Brain Research Bulletin 75 (2008) 138–145
Paracrine-like excitation of low-threshold mechanoceptive C-fibers
innervating rat hairy skin is mediated by substance P via NK-1 receptors
Shi-Hong Zhanga,b,∗, Qi-Xin Sunb, Ze’ev Seltzerc, Dong-Yuan Caob,
Hui-Sheng Wangb, Zhong Chena, Yan Zhaob,∗∗
aFaculty of Pharmacology, School of Medicine, Zhejiang University, Hangzhou, Zhejiang 310058, PR China
bDepartment of Physiology and Pathophysiology, School of Medicine, Xi’an Jiaotong University, Xi’an, Shaanxi 710061, PR China
cThe University of Toronto Centre for the Study of Pain, Toronto, Ontario M5G 1G6, Canada
Received 25 June 2007; received in revised form 7 August 2007; accepted 16 August 2007
Available online 7 September 2007
We reported previously that C-fibers innervating rat skin can be excited by short trains of electrical shocks (‘tetanus’) applied to neighboring
nerves. Since these nerves were disconnected from the CNS, the cross-talk is located peripherally. Here we tested if low-threshold mechanoceptive
(LTM) C-fibers can be excited by this cross-talk and if this process is mediated by substance P (SP) via neurokinin-1 (NK-1) receptors. In urethane
anesthetized rats we found that 80% (56/71) of LTM C-fibers, recorded in the lateral cutaneous branch of the dorsal ramus (CBDR) of T10 spinal
declined to the baseline frequency thereafter. When injected into their receptive fields, SP mimicked the tetanically induced increase of firing rate,
whereas the NK-1 receptor antagonist WIN 51708 blocked the excitation in most fibers. The excitation was significantly diminished in adult rats
afferents are functionally linked with adjacent LTM C-fibers in a non-synaptic, paracrine-like signaling pathway via SP and NK-1 receptors, and
perhaps also other agents as well. We propose that this cross-talk has evolved as a mechanism regulating the mechanoceptive characteristics of
LTM C-fibers, presumably contributing to pain sensation elicited by tactile stimuli (‘allodynia’).
© 2007 Elsevier Inc. All rights reserved.
Keywords: Substance P; NK-1 receptors; Low-threshold mechanoceptive C-fibers; Cross-talk
Some neuropeptides, such as substance P (SP), are richly
ents (A?- and C-fibers, respectively) . When released from
peripheral terminals following antidromic electrical stimula-
tion , noxious stimulation  or some chemical stimulants
, these neuropeptides, especially SP and calcitonin gene-
related peptide (CGRP), diffuse through the surrounding tissues
∗Corresponding author at: Faculty of Pharmacology, School of Medicine,
Zhejiang University, Hangzhou, Zhejiang 310058, PR China.
Tel.: +86 571 88208227; fax: +86 571 88208228.
∗∗Corresponding author. Tel.: +86 29 82655171; fax: +86 29 82655160.
E-mail addresses: firstname.lastname@example.org (S.-H. Zhang),
email@example.com (Y. Zhao).
and contribute to the occurrence of neurogenic inflammation
[20,29], and the sensitization of sensory receptors, producing
mechanism, nociceptive afferent nerve fibers carry out efferent
functions that affect targets carrying receptors for these neu-
ropeptides in the periphery.
Previous investigations in monkeys, rats or rabbits showed
that electrical antidromic stimulation of a nerve trunk, or perfu-
sion with SP, does not change the sensitivity of C-nociceptors
to heat or mechanical stimuli [10,32,37]. However, Carlton et
al. reported that the neurokinin-1 (NK-1) receptors, to which
SP shows high affinity, are present on peripheral unmyelinated
afferent axons in the rat . This implies that SP, released from
peripheral terminals of activated primary afferents, may affect
0361-9230/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145
activation of cutaneous NK-1 receptors excites nociceptive C-
fibers in the sural nerve of the rat and sensitizes them to heat
stimuli , further investigation is needed to clarify whether SP
can activate other types of unmyelinated afferent nerves.
Low-threshold mechanoceptive (LTM) C-fibers are a sub-
type of primary afferents that innervate the skin of mammals
and respond to light touch, pressure and stretching of the skin
[3,23,26,31,35,40,43,44]. Activation of sympathetic efferents
can excite these afferents transiently [2,12,38]. Their functional
C-fibers may be involved in itch , Roberts and Elardo 
proposed that the primary function of these receptors is to sig-
nal conditions within the skin for regulatory purposes, rather
than to encode external mechanical events. Additionally, it has
been proposed that these receptors convey sensory information
from the skin to limbic structures that elicit a pleasant feeling of
painful stimuli [43,44]. It remains unknown, however, whether
the receptors of LTM C-fibers respond to SP and whether SP
controls their excitability through NK-1 receptors.
