Pain 125 (2006) 125135

www.elsevier.com/locate/pain

Tonic inhibitory role of a4b2 subtype of nicotinic acetylcholine

receptors on nociceptive transmission in the spinal cord in mice
Md Harunor Rashid
b

a,b,*

, Hidemasa Furue a, Megumu Yoshimura a, Hiroshi Ueda

a
Department of Integrative Physiology, Kyushu University Graduate School of Medical Sciences, Fukuoka 812-8582, Japan
Division of Molecular Pharmacology and Neuroscience, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki 852-8521, Japan

Received 23 November 2005; received in revised form 6 April 2006; accepted 3 May 2006

Abstract
In the spinal dorsal horn, activation of the nicotinic acetylcholine receptors (nAChR) by exogenously applied agonists is known
to enhance inhibitory synaptic transmission, and to produce analgesia. However, it is still unknown whether endogenously released
acetylcholine exerts a tonic inhibition on nociceptive transmission through the nAChRs in the spinal dorsal horn. Here, we report
the presence of such a tonic inhibitory mechanism in the spinal dorsal horn in mice. In behavioral experiments, intrathecal (i.t.)
injection of non-selective nAChR antagonist mecamylamine and a4b2 subtype-selective antagonist dihydro-b-erythroidine (DHbE)
dose-dependently induced thermal and mechanical hyperalgesia in mice while the a7-selective antagonist methyllycaconitine (MLA)
had no eect. Similarly, antisense knock-down of a4 subunit of nAChR, but not a7 subunit, in spinal cord induced thermal and
mechanical hyperalgesia. In whole-cell patch-clamp experiments in spinal cord slice preparation from adult mice, the frequency
of miniature inhibitory postsynaptic currents (mIPSCs) observed in substantia gelatinosa (SG) neurons was decreased by mecamylamine and DHbE, but not by MLA. The amplitudes of the mIPSCs were not aected. The nicotinic antagonists decreased the
frequency of both GABAergic and glycinergic IPSCs. On the other hand, the nicotinic antagonists had no eect on the excitatory
postsynaptic currents (EPSCs). Finally, acetylcholine-esterase inhibitor neostigmine-induced facilitation of IPSC frequencies in SG
neurons was inhibited by mecamylamine and DHbE. Altogether these ndings suggest that nicotinic cholinergic system in the spinal
dorsal horn can tonically inhibit nociceptive transmission through presynaptic facilitation of inhibitory neurotransmission in SG via
the a4b2 subtype of nAChR.
2006 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Spinal dorsal horn; Nociceptive transmission; Nicotinic receptors; Tonic inhibition; Whole cell patch clamp; Substantia gelatinosa;
Inhibitory postsynaptic currents

1. Introduction
In the spinal cord dorsal horn, which is a major site for
modulation of sensory information, cholinergic interneurons are present (Ribeiro-da-Silva and Cuello, 1990; Barber et al., 1984; Borges and Iversen, 1986; Olave et al.,
2002). Their cell bodies are found in lamina IIIV and
*

Corresponding author. Present address: Anesthesia Research Unit,

McGill University, McIntyre Medical Building Room 1207, 3655
Promenade Sir William Osler, Montreal, Quebec H3G 1Y6, Canada.
Tel.: +1 514 398 4565; fax: +1 514 398 8241.
E-mail address: mdharunor.rashid@mcgill.ca (M.H. Rashid).

form a plexus of axon terminals in the lamina III and inner

part of lamina II (Olave et al., 2002). Endogenous acetylcholine released by these interneurons might act as a
major neuromodulatory transmitter in the spinal dorsal
horn since receptors for acetycholine are present in lamina
II and III, where they are likely located on terminals of
primary aerents, spinal interneurons as well as on
descending monoaminergic terminals (Coggeshall and
Carlton, 1997; Baba et al., 1998; Cordero-Erausquin
and Changeux, 2001; Khan et al., 2003; Zhang et al.,
2005). Although the eects of exogenous cholinomimetic
drugs on nociceptive transmission in the spinal cord have

0304-3959/$32.00 2006 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.pain.2006.05.011

126

M.H. Rashid et al. / Pain 125 (2006) 125135

been extensively studied (Iwamoto and Marion, 1993;

Damaj et al., 1998; Khan et al., 1998), the role of endogenous acetylcholine on nociceptive transmission remains
largely unexplored. A tonic inhibitory role of the spinal
muscarinic cholinergic system on mechanical nociceptive
transmission had been reported earlier (Zhuo and Gebhart, 1991; Zhuo et al., 1993). However, such a tonic
inhibitory role of the spinal nicotinic cholinergic system
on pain transmission is still unclear. In a previous study,
we demonstrated that intrathecal injections of nicotinic
antagonist produce thermal hyperalgesia in normal mice,
suggesting the presence of a tonic inhibitory mechanism
through nicotinic receptors in the spinal cord (Rashid
and Ueda, 2002).
Nociceptive information from the periphery is subjected to inhibitory modulation in the spinal dorsal horn,
specically in the substantia gelatinosa (SG; lamina II of
Rexed) which is very rich in inhibitory GABAergic
and glycinergic interneurons (Cervero and Iggo, 1980;
Narikawa et al., 2000; Furue et al., 2004). Both GABAergic and glycinergic synaptic transmissions in the SG
are reported to be enhanced by nicotinic agonists
through dierent subtypes of nAChRs (Kiyosawa
et al., 2001; Takeda et al., 2003; Genzen and McGehee,
2005). By using single-cell RT-PCR, Changeux and colleagues (Cordero-Erausquin et al., 2004) reported that
the majority of inhibitory GABAergic and/or glycinergic
interneurons in the dorsal horn preferentially express
a4a6b2 subunits whereas excitatory or NK1-expressing
neurons mainly express a7a3b2 subunits. Nevertheless,
it has been proposed that endogenous acetylcholine
may tonically activate the nAChRs located on inhibitory
interneurons in the dorsal horn (Cordero-Erausquin and
Changeux, 2001). This, in conjunction with our previous
speculation for the presence of a tonic nAChR-mediated
inhibitory mechanism in the spinal cord in mice (Rashid
and Ueda, 2002), prompted us to further investigate the
matter. In the present study, intrathecal injections of
non-specic nAChR antagonist mecamylamine and
a4b2-selective nAChR antagonist dihydro-b-erythroidine drastically reduced nociceptive thermal and
mechanical withdrawal thresholds in mice. Moreover,
frequencies of miniature inhibitory postsynaptic currents
(mIPSCs) observed in SG neurons were decreased by
these antagonists. Our combined behavioral and electrophysiological data demonstrate the presence of a tonic
inhibitory mechanism through a4b2 subtype of nAChR
on nociceptive transmission in the spinal cord.
Materials and method
1.1. Experimental animals
Male ddY mice weighing 2530 g (68 weeks old) were used
in the present study. All procedures throughout the present
study were approved by the Nagasaki University Animal Care

