Download 1 Title: Single low threshold afferents innervating the skin of the

Articles in PresS. J Neurophysiol (December 28, 2012). doi:10.1152/jn.00608.2012
1
1
2
3
4
5
6
7
8
9
10
11
Title: Single low threshold afferents innervating the skin of the human foot modulate on going
muscle activity in the upper limbs.
12
Leah R. Bent, Ph.D.
13
Assistant Professor
14
Dept. Human Health and Nutritional Sciences
15
University of Guelph
16
Guelph, Ontario, N1G 2W1
17
Canada
18
Phone: 1-519-824-4120 x 56442
19
Fax: 1-519-763-5902
20
21
22
23
Email: [email protected]
Authors: Leah R Bent1, Catherine R Lowrey1
Affiliations: 1. Department of Human Health and Nutritional Sciences, University of Guelph,
Guelph, Ontario, Canada, N1G 2W1
Running Title: Skin on the foot influences upper limb muscle activity
Corresponding Author:
Copyright © 2012 by the American Physiological Society.
2
24
25
Abstract
We have shown for the first time that single cutaneous afferents in the foot dorsum have
26
significant reflex coupling to motoneurons supplying muscles in the upper limb, particularly
27
posterior deltoid and triceps brachii. These observations strengthen what we know from whole
28
nerve stimulation that skin on the foot and ankle can contribute to the modulation of interlimb
29
muscles in distant innervation territories. The current work provides evidence of the
30
mechanism behind the reflex, where one single skin afferent can evoke a reflex response,
31
rather than a population. Nineteen of forty one (46%) single cutaneous afferents isolated in the
32
dorsum or plantar surface of the foot elicited a significant modulation of muscle activity in the
33
upper limb. Identification of single afferents in this reflex indicates the strength of the
34
connection and ultimately, the importance of foot skin in interlimb coordination. The median
35
response magnitude was 2.29% of background EMG and the size of the evoked response did
36
not significantly differ among the four mechanoreceptor classes (p>0.1). Interestingly, while the
37
distribution of afferents types did not differ across the foot dorsum, there was a significantly
38
greater coupling response from receptors located on the medial aspect of the foot dorsum
39
(p<0.01). Furthermore, the most consistent coupling with upper limb muscles was
40
demonstrated by Type I afferents (fast and slowly adapting). This work contributes to the
41
current literature on receptor specificity, supporting the view that individual classes of
42
cutaneous afferents may subserve specific roles in kinesthesia, reflexes and tactile perception.
3
43
Introduction
44
45
Recovery from a destabilization involves a multifaceted response of both lower (Zehr and Stein
46
1999; Haridas et al. 2005; Zehr et al. 1998) and upper limb (McIlroy and Maki 1995; Misiaszek
47
2003; Haridas et al. 2006) reflex activation. During locomotion, rapid onset of lower limb
48
muscles provides phase related excitation or inhibition responses to re-establish equilibrium
49
(Zehr et al. 1998; Eng et al 1994, Duysens et al. 1990; Van Wezel et al 1997). Similar responses
50
have been identified in the upper limb, where unexpected slips (Marigold et al. 2003) or
51
platform perturbations (McIlroy and Maki 1995) result in rapid and intentional arm movement
52
at latencies as early as 88ms.
53
There is much speculation as to the trigger for the initiation of the postural response in the
54
lower limb (muscle spindle, skin afferents, vestibular). Based on work in animals the onset of
55
the lower limb automatic postural responses are unlikely to be related to vestibular
56
contributions (Inglis and Macpherson1995). This is corroborated by research investigating
57
vestibular loss patients during whole body platform perturbations, where the initiation of the
58
response remains intact (Horak et al. 1990). In contrast, a reduction in large diameter
59
somatosensory input (skin and muscle), following changes related to diabetic neuropathy, is
60
correlated with considerable delays in the onset latency of postural responses evoked in the
61
lower limb, suggesting a role for these afferents in the postural response (Inglis et al. 1994).
62
Further segregation between muscle spindle and skin contributions has been investigated
63
through the removal of skin information in perturbation studies. In particular, changes in
64
dynamic postural recovery highlight a specific role for skin in automatic postural adjustments
65
(Meyer et al. 2004; Perry et al. 2000). The removal of skin information through anesthesia
4
66
(Meyer et al. 2004) and foot sole cooling (Magnusson et al 1990 a,b; Perry et al. 2000) has
67
been shown to lead to large deficits in postural control and dynamic balance; with specific
68
reductions in the occurrence and timing of lower limb responses to perturbations.
69
While rapid initiation of upper arm movement has also been reported in response to
70
unexpected perturbations (McIlroy and Maki 1995; Misiaszek 2003; Ghafouri et al. 2004;
71
Marigold et al. 2003) such a standard dissection of sensory contributions to the onset of this
72
upper limb response is lacking. So the question remains; ‘what sensory information is used to
73
trigger the rapid onset of responses in the upper limb?’
74
Early work has identified the presence of upper limb reflex responses to high intensity
75
stimulation over the foot sole skin (Kearney and Chan 1979), or ankle movement (Kearney and
76
Chan 1981) leading to speculation that propriospinal pathways ensure limb coordination during
77
specific tasks. Interlimb reflexes have also been evoked during electrical or mechanical
78
perturbations of the foot (Dietz et al. 2001; Haridas and Zehr 2003). Following superficial
79
peroneal stimulation (SP; a primarily cutaneous nerve at the ankle) in sitting (Zehr et al. 2001)
80
cycling (Balter and Zehr 2007) and during locomotion (Lamont and Zehr 2006, Lamont and Zehr
81
2007) upper limb musculature has been shown not only to be influenced, but also phase
82
dependently modulated (Lamont and Zehr 2007) as a result of lower limb cutaneous activation.
83
Evidence of these cutaneous-evoked interlimb responses support the hypothesis that skin on
84
the foot may serve as a potential trigger for rapid upper limb reflexes.
85
Further support is provided by evidence of strong synaptic intralimb connections from single
86
cutaneous afferents to motoneurons, which have been shown within the upper limb (McNulty
87
and Macefield 2001; McNulty et al. 1999) and within the lower limbs (Fallon et al. 2005),
5
88
demonstrating an important role for skin to evoke, or modulate muscle activity within its own
89
innervation territory. What remains to be determined is the specificity of this synaptic coupling.
90
Can individual skin afferents from the foot modulate the upper limb motor neuron pool and
91
ultimately muscle activity?
92
The current work aims to address whether single low threshold mechanoreceptors in the skin
93
of the foot modulate on-going muscle activity in the upper arm. The purpose of the work is to
94
determine the strength of the synaptic connection; whether the synaptic coupling from one
95
receptor is sufficient, or whether a population of cutaneous receptors, as supported through
96
digital stimulation, is necessary to elicit this interlimb reflex response. By examining individual
97
cutaneous contributions we can also gain insight into receptor-specific contributions to the
98
reflex response.
99
Materials and methods
100
Subjects
101
Eighteen healthy volunteers participated in 26 recording sessions (7 females (average height
102
and weight 168cm, 60kg) and 11 males (average height and weight 180cm, 78.5kg); aged 20 to
103
38 years, average 25 years). None of the participants reported any neurological or skeletal-
104
motor deficiencies. All subjects gave written informed consent as approved by the University of
105
Guelph human ethics committee, and carried out in accordance with the principles of the
106
Declaration of Helsinki.