We reported previously that the firing frequency of A?-
and C-fibers in the cutaneous branch of the dorsal ramus
(CBDR) is significantly increased following a repetitive elec-
trical stimulation (‘tetanization’) of the neighboring CBDR .
This indicates the existence of a functional coupling between
neighboring afferent nerve fibers coming from different spinal
segments (heretofore ‘cross-talk’). However, it is not clear as of
yet whether this phenomenon also includes LTM C-fibers and
whether SP and its NK-1 receptor mediate this functional link.
To this end, in the present study we used single unit record-
ings of LTM C-fibers and intracutaneous microinjection of SP
or an antagonist of NK-1 receptors into the receptive fields
of such units, to elucidate whether SP released from tetanized
afferent nerve fibers can affect the frequency of spontaneous fir-
ing of recorded fibers via NK-1 receptors. We also investigated
whether such cross-talk can be demonstrated in adult rats that
were treated neonatally with capsaicin (‘NN-CPSN’), a treat-
ment that irreversibly destroys a large number of C-fibers and
some A?-fibers, including many SP-containing fibers.
2. Materials and methods
This study reports on experiments that were carried out on 88 adult male
Xi’an Jiaotong University. They were housed under a 12–12h light–dark cycle
(lights switched on at 7:00am). Standard food and water were available ad
libitum. This study followed a protocol approved by the Institutional Animal
Care Committee of the university and adhered to the Principles of Laboratory
Animal Care (NIH publication No. 80-23, revised 1996).
Rats were deeply anesthetized with urethane (1.0g/kg, i.p.). Supplemental
flexia. The fur on the back was clipped and a longitudinal incision was made
of the right T9 and T10 spinal nerves. This approach left the receptive fields
(RFs) of the studied nerves intact. An oil pool was constructed with skin flaps
and filled with liquid paraffin kept at 36◦C. Animals were placed on a thermo-
of 37.5◦C via feedback from a rectal thermometer.
2.3. Single unit recording
A schematic drawing of the experimental set-up is shown in Fig. 1. Under
a dissecting microscope at 6× magnification, a length of ∼3cm of T9 and T10
CBDRs was isolated carefully from the surrounding tissues. These nerves were
of ∼3cm of these nerves was still connected to the periphery but disconnected
from the CNS. The cut T9 CBDR was placed on a pair of platinum stimulating
electrodes. Using a dissecting microscope at 25× magnification, a thin micro-
filament was teased out from the distal cut end of the T10 CBDR with a pair of
sharply honed jewellers’ forceps, and adhered to a pair of platinum recording
electrodes. The nerves were fully submerged in the oil pool to prevent drying.
Amplified and filtered electrical activity was displayed on an oscilloscope (VC-
11, Nihon Kohden, Japan). The signals were also fed to a computerized window
of cross-talk between two neighboring nerves. The firing of single C-fibers was
recorded in the lateral cutaneous branch of the dorsal ramus (CBDR) of the T10
spinal nerve, while the CBDR of T9 spinal nerve was electrically stimulated
(stimulus intensity 1.0mA; pulse width 0.2ms; shock frequency 20Hz; and
tetanus duration 10s). Since both CBDR nerves were cut centrally the increased
frequency of the recorded firing must be due to a cross-talk occurring in the
periphery. The scheme shows the hypothesis we tested here, suggesting that
the tetanus activated action potentials in T9 CBDR nerve fibers (clear arrow)
that propagated antidromically, causing the release of SP (filled circles) from
peripheral terminals of the tetanized peptidergic fibers. SP diffused in the skin
(semicircular black arrows) to neighboring low-threshold mechanoceptive C-
fibers (LTM C-fibers), binding to NK-1 receptors (triangles), sensitizing the
terminals of LTM C-fibers in T10 CBDR and causing them to increase their
spontaneous firing frequency or respond by firing de novo (black arrow). DRG:
dorsal root ganglion.