Committee and the Kyushu University Guidelines for Animal

Experimentations, and adhered to the guidelines of the
Committee for Research and Ethical Issues of the International Association for the Study of Pain. In all circumstances,
maximum possible eorts were made to minimize animal sufferings and to reduce number of animals used in the
experiments.
1.2. Drugs
The following drugs were obtained from SigmaAldrich
(St. Louis, MO, USA): mecamylamine hydrochloride, dihydro-b-erythroidine hydrobromide (DHbE), methyllycaconitine
citrate (MLA), neostigmine bromide. All drugs were dissolved
in deionized water to make the stock solution. The nal dilution was made with saline for behavioral studies and with
Krebs solution for electrophysiological studies.
1.3. Intrathecal injection
Intrathecal injections were performed free-hand between L5
and L6 lumber space in unanesthetized mice using a 30-gauge
needle attached to a Hamilton microsyringe according to the
methods of Hylden and Wilcox (Hylden and Wilcox, 1980).
The accurate placement of the needle tip in the subarachnoid
space was veried by a quick icking of the mouses tail
immediately upon entry of the needle. The injection was given
slowly in a volume of 5 ll.
1.4. Nociceptive tests
Nociceptive tests were performed with either thermal or
mechanical stimulus. In thermal paw withdrawal test, the
latency to withdrawal evoked by exposing the right hind paw
to a thermal stimulus was measured. Unanesthetized animals
were placed under Plexiglas cages on top of a glass sheet and
adapted in the testing environment for about an hour. The
thermal stimulus (IITC, Woodland Hills, CA, USA) was then
positioned under the glass sheet to focus the projection bulb
exactly on the middle of plantar surface of the mice. Time to
withdraw the paw from the thermal stimulus was then automatically measured. In mechanical paw pressure test, mice
were placed under Plexiglas cages on top of a wire mesh grid,
adapted for 1 h, and mechanical stimulus was delivered on the
plantar surface of right hand paw using an automated Transducer Indicator (IITC, Woodland Hills, CA, USA), and the
withdrawal thresholds were measured. A cut-o time or pressure of 15 s or 15 g was used to minimize tissue damage.
1.5. Antisense oligonucleotides and Western blotting
The antisense oligodeoxynucleotides and its mismatch for
a4 (AS-ODN; 5 0 CCC CCG ATC TCC ATG GCT 3 0 ;
MS-ODN 5 0 GCC GCG TTC ACC TTG CCT 3 0 ) and a7
(AS-ODN; 5 0 GCC GCG CAT GTC GCC GGA 3 0 ; MSODN 5 0 CCC CCA GAT CTC CCC CGA 3 0 ) subunits of
the nicotinic acetylcholine receptor were synthesized, freshly
dissolved in physiological saline and injected in mice in a volume of 2 ll (10 lg) on 1st, 3rd and 5th day, and nociceptive
tests were performed on the 6th day. To conrm the downregulation of the target receptor by the AS-ODN, Western blot-

