly (recurrent inhibition) or as a result of active muscle contraction (Ib reflex feedback). The increased peripheral feedback from the small-diameter muscle afferents associated with metabolic and thermal changes in the fatiguing muscle would increase the input to both systems. The result would be a general inhibition of the involved motoneuron pools and an overall decline in motoneuron firing rate and increase in motoneuron recruitment threshold. However, if the reflex pathway’s main function is to match motoneuron output to the changing contractile properties of fatiguing muscle fibers, then a case could be made that discharge rate should be controlled explicitly, rather than via changes in motoneuron excitability, which provide only indirect regulation of rate. Of the two proposed reflex pathways, the Renshaw recurrent inhibitory pathway is the most promising for specifically regulating motoneuron discharge rate, because activity of Renshaw neurons is uniquely related to the discharge rate of motoneurons. The level of recurrent inhibition could be enhanced as feedback from the muscle signals peripheral changes, progressively increasing the hyperpolarization after each spike and delaying the initiation of the next action potential. Recent results from McNabb and co-workers (McNabb et al. 1988) support the hypothesis that the recurrent inhibitory pathway may be excited during fatiguing voluntary contractions. However, little is known about how recurrent inhibition functions under conditions of natural activation. So whether this is an appropriate system to be involved in reflex inhibition of motoneurons during fatigue remains to be investigated. The authors thank J. Schotland for valuable comments on the manuscript and M. Munson for providing excellent technical assistance. This work was supported by National Institute of Neurological Disorders and Stroke (NINDS) Grant PO 1-NS- 17489, Veterans Administration Merit Review (W. Z. Rymer), and NINDS Grant T32-NS-07243 (L. Hayward). Address for reprint requests: W. Z. Rymer, Rm. 1406, Rehabilitation Institute of Chicago, 345 E. Superior Ave., Chicago, IL 606 11. Received

26 February

1990; accepted

in final

form

22 October

1990.

REFERENCES BALDISSERA, F., HULTBORN, H., AND ILLERT, M. Integration in spinal neuronal systems. In: Handbook of Physiology. The Nervous System. Motor Control. Bethesda, MD: Am. Physiol. Sot., 198 1, sect. 1, vol. II, part 1, p. 505-595. BIGLAND-RITCHIE, B.R., DAWSON,N. J., JOHANSSON, R.S., ANDLIPPOLD, 0. C. J. Reflex origin for the slowing of motoneurone firing rates in fatigue of human voluntary contractions. J. Physiol. Land. 379: 45 l459, 1986. BIGLAND-RITCHIE, B. R., JOHANSSON, R. S., LIPPOLD,O.C. J., SMITH,!%, AND WOODS, J. J. Changes in motoneuron firing rates during sustained maximal voluntary contractions. J. Physiol. Land. 340: 335-346, 1983a.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.