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145
experiment. Spontaneous and peri-stimulus firing of studied units was recorded
and analyzed offline thereafter.
applied by an ascending set of calibrated von Frey hair that ranged from 0.02
to 17mN, and the center of RF was marked with black color. No other stim-
ulus modalities were applied. Next, 3min of baseline spontaneous firing was
recorded, from the LTM unit, followed by tetanic stimulation of the T9 CBDR,
investigation was continuously followed for 3min thereafter, by counting the
number of impulses per 30s periods (i.e., 0–30, ..., 150–180s), comparing the
resulting firing frequency to the pre-tetanic baseline frequency. The baseline
frequency was expressed as the average number of impulses per 30s over a
period of 3min preceding the tetanus. If following the antidromic stimulation of
T9 CBDR, the firing frequency increased to ≥1.5 times more than the baseline
rate we considered this unit excited by the T9 CBDR tetanus. Then, a single
0.5ms suprathreshold electrical stimulus was applied to the center of the RF
by a pair of stimulating needle electrodes, concurrently triggering the sweep of
the oscilloscope. The conduction latency was measured from shock onset to the
rising phase of the action potential, without allowance for utilization time. The
conduction distance was measured at the end of the experiment as the length of
a 3–0 suture thread placed along the contour of the recorded nerve. The conduc-
tion velocity (CV) was calculated by dividing the distance from the stimulating
electrodes to the recording electrodes by the conduction latency. Units with a
CV lesser than 2.0m/s were considered C-fibers [21,26].
2.4. Intracutaneous microinjections
If the unit under investigation was cross-excited by the tetanization of
T9 CBDR, it was further studied as follows. Thirty minutes after the stim-
ulation, a 36-gauge needle connected via a polyethylene catheter (PE10)
to a Hamilton 50?L microsyringe, was inserted carefully intracutaneously
into the center of the unit’s RF. A volume of 10?L of the following com-
pounds (or their vehicles) was injected slowly over a period of 5min using
a syringe pump (WZ-50, Medical Research Apparatus, Zhejiang University,
China). Injected chemicals included SP (10−7M, Sigma, St. Louis, MO, USA)
and the non-peptide NK-1 receptor antagonist WIN 51,708 (10−6M, Sigma,
St. Louis, MO, USA). SP was dissolved in 0.01M acetic acid and artificial
cerebrospinal fluid to 10−4M, and WIN 51,708 – in ethanol and propane-
diol (ethanol:propanediol:water=1:1.5:7.5) to 10−3M as stock solutions. Just
before the microinjection they were diluted with saline to the concentration
The firing frequency of units whose RFs were injected with SP or its vehicle
frequency in the period of 3min preceding the injections. A second tetanus
was not applied to such units. Both WIN 51,708 and its vehicle excited the
units. When the firing frequency stabilized, and not earlier than 15min after the
injection, a second tetanic stimulation was delivered to T9 CBDR and the effect
on the firing frequency of the units under investigation was followed for 3min,
using the same process as described above.
2.5. Neonatal treatment with capsaicin
in ethanol:Tween-80:saline=1:1:8, and diluted with saline to 1%) was injected
C-fibers and some A?-fibers in peripheral nerves, including fibers expressing
SP [15,16,33,34]. Artificial respiration was applied to the pups in respiratory
distress. At the age of 8–9 weeks these animals were tested as described above
for intact animals.
2.6. Tail flick test
Success of the capsaicin treatment can be demonstrated by a significant
increase in the tail flick latency to noxious heat stimuli of NN-CPSN ani-
mals compared to intact control rats [15,16,34]. The rat was held gently with
a towel and a length of 3cm of the tip of tail was immersed in hot water
(50±0.5◦C). The latency to flicking the tail was determined with a stop-
watch at a resolution of 0.1s. Cutoff time was set at 8s to avoid heat injury to
2.7. Data analysis
To observe the time course of excitation of LTM C-fibers closely following
the tetanus of T9 CBDR, the impulses during an interval of 30s (impulses/30s)
(impulses/min). The data were expressed as means±S.E.M. Statistic analysis
was performed with SPSS Ver. 12.0 for Microsoft Windows. One-way ANOVA
followed by paired t-test or Wilcoxon signed rank test were used to compare
the average firing frequencies before and after the tetanization of T9 CBDR,
as well as before and after the intracutaneous injection of SP and its vehicle.
Student’s t-test and Fisher’s exact test were used to compare group differences
across intact and NN-CPSN animals, in the frequency of spontaneous firing, tail
flick latency, the proportion of units with spontaneous firing and units excited,
respectively. Correlation analysis was done by Spearman rank correlation test.
P<0.05 was considered significant in all tests.
3.1. Excitation of LTM C-fibers by activity in neighboring
spinal nerve in intact animals. They were all mechanoceptive
single units, responding to punctate stimulation with von Frey
hairs. The RF to mechanical stimulation was usually rounded
or oval shaped, ranging in size from a single small point to an
area covering 12.6mm2, located in an area spanning from 0.5 to
firing thresholds of the studied units in response to stimulation
units were LTM C-fibers, having activation thresholds ≤3.5mN
at a low frequency of 3.29±0.60impulses/30s, while the rest
of the units (37/71, 53%) had no spontaneous activity. The
overall average firing frequency was 1.62±0.40impulses/30s.