M.H. Rashid et al. / Pain 125 (2006) 125135

ting experiments in freshly isolated spinal cord samples of the

treated animals were performed using standard protocol. The
rabbit polyclonal antibodies raised against the a4 and a7 subunits of nAChR were used in immunoblotting studies (1:500;
Santa Cruz Biotechnology, CA, USA).
1.6. Electrophysiological studies
Blind whole-cell patch-clamp recordings were made from
substantia gelatinosa (SG) neurons in a transverse slice from
lumbar region of the spinal cord of mice. The method used
for obtaining the transverse slice preparation was as described
previously in the rat (Yoshimura and Jessell, 1989; Yoshimura
and Nishi, 1993; Miyakawa et al., 2005). Briey, mice were
anesthetized with urethane (1.5 g/kg, i.p.), and a thoraco-lumbar laminectomy was performed. A portion of the lumbar
L4L6 spinal cord was removed and submerged in ice-cold,
pre-oxygenated Krebs solution (in mM: NaCl 117, KCl 3.6,
CaCl2 2.5, MgCl2 1.2, NaH2PO4 1.2, NaHCO3 25, and D-glucose 11). After removal of dura mater, the dorsal and ventral
roots were cut and then pia-arachnoid membrane was
removed. A 500550 lm thick transverse slice was cut on a
vibratome microslicer. The slice was then placed in the recording chamber and perfused continuously at a rate of 1015 ml/
min with Krebs solution which was equilibrated with 95% O2
and 5% CO2 at 36 1 C. Blind whole-cell recordings were
then made in SG with patch pipettes (612 MX) lled with
internal solution containing (in mM): Cs2SO4 110, CaCl2 0.5,
MgCl2 2, TEA-Cl 5, EGTA 5, Hepes 5, and Mg-ATP 5. The
inhibitory postsynaptic currents were recorded at 0 mV and
amplied with an Axopatch 200B amplier (Axon Instruments, CA, USA). Signals were ltered at 5 kHz and digitized
with an A/D converter. The acquired data were analyzed with
a personal computer using pClamp version 8.1 (Axon Instruments) and Mini-analysis program version 6.0.3 (Synaptosoft,
Decatur, GA, USA). Drugs were applied by exchanging perfusion solution containing a known drug concentration without
altering the perfusion rate and temperature. We assumed the
eects of the drugs as decrease, increase or no eect if
changes compared with control were less than 80%, more
than 120%, and between 80 and 120%, respectively.
1.7. Statistical analysis
The data were analyzed by either Students t-test or oneway analysis of variance with suitable post hoc tests. The electrophysiological cumulative histogram data were analyzed by
KolmogorovSmirnov test. The criterion of signicance was
set at p < 0.05.

2. Results
2.1. Induction of thermal and mechanical hyperalgesia in
nave mice by intrathecal injection of non-specic nicotinic
antagonist mecamylamine
In a previous study with peripheral nerve injury
model (Rashid and Ueda, 2002), we observed that
intrathecal (i.t.) injections of nicotinic antagonist mecamylamine produced thermal hyperalgesia in control

127

sham-operated mice, suggesting the presence of a tonic

inhibitory mechanism through nicotinic receptors in
the spinal cord. In the present study also, we found that
i.t. injection of non-specic nicotinic receptor antagonist,
mecamylamine, dose-dependently (0.1 10 nmol, i.t.)
produced thermal and mechanical hyperalgesia in mice
(Fig. 1A and B). When we measured the area under the
time-course curves (AUC) in Fig. 1A and B, it was
observed that mecamylamine signicantly decreased
thermal and mechanical withdrawal thresholds in mice.
The induction of hyperalgesia was rapid and long-lasting. With i.t. 10 nmol of mecamylamine, the paw withdrawal latency or threshold was signicantly decreased
at 10 min after injection and the eects continued before
returning to baseline level at 90 min after the injection
(Fig. 1A and B).
2.2. Involvement of a4b2 subtype of nicotinic receptors for
the induction of hyperalgesia
To further identify the subtype of nAChR that
mediates the endogenous acetylcholine-mediated tonic
inhibition on nociceptive transmission in the spinal
cord, we used subtype-selective antagonists. We used
antagonists for a4b2 and a7 subtypes of nAChR,
the two major nAChR subtypes that had been reported to be involved in nicotinic agonists-mediated antinociception in the spinal cord in mice (Damaj et al.,
1998; Khan et al., 1998; Marubio et al., 1999). Similar
to mecamylamine, intrathecal injection of a4b2
nAChR antagonist dihydro-b-erythroidine (DHbE)
dose-dependently (110 nmol, i.t.) produced thermal
and mechanical hyperalgesia (Fig. 2A and B). However, i.t. injection of 10 nmol of a7 nAChR antagonist
methyllycaconitine (MLA) did not induce any thermal
or mechanical hyperalgesia (Fig. 2A and B). Plotting
the data as AUC also indicates that DHbE signicantly decreased thermal and mechanical withdrawal
thresholds while MLA had no signicant eects. Similarly, antisense knockdown of a4 subunit of nAChR
induced thermal and mechanical hyperalgesia
(Fig. 2C and D) while the antisense knockdown of
a7 subunit had no eect (data not shown). We further
examined whether antisense knockdown of a specic
nicotinic receptor subunit caused loss of function
through that specic subtype of nAChR in the spinal
cord in mice. For this purpose, we examined the
eects of i.t. a4b2-selective nAChR antagonist DHbE
on the a4 knockdown mice. As shown in Fig. 2E, i.t.
injection of 10 nmol of DHbE did not further decrease
the thermal latency in the AS-ODN-treated mice,
while DHbE decreased thermal latency in the salineor MS-ODN-treated mice (Fig. 2E), suggesting loss
of functional a4b2 nAChR by the AS-ODN. Similar
results were obtained with the paw pressure test (data
not shown).

128

M.H. Rashid et al. / Pain 125 (2006) 125135

Fig. 1. Induction of thermal and mechanical hyperalgesia in mice by intrathecal (i.t.) nicotinic antagonist. (A and B) Intrathecal injections of nonspecic nicotinic acetylcholine receptor antagonist, mecamylamine (Meca), dose-dependently induced thermal (A) and mechanical (B) hyperalgesia in
mice. The paw withdrawal latency (A) or thresholds (B) were drastically reduced by i.t. mecamylamine that persisted for more than an hour. The
right side bar graphs show the area under the curves (AUC) in (A and B), respectively. Each data point represents mean SEM from 6 to 8 mice. *
indicates statistically signicant dierence compared with the vehicle (Veh) saline injection at p < 0.05.