107
Experimental set-up
108
Subjects either sat in an adjustable chair, with both legs slightly flexed (common
109
peroneal/fibular recordings), or lay prone with the chair in a treatment table position (tibial
6
110
nerve recordings). In the sitting position the legs were supported with a piece of foam under
111
the distal thigh to maintain an adequate angle at the knee for palpation of the common
112
peroneal (fibular; CP) nerve and electrode insertion. In this position the knee was at an angle of
113
approximately 120° flexion (180° as horizontal) and the ankle relaxed at 100° extension. Versa
114
Form pillows were used to ensure subject comfort and subject position was adjusted to
115
maximize muscle relaxation to improve the recording success. This prevented disturbances to
116
neural sensory recordings via activation of motor units. In the prone position, subjects lay with
117
their head facing forward, legs extended and knees slightly flexed (10°). During data collection
118
from either nerve surface electromyography (EMG) recordings were collected using disposable
119
surface electrodes (Ag-AgCl bipolar electrodes, Kendall LTP, Chicopee, MA) placed over the
120
bellies of posterior deltoid (PD) and the lateral head of the triceps brachii (Tri). EMG signals
121
were differentially amplified (gain 500, band pass 10-1000Hz, AMT-8 Bortec Biomedical Ltd,
122
Calgary, Canada) and digitally sampled at 2000Hz (Model 1401 DAQ system, CED, Cambridge
123
UK).
124
To locate the site of the CP nerve at level of the fibular head or the tibial nerve in the popliteal
125
fossa, transdermal electrical stimulation (0.2 ms, 0 – 10 mA, 1 Hz: ) was performed using a
126
Grass SIU-C constant current stimulus isolation unit, and a S88X Grass stimulator (Astro-Med,
127
West Warwick, RI, USA). Successful stimulation of the CP nerve was demarcated by twitches in
128
the peronei muscles (Peroneus brevis and longus) as well as in extensor muscles of the ankle
129
and digits; extensor hallucis longus( EHL), extensor digitorum longus (EDL), tibialis anterior (TA).
130
Additionally paraesthesiae in the skin of the foot dorsum and front of the leg indicated CP
131
location. For the tibial nerve, the optimal site was chosen once paraesthesiae was experienced
7
132
into the heel and sole of the foot, accompanied by twitches in the triceps surae muscles
133
(soleus, lateral and medial gastrocnemius). The site for electrode insertion was chosen when
134
the greatest twitch/sensation could be elicited with the smallest current. Once the site was
135
marked, a low impedance reference electrode was placed just under the skin, roughly 20-30
136
mm distal (CP) or medial (tibial) to the insertion site of the recording microelectrode. The
137
recording electrode (insulated tungsten microelectrode, 200μm diameter, 30-35 mm length
138
(CP), 55mm length (tibial) , Frederick Haer Inc., ME, USA) was then inserted at the marked
139
location. Manipulation of the electrode was done using auditory feedback of the neural activity.
140
The neural signal was amplified (gain 104, bandwidth 300 Hz – 3 kHz: ISO-80, World Precision
141
Instruments, USA) to locate the nerve. Once the nerve was located, further fine manipulation of
142
the microelectrode was performed to isolate single cutaneous afferents while providing
143
mechanical stimuli to the skin in the area of interest (foot dorsum; CP, foot sole; tibial). Neural
144
activity was then digitized at 20 kHz and stored for subsequent analysis (Spike 2 version 6
145
software; Cambridge Electronics Design, UK).
146
Experimental procedure
147
Once located, individual afferents were classified as innervating either fast-adapting (FAI or
148
FAII) or slowly- adapting (SAI or SAII) low threshold mechanoreceptors using criteria that has
149
been described previously (Johansson 1978, Vallbo and Johansson 1984; Kennedy and Inglis
150
2002; Edin 2001; Fallon et al 2005). The receptive field size, mechanical threshold and location
151
of ‘hot spots’ within the receptive field were determined using calibrated monofilaments
152
(Semmes-Weinstein monfilaments, North Coast, San Jose CA). The monofilament which exerted
153
a force 4-5 times the threshold force was used to identify the receptive field. Mechanical
8
154
activation of the identified low threshold mechanoreceptor was then initiated using the handle
155
of a paint brush over the receptive field (tip diameter approximately 2mm). For Ectopic units,
156
the general location of the skin receptor was classified based on the location of the last fascicle
157
from which cutaneous multiunit action potentials were elicited. For example, for the two
158
ectopic units recorded palpation of the skin 1) near the great toe and 2) at the ankle midline
159
elicited multiunit cutaneous action potentials from the fascicle prior to the single afferent
160
recording. It is important to add here that we cannot say with 100% certainty that the single
161
afferents were cutaneous due to inability to verify via receptive field stroking, however, we are
162
confident based on auditory feedback that we remained within the same fascicle when the
163
single afferent became apparent. Mixed afferents (skin and muscle) are found infrequently in
164
the CP nerve.
165
Based on the receptor type, different receptor activation methods were used; for identified FAI
166
receptors a rapid ‘stroking’, for FAIIs blowing across the receptive field was first employed (and
167
activated three of four initially), however, all four FAIIs located here were also activated with
168
rapid stroking as blowing was unsuccessful to continuously activate the receptor. SAIs were
169
activated using a constant indentation. When the receptor adapted to the stimulus, the
170
indentation was reinitiated. SAIIs were activated using skin stretch along the axis that evoked
171
the greatest response or indentation. Occasionally firing from the SAIs and SAIIs had to be
172
supplemented with some stroking to ensure adequate collection of spikes. Additional stroking
173
only occurred in 4 units (of 19) during the last minute of a 5 or 6 minute data collection. Of
174
note, 2 of the 4 units which were supplemented by stroking did NOT demonstrate reflex
175
coupling.
9
176
During mechanical activation of the cutaneous receptor participants were instructed to
177
isometrically activate their deltoid muscle and lateral triceps simultaneously at approximately
178
10% (light contraction). A researcher gently supported the arm to facilitate controlled muscle
179
activation. Contractions were held for 2-3 minutes before relaxation. An additional 2 minutes of
180
contraction was then completed. These short duration contractions were designed to minimize
181
the risk of fatigue. The total duration of contraction enabled adequate time to elicit a sufficient
182
number of action potentials from the cutaneous unit for spike triggered averaging (average
183
number of sweeps 4598 range 364-23,595 sweeps). Contraction level was calculated as a
184
percentage of maximum voluntary effort(MVE), which was established as the peak EMG over
185
100ms across three trials of maximal effort. MVE was performed at the onset of the experiment
186
prior to the microneurography session. During the experiment subjects were verbally instructed
187
by the experimenter to maintain the 10% level.
188
For three receptors (SAI, FAI, FAII), in addition to the isometric contraction subjects were
189
required, in a second data collection, to grasp a vertical pole to activate triceps and deltoid
190
muscles. This provided a more posturally-relevant context for the activation of the upper limb
191
muscles (Lamont and Zehr 2007). Of note, subjects were instructed not to lean against the back
192
rest of the chair in order to increase their reliance on the arm musculature for postural support.
193
Lamont and Zehr (2007) found that when subjects held on to an earth-referenced pole during
194
locomotion, the cutaneous reflex response evoked in upper limb muscles was increased
195
(specifically triceps brachii and posterior deltoid), compared to unsupported walking. They
196
concluded that holding a pole results in an overall decrease in cutaneous threshold, which may
197
facilitate the use of the upper limbs in a functional balance response. In the current study a
10
198
position was assumed that was identical to the isometric contraction and both muscles
199
(posterior deltoid and triceps) were monitored for comparable levels of background
200
contraction (10%).
201
Data analysis
202
Identification of unitary spikes was performed using Spike 2 software (Spike 2, version 6,
203
Cambridge Electronics, UK England). Spike morphology was used to generate a template and
204
spikes throughout the recording were classified based on shape. If a recording was found to
205
include multiple cutaneous afferent contributions, the recording was not included for further
206
analysis.
207
Surface EMG was analysed using a 5ms sliding window root mean square (RMS) using Spike 2
208
software. The RMS EMG was then spike-trigger averaged to the afferent firing to identify any
209
time locked modulation of upper limb activity related to cutaneous activation. A significant
210
response was identified when the EMG modulation surpassed two and a half standard
211
deviations (SD) about the mean for at least 5ms, which is equivalent to a 99% confidence
212
interval. Due to the cyclic nature of several recorded afferents it was necessary to generate a
213
control random spike. The random spike was generated using a custom made program (spike 2
214
software). The mean interval of the ‘real’ spike train was used to generate a Poisson
215
distribution of random intervals with the same mean interval. This random spike was then used
216
to generate the confidence interval, which determined significance.