The conduction velocity of all studied units ranged from 0.5 to
1.9m/s, 1.21±0.04m/s on average.
During the 10s, 20Hz tetanization of the T9 CBDR, none
of the recorded fibers responded to the 200 shocks in a fixed
Fig. 2. Histogram showing the frequency of units having firing thresholds in
response to tactile stimulation of their RFs with an ascending set of calibrated
stimuli, below the noxious range. Thus, this report is on LTM C-fibers.
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145
Fig. 3. (A) Trace showing the firing of a LTM C-fiber in T10 CBDR, with
mechanoreceptive threshold of 0.44mN, that was excited by tetanization of T9
CBDR. The black bar indicates the tetanus duration (stimulus intensity 1.0mA;
showing the response of the unit to a single suprathreshold electrical shock to
the cutaneous receptive field. The delay from shock onset to the action potential
(43ms) and the measured conduction distance (37mm) enabled us to determine
that the conduction velocity of this unit was 0.86m/s. The arrow points to the
evoked action potential of the unit.
latency, suggesting that the shocks did not directly excite them.
However, after the tetanization ended, 80% of the LTM C-fibers
(56/71) recorded in T10 CBDR increased their firing rate indi-
cating they were excited indirectly by the electrical stimulation
of T9 CBDR (Fig. 3A and B). Units with spontaneous firing as
well as units without spontaneous firing showed this excitation.
The percent of units that were excited by the tetanus in these
two groups was not significantly different (74%, 25/34 versus
84%, 31/37, respectively) (P>0.05, chi-square test). Overall,
in the first period of 30s after the electrical stimulation the
average firing frequency significantly increased from a base-
line of 1.62±0.40impulses/30s to 3.74±0.99impulses/30s
(P<0.01, Wilcoxon signed rank test). The firing frequency
during the second and third periods of 30s post-tetanus were
level (3.17±0.69 and 2.92±0.63impulses/30s, P<0.01 and
P<0.05, respectively, paired t-test, Fig. 4). Thereafter, the fir-
rate was during the first period of 30s post-tetanus and was not
correlated significantly to the firing frequency during the pre-
tetanus baseline period (Spearman rank correlation, rs=0.15,
3.2. Do excited LTM C-fibers respond to SP?
Sixteen of the 56 T10 CBDR LTM C-fibers that responded
to tetanization of T9 CBDR were further tested to see if they
respond to SP (n=8) or the vehicle (n=8). Following intracu-
taneous microinjection of SP into the center of their RFs, the
spontaneous firing frequency of all 8 LTM C-fibers was sig-
nificantly augmented for a period ranging from 2 to 10min (P
values ranged from <0.05 to <0.01, paired t-test or Wilcoxon
signed rank test, Fig. 5). In contrast, injecting the vehicle had
no significant effect on the firing frequency of other eight LTM
C-fibers (Fig. 5).
Fig. 4. The time course of the average firing frequency (expressed as
impulses/30s) of LTM C-fibers in T10 CBDR before, and after the tetanization
treated rats (‘NN-CPSN’, n=17units). The break on the abscissa indicates the
respectively, when compared to the pre-tetanic baseline firing frequency (‘Con-
trol’). This frequency is expressed as the average impulses recorded per 30s
over a period of 3min preceding the first tetanus.
3.3. Does the excitation of LTM C-fibers depend on
activation of NK-1 receptors?
excited by a tetanus of the T9 CBDR were further studied by
testing whether the cross-talk could still be demonstrated under
NK-1 receptor blockade by WIN 51,708 (n=20), or after the
vehicle injection (n=12). Inserting the needle to the center of
RF, followed by injection of WIN 51,708 or its vehicle, excited
all 32 studied units, for a period lasting more than ten minutes
(data not shown). When the firing stabilized at the new plateau,
and not shorter than 15min after the injection, a second tetanic
stimulation was delivered to the T9 CBDR and the effect on the
Fig. 5. The time course of the average firing frequency (expressed as
impulses/min) of LTM C-fibers in the T10 CBDR following a local intracuta-
neous microinjection of substance P and its vehicle (n=8units for each group)
and its duration (5min). ** and * represent P<0.01 and P<0.05, respectively,
when compared to the baseline frequency, which is expressed as the average
impulses per minute recorded over a period of 3min preceding the injections of
SP or its vehicle (denoted by ‘0’).
