Imbalance between excitatory and inhibitory amino acids at spinal level is associated with maintenance of persistent pain-related behaviors
Although the postsynaptic events responsible for development of pathological pain have been intensively studied, the relative contribution of presynaptic neurotransmitters to the whole process remains less elucidated. In the present investigation, we sought to measure temporal changes in spinal release of both excitatory amino acids (EAAs, glutamate and aspartate) and inhibitory amino acids (IAAs, glycine, ?-aminobutyric acid and taurine) in response to peripheral inflammatory pain state. The results showed that following peripheral chemical insult induced by subcutaneous bee venom (BV) injection, there was an initial, parallel increase in spinal release of both EAAs and IAAs, however, the balance between them was gradually disrupted when pain persisted longer, with EAAs remaining at higher level but IAAs at a level below the baseline. Moreover, the EAAs-IAAs imbalance at the spinal level was dependent upon the ongoing activity from the peripheral injury site. Intrathecal blockade of ionotropic (NMDA and non-NMDA) and metabotropic (mGluRI, II, III) glutamate receptors, respectively, resulted in a differential inhibition of BV-induced different types of pain (persistent nociception vs. hyperalgesia, or thermal vs. mechanical hyperalgesia), implicating that spinal antagonism of any specific glutamate receptor subtype fails to block all types of pain-related behaviors. This result provides a new line of evidence emphasizing an importance of restoration of EAAs-IAAs balance at the spinal level to prevent persistence or chronicity of pain.
Pharmacological Research 59 (2009) 290–299
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/yphrs
Imbalance between excitatory and inhibitory amino acids at spinal level is
associated with maintenance of persistent pain-related behaviors
, Jun-Feng Hou
, Ming-Gang Liu
, Meng-Meng Li
, Xiu-Yu Cui
, Zhuo-Min Lu
, Yang-Yuan An
, Lin Shi
, Jun Chen
Institute for Biomedical Sciences of Pain, Capital Medical University, Beijing 100069, PR China
Institute for Biomedical Sciences of Pain and Institute for Functional Brain Disorders, Tangdu Hospital, Fourth Military Medical University, Xi’an 710038, PR China
Department of Anesthesiology, 304th Clinical Division of PLA General Hospital, Beijing 100037, PR China
Department of Neuroscience, Janssen Pharmaceutica, Johnson & Johnson, Turnhoutseweg 30, B-2340 Beerse, Belgium
Received 3 December 2008
Received in revised form 16 January 2009
Accepted 27 January 2009
Excitatory amino acids
Inhibitory amino acids
Although the postsynaptic events responsible for development of pathological pain have been intensively
studied, the relative contribution of presynaptic neurotransmitters to the whole process remains less
elucidated. In the present investigation, we sought to measure temporal changes in spinal release of
both excitatory amino acids (EAAs, glutamate and aspartate) and inhibitory amino acids (IAAs, glycine,
␥-aminobutyric acid and taurine) in response to peripheral inﬂammatory pain state. The results showed
that following peripheral chemical insult induced by subcutaneous bee venom (BV) injection, there was
an initial, parallel increase in spinal release of both EAAs and IAAs, however, the balance between them
was gradually disrupted when pain persisted longer, with EAAs remaining at higher level but IAAs at a
level below the baseline. Moreover, the EAAs–IAAs imbalance at the spinal level was dependent upon
the ongoing activity from the peripheral injury site. Intrathecal blockade of ionotropic (NMDA and non-
NMDA) and metabotropic (mGluRI, II, III) glutamate receptors, respectively, resulted in a differential
inhibition of BV-induced different types of pain (persistent nociception vs. hyperalgesia, or thermal vs.
mechanical hyperalgesia), implicating that spinal antagonism of any speciﬁc glutamate receptor subtype
fails to block all types of pain-related behaviors. This result provides a new line of evidence emphasizing
an importance of restoration of EAAs–IAAs balance at the spinal level to prevent persistence or chronicity
© 2009 Elsevier Ltd. All rights reserved.
There is substantial evidence indicating that the excitatory
amino acids (EAAs), namely glutamate (Glu) and aspartate (Asp),
are critically important for normal transmission of nociceptive
information from the periphery to the spinal cord [1,2]. They are
major neurotransmitters released from the central and peripheral
terminals of nociceptive primary afferent ﬁbers [3,4]. Numerous
studies have been associated with estimating spinal release of
EAAs in response to acute mechanical, chemical and electrical noci-
ceptive stimuli . Moreover, intradermal injection of formalin
[6–8] or capsaicin [9,10] could also produce certain amounts of
Corresponding author at: Institute for Biomedical Sciences of Pain and Institute
for Functional Brain Disorders, Tangdu Hospital, Fourth Military Medical University,
#1 Xinsi Road, Baqiao, Xi’an 710038, PR China. Tel.: +86 29 84777942/10 83911513;
fax: +86 29 84777945/10 83911491.
E-mail addresses: firstname.lastname@example.org, email@example.com (J. Chen).
These authors contributed equally to this work.