217
Latency and amplitude
218
The latency of the reflex coupling was determined as the duration from spike onset to the time
219
the RMS EMG surpassed the confidence interval (99%). Significant reflex coupling between
11
220
single low threshold mechanoreceptors and the muscles of the upper limb were classified into
221
three categories based on latency. Reflex responses occurring at a latency between 25ms-70ms
222
were classified as short latency, 70-120ms as medium latency and >120ms as long latency
223
(Fallon et al. 2005; Aniss et al. 1992; Brooke et al. 1997; Burke et al. 1991; Gibbs et al. 1995;
224
Kukulka 1994; Zehr et al. 2001). The current paper focused on reflexes occurring in the early
225
and mid time frame (25ms to 120ms) to highlight responses with a greater probability of
226
traveling within a spinal reflex loop (early latency), spinal oligosynaptic or transcortical
227
(medium latency) (Nielsen et al. 1997). Activation of several of the single units was performed
228
by manually scratching across the surface. Due to the cyclical nature of the afferent response it
229
was difficult to establish the latency of onset (ie within an analysis window the afferent may
230
respond several times to activation (based on frequency of stroking) making it difficult to
231
establish the timing of the response relative to the stimulus). In these afferents the unit was
232
classified as significant, with the shortest onset latency, and the cyclical nature recorded. The
233
amplitude of the reflex response was calculated as the peak percent change from background
234
RMS EMG and was identified as a reflex response if the latency was shorter than 120ms (Fallon
235
et al 2005, ). ‘Cyclic’ onsets latencies were not included in the overall average for onset latency
236
calculations.
237
Statistics
238
χ2 tests were used to examine the distribution of receptor types across the foot and the
239
number of receptors in each category that exhibited a significant reflex coupling versus those
240
that did not. Receptive field and threshold were assessed using t-tests to compare receptor
241
characteristics of receptors that demonstrated a significant EMG modulation versus those that
12
242
did not. Reflex magnitude (measured as a % of background EMG) was examined between the
243
four receptor types using a Kruskal-Wallis one-way ANOVA with Bonferroni adjustments for
244
multiple comparisons.
245
Results
246
Forty one low threshold mechanoreceptors were sampled from the skin overlying the dorsum
247
(32) and sole (9) of the foot. Of these units there were 15 FAIs (37%), 4 FAIIs (10%), 9 SAIs
248
(22%), 10 SAIIs (24%), 2 ectopic units (5%) and one hair unit (2%). Following spike triggered
249
averaging of upper limb muscles a total of 18 of 41 (44%) single low threshold
250
mechanoreceptors in the foot demonstrated significant synaptic coupling with motoneurons
251
supplying muscles in the upper limb (Figure 1). The afferents that exhibited a significant
252
modulation of the ongoing upper limb EMG activity included 10/15 (67%) innervated by FAI,
253
1/4 (25%) innervated by FAII, 4/9 (44%) innervated by SAI, and 2/10 (20%) innervated by SAII
254
low threshold mechanoreceptors. Both ectopic units demonstrated a significant coupling
255
response represented as a modulation of the upper limb muscle activity, while the hair
256
receptor did not (Figure 1). An equal number of units demonstrated significant responses in
257
the upper limb, whether they were located on the dorsal (47%) or plantar (50%) surface of the
258
foot. One additional unit (SAI) did not show significant coupling during a ‘free’ isometric
259
contraction, however, when an earth referenced pole was held during muscle activation
260
significant modulation of both deltoid and triceps was observed (Figure 6 and earth vertical
261
reference, below) increasing the total to 19 of 41 units with a significant coupling (46%).
262
Receptor characteristics and distribution
13
263
The average receptive field (RF) size and threshold for each of the four classes of receptor
264
closely resembles current literature on cutaneous mechanoreceptors in the foot (Table 1). To
265
assess differences relative to the literature data were compared using chi square analyses.
266
Receptive field size for units on the dorsal or plantar surface of the foot did not differ
267
significantly from those reported previously by Fallon et al (2005) on the plantar surface of the
268
foot (p>0.05), nor did they differ significantly from foot dorsum receptors in a recent study by
269
Aimonetti and colleagues (2007) (p>0.1). With respect to the proportions of afferents
270
distributed among the four mechanoreceptor classes, the 41 single afferents recorded in the
271
current study did not differ significantly from those recorded in the plantar surface of the foot
272
(p>0.05; Fallon et al 2005). There was a predominance of afferents innervating fast adapting
273
receptors identified in both the current study (37%) and Fallon et al. (40%). In contrast, the
274
distribution of afferents among the four different classes did differ significantly from those
275
reported by Aimonetti et al (2007) in the foot dorsum (p<0.01), where the greatest number of
276
afferents were innervated by slowly adapting Type II (30%) mechanoreceptors (followed by FAI;
277
27%). The large representation of SAII afferents (with Aimonetti et al 2007) may result from
278
the location of the recorded afferent RF; predominantly on the anterior surface of the leg,
279
where no receptors were recorded in this location in the current data pool.
280
To identify the distribution of receptors that demonstrated a significant coupling response the
281
foot was divided into quadrants (Figure 2). Twenty two and a half of the thirty two units on the
282
dorsum of the foot were found on the medial aspect, with seven and a half on the lateral side.
283
Within each quadrant reflex coupling with the upper limb was assessed to identify whether a
284
specific location on the foot had a higher probability of eliciting a response. While the total
14
285
number of recorded afferents did not differ significantly between the four quadrants (p>0.05),
286
the distribution of receptors that demonstrated a significant reflex response were found to
287
differ across the foot dorsum, with significantly greater coupling shown in the two medial
288
quadrants (p<0.01). Additionally, the RF size was significantly different between receptors
289
demonstrating a coupling with upper limb motoneurons (smaller) and those which did not
290
modulate upper limb activity (larger) (p<0.001), whereas threshold levels of these receptor
291
groups did not differ (P>0.1).
292
293
Reflex response
294
The most common reflex coupling was seen between FAI mechanoreceptors and motoneurons
295
innervating muscles of the upper limb, where 67% of the recorded FAIs demonstrated a
296
significant modulation. This prevalence of coupling in the FAI receptors supports findings by
297
Fallon et al. (2005), where 81% of the FAIs in the foot sole demonstrated the ability to
298
modulate ongoing muscle activity in the lower limb. The second highest prevalence of coupling
299
in the current study was seen with SAI units (44%), also consistent with previous reports of
300
intralimb responses in the lower limb (Fallon et al 2005). In fact the proportion of units in each
301
class that demonstrated significant coupling did not differ from the results obtain by Fallon et al
302
(2005) (p>0.1), which is the only other study to date that has examined these reflex responses
303
from cutaneous receptors in the foot.
304
The average magnitude and latency of the reflex responses are given in Table 2. A total of 19
305
single units are represented including one unit in which a reflex was demonstrated only after
306
the subject held onto a vertical reference pole (see below). Six of the 19 units (including both
15
307
ectopic units) demonstrated synaptic coupling with motoneurons innervating both triceps and
308
deltoid muscles. The median response magnitude was found to be 2.29% of background EMG
309
(2.59% including the SAI pole unit) and did not significantly differ among the four receptor
310
classes (p>0.1). Interestingly, the reflex strength (amplitude of response) was similar to the
311
reflex magnitude reported by Fallon and colleagues for ‘within’ lower limb responses (2.3%)
312
and for responses ‘within’ the upper limb of 2.05% reported by McNulty et al. (2001).