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145
Fig. 6. The time course of the average firing frequency (expressed as
impulses/30s) of LTM C-fibers in T10 CBDR following tetanization of the
T9 CBDR, before (‘Control’) and after a local intracutaneous microinjection of
duration of the tetanus (stimulus intensity 1.0mA; pulse width 0.2ms; shock
frequency 20Hz; and tetanus duration 10s). ** and * represent P<0.01 and
P<0.05, respectively, when compared to the pre-tetanic baseline. This baseline
(‘Control’) was the average firing rate (expressed as impulses/30s) recorded
over a period of 3min preceding the second tetanus, and not earlier than 15min
after the injection of WIN 51,708 or the vehicle.
firing frequency of the units under investigation was followed
for 3min. In all 12units treated with vehicle alone, the second
tetanus produced an excitation that manifested in a significant
increase in the firing frequency, above that of the pre-tetanic fir-
ing plateau (P<0.05 or 0.01, paired t-test or Wilcoxon signed
rank test, Fig. 6). However, in WIN 51,708 treated units, the
firing frequency following the second tetanus was not signif-
icantly different from the frequency in the pre-tetanic firing
plateau (P>0.05, paired t-test, Fig. 6). Nine of the 20units
(45%) showed no excitation after the second tetanus, that is,
WIN 51,708 entirely blocked the excitation. In 6/20units (30%)
the increase of firing rate after the second tetanus was markedly
diminished compared to that after the first tetanus, and in the
remainder of the units (5/20, 25%) WIN 51,708 did not block
the excitation at all, since the increase of firing rate was similar
to, or even greater than that after the first tetanus.
3.4. Cross-talk in rats neonatally treated with capsaicin
CPSN rats was significantly delayed from 4.07±0.1s (n=19)
to 5.79±0.42s (n=17, P<0.01, t-test). This expected result
indirectly verified that NN-CPSN eliminated from peripheral
nerves a subpopulation of heat-nociceptive primary afferent
fibers, many of which are known to express SP.
We recorded in NN-CPSN animals from 17 T10 CBDR
LTM C-fibers. Their RF shape and location were like those
of units in intact animals. Five out of seventeen of these units
(29.4%) spontaneously fired action potentials. The proportion
of units expressing spontaneous firing was lower than that of
intact animals (34/71, 47%) but this reduction was not signifi-
cant (P>0.05, Fisher’s exact test). However, the overall average
spontaneous firing frequency of these units was significantly
lower than that of intact animals (P<0.05, t-test).
A tetanus of T9 CBDR in NN-CPSN animals increased the
firing frequency in 5/17units (29.4%) only slightly above the
baseline level, indicating the presence of a very weak form of
fibers (56/71) in intact animals. The same tetanus in NN-CPSN
In 1/17units (5.9%) the tetanus actually caused a decrease in
spontaneous firing. The proportion of units showing cross-talk
in NN-CPSN animals was significantly lower than that of intact
animals (P<0.001, Fisher’s exact test). Overall, the average fir-
ing rate after the tetanus was not significantly different from the
baseline (P>0.05, paired t-test or Wilcoxon signed rank test,
skin in the back of the rat fire spontaneously at a low frequency
and that this spontaneous activity can be enhanced by neural
activity in neighboring primary afferents. The results of several
experiments in this study converge on the same conclusion that
like signaling pathway that is mediated in part by SP via NK-1
4.1. SP via NK-1 is the mediating link of the cross-talk
Since both nerves that we studied in this preparation were
disconnected from the CNS, the functional link must be located
in the periphery. It is well known that SP could be released from
stimulation . We propose that SP via NK-1 receptors medi-
results of the present study: (i) all of the T10 CBDR LTM C-
fibers that were excited by tetanic stimulation of the T9 CBDR
also responded to SP, (ii) significantly fewer units expressed
cross-talk when NK-1 receptors were blocked, and (iii) fewer
units showed cross-talk in NN-CPSN rats.
It could be argued that the increased activity in the recorded
LTM C-fibers resulted from current spread from the stimulat-
ing electrodes to the recorded fibers, directly stimulating the
fibers. However, for the following reasons this explanation is
highly unlikely. First, current spread is expected to excite the
recorded units to fire during the tetanus, and not thereafter as
was the case in the activity reported here. Second, since the
stimulated and recorded nerves were attached to electrodes that
were immersed in a pool of electrically isolating mineral oil,
the current could only spread along the stimulated nerve. Third,
at the end of the experiment, before sacrificing the animal, in a
few cases we crushed the stimulated nerve distal to the stimulat-
ing electrodes, near the skin. In all cases crushing the nerve
prevented the increased firing rate of the tetanized LTM C-
fibers. Taken together, we conclude that antidromic conduction
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145
It could be argued that the excitation observed following the
injection of SP was caused by activation of acid-sensing ion
channels (ASICs) by the acidic injected solutions. Indeed, SP
and its vehicle were both weakly acidic (pH ∼6). But since the
vehicle alone did not cause excitation of the units under investi-
over, ASICs were demonstrated on high-threshold, nociceptive
C-fibers, but not LTM C-fibers . Therefore, the cross-talk is
most likely due to a specific effect of SP at NK-1 receptors.