Asp and Glu release in the spinal cord, although these changes
may speciﬁcally vary in amplitude and duration with different pain
models . Inhibitory amino acids (IAAs), including glycine (Gly),
␥-aminobutyric acid (GABA) and taurine (Tau), are believed to work
as counterparts of EAAs at the spinal level and play very important
roles in keeping tonic inhibition of nociceptive input under physi-
ological state. However, whether EAAs–IAAs balance is changed or
not in response to peripheral persistent nociception remains less
clear. Thus, a parallel measurement of spinal release of both EAAs
and IAAs is essential for a better understanding of the roles of EAAs
and IAAs in development of pain persistence or chronicity as well.
In the past several decades, pharmacological blockade of
ionotropic glutamate receptors (iGluRs) activation at the spinal
level has been demonstrated to be effective in inhibition of
both thermal and mechanical hyperalgesia [12–17] as well as the
capsaicin- or formalin-induced spontaneous nociception [18,19].
Moreover, the antagonism against metabotropic glutamate recep-
tors (mGluRs) at the spinal level was also demonstrated to be
effective in anti-nociception [1,20–24]. However, discrepancies
were frequently observed across uses of different animal models
1043-6618/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299 291
and nociceptive stimulus modalities (electrical, thermal, mechan-
ical and chemical) . Therefore, effects of spinal antagonism at
speciﬁc glutamate receptor subtypes (iGluRs vs. mGluRs) are also
required to be re-evaluated by the use of an animal model display-
ing multiple “phenotypes” of pain.
The bee venom (BV) test, a well-established experimental
animal model mimicking honeybee sting-induced natural tissue
injury, is produced by subcutaneous (s.c.) injection of a given dose
of honeybee venom into one hind paw of rats [25–28]. This model
of clinically relevant pathological pain is behaviorally characterized
by persistent spontaneous nociception (PSN) as well as prolonged
thermal and mechanical hyperalgesia (primary or secondary)
related to peripheral inﬂammation [12,25–27,29,30]. Previous elec-
trophysiological studies have shown that BV-elicited PSN and
hypersensitivity are mediated by a long-lasting increase in spon-
taneous discharges and subsequent enhanced heat/mechanical
responsiveness of wide dynamic range neurons in the spinal dorsal
horn [31–35]. All of these BV-induced behavioral and electrophysio-
logical changes have been proved to be peripheral-dependent, since
either peripheral sciatic nerve blockade or local administration of
N-methyl-D-aspartate (NMDA) receptor antagonists or destruction
of capsaicin-sensitive primary afferent ﬁbers could effectively elim-
inate them [27,31,32,35,36]. In addition, a wealth of evidence has
been accumulated supporting the viewpoint that BV-evoked dif-
ferent “phenotypes” of pain might be mediated by different spinal
signaling pathways [29,30,37–40]. Thus, the BV test is the most
appropriate model for examining the roles of spinal amino acids
and their receptors in mediating different “phenotypes” of pain.
The present study was designed to assess the time course and
extent of spinal EAAs (Glu and Asp) release after s.c. BV injection
as well as its peripheral-dependence. Changes in IAAs (Gly, GABA
and Tau) and other amino acids or metabolites, including threo-
nine (Thr), arginine (Arg), alanine (Ala) and glutamine (Gln), at
spinal level, were also quantiﬁed. Besides, we further examined
putative roles of central iGluRs (NMDA and non-NMDA receptors)
and mGluRs in induction and/or maintenance of BV-induced PSN,
primary thermal and mechanical hyperalgesia.
2. Materials and methods
The experiments were performed on male Sprague–Dawley
albino rats (provided by Experimental Animal Center of the Cap-
ital Medical University, CCMU) weighing from 200 to 250 g. All
experiments were carried out with the approval of the Institutional
Animal Care and Use Committee at the CCMU. The animals were
maintained and cared for in accordance with the guidelines set forth
by the International Association for the Study of Pain . All rats
were housed and maintained in plastic boxes on a 12 h light/dark
cycle at 24–26
C with food and water available ad libitum. The rats
were acclimatized to the laboratory and habituated to the test boxes
for at least 30 min each day for 5 days before testing. All efforts were
made to minimize animal suffering and to reduce the number of
2.2. Experimental groups
To examine whether BV-induced release of spinal amino acids
and PSN were dependent upon the ongoing primary afferent inputs
from the periphery, rats were randomly divided into four groups:
(1) Saline-treated group: rats with s.c. injection of 50 l 0.9% sterile,
isotonic saline solution into the hind paw; (2) BV-inﬂamed group:
rats with s.c. injection of 50 l solution of whole BV (200 g BV dis-
solved in 0.9% sterile saline, Apis mellifera, Sigma); (3) Bup
group: rats with s.c. injection of 200 l bupivacaine (0.75%, 10 min
before BV treatment) into the same hind paw (identical to the BV
injection site); (4) Bup
+ BV group: rats with s.c. injection of
200 l bupivacaine (0.75%, 10 min before BV treatment) into the
corresponding site of the contralateral hind paw (symmetrical to
the BV injection site). Notably, the Bup
+ BV group of rats was
only used in behavioral assays of PSN in the present study.