313
314
315
Cyclic response
316
All 15 of the FAI afferents were activated by stroking across the receptive field. Rapid stroking
317
increases the number of action potentials during the data collection phase to enable
318
observation of any present EMG modulation after spike triggered averaging. Figure 3 illustrates
319
one FAI afferent that evoked cyclic activation in both PD and Tri EMG following high frequency
320
activation across its receptive field (Figure 3 and Table 2). This single FAI afferent was located
321
along the dorsal/medial border of the foot slightly proximal to the 1st metatarsal head. Its RF
322
was small (2mm x 3mm) and threshold was 0.085g, which was low relative to the other
323
recorded FAIs. This particular FAI was able to evoke significant and periodic modulation in both
324
deltoid and triceps that followed the 10Hz activation of the afferent over its receptive field
325
(Figure 3). As a result of the cyclic nature of this unit, the onset latency of the muscle response
326
was calculated as 11ms relative to ‘time zero’. Clearly this muscle onset latency is too early to
327
be associated with the spike trigger averaged units at time zero, and is more likely
328
representative of a response to an earlier cycle of activation of the FAI unit (resulting in a reflex
16
329
latency closer to 110ms). The cyclic nature itself makes it impossible to predict which of the
330
cycles of afferent firing corresponds to the muscle modulation. As a result onset latencies
331
calculated in afferents that clearly exhibited a cyclic response in nature were not included in
332
the overall latency calculation. The remaining nine FAI units were all activated at a frequency
333
which generated the periodic bursts at a frequency outside of the latency of interest for reflex
334
coupling (periodicity>300ms, reflex coupling <120ms). All other types of receptors (FAII, SAI,
335
SAII) can be activated by stroking across the receptive field, however, this type of activation
336
was only applied if a response was not apparent using conventional approaches (only used for 4
337
additional units- see methods). Figure 4 illustrates a FAII mechanoreceptor (pacinian corpuscle)
338
that was also activated through stroking, which resulted in a cyclical response from the afferent
339
and a time locked response in the PD EMG following each afferent burst. Activation of this
340
particular FAII was not possible through rigorous blowing across the RF. A strong correlation
341
was found (R2=0.997) between the mean inter-spike interval of the FAII afferent activation, and
342
the reflex periodicity of the EMG response in Deltoid.
343
Single afferent response
344
The probability of representing reflex coupling related to one single afferent is increased in
345
afferents that generate their own unique, and thus asynchronous firing pattern relative to
346
other skin receptors located in the immediate anatomical region. SAI and SAII receptors
347
demonstrate unique firing to constant indentation, and SAII also respond to skin stretch. When
348
natural activation is used to generate action potentials from which to spike trigger average
349
ongoing EMG there is an increased confidence that coupling is related to single afferent
350
activation. Figure 5 is an example of an SAII receptor located slightly proximal to the ankle that
17
351
was activated by skin stretch along the ankle dorsum in a proximal-distal direction (Figure 5,
352
inset). This single afferent demonstrated significant modulation of ongoing deltoid activity
353
roughly every 60ms, which correlates well with the regular firing of the SAII afferent (16.7Hz).
354
Earth-referenced pole
355
In two participants, following the isometric contraction protocol, activation over the receptive
356
field was repeated while subjects generated muscle activity by holding onto an earth
357
referenced pole. The amplitude of the background EMG activity (100ms) was not statistically
358
different between the two trials (isometric contraction vs. muscle activation to grasp the pole)
359
(P=0.13). Both units (FAII and SAI) demonstrated significant coupling to motoneurons in the
360
upper limb muscles during trials with the earth referenced pole. For the SAI unit, grasping the
361
earth referenced pole resulted in a significant inhibition of ongoing muscle activity in both
362
deltoid (61ms onset, 7.65% amplitude) and triceps (88ms onset, 6.38% amplitude) (Figure 6).
363
This afferent was located in the glabrous skin on the pad (plantar surface) of digit IV. The
364
receptive field size (42 mm²) and threshold level (0.4g) were both low relative to the recorded
365
population. This SAI unit was activated using constant indentation over its receptive field to
366
avoid cyclical activation of the afferent, and therefore reduce the chances of contributions from
367
neighbouring skin receptors (corroborated by the autocorrelogram). Also of note is that the
368
number of sweeps during isometric contraction included 1000 more spikes than the trial with
369
the pole, supporting an overt modulation when the arm grasped onto an earth vertical
370
reference. The second unit (FAII) also demonstrated a significant inhibitory modulation; this
371
response was in the triceps brachii (155ms, 2.27%). However, the modulation fell outside of our
372
criteria for a short or medium latency response. This unit was located on the plantar surface of
18
373
the foot between digits III and IV. Interestingly, neither of these units (SAI, FAII) were able to
374
initially (with isometric contraction alone) modulate the motor neuron pool of either upper
375
limb muscle.
376
Discussion
377
We have, for the first time, demonstrated a reflex connection between single low threshold
378
mechanoreceptors in the skin on the foot dorsum with motoneurons supplying muscles of the
379
upper limb. Unique to other studies that have examined individual cutaneous contributions to
380
muscle modulation, we have identified the presence of interlimb reflex responses. Similar to
381
the foot sole, but in contrast to the hand we have shown that strong reflexes can be generated
382
from all four types of cutaneous receptors in this hairy skin (including SAI).
383
Afferent response
384
The most commonly observed reflex coupling in the upper limb muscles was through activation
385
of FAIs (67%) followed by SAIs (44%). Coupling within the remaining afferent classes was
386
considerably less frequent (20% SAII, 25% FAII). FAIs from the foot sole in a previous study were
387
also shown to elicit a strong coupling response within the lower limb (Fallon et al 2005) but
388
were not as prominently featured in the reflex response within the arm (McNulty et al 1999).
389
FAI afferents are believed to innervate Meissner corpuscles, which are most sensitive to forces
390
tangential to the skin (shear) and are known to respond to incipient slips within the receptive
391
field, primarily signaling external perturbation forces (Macefield 1998; Johansson et al. 1982).
392
Based on their rapidly adapting nature, they are also ideal for providing dynamic input
393
(velocity). Skin on the foot is our direct interface with the environment and can provide
394
valuable information regarding the support surface and obstacles in our path. Contact signaled
19
395
by the foot dorsum skin during dynamic tasks (tripping) provides critical and timely information
396
that can influence upper limb responses to help move the centre of mass forward (Marigold et
397
al. 2003), or grab for a support hold (McIlroy and Maki 1995).
398
It is notable that SAI receptors demonstrated strong reflex coupling between foot dorsum skin
399
and muscles of the upper limb (second only to the frequency of FAI responses). In combined
400
data across two papers, McNulty et al. (1999; 2001) reported a distinct lack of response from
401
SAI receptors in reflex coupling within the upper limb. While FAIs can provide some input
402
regarding shape (Philips et al. 1992), SAI receptors are known to be important in tactile
403
discrimination, including the ability to encode raised dots, edges, curves and object orientation
404
in the hand (Johansson and Vallbo 1980; Goodwin et al 1995; LaMotte and Srinivasan 1996;
405
Johansson et al. 1982) and respond with high resolution to changes in pressure (Mountcastle
406
1966). Input from SAIs regarding shape and indentation may not feature prominently for reflex
407
responses when grasping objects (McNulty et al 1999), however, such input may be critical for
408
facilitating appropriate cautionary strategies during locomotion. The ability of SAIs to faithfully
409
code compression may specifically highlight their role on the foot dorsum, to signal shape and
410
size of contacted objects or changes in (expected) compression of the foot skin during gait.
411
SAIIs
412
Previous reports in the hand have indicated that SAII afferents respond to movement of the
413
digits (Burke et al. 1988) as well as skin stretch around joints (Knibestol and Vallbo 1970)
414
suggesting a role in kinesthesia. Aimonetti et al. (2007) have also demonstrated pronounced
415
firing of SAII afferents in the skin of the ankle following movement of the foot. SAIIs in the
416
current study, were the group of receptors to show the least amount of coupling (20%). It is
20
417
feasible that SAII information regarding skin stretch and potentially ankle joint position are
418
most important in eliciting intralimb responses to activate muscles at the ankle, and not
419
interlimb responses to distant arm muscles. Previous work, both within the upper limb
420
(McNulty 1999; 2001) and lower limb (Fallon et al. 2005) have shown a strong coupling
421
response of SAIIs within their own innervation territory (52% coupling within the upper limb,
422
40% coupling within the lower limb). Therefore, as a proprioceptor, SAII input can provide
423
information regarding local skin stretch and is therefore ideal to modulate joint position for
424
successful grasp, or foot placement around the local joint. Such modulations may not be
425
appropriate or necessary for interlimb control.