2.3. Reduction in mIPSCs frequency in dorsal horn SG

neurons by nicotinic antagonists
In whole-cell patch-clamp experiments, we recorded
the inhibitory postsynaptic currents (IPSCs) in substantia gelatinosa (SG) neurons in adult mice spinal cord
slice preparations, and examined the eects of nicotinic
antagonists thereon. The cells were held at 0 mV where
the IPSCs were observed as outward currents since
inward glutamatergic currents were all minimized at this
potential. As shown in Fig. 3A and B, the frequencies of
miniature inhibitory postsynaptic currents (mIPSCs)
observed in presence of TTX (1 lM) were reduced by
bath application of 10 lM of mecamylamine
(61.59 6.41% of control; n = 6). While mecamylamine
signicantly decreased the cumulative probability of
mIPSC frequencies, it had no eect on the mIPSC
amplitudes (Fig. 3C and D). Fig. 3E shows the time
course for the eects of mecamylamine on mIPSC frequency in a SG neuron. Out of total 62 SG neurons tested, application of 10 lM of mecamylamine decreased

the frequency of IPSCs in 20 neurons (control:

9.07 0.03 Hz vs mecamylamine: 5.37 0.23 Hz). The
IPSCs amplitudes were unaected by mecamylamine
(control: 29.4 1.1 pA vs mecamylamine: 26.7 0.7
A). In three neurons, mecamylamine increased the IPSC
frequency but did not aect the IPSC amplitude (control: 5.74 1.13 Hz vs mecamylamine: 9.24 0.98 Hz;
and control: 16.7 0.6 pA vs mecamylamine:
18.5 0.8 pA). The rest of the neurons did not respond
to mecamylamine (% change in IPSC frequency and
amplitudes was between 80% and 120% of control; data
not shown). On the other hand, mecamylamine (10 lM)
had no eects on the excitatory postsynaptic currents
(EPSCs) recorded at a holding potential of 70 mV
(Fig. 4A; n = 7).
Next, we examined the eects of subtype-selective
nAChR antagonists on the IPSCs in SG neurons that
were responsive to mecamylamine. Out of 15 mecamylamine-sensitive neurons, antagonist of the a4b2 subtype
of nAChR dihydro-b-erythroidine (DHbE; 10 lM)
decreased the mIPSC frequency in 11 neurons

M.H. Rashid et al. / Pain 125 (2006) 125135

129

Fig. 2. Involvement of a4b2 subtype of nicotinic receptor for the induction of hyperalgesia in mice. (A and B) Intrathecal injections of antagonist of
the a4b2 subtype of nicotinic receptor dihydro-b-erythroidine (DHbE), but not a7 subtype antagonist methyllycaconitine (MLA), induced thermal
(A) and mechanical (B) hyperalgesia in the mice. The right side bar graphs show the AUC from (A and B), respectively. * indicates statistically
signicant dierence compared with the vehicle (Veh) saline injection at p < 0.05. (C and D) Involvement of a4b2 subtype of nAChR for the
induction of such hyperalgesia was further conrmed by antisense blockade of expression of a4 subunit. Pretreatments with antisense
oligodeoxynucleotides for a4 subunit induced thermal and mechanical hyperalgesia in mice. Mice were treated with vehicle saline (Veh), missense
oligodeoxynucleotides (MS) and antisense oligodeoxynucleotides (AS) at day 1, day 3 and day 5, and paw withdrawal latency or thresholds were
measured at day 6. The upper panel shows Western blots for a4 subunit of nAChR in spinal cord tissues from the dierent treatment groups. (E) No
further decrease in thermal paw withdrawal latency by DHbE in a4 antisense oligodeoxynucleotides (AS)-treated mice. Each data point represents
mean SEM from 6 to 8 mice. *p < 0.05.

(Fig. 4E, control: 6.68 0.23 Hz vs DHbE:

3.22 0.8 Hz; n = 15). Moreover, DHbE decreased the
mIPSC frequency in a dose-dependent manner

(Fig. 4F). On the other hand, application of a7-selective

nAChR antagonist methyllycaconitine (MLA; 100 nM)
decreased the mIPSC frequency only in 1 cell out of

130

M.H. Rashid et al. / Pain 125 (2006) 125135

Fig. 3. Reversible decrease in miniature inhibitory postsynaptic currents (mIPSC) frequency in SG neurons by nicotinic antagonist. (A) In whole-cell
patch-clamp experiments in substantia gelatinosa (SG) of spinal dorsal horn of mice, the frequency of mIPSCs observed in presence of 1 lM of TTX
was reversibly decreased by bath application of 10 lM of non-specic nicotinic antagonist, mecamylamine. (B) Traces of mIPSCs on an extended
time-scale in absence or presence of mecamylamine. (C and D) Analyses of cumulative frequency and amplitude histograms indicate that
mecamylamine signicantly decreased the frequency of mIPSCs in SG neurons (C; KolmogorovSmirnov test, p < 0.05) while the eect on mIPSC
amplitudes was not statistically signicant (D; KolmogorovSmirnov test, p < 0.05). (E) Time-course for the eects of mecamylamine on mIPSC
frequency in the neuron in (A).

six mecamylamine-sensitive cells tested (Fig. 4E, control:

9.27 1.13 Hz vs MLA: 8.4 1.05 Hz; n = 6). Both
DHbE and MLA did not aect the amplitude of the
mIPSCs. Traces in Fig. 4BD showed a representative
neuron that was responsive to mecamylamine and
DHbE, but not to MLA.
2.4. Nicotinic antagonist decreased frequency of both
GABAergic and glycinergic IPSCs
Two types of IPSCs are usually observed in the SG
neurons according to their decay time. One type has a
shorter duration and is blocked by strychnine (2 lM),
and thus referred to as glycinergic IPSCs; the other type
has a relatively longer duration and is antagonized by
bicuculline (20 lM), and thus referred to as GABAergic
IPSCs (Yoshimura and Nishi, 1993). Out of 15 neurons
that were tested for the eects of DHbE, 8 had