In experiments regarding pharmacological examination of roles
of Glu receptor types in BV-induced nociception and hyperalge-
sia, two major groups of animals were assigned: (1) pre-treatment
group: CNQX (6-cyano-7- nitroquinoxaline-2,3-dione, a competi-
tive non-NMDA receptor antagonist, dissolved in dimethyl sulfoxide
(DMSO) at 3.5 nmol/10 l, 35 nmol/10 l, or 100 nmol/10 l, Sigma)
or MK-801 (a non-competitive NMDA receptor channel blocker,
dissolved in 0.9% sterile saline at 1.2 nmol/10 l or 24 nmol/10 l,
Sigma) or vehicle (DMSO or saline) was intrathecally adminis-
tered 5 min prior to s.c. BV injection for studying the induction
of PSN; (2) post-treatment group: intrathecal (i.t.) administration
of 10 l CNQX (100 nmol) or 10 l MK-801 (24 nmol) or vehicle
(DMSO or saline) was performed at either 5 min or 2 h after s.c.
BV injection for testing maintenance of PSN or primary hyperalge-
sia, respectively. In a separate set of experiments, the same volume
of AIDA (1-aminoindan-1,5-dicarboxylic acid, a group I mGluR
antagonist, dissolved in DMSO at 10 nmol/10 l, Sigma), EGLU ((S)-
a-ethylglutamic acid, a group II mGluR antagonist, dissolved in
DMSO at 20 nmol/10 l, Sigma) or MSOP ((RS)-␣-methylserine-
O-phosphate, a group III mGluR antagonist, dissolved in DMSO
at 10 nmol/10 l, Sigma) was intrathecally applie d at 2 h after s.c
BV injection to explore their possible roles in mediating thermal
and mechanical hypersensitivity. The doses of all drugs used in
the present study were determined according to our preliminary
data. It should be noted that only the post-treatment paradigm
was adopted in behavioral assays of primary hyperalgesia in the
current study. It is also important to mention here that our prelimi-
nary experiments found no signiﬁcant effects of each drug on basal
thermal latency or mechanical threshold under normal conditions
(without BV-initiated inﬂammation), so the single drug group was
not incorporated in the present experimental design.
For the in vivo i.t. microdialysis of spinal amino acid release
in response to various experimental manipulations, a spinal triple
lumen catheter with a single loop (Marsil Scientiﬁc, San Diego, USA)
was chronically implanted into the subarachnoid space of each rat
as described previously [7,42]. In brief, rats were ﬁrst anesthetized
with 7% trichloroacetaldehyde (1.2–1.5 g/kg, i.p.), and then placed
properly in a stereotaxic apparatus. After shaving off the air on the
top of the head, a 1-cm long midline incision was made across the
occipital bone edge, and cervical muscles were split by using a blunt
forceps, so the atlanto-occipital membrane was fully exposed. Sub-
sequently, a triple dialysis catheter was carefully implanted into
the subarachnoid space through an incision at the atlanto-occipital
membrane, with the loop membrane of the catheter eventually set-
tled at the rostral margin of the lumbar enlargement. Af ter closing
the incision, sterile normal saline was ﬂushed through the catheter,
and then the three free ends of the microdialysis catheter were
externalized through the skin at the top of the head and squeezed
tightly. At last, 5 ml of normal saline was subcutaneously injected
into the abdominal area of each rat to replenish the lost body ﬂuid.
Following recovery from anaesthesia, the rat was returned to its
cage and monitored for 5 days before further experiments were
performed. All those animals showing motor weakness or signs
of paralysis or sensory impairment during the observation period
For i.t. administration of drugs in pharmacological experiments,
chronic i.t. catheterization was performed according to our modi-
292 L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299
ﬁed methodology [37,38,40]. Brieﬂy, a 2-cm-long skin incision was
made and the muscles were separated from the C7 to T4 verte-
brae under ketamine anesthesia (50 mg/kg, i.p.). A laminectomy
was then performed at T2 or T3 level and the dura was opened. To
prevent the inset tubing from moving, a polyethylene (PE)-8 tube
(0.2 mm i.d., 0.5 mm o.d.) was passed through a 1-cm-long muscle
tunnel and then advanced caudally (3–5 cm distance between the
entry and the target level) through the subarachnoid space to the
rostral side of the lumbar enlargement (namely, the caudal tip of
the catheter ended between spinal levels L3 and L4). The outer end
of the PE-8 tubing was ﬁrmly ﬁxed to the paravertebral muscles.
The wound was washed with sterile saline, treated with antibiotics
and the muscles and skin were sutured by layers. The whole opera-
tion was performed in strictly sterile conditions. Rats showing any
neurological deﬁcits postoperatively were sacriﬁced. After testing,
placement of the inner end of the tube was veriﬁed, and animals
with the tube in the wrong place or with local pathological changes
were also excluded from the ﬁnal analysis.