426
Reflex amplitudes
427
We compared the reflex magnitudes of our upper limb responses to those reported in two
428
previous papers, which examined coupling with motoneurons acting on muscles of a
429
homonymous innervation territory (Fallon et al. 2005; McNulty et al. 1999, 2001). Our
430
amplitude calculation was based on peak responses, and therefore we halved the amplitude
431
responses reported previously by McNulty et al (2001; who used peak to peak) to enable a
432
comparison. Perhaps surprisingly, the magnitude of the median reflex response in the current
433
study (2.29% of background) did not differ from previous intralimb findings (2.3%; Fallon 2005
434
and 2.05%; McNulty 2001), despite our modulation of muscles of a distant innervation territory.
435
These previous intralimb reflex magnitudes can no longer be attributed to a general cutaneous
436
coupling response within a limb, rather it appears that interlimb cutaneous input displays an
437
equally potent impact on muscle activity. Zehr and colleagues (2001), were the first group to
438
report significant inter and intralimb reflexes following whole nerve stimulation of SP. Our
21
439
current observations from single cutaneous afferents, of prominent interlimb reflex coupling,
440
adds further support that such links from physically distant cutaneous territories can influence
441
and shape muscle activity for the reflex coordination of movement.
442
Our results show EMG responses that are clearly time-locked to action potentials from a single
443
afferent providing a strong indication that the recorded afferent is the primary contributor to
444
the reflex. The possibility cannot be overlooked, however, that in some cases other afferents
445
may contribute. Indeed, FAIs are often found in high densities in the skin and stroking over the
446
receptive field of one receptor could simultaneously activate others, contributing to the overall
447
EMG response. With SAI and SAII afferents, indentation and stretch activation of these
448
afferents are more likely to segregate single receptor contribution, given that each receptor has
449
a unique firing response to these natural stimuli. While the EMG response is time-locked to the
450
recorded spike, increasing the probability that this single afferent is having the greatest (and
451
potentially only) impact, it is also noteworthy that additional afferents may respond somewhat
452
to these natural stimuli and fire at sub-harmonics to contribute to the EMG reflex response.
453
Ultimately, the activation strategies used targeted skin over the receptive field and localized
454
the response as much as possible and, we feel, succeeded in segregating individual afferent
455
types and their responses. Compelling evidence that single afferents contribute to the reflex
456
response is demonstrated by our data on ectopic units; where no external stimulation is
457
applied. The lack of evoked stimulation argues definitively for single afferent contribution to
458
the reflex response.
459
Pathways
22
460
Latencies for the interlimb reflex response evoked via single cutaneous afferents ranged from
461
38ms to 120ms, with a median of 59ms (mean 71ms). Interestingly, no differences were found
462
in onset latency in our upper limb responses when compared to single afferent reflexes within
463
the lower limb (Fallon et al 2005), where a median onset latency of 57.4ms was reported across
464
all muscles and all receptor types. We propose that the reflexes evoked in the current study
465
involve propriospinal pathways, however, the involvement of subcortical and even cortical
466
structures cannot be ruled out. Conduction velocities in the upper and lower limb are known to
467
vary considerably from 20m/s to 60m/s (Mackel et al. 1988) 26m/s to 91m/s, (Knibestol 1973)
468
and 52m/s to 89m/s (Macefield et al 1989). Based on our shortest non-cyclical response
469
(38ms) and a travel distance of roughly 1m (CP at knee, to shoulder), the conduction velocity of
470
our fastest afferent is estimated at 30m/s, falling within the reported velocities for cutaneous
471
afferents. Whole nerve SP, or foot skin stimulation to date has reported early, 45-50 ms (Zehr
472
et al. 2001; Pieseiur-Strehlow and Meinck 1980) or middle, 75ms (Zehr et al 2001), 91-92ms
473
(Kearney and Chan 1979) reflex latencies in the upper arm with these authors purporting
474
connections via propriospinal pathways. Previous work has provided evidence that long-latency
475
intralimb reflexes via SP stimulation (70-95ms) are at least partly mediated by a transcortical
476
pathway (Nielsen et al 1997), however, it is noteworthy that these reflexes are within the lower
477
limb. The range of our reflex latencies suggest they may fall into several categories, with
478
evidence of both propriospinal (nine SL responses; 38ms-62ms) as well as coupling which may
479
involve subcortical or cortical loops (eight ML responses; 71-120ms). Cutaneous reflexes
480
reported by Pruszynski et al (2008) indicated the fastest voluntary activity in muscles across the
23
481
elbow to begin at 120ms, indicating our responses, for the most part, fall within a reflex
482
latency.
483
The question remains, are these reflex responses fast enough to be involved in upper limb
484
perturbation responses? Onset of activity in deltoid has been reported at 88ms following whole
485
body perturbations (McIlroy and Maki 1995) and 143-150ms following slips (Marigold et al.
486
2003). With our median latency of 59ms, and maximum latencies of 120ms, it is feasible that
487
skin receptors on the dorsum of the foot can contribute to these rapid upper limb responses,
488
with FAIs and SAIs featuring prominently.
489
Regional differences and function
490
Fifteen FAI units were identified in the dorsum of the foot in the current study, despite a lack
491
of evidence of fast adapting types of receptors in previous studies of hairy skin; in the face,
492
forearm and thigh (Edin 2001; Vallbo et al. 1995). The authors suggested that receptor
493
distribution (fewer FA I, more SA and hairy receptors) reflected regional differences in
494
functional roles. SA receptors are known to provide rich information for proprioception,
495
especially for joint position (Johnson KO 2001; Edin and Johansson 1995), which makes them
496
suitably located across joints in the hairy skin of the body. FA receptors are able to provide
497
information regarding contact timing and velocity of objects as they move along the skin (in
498
grip or in slip), which is pertinent for the glabrous skin of the palms of the hands and soles of
499
the feet. Here we found a large distribution of FAIs in the foot dorsum hairy skin (15 FAIs, 37%
500
of all collected) as compared to 5% in knee and thigh (Edin 2001) and 0% in the forearm (Vallbo
501
et al. 1995). In contrast, our distribution was very similar to reports in the glabrous skin (Fallon
502
et al 2005; Kennedy and Inglis 2001), where the largest population of afferents innervated FAI
24
503
end organs. Aimonetti and colleagues (2007), who also sampled from skin on the foot dorsum
504
and leg found the majority of afferents to innervate Type SAII. This may be due to receptor
505
location, where 72% of the identified afferents were located on the leg and not the dorsum of
506
the foot. Interestingly, when considering only receptors on the foot dorsum (25), Aimonetti
507
found the greatest density of FAIs (nine; 36%), which corresponds well with the current study
508
(37% FAIs). What appears to be emerging here is a regional difference across the foot dorsum
509
from other regions of hairy skin. FAIs on the foot dorsum signal unexpected contact with
510
objects in the path trajectory, and when coupled to motoneurons supplying muscles in the
511
upper limb can subserve a protective balance role.
512
Earth referenced contact
513
Task dependent reflex modulation is not unique to a particular system. Coupling between
514
cutaneous receptors and muscle spindles of the lower limb has been shown to be task
515
dependent, with responses only emerging in a balance context (standing, Aniss et al. 1992). The
516
technique of microneurography poses limitations on the ability to assess afferent responses in
517
a posturally relevant position, as for the most part experiments are performed sitting or lying
518
down. To enhance the balance context in the current study we provided an earth referenced
519
pole.