4.2. Proposed mechanisms for the cross-talk
Carlton et al. reported that NK-1 receptors are expressed
on unmyelinated fibers . This finding leads to the possibil-
ity that SP released from the peripheral peptidergic afferent
terminal endings may bind to these NK-1 receptors and con-
sequently affect the activity of adjacent C-fibers. Indeed, it has
been reported that SP can sensitize nociceptive C-fibers via
direct activation of NK-1 receptor . Arguably, LTM C-fibers
from the antidromically stimulated T9 CBDR peptidergic pri-
mary afferents. SP activates NK-1 receptors on LTM C-fibers in
T10 CBDR, which in turn excites the LTM C-fibers (as shown
in Fig. 1).
Since the peak increase in spontaneous activity was as short
as 30s after tetanus onset, we conclude that SP released from
the tetanized primary afferents had to diffuse in the tissues over
a short distance until reaching NK-1 receptors on the recorded
C-fibers. Since there is no morphological evidence that sug-
gests the existence of synaptic contacts between neighboring
afferent nerve fibers in the periphery, we propose that the cross-
talk we observed is like a non-synaptic, paracrine-like signaling
Cross-talk induced by tetanic stimulation was significantly
diminished in NN-CPSN-treated animals, supporting the role
putative signaling route. In spite of the loss of some peptidergic
afferents in NN-CPSN rats, we could still demonstrate a cross-
CBDR of T10. It is possible that other neuroactive substances
may be involved in the cross-talk as well.
It is interesting that LTM C-fibers of NN-CPSN-treated rats
fired spontaneously at a lower rate than that of intact animals.
fibers is affected by a low level release of SP from neighboring
of LTM C-fibers to natural mechanical stimuli. This proposed
concept needs further investigation.
4.3. The possible role of other neuroactive substances in
In adult rats that underwent a neonatal capsaicin treatment,
in the presumed absence of most SP-expressing afferent fibers,
capsaicin-resistant fibers that survived the NN-CPSN treatment
using a different mediator than SP. Indeed, in the present study,
55% of LTM C-fibers in the CBDR of T10, that were activated
by tetanizing the T9 CBDR, could still demonstrate cross-talk
with the presence of other signaling pathways that mediate the
cross-talk, in addition to SP via NK-1 receptors. Future studies
that would manipulate other fiber classes in NN-CPSN animals,
are needed to better clarify this mechanism.
There are several candidates who these additional mediators
in the skin. Following antidromic stimulation, SP released from
peptidergic afferent terminal endings could act on NK-1 recep-
the latter [36,41] which could induce the release of histamine
from mast cells [13,14,36,41]. Most C-fibers can be activated
by histamine, including LTM C-fibers [22,24,30].
Another candidate mediator is glutamate. Antidromic acti-
vation of primary afferents by electrical stimulation causes the
release of glutamate from their peripheral terminals . More-
over, glutamate receptors have been identified on C-fibers in the
plantar nerve . Glutamate is able to excite primary sensory
nerves [1,4]. Furthermore, glutamate injected to the periphery
interacts with NK-1 receptors [8,47]. Therefore, it is possible
that the tetanus of T9 CBDR triggered the release of glutamate
in the periphery, which participated in the process of cross-talk
carry NK-1 receptors.
Noradrenalin, ATP and adenosine are additional candidate
mediators. Peripheral nerves contain sympathetic postgan-
glionic fibers. It has been demonstrated that LTM C-fibers can
be excited by sympathetic activation or by administration of
noradrenalin (NA), but this transient activation lasted not more
than 30s after the stimulus [2,12,38], in contrast to the longer
lasting cross-excitation reported here. In addition, adenosine
and ATP are released from afferents and sympathetic postgan-
glionic fibers when stimulated electrically. Both adenosine and
ATP affect nociception [39,46,48]. The possible contribution of
NA, adenosine and ATP to the non-synaptic cross-talk between
adjacent afferent nerve fibers needs more research.
experiments suggesting that there is a non-synaptic, paracrine-
like excitatory functional link between afferent nerve fibers that
belong to different spinal segments, involving SP and NK-1
receptors, and perhaps other neuroactive substances as well.
More work is needed to elucidate the role of this signaling path-
way, testing the hypothesis that it serves an adaptive function
in modulating the excitability of low-threshold mechanoceptive
C-fibers in normal conditions, after nerve injury, inflammation
and diseases. This mechanism could be part of the appearance
after nerve injury, certain diseases and inflammation.
Conflict of interest
There is no conflict of interest about this paper.