2.4.1. Perfusion of catheters and sample collection
Spinal microdialysis was carried out in freely moving rats fol-
lowing 5 days of recovery from surgery . First, the rat was put
into a round Lucite cubicle (CMA/120, CMA/Microdialysis, Sweden,
designed to limit the mobility of the animal) for 15 min before the
probes were connected. To initiate dialysis, one of the external-
ized PE connections was attached to a 30 cm length of PE tubing
(inﬂow) and the other arm to a 25 cm length of PE tubing (outﬂow).
Then the dialysis tubing, connected to a microdialysis pump (CMA
100, CMA/Microdialysis, Sweden) with a 2.5 ml plastic syringe, was
perfused with modiﬁed Ringer’s solution (NaCl 147 mM, KCl 4 mM,
2.3 mM) at a ﬂow rate of 5 l/min. After a 2 h wash-out
period to establish a diffusion equilibrium, 3 baseline dialysates
were collected sequentially by a microfraction collector (CMA/142,
CMA/Microdialysis, Sweden) at 20-min intervals and then the fol-
lowing experiments were performe d separately:
1. To investigate possible changes in spinal amino acid release after
inﬂammatory pain, rats were subjected to s.c. injection of 50 l
BV solution after establishing the baseline. Then, samples were
collected every 20 min for 2 h.
2. As a control group, rats were treated with s.c. injection of equal
volume of normal saline in the hind paw after baseline testing.
Then samples were collected as described above.
3. To examine whether amino acid release was dependent upon the
ongoing primary afferent input from the site of peripheral injury,
rats were injected with 0.75% bupivacaine 10 min prior to s.c. BV
treatment in the same hind paw. Dialysates were then collected
at the same time points as mentioned above.
After completion of the microdialysis experiment, all rats were
euthanized with an overdose of trichloroacetaldehyde and anatom-
ical observation was performed by an observer unaware of the rats’
treatments to check the location of the microdialysis membrane.
The majority of the dialysis catheter sites were in the dorsal horn of
L5–L6 spinal segments (lumbosacral enlargement). If it was not in
the proper location, the rat was excluded from the further analysis.
2.5. Amino acids analysis of dialysates
Dialysate samples were collected on ice and immediately ana-
lyzed for amino acid concentrations by using the high performance
liquid chromatography (HPLC) method [43,44].
2.5.1. Reagents and equipments
Standard amino acid reagents (including Asp, Glu, Gly, Tau, Ala,
Gln, Thr, and Arg) and phthaldialdehyde (derivatization reagent)
were purchased from Sigma Chemical Co. (USA). Methanol (chro-
matographic grades) and tetrahydrofuran (THF, chromatographic
grades) were purchased from Fisher Scientiﬁc Co. (USA). Sodium
acetate anhydrous (NaAc) was obtained from Beijing Chemical
Industry (Beijing, P.R. China).
HPLC was performed by using an Aglient 1050 series chromato-
graphic system which consisted of a 1050 series Quaternary pump,
a 1050 series programmable ﬂuorescence detector, a 1050 series
auto sample injector, a 1100 series online degasser, an analysis col-
umn temperature controlling chamber and a Aglient Pheonix DOS
chromatographic workstation (Agilent, USA).
2.5.2. Mobile phase
Mobile phase “A”: Mobile phase “A” consisted of 410 ml 0.5 M
NaAc, 85 ml methanol, and 5 ml THF. This solution was adjusted to
pH 6.8 with acetic acid (10 N) and ﬁltered through 0.4-m pore
Mobile phase “B”: Mobile phase “B” consisted of 110 ml 0.5 M
NaAc, 385 ml methanol, and 5 ml THF. This solution was also
adjusted to pH 6.8 and ﬁltered as above.
First, a mixture of 95% A: 5% B was used at a ﬂow rate of
0.8 ml/min. Then the gradient was gradually changed to 10 0% B
for about 3 min. At 35 min the initial conditions were returned and
maintained for at least 15 min to stabilize the column for the next
A Waters C18 column (3.9 mm × 150 mm, 5 m) was set at a con-
stant temperature of 40
C. A 20-l portion of sample (dialysate
or standard) was taken up and 20 l phthaldialdehyde was then
added. Pre-column derivatization was completed by the auto sam-
ple injector. Excitation and emission wavelengths were selected at
232 and 440 nm, respectively. The order of elution of the amino
acids was Asp, Glu, Gln, Gly, Thr, Arg, Tau, and Ala in a typical HPLC
trace obtained in these experiments.
2.6. Assessment of spontaneous pain-related behaviors
As previously described [26,27], a 30 cm × 30 cm × 30 cm trans-
parent plexiglas test box with a transparent glass ﬂoor was placed
on a supporting frame of 30 cm high above the experimental table to
allow the experimenters to observe the paws of the animals without
obstruction. The rat was placed in the test box for at least 30 min
before administration of any chemical agents. After the acclima-
tion period, s.c. injection of BV or saline was made into the center
of the plantar surface of one hind paw with slight restraint. The rat
was then returned to the test box, and pain-related spontaneous
behavioral responses were determined by counting the number of
paw ﬂinches occurring during 5 min intervals for 1 h (pharmaco-
logical experiments) or 2 h (peripheral blockade studies) following
2.7. Behavioral assays of pain sensitivity
2.7.1. Quantitative measurement of thermal pain sensitivity
Thermal pain sensitivity of rats was determined by testing
paw withdrawal thermal latency (PWTL, s) in response to heat
stimuli applied onto the injection site of the inﬂamed hindpaw.