520
Two subjects were instructed to hold an earth referenced rail while performing a level of
521
contraction comparable to isometric contraction in a previous trial. Spike triggered averaging
522
revealed no significant modulation during isometric contraction without the pole, however,
523
while holding the vertical pole, there was an emergence of a significant response. These
524
findings corroborate previous data of enhanced reflex responses in PD and Tri during a
25
525
locomotor study that examined gating of interlimb reflexes with an earth referenced handrail
526
(Lamont and Zehr 2007). The conclusion from these observations combined, is that a posturally
527
relevant context enhances the interlimb reflex response.
528
Conclusions
529
We have shown for the first time that single low threshold tactile afferents from the foot, and
530
particularly from the foot dorsum can modulate ongoing muscle activity in the upper limb.
531
Identification of each class of skin receptor also enables insight into those most closely coupled
532
to this interlimb reflex response (FAI, SAI). While our current data cannot determine whether
533
skin is the trigger for rapid upper limb responses (or simply a modulator) we have
534
demonstrated strong coupling that indicates the substrate does exist for skin to have a role in
535
these rapid perturbation responses. Modulation of ongoing muscle activity generated by the
536
activation of a single low threshold cutaneous afferent has previously been shown within the
537
upper limb (McNulty et al, 2001, 1999) and within the lower limb (Fallon et. al. 2005). The
538
current work extends these findings to highlight cutaneous modulation of muscle activity across
539
innervation territories, arguing that cutaneous input is an important component in the
540
coordination of interlimb activity. Furthermore, testing of reflexes in a balance relevant context
541
(reference pole) demonstrated even larger responses, supporting a functional role for foot skin
542
in balance related upper limb reflexes.
543
544
545
546
547
548
549
550
Acknowledgements
Special thanks to Stephanie Muise for her contributions to this work. This work was supported
by funding from the Natural Science and Engineering Research Council of Canada Discovery
Grant to L.R.B.
Grants: This work was supported by funding from the Natural Science and Engineering
Research Council of Canada Discovery Grant to L.R.B.
26
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
Author contributions:
L.R.B. research conception and design; C.R.L and L.R.B. performed experimental procedures;
L.R.B and C.R.L. analyzed data. L.R.B. prepared figures; L.R.B. and C.R.L. interpreted results of
experiments; L.R.B. drafted manuscript; L.R.B. and C.R.L. edited and revised manuscript; L.R.B.
and C.R.L. approved final version of manuscript.
Disclosures:
No conflicts of interest, financial or otherwise, are declared by the author(s)
References
Aimonetti JM, Hospod V, Roll JP, Ribot-Ciscar E. Cutaneous afferents provide a neuronal
population vector that encodes the orientation of human ankle movements. Journal of
Physiology 580: 649-58, 2007.
Aniss AM, Gandevia SC, Burke D. Reflex responses in active muscles elicited by stimulation of
low-threshold afferents from the human foot. Journal of Neurophysiology 67: 1375-84, 1992.
Balter JE, Zehr EP. Neural coupling between the arms and legs during rhythmic locomotor-like
cycling movement. J Neurophysiology 97(2):1809-18, 2007.
Brooke JD, Cheng J, Collins DF, McIlroy WE, Misiaszek JE, Staines WR. Sensori-sensori :
afferent conditioning with leg movement: gain control in spinal reflex and ascending paths.
Progress Neurobiology. 51:393-421, 2007.
Burke D, Gandevia S C, Macefield G. Responses to passive movement of receptors in joint, skin
and muscle of the human hand. Journal of Physiology. 402:347–361, 1988
Burke D, Dickson HG, Skuse NF. Task-dependent changes in the responses to low-threshold
cutaneous afferent volleys in the human lower limb. Journal of Physiology. 432:445-58, 1991.
Duysens J, Trippel M, Horstmann GA, Dietz V. Gating and reversal of reflexes in ankle muscles
during human walking. Experimental Brain Research. 82 (2):351-8, 1990.
Edin B. Cutaneous afferents provide information about knee joint movements in humans.
Journal of Neurophysioogy. 531.1:289-297, 2001.
Edin BB, Johansson N. Skin strain patterns provide kinaesthetic information to the human
central nervous system. Journal of Physiology. 487: 243-51, 1995.
Eng JJ, Winter DA, Patla AE. Strategies for recovery from a trip in early and late swing during
human walking. Experimental Brain Research. 102 (2):339-49, 1994.
27
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
Fallon JB, Bent LR, McNulty PA, Macefield VG. Evidence for strong synaptic coupling between
single tactile afferents from the sole of the foot and motoneurons supplying leg
muscles. Journal of Neurophysiology. 94(6):3795-804, 2005.
Ghafouri M, McIlroy WE, Maki BE. Initiation of rapid reach-and-grasp balance reactions: is a
pre-formed visuospatial map used in controlling the initial arm trajectory? Experimental Brain
Research. 155(4):532-6, 2004.
Gibbs J, Harrison LM, Stephens JA. Cutaneomuscular reflexes recorded from the lower limb in
man during different tasks. Journal of Physiology 487: 237-42, 1995.
Goodwin AW, Browning AS, Wheat HE. Representation of curved surfaces in responses of
mechanoreceptive afferent fibers innervating the monkey's fingerpad. Journal of Neuroscience
15:798-810, 1995.
Goodwin AW, Macefield VG, Bisley JW. Encoding of object curvature by tactile afferents from
human fingers. Journal of Neurophysiology. 78(6): 2881-8, 1997.
Haridas C, Zehr EP. Coordinated interlimb compensatory responses to electrical stimulation of
cutaneous nerves in the hand and foot during wlaking. Journal of Neurophysiology 90: 2850-61,
2003.
Haridas C, Zehr EP, Misiaszek JE. Postural uncertainty leads to dynamic control of cutaneous
reflexes from the foot during human walking. Brain Research. 1062(1-2):48-62, 2005.
Haridas C, Zehr EP, Misiaszek JE. Context-dependent modulation of interlimb cutaneous
reflexes in arm muscles as a function of stability threat during walking. Journal of
Neurophysiology. 96(6): 3096-103, 2006.
Horak F, Nashner L, Diener H. Postural strategies associated with somatosensory and vestibular
loss. Experimental Brain Research. 82:167-177, 1990.
Inglis JT, Macpherson JM. Bilateral labyrinthectomy in the cat: effects on the postural response
to translation. Journal of Neurophysiology. 73:1181–1191, 1995.
Inglis JT, Horak FB, Shupert CL, Jones-Rycewicz C. The importance of somatosensory
information in triggering and scaling automatic postural responses in humans. Experimental
Brain Research. 101:159-164, 1994.
Johansson RS. Tactile sensibility in the human hand: receptive field characteristics of
mechanoreceptive units in the glabrous skin area. Journal of Physiology. 281:101-25, 1978.
28
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
Johansson RS, Landstrom U, Lundstrom R. Respones of mechanoreceptive afferent units in the
glabrous skin of the human hand to sinusoidal skin displacements. Brain Research. 244:17-25,
1982.
Johansson RS, Vallbo AB. Spatial properties of the population of mechanoreceptive units in the
glabrous skin of the human hand. Brain Research. 184(2):353-66,1980.
Johnson, KO. The roles and functions of cutaneous mechanoreceptors. Current Opinions in
Neurobiology. 11:455-61,2001.
Kennedy PM, Inglis JT. Distribution and behaviour of glabrous cutaneous receptors in the
human foot sole. Journal of Physiology. 538:995-1002, 2002.
Knibestol M, Vallbo Å B. Single unit analysis of mechanoreceptor activity from the human
glabrous skin. Acta Physiologica Scandinavica. 80:178–195, 1970.
Knibestöl M.. Stimulus-response functions of rapidly adapting mechanoreceptors in human
glabrous skin area. Journal of Physiology. 232(3):427-52, 1973.
Kukulka CG. The reflex effects of nonnoxious sural nerve stimulation on human triceps surae
motor neurons. Journal of Neurophysiology. 71(5):1897-906, 1994.
Kearney RE and Chan CW. Reflex response of human arm muscles to cutaneous stimulation of
the foot. Brain Research. 170:214-217, 1979.