GABAergic IPSCs, 5 had glycinergic IPSCs and 2 had

mixed IPSCs as determined by decay time and responsiveness to specic antagonists. As shown in Fig. 5A
and B, the a4b2-selective antagonist DHbE decreased
the frequency of both GABAergic (control: 7.08
1.18 Hz vs DHbE: 3.41 0.41 Hz; n = 7) and glycinergic IPSCs (control: 5.4 0.46 Hz vs DHbE: 2.73
0.03 Hz; n = 3). DHbE did not aect the amplitudes of
GABAergic (control: 14.4 1.3 pA vs DHbE: 12.9
1.2 pA) or glycinergic IPSCs (control: 32.4 3.5 pA vs
DHbE: 29.5 3.0 pA).
2.5. Nicotinic antagonists blocked the enhancement of
IPSC frequency induced by acetylcholinesterase inhibitor,
neostigmine
The acetylcholinesterase inhibitor neostigmine is
known to enhance the level of endogenous ACh by

M.H. Rashid et al. / Pain 125 (2006) 125135

131

Fig. 4. Nicotinic antagonists dose-dependently decreased IPSC frequencies and had no eect on EPSC frequencies. (A) The EPSC frequencies
observed at 70 mV were unaected by nicotinic antagonist mecamylamine. (BD) Bath application of non-specic nicotinic antagonist
mecamylamine (10 lM) and a4b2 nicotinic antagonist DHbE 10 lM, but not a7 antagonist MLA (100 nM) reversibly decreased the IPSC frequencies
in SG neurons. The traces in (AD) were taken from the same neuron. (E) Analyses of percent decrease in IPSC frequency in SG neurons by dierent
nicotinic antagonists. (F) Dose-dependent decrease in IPSC frequency by DHbE. The number of neurons is indicated above each column. *p < 0.05.

blocking its metabolism. Previous report demonstrates

that neostigmine can increase the frequency of IPSCs
in the spinal dorsal horn in the rat (Baba et al.,
1998). In the present report, to strengthen our proposition for the presence of an endogenous ACh-mediated tonic inhibitory mechanism through nAChR in
the spinal cord, we sought to know whether the neostigmine-induced increases in the IPSCs frequencies
in SG neurons are blocked by the nicotinic antagonists. As shown in Fig. 6A, bath application of
10 lM of neostigmine markedly increased the IPSC
frequency
in
the
SG
neurons
in
mice
(303.98 16.25% of control; n = 14). We observed
that pre-application of nAChR antagonists mecamylamine and DHbE signicantly inhibited the increase
in IPSC frequencies that was induced by neostigmine
(Fig. 6BE), further suggesting that the spinal nicotinic

cholinergic system had a tonic inhibitory role on dorsal horn synaptic transmission.
3. Discussion
Consistent with our previous observation (Rashid
and Ueda, 2002), in the present study, intrathecal (i.t.)
injection of the nicotinic antagonists induced thermal
and mechanical hyperalgesia in nave mice. In several
previous studies with the normal rat, i.t. nicotinic antagonists alone had no eects while they blocked the nociceptive or antinociceptive eects of the i.t. nicotinic
agonists (Khan et al., 1998; Rueter et al., 2000). These
dierences in the eects of i.t. nicotinic antagonists in
the normal mice and rat might be due to species dierences and/ or dierential expression of nicotinic receptor
subtypes or dierential tonic release of acetylcholine in

132

M.H. Rashid et al. / Pain 125 (2006) 125135

Fig. 5. Nicotinic antagonists decreased the frequency of both GABAergic and glycinergic IPSCs. (A) Frequencies of GABAergic IPSCs observed in
the presence of glycine receptor antagonist strychnine were decreased by a4b2 nicotinic antagonist, DHbE. (B) Frequencies of glycinergic IPSCs
observed in the presence of GABA-A receptor antagonist bicuculline were also decreased by DHbE. The lower panels show traces from indicated
parts in Fig. 5 A and B, respectively, on an extended time scale.

the spinal cord. Nevertheless, in our present experiments, i.t. injection of non-specic nicotinic antagonist
mecamylamine and a4b2-selective antagonist dihydrob-erythroidine (DHbE) caused drastic reduction in the
thermal and mechanical nociceptive thresholds in mice
while a7 antagonist methyllycaconitine (MLA) had no
eect. In electrophysiological experiments, the frequency
of miniature inhibitory postsynaptic currents (mIPSCs)
observed in SG neurons was decreased by mecamylamine and DHbE, but not by MLA. Mecamylamine
and DHbE did not aect the amplitude of the mIPSCs,
suggesting that the eects were mediated through blockade of presynaptic nicotinic receptors. Moreover, the
increases in IPSC frequencies induced by the cholinesterase inhibitor neostigmine were blocked by both mecamylamine and DHbE. All these results suggest that
cholinergic neurons in the spinal dorsal horn tonically
activate the inhibitory GABAergic and glycinergic interneurons to produce a constant and sustained endogenous inhibition on nociceptive transmission.
It is well known that nociceptive transmission in the
spinal dorsal horn is subjected to tonic modulation by
local interneurons and descending bers. Behavioral
and biochemical studies suggest that GABA and glycine released by the local GABAergic and glycinergic
neurons in the dorsal horn may tonically inhibit nociceptive transmission (Ishikawa et al., 2000; Cronin