As described previously [26,27], the rat was placed on the sur-
face of a 2 mm thick glass plate covered with a plastic chamber
(20 cm × 20 cm × 25 cm) to measure its sensitivity to heat stim-
uli with a TC-1 radiant heat stimulator (new generation of RTY-3
made in Xi’an Bobang Technologies of Chemical Industry Co. Ltd.,
L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299 293
P.R. China). The radiant heat source was a high intensity halo-
gen lamp bulb (150 W) positioned under the glass ﬂoor directly
beneath the target area on the hind paw. The rate of heating
C per 1 s from pre-set baseline temperature (about 35
human body temperature). The distance between the projector
lamp bulb and lower surface of the glass ﬂoor was adjusted to pro-
duce a light spot on the ﬂoor surface with 5 mm diameter. The
heat stimulus was applied to the same site for 5 times and the
latter three values were averaged as the mean PWTL. The inter-
stimulus interval was between 10 and 15 min. The thermal latency
was determined as the duration from the onset of heat stimulus to
the occurrence of hind paw withdrawal reﬂex. The stimulus was
stopped if the latency exceeded 30 s so as to avoid excessive tissue
injury and the region was considered to be completely unrespon-
2.7.2. Quantitative measurement of mechanical pain sensitivity
Mechanical pain sensitivity of rats was determined by testing
paw withdrawal mechanical threshold (PWMT, mN) in response
to mechanical stimuli applied in ipsilateral hindpaw produced by
ascending graded individual von Frey monoﬁlaments with bend-
ing forces of 21.07, 27.44, 34.30, 44.10, 53.90, 76.44, 107.80, 147.00,
196.00, 245.00, 294.00, 392.00, 490.00, and 588.00 mN. The rat
was placed on a metal mesh ﬂoor covered with the same plastic
chamber and von Frey ﬁlaments were applied from underneath
the metal mesh ﬂoor to the testing site of the target hind paw. A
von Frey ﬁlament was applied 10 times (several seconds for each
stimulus) to each testing area. The bending force of the von Frey ﬁl-
ament being able to evoke an approximate 50% occurrence of paw
withdrawal reﬂex was expressed as the PWMT. The stimulus was
stopped if the threshold exceeded 588.00 mN (cutoff value). Notice-
ably, it was not really the case that all of the 14 forces were applied
to the same testing site in the experiment. Contrarily, we tended
to select a range of von Frey ﬁlament forces (245.00–490.00 mN)
as the priority over other forces according to our previous experi-
ences in performing such behavioral testing. If the ﬁrst ﬁlament
used could elicit an appropriate 50% occurrence of paw with-
drawal reﬂex across ten times of stimulation, then its force was
considered as the PWMT value at that time point. If the selected
ﬁlament force could not, the next higher force was tested. Simi-
larly, if the ﬁlament elicited a more than 50% occurrence of paw
withdrawal, the next lower force was used. The whole process
continued until the optimal intensity of the ﬁlament was found.
What we did in the testing avoided excessive poking of the hind-
2.8. Experimental protocol
For evaluating the effects of MK-801 and CNQX on initiation
and maintenance of PSN, the two drugs (for doses, see above) were
intrathecally administered 5 min prior to or 5 min after s.c. BV injec-
tion and the testing of PSN began immediately after BV or drug
treatment for pre- and post-administration paradigm, respectively.
For examining effects of ionotropic and metabotropic receptor
antagonists on established thermal and mechanical hyperalgesia,
all those compounds were injected at 2 h after s.c. BV injection and
testing of thermal or mechanical hypersensitivity was performed
between 2 and 4 h after s.c. BV injection. In addition, for each group
of animals, the baseline values of PWTL and PWMT were measured
prior to any treatment. The mechanical and thermal testing were
done in the same group of animals in the present study. In general,
the experimenting orders of these two measurements were ran-
domly arranged to rule out the effects of expectation of rats. It is
noteworthy that all behavioral testing procedures were performed
blind to the treatment of the animals.
2.9.1. Microdialysis data
Concentrations of amino acids were calculated base d on the
normalized peak areas with the external standards. Otherwise
stated, all sample values were expressed as a percentage of
baseline concentration (mean ± S.E.M.). Normalized data in one
group across time and at each time point in different groups
were analyzed by one way analysis of variance (ANOVA) fol-
lowed by individual post hoc multiple comparisons (Fisher’s
PLSD test). A statistical difference was accepted as signiﬁcant at
P < 0.05.