Kearney R.E. and Chan C.W.. Interlimb reflexes evoked in human arm muscles by ankle
displacement. Electroencephalography and Clinical Neurophysiology. 52:65-71, 1981.
Lamont EV, Zehr EP. Task-specific modulation of cutaneous reflexes expressed at functionally
relevant gait cycle phases during level and incline walking and stair climbing. Experimental
Brain Research. 173(1):185-92, 2006.
Lamont EV, Zehr EP. Earth-referenced handrail contact facilitates interlimb cutaneous reflexes
during locomotion. Journal of Neurophysiology. 98(1):433-42, 2007.
LaMotte RH, Srinivasan MA. Neural encoding of shape: responses of cutaneous
mechanoreceptors to a wavy surface stroked across the monkey fingerpad. Journal of
Neurophysiology. 76(6):3787-97, 1996.
Mackel R. Conduction of neural impulses in human mechanorecpetive cutaneous afferents
Journal of Physiology 401:597-615, 1988.
29
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
Macefield VG. The signalling of touch, finger movements and manipulation forces by
mechanoreceptors in human skin. In Neural Aspects of Tactile Sensation, ed. MORLEY, J. W., pp.
89–87. Elsevier Science Publishers, Amsterdam. 1998.
Macefield G, Gandevia SC, Burke D. Conduction velocities of muscle and cutaneous afferents in
the upper and lower limbs of human subjects. Brain. 112:1519-32, 1989.
Magnusson M, Enbom H, Johansson R, Pyykko I. Significance of the pressor from the human
feet in anterior-posterior postural control. The effect of hypothermia on vibration-induced
body-sway. Acta Otolaryngolica. 110(3-4):182-8, 1990a.
Magnusson M, Enbom H, Johansson R, Wiklund J. Significance of the pressor from the human
feet in medial-lateral postural control. The effect of hypothermia on galvanically induced body
sway. Acta Otolaryngolica. 110(5-6):321-327,1990b.
Marigold DS, Bethune AJ, Patla AE. Role of unperturbed limb and arms in the reactive recovery
response to an unexpected silp during locomotion. J Neurophysiology. 89:1727-1737, 2003.
McIlroy WE and Maki BE. Early activation of arm muscles follows external perturbation of
upright stance. Neuroscience Letters. 184(3):177-80, 1995.
McNulty PA, Macefield VG. Modulation of ongoing EMG by different classes of low-threshold
mechanoreceptors in the human hand. Journal of Physiology. 537:1021-32, 2001.
McNulty PA, Türker KS, Macefield VG. Evidence for strong synaptic coupling between single
tactile afferents and motoneurones supplying the human hand. Journal of Physiology. 518: 88393,1999.
Meyer PF, Oddsson LI, De Luca CJ. Reduced plantar sensitivity alters postural responses to
lateral perturbations of balance. Experimental Brain Research. 157(4):526-36, 2004.
Misiaszek JE. Early activation of arm and leg muscles following pulls to the waist during
walking. Experimental Brain Research. 151(3):318-29, 2003.
Mountcastle VB, Talbot WH, Kornhuber HH. The neural transformation of mechanical stimuli
delivered to the monkey’s hand. In: AVS de Reuck, J Knight (eds). Ciba Foundation Symposium:
touch, heat and pain, pp. 325-351. London: Churchill. 1966.
Nielsen JB, Petersen N, Fedirchuk B. Evidence suggesting a transcortical pathway from
cutaneous foot afferents to tibialis anterior motoneurones in man. Journal of Physiology.
501:473-84, 1997.
30
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
Perry SD, McIlroy WE, Maki BE. The role of plantar cutaneous mechanoreceptors in the control
of compensatory stepping reactions evoked by unpredictable, multi-directional perturbation.
Brain Research. 877:401-406, 2000.
Phillips JR, Johansson RS, Johnson KO. Responses of human mechanoreceptive afferents to
embossed dot arrays scanned across fingerpad skin. Journal of Neuroscience. 12(3):827-39,
1992.
Pieseiur-Strehlow B, Meinck HM. Response patters of human lumbo-sacral motoneurone
pools to distant somatosensory stimuli. Electroencephalography Clinical Neurophysiology.
48:673-682, 1980.
Pruszynski JA, Kurtzer I, Scott SH. Rapid motor responses are appropriately tuned to the
metrics of a visuospatial task. Journal of Neurophysiology. 100:224-38, 2008.
Vallbo AB, Johansson RS. Properties of cutaneous mechanoreceptors in the human hand
related to touch sensation. Human Neurobiology. 3(1):3-14, 1984.
Vallbo A B, Olausson H, Wessberg J and Kakuda N. Receptive field characteristics of tactile
units with myelinated afferents in hairy skin of human subjects. Journal of Physiology.483:783–
795, 1995.
Van Wezel BM, Ottenhoff FA, Duysens J. Dynamic control of location-specific information in
tactile cutaneous reflexes from the foot during human walking. Journal of Neuroscience.
17(10):3804-14, 1997.
Zehr EP, Stein RB, Komiyama T. Function of sural nerve reflexes during human walking. Journal
of Physiology. 507:305-14, 1998.
Zehr EP, Stein RB. What functions do reflexes serve during human locomotion? Progress in
Neurobiology 58(2):185-205, 1999.
Zehr EP, Collins DF, Chua R. Human interlimb reflexes evoked by electrical stimulation of
cutaneous nerves innervating the hand and foot. Experimental Brain Research. 140(4):495-504,
2001.
31
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
Figure Captions
Figure 1: Distribution of identified low threshold cutaneous afferents in the dorsum and plantar
surface skin of the foot. Representative receptive field size and location is illustrated.
Cutaneous receptors that demonstrated a significant coupling response are shaded.
Figure 2: Distribution and reflex coupling response for cutaneous receptors A) on the dorsum
and B) on the plantar surface of the foot. Afferents that demonstrated a significant coupling
with muscles of the upper limb are represented by shaded receptive fields. Those showing no
significant coupling remain unfilled. The foot is divided into quadrants to highlight receptor
density and the relation of foot location to reflex coupling in the upper limb. Bar graphs
indicate the percentage of each type of receptor with a coupling response in each quadrant.
Figure 3: Reflex coupling demonstrated with a fast adapting FAI afferent with receptive field
located on the dorsal aspect of the foot proximal to digit I. This afferent was activated using
rapid strokes across the receptive field. A significant reflex coupling was demonstrated in the
triceps brachii muscle at 96ms. A) The cyclic nature of activation resulted in a cyclic response in
the afferent autocorrelation as well as a reflex periodicity in both Deltoid and Triceps EMG
response at similar latencies. No significant responses were shown in response to the randomly
generated spike (B). The spike overlay is presented to illustrate unitary activity in the FAI.
Figure 4: Reflex coupling demonstrated by an FAII afferent. A) autocorrelation of the single
afferent showing a 10Hz firing rate (generated by stroking across the receptive field) and
significant reflex coupling in deltoid at 66, 166 and 263ms. B) Autocorrelation of random
generated spike and absence of significant coupling with Deltoid. C) Correlation between reflex
periodicity in Deltoid EMG and the interspike interval of the periodically activated FAII,
R2=0.997. The spike overlay is presented to illustrate unitary activity in the FAII.
Figure 5: SAII unit found on the front aspect of the ankle (see photo inset). The SAII was
activated by stretching along its preferred direction (proximal-distal along the leg). From
bottom to top: a Raw neurogram of the SAII action potential, Interstimulus interval
demonstrating an inherent firing frequency of 16.7 Hz in response to stretch, the spike trigger
averaged Deltoid EMG with responses approximating a 16 Hz cycle (every 60ms) and an
autocorrelogram of the SAII demonstrating a firing rate that corresponds (approximately 16
Hz). The spike overlay is presented to illustrate unitary activity in the SAII.