370

HAYWARD,

WESSELMANN,

B. R., JOHANSSON, R., LIPPOLD, 0. C. J., AND WOODS, J. J. Contractile speed and EMG changes during fatigue of sustained maximum voluntary contractions. J. Neurophysiol. 50: 3 13-324, 1983b. BROWNSTONE, R. M., JORDAN, L. M., AND KRIELLAARS, P. J. Evidence for a spinal mechanism regulating the afterhyperpolarization amplitude in lumbar motoneurons during fictive locomotion in cat. Sot. Neurosci. Abstr. 13: 875, 1987. CHRISTAKOS, C. N. AND WINDHORST, U. Spindle gain increases during muscle unit fatigue. Brain Res. 365: 388-392, 1986. CLELAND, C., HAYWARD, L., AND RYMER, W. Z. Neural mechanisms underlying the clasp-knife reflex in the cat. II. Stretch-sensitive muscularfree nerve endings. J. Neurophysiol. 64: 13 19- 1330, 1990. ECCLES, R. M. AND LUNDBERG, A. Synaptic actions in motoneurones by afferents which may evoke the flexion reflex. Arch. Ital. Biol. 97: 199221, 1959. EDWARDS, R. H. T. Hypotheses of peripheral and central mechanisms underlying occupational muscle pain and injury. Eur. J. Appl. Physiol. 57: 275-281, 1988. GARLAND, S. J., GARNER, S. H., AND MCCOMAS, A. J. Reduced voluntary electromyographic activity after fatiguing stimulation of human muscle. J. Physiol. Lond. 401: 547-556, 1988. HARRISON, P. J. AND JANKOWSKA, E. Sources of input to interneurons mediating group I non-reciprocal inhibition of motoneurons in the cat. J. Physiol. Land. 36 1: 379-40 1, 1985. HAYWARD, L., BREITBACH, D., AND RYMER, W. Z. Increased inhibitory effects on close synergist during muscle fatigue in the decerebrate cat. Brain Res. 440: 199-203, 1988a. HAYWARD, L., WESSELMANN, U., AND RYMER, W. Z. The effects of muscle fatigue on small diameter afferents in anesthetized cat. Sot. Neurosci. Abstr. 14: 794, 1988b. HERBACZYNASKA-CEDRO, K., STASZEWSKA-BARCZAK, J., AND JANCZEWSKA, H. Muscular work and the release of prostaglandin-like substances. Cardiovasc. Res. 10: 4 13-420, 1976. HERTEL, H. L., HOWALDT, B., AND MENSE, S. Responses of group IV and group III muscle afferents to thermal stimuli. Brain Res. 113: 20 l-205, 1976. HUTTON, R. S. AND NELSON, D. L. Stretch sensitivity of Golgi tendon organs in fatigued gastrocnemius muscle. Med. Sci. Sports Exercise 18: 69-74, 1986. KAUFMAN, M. P., IWAMOTO, G. A., LONGHURST, J. C., AND MITCHELL, J. H. Effects of capsaicin and bradykinin on afferent fibers with endings in skeletal muscle. Circ. Res. 50: 133- 139, 1982. KAUFMAN, M. P., LONGHURST, J. C., RYBICKI, IS. J., WALLACH, J. H., AND MITCHELL, J. H. Effects of static muscular contraction on impulse activity of groups III and IV afferents in cats. J. Appl. Physiol. 55: 105112, 1983. KAUFMAN, M. P., WALDROP, T. G., RYBICKI, K. J., ORDWAY, G. A., AND MITCHELL, J. H. Effects of static and rhythmic twitch contractions on the discharge of group III and IV muscle afferents. Cardiovasc. Res. 18: 663-668, 1984. KERNELL, D. AND MONSTER, A. W. Time course and properties of late adaptation in spinal motoneurones of the cat. Exp. Brain Res. 46: 19 l196, 1982. KNIF~u, K. D., MENSE, S., AND SCHMIDT, R. F. Responses of group IV afferent units from skeletal muscle to stretch, contraction and chemical stimuli. Exp. Brain Res. 3 1: 5 1 l-522, 1978. MATTHEWS, P. B. C. Mammalian Muscle Receptors and Their Central Actions. London: Arnold, 1972. MCCLOSKEY, D. I. AND MITCHELL, J. H. Reflex cardiovascular and respiratory responses originating in exercising muscle. J. Physiol. Lond. 224: 173-186, 1972. BIGLAND-RITCHIE,