et al., 2004). Electrophysiological studies also showed

that excitatory inputs in spinal dorsal horn are under
the control of inhibitory interneurons that may mediate
presynaptic inhibition through axo-axonic synapses
and postsynaptic inhibition through axo-dendritic or
axo-somatic synapses (Todd, 1990, 1996; Yoshimura
and Nishi, 1995). Our present results provide evidence
that this inhibitory system is further controlled by cholinergic inputs through spinal nicotinic receptors. Our
behavioral results of induction of hyperalgesia in mice
by i.t. injection of nicotinic antagonists clearly indicate
the presence of a tonic inhibitory mechanism through
the spinal nicotinic cholinergic system. A similar tonic
inhibitory mechanism through spinal muscarinic cholinergic system on mechanical nociceptive transmission
had already been reported in behavioral studies (Zhuo
and Gebhart, 1991; Zhuo et al., 1993; Baba et al.,
1998). Our electrophysiological data also strongly support our behavioral results of tonic nicotinic inhibition
of nociceptive transmission in the spinal cord. The fact
that nicotinic antagonists can decrease the frequency of
both GABAergic and glycinergic mIPSCs in SG indicates that endogenous acetylcholine spontaneously
released at the presynaptic terminals of inhibitory neurons can increase the release probability of GABA and
glycine to modulate nociceptive transmission. A
decrease in release probability of inhibitory neurotrans-

M.H. Rashid et al. / Pain 125 (2006) 125135

133

Fig. 6. Nicotinic antagonists inhibited the neostigmine-induced facilitation of IPSC frequencies in SG neurons. (A) Bath application of 10 lM of
acetylcholinesterase inhibitor neostigmine increased IPSC frequencies in SG neurons. (B and C) In presence of nicotinic receptor antagonists
mecamylamine and DHbE, neostigmine did not increase the IPSC frequencies. (D) Cumulative frequency histogram indicates signicant reduction in
neostigmine-induced increase in IPSC frequencies by mecamylamine and DHbE (KolmogorovSmirnov test, p < 0.05). (E) Analyses of percent
increase in IPSC frequencies by neostigmine (Neostg) and their blockade by nicotinic antagonists mecamylamine (Meca) and DHbE. The number of
neurons is indicated above each column. *p < 0.05.

mitters in the SG has been implicated in pathological

painful conditions like neuropathic pain (Moore
et al., 2002). Moreover, the phenomenon of central
sensitization might involve a reduced GABAergic or
glycinergic tone on dorsal horn neurons (Sivilotti and
Woolf, 1994). As already mentioned, in a previous
study we also speculated that a reduction in nicotinic
receptor-mediated tonic GABAergic inhibitory tone
on nociceptive transmission might cause neuropathic
pain in partial sciatic nerve injury model mice (Rashid
and Ueda, 2002). Our current results of the presence of
a nicotinic receptor-mediated tonic inhibitory tone on
nociceptive transmission further prove our previous
speculations.

Our behavioral and electrophysiological data also suggest that endogenous acetylcholine mediates the tonic
inhibition on spinal nociceptive transmission through
the a4b2 subtype of nicotinic receptors. Multiple subtypes
of nAChRs are expressed in the spinal dorsal horn (Wada
et al., 1989; Khan et al., 2003; Cordero-Erausquin et al.,
2004). Their exact location and functional role in spinal
synaptic transmission are largely unknown. Damaj et al.
(2000) reported that intrathecal a7 nicotinic agonist produces antinociception in acute pain models. However,
electrophysiological experiments show that a7 nAChRs
expressed on central terminals of primary aerents
enhance short and long-term glutamatergic transmission
in the spinal dorsal horn. Nevertheless, many previous

134

M.H. Rashid et al. / Pain 125 (2006) 125135

behavioral as well as gene knock-out experiments strongly suggest a role for mainly a4b2 subtype of nAChR in
spinal nociceptive inhibition (Khan et al., 1998; Marubio
et al., 1999; Rashid and Ueda, 2002). The a4b2 subtype of
nAChRs is also known to be involved in enhanced
GABAergic and glycinergic transmission in the dorsal
horn in newborn rat (Kiyosawa et al., 2001; CorderoErausquin et al., 2004; Genzen and McGehee, 2005).
Our results for the involvement of a4b2 subtype of
nAChR in GABAergic and glycinergic synaptic transmission in the SG in adult mice are consistent with these lines
of evidence. On the other hand, Takeda et al. (2003)
reported the involvement of a non-a4b2, non-a7 subtype
of nAChR for the enhancement of GABAergic transmission in SG in the adult rat. These dierences might be due
to use of dierent species of animals and dierent experimental conditions.
The reduction in mIPSCs frequency in SG neurons by
the nicotinic antagonist DHbE in our experiments suggests that a4b2 receptors located on the presyanptic terminals might be involved in tonic nicotinic inhibition of
spinal nociceptive transmission through enhanced
release of GABA or glycine. In many brain regions,
nAChRs expressed on presynaptic terminals are known
to enhance neurotransmitter release (McGehee et al.,
1995; Alkondon et al., 1997). In the spinal cord also, nicotinic agonists are known to increase the mIPSC frequency by acting on presynaptic a4b2 nAChRs
(Kiyosawa et al., 2001). Recently, Vincler and Eisenach
(2004) immunohistochemically labeled various subtypes
of nAChRs in the spinal cord in normal and neuropathy
state rat. The immunostaining for a4 subunit was
observed in lamina IIV of spinal dorsal horn where
they were located on round cells as well as on punctate
bers, suggesting their presence on soma and axon terminals in dorsal horn neurons (Vincler and Eisenach,
2004). This is consistent with previous electrophysiological data where nicotinic agonists mainly increased the
frequency of mIPSCs and in some cases induced postsynaptic inward current in some SG neurons in the rat
(Kiyosawa et al., 2001; Takeda et al., 2003). In the present study with adult mouse spinal cord slice preparation,
our data suggests that the endogenous ACh may induce
tonic release of GABA or glycine in the SG through the
presynaptic nicotinic receptors. However, a postsynaptic
component for the eects of acetylcholine in tonic nicotinic receptor-mediated inhibitory eects cannot be
excluded since previous studies with muscarinic cholinergic system suggest involvement of such mechanisms
(Baba et al., 1998). Moreover, the possibility of an indirect eect for the tonic nicotinic inhibition cannot be
excluded since cholinergic system is also known to activate other inhibitory mechanisms in the spinal cord such
as 5HT and nitric oxide system. It has been reported
that endogenous acetylcholine may tonically induce
release of serotonin by directly acting on the nicotinic