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145
This project was supported by the National Natural Science
Foundation of China (Nos. 30701015, 30600219, 30400131,
and 30572176). ZS was supported by the Canada Research
Chair Program. We thank John Radcliffe for proof reading the
 B. Ault, L.M. Hildebrand, l-Glutamate activates peripheral nociceptors,
Agents Actions 39 (1993) 142–144.
and nociceptive afferent units from the rabbit pinna, Brain Res. 378 (1986)
 P. Bessou, P.R. Burgess, E.R. Perl, C.B. Taylor, Dynamic properties of
 D.Y. Cao, Y. Guo, Q. Zhang, Y.L. Tian, H.S. Wang, Y. Zhao, Effects of
glutamate on the afferent discharges of dorsal cutaneous sensory nerves in
rats, Neurosci. Bull. 21 (2005) 111–116.
 D.Y. Cao, H.J. You, Y. Zhao, Y. Guo, H.S. Wang, L. Arendt-Nielsen, H.L.
Wang, Q. Zhang, Involvement of peripheral ionotropic glutamate recep-
tors in activation of cutaneous branches of spinal dorsal rami following
antidromic electrical stimulation of adjacent afferent nerves in rats, Brain
Res. Bull. 72 (2007) 10–17.
glutamate receptors in unmyelinated axons of rat glabrous skin, Neurosci.
Lett. 197 (1995) 25–28.
 S.M. Carlton, S. Zhou, R.E. Coggeshall, Localization and activation of
substance P receptors in unmyelinated axons of rat glabrous skin, Brain
Res. 734 (1996) 103–108.
 S.M. Carlton, S. Zhou, R.E. Coggeshall, Evidence for the interaction of
glutamate and NK1 receptors in the periphery, Brain Res. 790 (1998)
 W.L. Chen, Y.Q. Zhang, Z.Q. Zhao, Neurokinin-1 receptor in peripheral
nerve terminals mediates thermal hyperalgesia, Biochem. Biophys. Res.
Commun. 339 (2006) 132–136.
 R.H. Cohen, E.R. Perl, Contributions of arachidonic acid derivatives and
substance P to the sensitization of cutaneous nociceptors, J. Neurophysiol.
64 (1990) 457–464.
paw following low and high intensity sciatic stimulation, NeuroReport 11
 E.D. Efes, Mechanism of effect of sympathetic nervous system on skin
receptors, Neurophysiology 24 (1992) 358–363.
D. Whitaker-Menezes, G.F. Murphy, Characterization of unmyelinated
axons uniting epidermal and dermal immune cells in primate and murine
skin, J. Cutan. Pathol. 25 (1998) 20–29.
 F. Erjavec, F. Lembeck, T. Florjanc-Irman, G. Skofitsch, J. Donnerer, A.
Saria, P. Holzer, Release of histamine by substance P, Naunyn Schmiede-
bergs Arch. Pharmacol. 317 (1981) 67–70.
 R. Gamse, Capsaicin and nociception in the rat and mouse. Possible
role of substance P, Naunyn Schmiedebergs Arch. Pharmacol. 320 (1982)
 D.L. Hammond, M.A. Ruda, Developmental alterations in nociceptive
threshold, immunoreactive calcitonin gene-related peptide and substance
P, and fluoride-resistant acid phosphatase in neonatally capsaicin-treated
rats, J. Comp. Neurol. 312 (1991) 436–450.
 R.D. Helme, G.M. Koschorke, M. Zimmermann, Immunoreactive sub-
stance P release from skin nerves in the rat by noxious thermal stimulation,
Neurosci. Lett. 63 (1986) 295–299.
 B. Heppelmann, M. Pawlak, Sensitisation of articular afferents in normal
and inflamed knee joints by substance P in the rat, Neurosci. Lett. 223
 M.K. Herbert, R.F. Schmidt, Sensitisation of group III articular afferents
to mechanical stimuli by substance P, Inflamm. Res. 50 (2001) 275–282.
 P. Holzer, Local effector functions of capsaicin-sensitive sensory nerve
endings: involvement of tachykinins, calcitonin gene-related peptide and
other neuropeptides, Neuroscience 24 (1988) 739–768.
 K.W. Horch, R.P. Tuckett, P.R. Burgess, A key to the classification of
cutaneous mechanoreceptors, J. Invest. Dermatol. 69 (1977) 75–82.
 B. Hutcheon, E. Puil, I. Spigelman, Histamine actions and comparision
with substance P effects in trigeminal neurons, Neuroscience 55 (1993)
 A. Iggo, Cutaneous mechanoreceptors with afferent C fibres, J. Physiol.
152 (1960) 337–353.