Figure 6: Low threshold cutaneous afferent innervating an SAI receptor located on the pad of
the fourth digit. SAI was activated during A) isometric contraction of the upper arm or B)
contraction of upper limb muscles while grasping an earth vertical referenced pole. The first
row illustrates the autocorrelogram (count: 5 ms bins) of the afferent demonstrating its
irregular firing behaviour (top; superimposed unitary spike in inset), followed by the spiketriggered averaged RMS EMG of deltoid and Triceps (2nd and bottom trace) (with 99%
confidence limits). Despite the fewer number of sweeps in the earth referenced condition
(666), there was a clear and significant modulation of both deltoid and triceps at latencies
32
800
801
802
803
indicating a potential spinal oligosynaptic, or subcortical reflex response. Average background
EMG did not differ between the isometric contraction and the EMG generated while holding
the earth reference (p=0.13)
33
804
805
806
807
808
Table 1: Summary of afferent median receptive field (mm2) and median threshold levels (N)
Median and Range are given for receptive field and activation threshold for each of the
receptor classes. N represents the number of afferents contributing and the percent of the
total population recorded (%).
n (%)
Threshold mN
Receptive Field mm2
Median
Range
Median
Range
FAI
15 (37)
7.31
0.83-81.89
75
32-144
FAII
4 (10)
0.73
0.63-0.83
180
42-330
SAI
9 (22)
25.3
0.63-50.01
73.5
24-266
SAII
10 (24)
1.42
0.63-32.36
200
70-567
34
809
810
811
812
813
814
815
816
Table 2. Summary details of cutaneous afferents showing significant coupling
Data from all units that demonstrated a significant coupling with motoneurons of upper limb
muscles are presented by afferent type. Latency (ms) and amplitude (peak as a % of
background EMG) are given for all units. Median latency is presented for all non-cyclical units.
One unit (39) showed a significant coupling only when the participant held onto an earth
referenced vertical pole (median in brackets for latency and amplitude includes this unit).
MUSCLE NERVE
UNIT
POLARITY
LATENCY (ms)
AMP (%background)
FAI
Deltoid
Cp
2
Inhibitory
72.3
3.5
Deltoid
Cp
5
Cyclic
11
2.86
Deltoid
Cp
9
Inhibitory
120
2.59
Deltoid
Cp
11
Inhibitory
62
1.12
Deltoid
Tibial
37
Inhibitory
105
3.66
Triceps
Cp
5
Cyclic
96
1.68
Triceps
Cp
8
Excitatory
120
1.21
Triceps
Cp
10
Excitatory
54
1.43
Triceps
Cp
13
Inhibitory
46
1.5
Triceps
Cp
16
Excitatory
38
3.48
Triceps
Tibial
40
Inhibitory
55
1.99
FAII
Deltoid
Cp
28
Excitatory
71
0.97
SAI
Deltoid
Cp
7
Inhibitory
48
3.16
Deltoid
Cp
22
Cyclic
35
1.85
Deltoid
Cp
25
Inhibitory
102
1.18
Deltoid
Tibial
39
Inhibitory
61
7.65 (pole)
Triceps
Cp
7
Inhibitory
56
2.6
Triceps
Tibial
39
Inhibitory
80
6.38 (pole)
SAII
Deltoid
Cp
6
Excitatory
43
6.1
Deltoid
Cp
35
Cyclic
30
1.21
Triceps
Cp
35
Cyclic
95
1.87
Ectopic Deltoid
Cp
20
Cyclic
98
2.03
Deltoid
Cp
21
Cyclic
65
2.99
Triceps
Cp
20
Cyclic
121
1.73
Triceps
Cp
21
Cyclic
70
3.2
Median
59 ms (61ms)
2.29 % (2.59%)
FAI
FAII
SAI
SAII
Ectopic
A
Hair
B
SAII
FAI
FAII
SAI
20.5cm
A
B
100
100
80
80
60
60
100
80
60
40
40
N/A
20
N/A
20
40
20
0
0
FAI
FAII
SAI
SAII
ectopic
FAI
FAII
SAI
SAII
ectopic
0
FAI
50%
Total:
FAI- 3
FAII- 1
SAI- 1
SAII- 1
17%
Significant:
Total:
FAI- 7
4
FAII- 0
0
SAI- 2.5
1
SAII- 1.5
0.5
SAI
SAII
Significant:
2
0
1
0
Total:
FAI- 0
FAII- 0
SAI- 0
SAII- 1
50%
Total:
Significant:
FAI- 1
0
FAII- 0
0
SAI- 1.5
0
SAII- 0.5
0.5
Total:
Significant:
FAI- 4
3
FAII- 3
1
SAI- 2
1
SAII- 1.5
1
Hair - 1
0
FAII
FAI
Total:
Significant:
FAI- 0
0
FAII- 0
0
SAI- 2
1
SAII- 2.5
0
FAII
SAI
SAII
Not significant
22%
52%
Significant:
Total:
FAI- 0
0
FAII- 0
0
SAI- 0
0
SAII- 1.5
0
100
100
80
80
60
60
40
40
N/A
N/A
FAI
FAII
20
20
0
0
FAI
FAII
SAI
SAII
ectopic
* two Ectopic not displayed due to uncertainty of location
SAI
SAII
ectopic
20.5cm
Significant:
Total:
FAI- 0
0
FAII- 0
0
SAI- 0
0
SAII- 0.5
0
Significant:
0
0
0
0
A
FAI
B
8506 sweeps
30mV
3500
3000
3000
1ms
~100ms
2500
2500
Count
Count
Random generated Spike
3500
2000
2000
1500
1500
1000
1000
500
500
0
0
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
-0.05
0.35
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
Time
Time
~100ms
25.4uV
23.4uV
Deltoid
25uV
22.4uV
-0.1
0
0.1
0.2
0.3
*96ms
0.4
-0.1
0
0.1
0.2
0.3
0.4
125.5uV
118.5uV
Triceps
115.2uV
-0.1
0
0.1
0.2
0.3
0.4
-0.1
Time (s)
0
0.1
0.2
0.3
123.7uV
0.4
10uV
A
B
1ms
F A II
10000
183 80 s w eeps
R a n d o m g e n e ra te d sp ike
10000
8000
C oun t
C oun t
8000
6000
6000
4000
4000
2000
2000
0
0
-0. 10 -0. 05
0. 00
0. 05
0. 10
0. 15
0. 20
66ms
0. 25
0. 30
0. 35
0. 40
-0. 10 -0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
263ms
166ms
131uV
111uV
129.2uV
0
0 9060 2 M C V U01 F AII
0 9060 2 M C V U01 F AII
0.1
0.1
0.2
0.2
0.3
0.3
0
0.1
eltoid
DDeltoid
300
250
200
150
100
R2=0.997
50
0
0
50
100
150
0.2
T im e (s)
Time (s)
D eltoid
D eltoid
C
-0.1
0.4
0.4
T im e (s)
ISI of FAII (ms)
-0.1
-0.1
110uV
200
250
Reflex periodicity (ms)
300
0.3
0.4
Spike
Overlay
180
~60ms
140
100
0 9 0 4423
EC
23 P
EC P
Count
60
20
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.20
0.25
0.30
0.35
~60ms
Deltoid
-0.05
0.00
0.05
0.10
0.15
20Hz
16.7Hz
ISI
Hz
25 mV
Ner ve
500ms
Time (S)
B
Isometric (10% MVC)
A
1662 sweeps
666 sweeps
10uV
120
Earth Reference (10% MVC)
120
Count
1ms
100
100
80
80
60
40
40
20
20
0
-0.10
-0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
240610
ECP U01
60
1
0.40
0
-0.10 -0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
30.9uV
99.8uV
28.5uV
88.6uV
Deltoid
-0.1
0
0.1
0.2
0.3
0.4
-0.1
0
0.1
0.2
0.3
0.4
61 ms
51uV
100uV
47uV
84uV
Triceps
-0.1
Time (s)
0
0.1
0.2
0.3
0.4
-0.1
Time (s)
0
0.1
88 ms
0.2
0.3
0.4