receptors located on spinal serotonergic projections

from the raphe magnus (Cordero-Erausquin and
Changeux, 2001). Similarly, endogenous acetylcholine
is known to induce nitric oxide (NO) synthesis through
both nicotinic and muscarinic receptors in the spinal
cord to produce inhibitory eects (Zhuo et al., 1993;
Xu et al., 2000). However, our results of blockade of
mIPSC frequency by the nicotinic antagonists as well
as other electrophysiological studies where nicotinic
agonists increased the mIPSC frequencies in presence
of TTX (Kiyosawa et al., 2001; Takeda et al., 2003;
Genzen and McGehee, 2005) suggest that nicotinic cholinergic system may also directly mediate tonic inhibition through presynaptic nAChRs in the spinal cord.
In conclusion, our behavioral and electrophysiological
data provide evidence that nociceptive transmission in the
spinal cord is under a tonic nicotinic cholinergic inhibition, and the a4b2 subtype of nicotinic receptors located
on the terminals of GABAergic and glycinergic inhibitory
interneurons in the SG may, at least in part, contribute to
this tonic nicotinic inhibitory mechanism. In continuation
of our previous ndings (Rashid and Ueda, 2002), we
speculate that a reduction in this tonic nicotinic receptor-mediated inhibitory tone might be one of the spinal
mechanisms for the induction of neuropathic pain.

Acknowledgements
Research described in this article was supported in
parts by funds from Philip Morris U.S.A. Inc. and Philip Morris International granted to H.U. Parts of this
study were also supported by grants from Japan Society
for the Promotion of Science (JSPS) to M.H.R. and
Grants-in-Aid from the Ministry of Education, Science,
Sports and Culture of the Govt. of Japan to M.Y. and
H.F., and the 21st Century Centre of Excellence
(COE) program to M.Y.

References
Alkondon M, Pereira EF, Barbosa CT, Albuquerque EX. Neuronal
nicotinic acetylcholine receptor activation modulates gamma-aminobutyric acid release from CA1 neurons of rat hippocampal slices.
J Pharmacol Exp Ther 1997;283:1396411.
Baba H, Kohno T, Okamoto M, Goldstein PA, Shimoji K, Yoshimura
M. Muscarinic facilitation of GABA release in substantia gelatinosa of the rat spinal dorsal horn. J Physiol 1998;508(Pt 1):8393.
Barber RP, Phelps PE, Houser CR, Crawford GD, Salvaterra PM,
Vaughn JE. The morphology and distribution of neurons containing choline acetyltransferase in the adult rat spinal cord: an
immunocytochemical study. J Comp Neurol 1984;229:32946.
Borges LF, Iversen SD. Topography of choline acetyltransferase
immunoreactive neurons and bres in the rat spinal cord. Brain
Res 1986;362:1408.
Cervero F, Iggo A. The substantia gelatinosa of the spinal cord: a
critical review. Brain 1980;103:71772.

M.H. Rashid et al. / Pain 125 (2006) 125135

Coggeshall RE, Carlton SM. Receptor localization in the mammalian
dorsal horn and primary aerent neurons. Brain Res Rev
1997;24:2866.
Cordero-Erausquin M, Changeux JP. Tonic nicotinic modulation of
serotoninergic transmission in the spinal cord. Proc Natl Acad Sci
USA 2001;98:28037.
Cordero-Erausquin M, Pons S, Faure P, Changeux JP. Nicotine
dierentially activates inhibitory and excitatory neurons in the
dorsal spinal cord. Pain 2004;109:30818.
Cronin JN, Bradbury EJ, Lidierth M. Laminar distribution of GABAAand glycine-receptor mediated tonic inhibition in the dorsal horn
of the rat lumbar spinal cord: eects of picrotoxin and strychnine
on expression of Fos-like immunoreactivity. Pain 2004;112:15663.
Damaj MI, Fei-Yin M, Dukat M, Glassco W, Glennon RA, Martin
BR. Antinociceptive responses to nicotinic acetylcholine receptor
ligands after systemic and intrathecal administration in mice. J
Pharmacol Exp Ther 1998;284:105865.
Damaj MI, Meyer EM, Martin BR. The antinociceptive eects of
alpha7 nicotinic agonists in an acute pain model. Neuropharmacology 2000;39:278591.
Furue H, Katafuchi T, Yoshimura M. Sensory processing and
functional reorganization of sensory transmission under pathological conditions in the spinal dorsal horn. Neurosci Res
2004;48:3618.
Genzen JR, McGehee DS. Nicotinic modulation of GABAergic synaptic
transmission in the spinal cord dorsal horn. Brain Res
2005;1031:22937.
Hylden JL, Wilcox GL. Intrathecal morphine in mice: a new
technique. Eur J Pharmacol 1980;67:3136.
Ishikawa T, Marsala M, Sakabe T, Yaksh TL. Characterization of
spinal amino acid release and touch-evoked allodynia produced by
spinal glycine or GABA(A) receptor antagonist. Neuroscience
2000;95:7816.
Iwamoto ET, Marion L. Characterization of the antinociception
produced by intrathecally administered muscarinic agonists in rats.
J Pharmacol Exp Ther 1993;266:32938.
Khan IM, Buerkle H, Taylor P, Yaksh TL. Nociceptive and
antinociceptive responses to intrathecally administered nicotinic
agonists. Neuropharmacology 1998;37:151525.
Khan I, Osaka H, Stanislaus S, Calvo RM, Deerinck T, Yaksh TL, et al.
Nicotinic acetylcholine receptor distribution in relation to spinal
neurotransmission pathways. J Comp Neurol 2003;467:4459.
Kiyosawa A, Katsurabayashi S, Akaike N, Pang ZP, Akaike N.
Nicotine facilitates glycine release in the rat spinal dorsal horn. J
Physiol 2001;536:10110.
Marubio LM, del Mar Arroyo-Jimenez M, Cordero-Erausquin M,
Lena C, Le Novere N, de Kerchove dExaerde A, et al. Reduced
antinociception in mice lacking neuronal nicotinic receptor subunits. Nature 1999;398:80510.
McGehee DS, Heath MJ, Gelber S, Devay P, Role LW. Nicotine
enhancement of fast excitatory synaptic transmission in CNS by
presynaptic receptors. Science 1995;269:16926.
Miyakawa A, Furue H, Katafuchi T, Jiang N, Yasaka T, Kato G,
et al. Action of neuropeptide Y on nociceptive transmission in
substantia gelatinosa of the adult rat spinal dorsal horn. Neuroscience 2005;134:595604.
Moore KA, Kohno T, Karchewski LA, Scholz J, Baba H, Woolf CJ.
Partial peripheral nerve injury promotes a selective loss of