2.9.2. Behavioral data
All data were expressed as mean ± S.E.M. Two-way repeated
measure ANOVA was used to analyze group differences in mean
time courses of PSN. One-way ANOVA (post hoc Fisher’s PLSD test)
was applied to compare differences in averaged mean number of
ﬂinching reﬂex per 5 min as well as the thermal or mechanical
hyperalgesia measurements. A level of P < 0.05 was accepted as
3.1. Peripheral-dependence of BV-induced spontaneous
Consistent with our previous reports [26,27], s.c. injection of
whole BV, but not normal saline solution, produced marked spon-
taneous pain-related behaviors characterized by increased paw
ﬂinch responses. The number of ﬂinches peaked at 10 min and
returned back to baseline levels before the 1 h time point (Fig. 1,BV-
inﬂamed). Local pre-administration of an anesthetic agent (0.75%
bupivacaine, 10 min prior to BV injection) resulted in an almost
complete blockade of PSN (Fig. 1, Bup
+ BV). To exclude a possi-
ble systemic effect, bupivacaine was also subcutaneously injected
into a region on the contralateral hindpaw symmetrical to the BV
injection site. It was found that contralateral bupivacaine applica-
tion did not signiﬁcantly affect the development of BV-evoked PSN
(Fig. 1, Bup
Fig. 1. Mean time courses of the spontaneous paw ﬂinching responses to subcuta-
neous bee venom injection. The mean number of ﬂinchescounted at each 5 min block
of a 2 h observation period is shown. The rats were treated with subcutaneous injec-
tion of saline, bee venom, Bup
+ BV (local administration of 0.75% bupivacaine
10 min prior to BV injection in the ipsilateral paw) and Bup
+ BV group (local
administration of 0.75% bupivacaine 10 min prior to BV injection in the contralateral
paw). Vertical bars: ±S.E.M. (n = 6 for each group).
294 L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299
3.2. Activity-dependent release of spinal amino acids following
BV-evoked persistent nociception
3.2.1. Changes in extracellular concentrations of Glu, Asp and Gln
After collecting the baseline for 60 min, s.c. injection of BV into
the rat hind paw elicited an apparent increase in spinal release of
Glu (Fig. 2A) and Asp (Fig. 2B), whereas no appreciable changes
were observed in the saline control group. It appeared clear that
both Glu and Asp concentrations began to rise immediately after
BV injection, with a peak attained at about 20 min (174% and 204%
of baseline, n = 6 for each group, P < 0.05 and P < 0.001 vs. saline con-
trol for Glu and Asp respectively), and then declined gradually at
40 min. Increased responses recovered to the baseline at 60 min
after s.c. BV treatment (P > 0.05 vs. saline control for both Glu and
Asp). Peripheral pre-administration of bupivacaine (0.75%, 10 min
prior to BV injection) completely inhibited BV-evoked increase in
release of Glu and Asp, even the levels of these two EAAs being
much lower than those of baseline (Fig. 2A and B).
The extracellular amount of Gln, a metabolic product of Glu
[45–47], also peaked at 20 min after BV injection (208% of baseline,
n =8,P < 0.05 vs. saline control), and returned to baseline at 40 min.
Interestingly, its level continually decreased and became even lower
than baseline at 60 min (48% of baseline, P < 0.05 vs. saline control).
This decrease of Gln release continued until 120 min after BV injec-
tion. The temporal proﬁle of Gln release in Bup
+ BV group was
almost the same as that of Glu and Asp (Fig. 2C).
Fig. 2. Temporal changes in extracellular concentrations of aspartate, glutamate and
glutamine at the spinal level. The rats were treated with subcutaneous injection of
saline (n = 6 for each), bee venom (n = 6–8 for each) and Bup
+ BV (local adminis-
tration of 0.75% bupivacaine 10 min prior to BV injection in the ipsilateral paw, n =4
for each). Big reverse arrows indicate the time of saline or BV injection, while small
reverse arrows denote the time of bupivacaine administration. Vertical bars: ±S.E.M.
P < 0.05,
P < 0.01,
P < 0.001 vs. BV-inﬂamed group;
P < 0.05,
P < 0.01,
P < 0.001
vs. saline-treated group.
3.2.2. Changes in extracellular concentrations of Gly, Tau and
BV-induced persistent pain elicited a signiﬁcant increase in the
amount of extracellular Gly (Fig. 3A) and Tau (Fig. 3B), with a maxi-
mum level reaching at 20 min (136% and 270% of baseline, n = 7 and
8, P < 0.05 and P < 0.01vs. saline control for Gly and Tau respectively).
Then increased concentrations of the two IAAs went back to the
baseline at 40 min after BV injection (Fig. 3A and B). Subsequently,
the concentration of Tau remained constant (P > 0.05 vs. saline con-
trol, Fig. 3B), while the level of Gly further decreased and became
lower than baseline at 60 min (62% of baseline, P < 0.05 vs. saline
control). Peripheral bupivacaine pre-treatment fully abolishe d BV-
evoked release of Gly and Tau (Fig. 3A and B). While the baseline
level of Gly release was also reduced by bupivacaine-induced local
anesthesia, that of Tau remained less inﬂuenced.
Here, it is noteworthy that ␥-aminobutyric acid (GABA), another
principal inhibitory neurotransmitter involved in nociceptive mod-
ulation , was below detection limits under baseline conditions.