AND RYMER

MCNABB, M., FRANK, J. S., AND GREEN, H. J. Recurrent inhibition during sustained submaximal contractions in humans. Sot. Neurosci. Abstr. 14: 948, 1988. MENSE, S. Sensitization of group IV muscle receptors to bradykinin by 5-hydroxytryptamine and prostaglandin E2. Brain Res. 225: 95-105, 1981. MENSE, S. Slowly conducting afferent fibers from deep tissues-neurobiological properties and central nervous actions. In: Progress in Sensory Physiology, edited by D. Ottoson. Berlin: Springer-Verlag, 1986, vol. 6, p. 139-219. MENSE, S. AND MEYER, H. Different types of slowly conducting afferent units in cat skeletal muscle and tendon. J. Physiol. Lond. 363: 403-4 17, 1985. MENSE, S. AND MEYER, H. Bradykinin-induced modulation of the response behavior of different types of feline group III and IV muscle receptors. J. Physiol. Land. 398: 49-63, 1988. MENSE, S. AND STAHNKE, M. Responses in muscle afferent fibers of slow conduction velocity to contractions and ischemia in the cat. J. Physiol. Land. 342: 383-397, 1983. MILLER, R. G., GIANNINI, D., MILNER-BROWN, H. S., LAYZER, R. B., KORETSKY, A. P., HOOPER, D., AND WEINER, M. W. Effects of fatiguing exercise on high-energy phosphates, force and emg: evidence for three phases of recovery. Muscle Nerve 10: 8 1O-82 1, 1987. MITCHELL, J. H. AND SCHMIDT, R. F. Cardiovascular reflex control by afferent fibers from skeletal muscle receptors. In: Handbook of Physiology. The Cardiovascular System. Peripheral Circulation and Organ Blood Flow. Bethesda, MD: Am. Physiol. Sot., 1983, sect. 2, vol. III, part 2, p. 623-258. NELSON, D. L. AND HUTTON, R. F. Dynamic and static stretch responses in muscle spindle receptors in fatigued muscle. Med. Sci. Sports Exercise 17: 445-450, 1985. PAINTAL, A. S. Functional analysis of group III afferent fibers of mammalian muscle. J. Physiol. Land. 152: 250-270, 1960. PIERCY, M. F. AND GOLDFARB, J. Discharge patterns of Renshaw cell evoked by volleys in ipsilateral cutaneous and high threshold muscle afferents and their relationship to reflexes recorded in ventral roots. J. Neurophysiol. 37: 294-302, 1974. POPPELE, R. E. AND BOWMAN, R. J. Quantitative description of linear behavior of mammalian muscle spindles. J. Neurophysiol. 33: 59-72, 1970. ROTTO, D. M. AND KAUFMAN, M. P. Effect of metabolic products of muscular contraction on discharge of group III and IV afferents. J. AppZ. Physiol. 64: 2306-23 13, 1988. M. P. ROTTO, D. M., MASSEY, K. D., BURTON, K. P., AND KAUMAN, Static contraction increases arachidonic acid levels in gastrocnemius muscles of cats. J. Appl. Physiol. 66: 272 l-2724, 1989. ROTTO, D. M., SCHULTZ, H. D., LONGHURST, J. C., AND KAUFMAN, M. P. Sensitization of group III muscle afferents to static contraction by arachidonic acid. J. Appl. Physiol. 68: 86 l-867, 1990. THIMM, F. AND BAUM, K. Response of chemosensitive nerve fibers of group III and IV to metabolic changes in rat muscles. Pfluegers Arch. 410: 143-152, 1987. VOLLESTAD, N. K. AND SEJERSTED, D. M. Biochemical correlates of fatigue. Eur. J. Appl. Physiol. 57: 336-347, 1988. WESTRA, H. G., DE HAAN, A., VAN DOORN, J. E., AND DE HAAN, E. J. Anaerobic chemical changes and mechanical output during isometric tetani of rat skeletal muscle in situ. Pfluegers Arch. 4 12: 12 l- 127, 1988. WOODS, J. J., FURBUSH, F., AND BIGLAND-RITCHIE, B. Evidence for a fatigue-induced reflex inhibition of motoneuron firing rates. J. Neurophysiol. 58: 125-137, 1987.

Downloaded from www.physiology.org/journal/jn by ${individualUser.givenNames} ${individualUser.surname} (192.236.036.029) on September 4, 2018. Copyright © 1991 American Physiological Society. All rights reserved.