135

GABAergic inhibition in the supercial dorsal horn of the spinal

cord. J Neurosci 2002;22:672431.
Narikawa K, Furue H, Kumamoto E, Yoshimura M. In vivo patchclamp analysis of IPSCs evoked in rat substantia gelatinosa
neurons by cutaneous mechanical stimulation. J Neurophysiol
2000;84:21714.
Olave MJ, Puri N, Kerr R, Maxwell DJ. Myelinated and unmyelinated
primary aerent axons form contacts with cholinergic interneurons
in the spinal dorsal horn. Exp Brain Res 2002;145:44856.
Rashid MH, Ueda H. Neuropathy-specic analgesic action of intrathecal nicotinic agonists and its spinal GABA-mediated mechanism. Brain Res 2002;953:5362.
Ribeiro-da-Silva A, Cuello AC. Choline acetyltransferaseimmunoreactive proles are presynaptic to primary sensory bres in the rat
supercial dorsal horn. J Comp Neurol 1990;295:37084.
Rueter LE, Meyer MD, Decker MW. Spinal mechanisms underlying
A-85380-induced eects on acute thermal pain. Brain Res
2000;872:93101.
Sivilotti L, Woolf CJ. The contribution of GABAA and glycine
receptors to central sensitization: disinhibition and touch-evoked
allodynia in the spinal cord. J Neurophysiol 1994;72:16979.
Takeda D, Nakatsuka T, Papke R, Gu JG. Modulation of inhibitory
synaptic activity by a non-alpha4beta2, non-alpha7 subtype of
nicotinic receptors in the substantia gelatinosa of adult rat spinal
cord. Pain 2003;101:1323.
Todd AJ. An electron microscope study of glycine-like immunoreactivity in laminae IIII of the spinal dorsal horn of the rat.
Neuroscience 1990;39:38794.
Todd AJ. GABA and glycine in synaptic glomeruli of the rat spinal
dorsal horn. Eur J Neurosci 1996;8:24928.
Vincler M, Eisenach JC. Plasticity of spinal nicotinic acetylcholine
receptors following spinal nerve ligation. Neurosci Res
2004;48:13945.
Wada E, Wada K, Boulter J, Deneris E, Heinemann S, Patrick J, et al.
Distribution of alpha 2, alpha 3, alpha 4, and beta 2 neuronal
nicotinic receptor subunit mRNAs in the central nervous system: a
hybridization histochemical study in the rat. J Comp Neurol
1989;284:31435.
Xu Z, Chen SR, Eisenach J, Pan HL. Role of spinal muscarinic and
nicotinic receptors in clonidine-induced nitric oxide release in a rat
model of neuropathic pain. Brain Res 2000;861:3908.
Yoshimura M, Jessell TM. Membrane properties of rat substantia
gelatinosa neurons in vitro. J Neurophysiol 1989;62:10918.
Yoshimura M, Nishi S. Blind patch-clamp recordings from substantia
gelatinosa neurons in adult rat spinal cord slices: pharmacological
properties of synaptic currents. Neuroscience 1993;53:51926.
Yoshimura M, Nishi S. Primary aerent-evoked glycine- and GABAmediated IPSPs in substantia gelatinosa neurones in the rat spinal
cord in vitro. J Physiol 1995;482:2938.
Zhang HM, Li DP, Chen SR, Pan HL. M2, M3, and M4 receptor
subtypes contribute to muscarinic potentiation of GABAergic
inputs to spinal dorsal horn neurons. J Pharmacol Exp Ther
2005;313:697704.
Zhuo M, Gebhart GF. Tonic cholinergic inhibition of spinal mechanical transmission. Pain 1991;46:21122.
Zhuo M, Meller ST, Gebhart GF. Endogenous nitric oxide is required
for tonic cholinergic inhibition of spinal mechanical transmission.
Pain 1993;54:718.