However, from 20 min towards 60 min after s.c. BV injection, a cer-
tain amount of extracellular GABA became detectable in some cases.
Because of the so small amount of GABA release, we quantiﬁed it
by measuring the peak area under the HPLC curve instead of con-
verting it to percentage of baseline. At 20 min after BV injection,
the extracellular GABA content raised from undetectable level to
0.9829 ± 0.4227 Norm. Then, the GABA release decreased, with the
peak area being 0.6100 ± 0.2672 Norm and 0.5100 ± 0.2261 Norm at
40 min and 60 min, respectively. After that, the GABA level became
undetectable again until the end of the detection period (2 h after
3.2.3. Changes in extracellular concentrations of Ala, Arg and Thr
The extracellular concentrations of Ala (Fig. 4A), Arg (Fig. 4B) and
Thr (Fig. 4C) increased to their maximum at 20 min after adminis-
tration of BV (153%, 127% and 157% of baseline, n = 7–8, P < 0.05 vs.
saline control for Ala, Arg and Thr respectively). Their concentra-
Fig. 3. Temporal changes in extracellular concentrations of glycine and taurine at
the spinal level. The rats were treated with subcutaneous injection of saline ( n =6for
each), bee venom (n = 7–8 for each) and Bup
+ BV (local administration of 0.75%
bupivacaine 10 min prior to BV injection in the ipsilateral paw, n = 4 for each). Big
reverse arrows indicate the time of saline or BV injection, while small reverse arrows
denote the time of bupivacaine administration. Vertical bars: ±S.E.M.
P < 0.05 vs.
P < 0.05,
P < 0.01,
P < 0.001 vs. saline-treated group.
L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299 295
Fig. 4. Temporal changes in extracellular concentrations of alanine, arginine and
threonine at the spinal level. The rats were treated with subcutaneous injection of
saline (n = 6 for each), bee venom (n = 7–8 for each) and Bup
+ BV (local adminis-
tration of 0.75% bupivacaine 10 min prior to BV injection in the ipsilateral paw, n =4
for each). Big reverse arrows indicate the time of saline or BV injection, while small
reverse arrows denote the time of bupivacaine administration. Vertical bars: ±S.E.M.
P < 0.05 vs. BV-inﬂamed group;
P < 0.05,
P < 0.01,
P < 0.001 vs. saline-treated
tions consistently returned to the basal levels at 40 min. Similar to
Gln and Gly, the levels of these amino acids decreased gradually
after recovering to baseline and became signiﬁcantly lower than
the basal levels at 60 min and other following time points (59%,
63% and 55% of baseline, P < 0.05 at 100 min, P < 0.05 at 60 min,
P < 0.05 at 100 min vs. saline control for Ala, Arg and Thr respec-
tively). Local anesthesia produced by bupivacaine pre-treatment
dramatically decreased the level of Ala released form the spinal
cord in response to BV-related persistent pain stimulation, even
becoming still lower than the baseline at later stages (Fig. 4A).
3.3. Effects of i.t. CNQX and MK801 on BV-induced spontaneous
nociception and hyperalgesia
3.3.1. Prevention and reversal of PSN by spinal blockade of NMDA
and non-NMDA receptors activation
In comparison with vehicle control, i.t pre-treatment with
CNQX (3.5 nmol, 35 nmol and 100 nmol) and MK-801 (1.2 nmol and
24 nmol) resulted in a dose-dependent suppression of BV-induced
persistent paw ﬂinching reﬂex over the 1 h time course of obser-
vation (Fig. 5A and B). The averaged mean number of ﬂinching
reﬂex per 5 min was shown in the left panel of Fig. 5D and E for
i.t. pre-treatment with CNQX and MK-801, respectively. The three
doses of CNQX produced 19.42% (3.5 nmol, n =8, P > 0.05), 37.27%
(35 nmol, n =8,P < 0.05) and 57.32% (100 nmol, n =8,P < 0.001) inhi-
bition of PSN, while the values of MK-801 were 32.42% (1.2 nmol,
n =8,P < 0.05) and 57.97% (24 nmol, n =8,P < 0.001), respectively.
Intrathecal post-treatment with CNQX and MK-801 at 5 min
after s.c. BV could also produce 31.98% (100 nmol, n =8, P < 0.001)
and 44.81% (24 nmol, n =8, P < 0.001) inhibition of the BV-induced
PSN in the subsequent 55 min period (Fig. 5C and the right panel of
Fig. 5D and E). Taken together, the above results suggest that acti-
vation of spinal NMDA and non-NMDA receptors plays crucial roles
in both inducing and maintaining processes of BV-evoked PSN.
3.3.2. Spinal inhibition of NMDA and non-NMDA receptors fails to
reverse primary thermal and mechanical hyperalgesia in the BV
Post-treatment with i.t. administration of the highest dose of
CNQX (100 nmol) or MK-801 (24 nmol) used in the current study
produced no signiﬁcant effect on basal pain sensitivity to either
thermal or mechanical stimuli (data not shown). There were still
no obvious changes in either PWTL or PWMT when the same doses
of CNQX and MK-801 were intrathecally applied in BV-inﬂamed rats
(Fig. 6), indicating no involvement of iGluRs in the maintenance of
BV-induced primary heat and mechanical hyperalgesia.