 M.S. Jafri, K.A. Moore, G.E. Taylor, D. Weinreich, Histamine H1 receptor
activation blocks two classes of potassium current, IK (rest) and IAHP, to
excite ferret vagal afferents, J. Physiol. 503 (1997) 533–546.
 S.N. Lawson, B.A. Crepps, E.R. Perl, Relationship of substance P to
afferent characteristics of dorsal root ganglion neurones in guinea-pig, J.
Physiol. 505 (Pt 1) (1997) 177–191.
 J.W. Leem, W.D. Willis, J.M. Chung, Cutaneous sensory receptors in the
rat foot, J. Neurophysiol. 69 (1993) 684–699.
 A. Leffler, B. Monter, M. Koltzenburg, The role of the capsaicin receptor
139 (2006) 699–709.
 F. Lembeck, Sir Thomas Lewis’s nocifensor system, histamine and
 F. Lembeck, J. Donnerer, L. Bartho, Inhibition of neurogenic vasodilation
and plasma extravasation by substance P antagonists, somatostatin and
[D-Met2, Pro5] enkephalinamide, Eur. J. Pharmacol. 85 (1982) 171–176.
 B. Lynn, Capsaicin: actions on C fibre afferents that may be involved in
itch, Skin Pharmacol. 5 (1992) 9–13.
 B. Lynn, S.E. Carpenter, Primary afferent units from the hairy skin of the
rat hind limb, Brain Res. 238 (1982) 29–43.
 R.A. Meyer, J.N. Campbell, S.N. Raja, Antidromic nerve stimulation in
monkey does not sensitize unmyelinated nociceptors to heat, Brain Res.
441 (1988) 168–172.
 J.I. Nagy, L.L. Iversen, M. Goedert, D. Chapman, S.P. Hunt, Dose-
dependent effects of capsaicin on primary sensory neurons in the neonatal
rat, J. Neurosci. 3 (1983) 399–406.
 J.I. Nagy, D. van der Kooy, Effects of neonatal capsaicin treatment on
nociceptive thresholds in the rats, J. Neurosci. 3 (1983) 1145–1150.
 M. Nordin, Low-threshold mechanoreceptive and nociceptive units with
unmyelinated (C) fibres in the human supraorbital nerve, J. Physiol. 426
neurokinin NK-1 receptor expression in rat peritoneal mast cells, Inflamm.
Res. 48 (1999) 274–279.
 P.W. Reeh, L. Kocher, S. Jung, Does neurogenic inflammation alter the
sensitivity of unmyelinated nociceptors in the rat? Brain Res. 384 (1986)
 W.J. Roberts, S.M. Elardo, Sympathetic activation of unmyelinated
mechanoreceptors in cat skin, Brain Res. 339 (1985) 123–125.
 J. Sawynok, Adenosine receptor activation and nociception, Eur. J. Phar-
macol. 347 (1998) 1–11.
vating the skin of the rabbit’s ear, J. Neurophysiol. 54 (1985) 491–501.
J. Bienenstock, Direct neurite-mast cell communication in vitro occurs via
the neuropeptide substance P, J. Immunol. 163 (1999) 2410–2415.
 E. Theriault, M. Otsuka, T. Jessell, Capsaicin-evoked release of substance
P from primary sensory neurons, Brain Res. 170 (1979) 209–213.
 A.B. Vallbo, H. Olausson, J. Wessberg, Unmyelinated afferents consti-
tute a second system coding tactile stimuli of the human hairy skin, J.
Neurophysiol. 81 (1999) 2753–2763.
 J. Wessberg, H. Olausson, K.W. Fernstrom, A.B. Vallbo, Receptive field
properties of unmyelinated tactile afferents in the human skin, J. Neuro-
physiol. 89 (2003) 1567–1575.
S.-H. Zhang et al. / Brain Research Bulletin 75 (2008) 138–145 Download full-text
 D.M. White, R.D. Helme, Release of substance P from peripheral nerve
terminals following electrical stimulation of the sciatic nerve, Brain Res.
336 (1985) 27–31.
 H. Yajima, J. Sato, R. Giron, R. Nakamura, K. Mizumura, Inhibitory, facil-
itatory, and excitatory effects of ATP and purinergic receptor agonists on
the activity of rat cutaneous nociceptors in vitro, Neurosci. Res. 51 (2005)
trophysiological evidence for the interaction of substance P and glutamate
on A? and C afferent fiber activity in rat hairy skin, Clin. Exp. Pharmacol.
Physiol. 33 (2006) 1128–1133.
 Q. Zhang, Y. Zhao, Y. Guo, D.Y. Cao, X.D. Tang, Y.L. Tian, F.R. Yao, H.S.
by P2X receptors in rat dorsal skin, Brain Res. 1102 (2006) 78–85.