3.4. Effects of i.t. mGluR antagonism on BV-evoked thermal and
To further elucidate the Glu receptor subtypes mediating its
putative inﬂuence on thermal and mechanical hyperalgesia, we
investigated effects of i.t. post-treatment with three kinds of mGluR
antagonists: AIDA for group I [49,50], EGLU for group II [21,51] and
MSOP for group III mGluRs [51,52]. As illustrated in Fig. 6, i.t. admin-
istration of AIDA (10 nmol) produced a partial reversal of BV-evoked
mechanical (but not thermal) hyperalgesia, whereas post-injection
of EGLU (20 nmol) and MSOP (10 nmol) elicited an almost full rever-
sal of BV-induced thermal (but not mechanical) hyperalgesia.
The present study demonstrated a rapid and signiﬁcant release
of EAAs (Asp and Glu), IAAs (Gly, Tau and GABA) and other amino
acids (Ala, Arg and Thr) in the spinal cord following BV-induced
persistent nociception. Peripheral afferent inputs from the injec-
tion site were found to be required for both behavioral PSN and
spinal amino acid release. Furthermore, our pharmacological exper-
iments showed that i.t. administration of CNQX or MK801 could
signiﬁcantly prevent the induction and reverse the maintenance
of BV-induced persistent spontaneous nociception, whereas i.t.
mGluRs antagonism resulted in an apparent reversal of BV-evoked
thermal or mechanical hypersensitivity.
One gain of this work lies in the ﬁnding that extracellular con-
centrations of spinal Glu and Asp profoundly increased in the
BV model of inﬂammatory pain. These results are in large accor-
dance with previous observations from other animal pain models
[6–11,53–55], although the speciﬁc release proﬁle of EAAs varies
a lot among each other (see  for a brief summary), possibly
reﬂecting inter-model differences in EAAs responses (magnitude or
duration). The divergence might also be explained by subtle differ-
ences in experimental conditions (such as the exact location of the
dialysis ﬁber, the accuracy or sensitivity of the detection method,
the animal species etc.). In the present study, the sample collection
interval was set at 20 min. Although one cannot gain any certain
knowledge of when the EAAs level began to rise substantially after
BV injection (maybe earlier than 20 min), the general release proﬁle
of spinal EAAs paralleled well with the time course of BV-induced
PSN, suggesting a strong correlation between them.
The present results also revealed a smaller and shorter increase
in spinal IAAs release following BV-induced persistent pain.
296 L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299
Fig. 5. Effects of intrathecal pre- and post-treatment with CNQX and MK801 on the induction and maintenance of persistent spontaneous ﬂinching reﬂex inducedby
subcutaneous BV injection. Curve graph (A, B, C) shows the time courses of CNQX and MK-801 effects on mean number of paw ﬂinches recorded at each 5 min time block. (A)
Pre-treatment with CNQX at three doses (3.5 nmol, 35 nmol and 100 nmol); (B) pre-treatment with MK-801 at two doses (1.2 and 24 nmol); (C) post-treatment with CNQX
(24 nmol) and MK-801 (100 nmol). Column graph (D, E) illustrates the mean number of paw ﬂinches per 5 min averaged from the observation period. (D) Effects of pre- and
post-treatment with CNQX; (E) effects of pre- and post-treatment with MK-801. Values are mean ±S.E.M. (n = 8 for each group).
P < 0.05,
P < 0.001 vs. vehicle group.
Enhanced release of IAAs under the pain state has also been
reported by a multiplicity of previous papers [7,9,53]. This transient
increase in concentrations of spinal IAAs may be interpreted as a
reﬂection of activation of an endogenous spinal analgesia system
or descending inhibitory response, based on the previous sugges-
tion that the large increase in EAAs was often accompanied by
a corresponding rise in IAAs level in an attempt to reinstate the
homeostasis [11,53]. A distinct feature of the present spinal IAAs
release proﬁle was that the level of Gly continued to decrease after
returning to the baseline, becoming much lower at the end of the
Fig. 6. Effects of intrathecal post-treatment with CNQX (100 nmol), MK801 (24 nmol), AIDA (10 nmol), EGLU (20 nmol), and MSOP (10 nmol) on maintenance of BV-induced
primary thermal (A) and mechanical (B) hyperalgesia. Vertical bars: ±S.E.M. (n = 8 for each). Baseline, averaged values of PWTL or PWMT obtained from the other groups of
animals prior to any treatment. PWTL, paw withdrawal thermal latency; PWMT, paw withdrawal mechanical threshold.
P < 0.05 vs. BV + vehicle group.
L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299 297
detection period. This interesting phenomenon is reminiscent of
our previous behavioral observationsshowingthat a plateau level of
hyperalgesia arrived at 2–6 h after s.c. BV administration [26,29,30].
This striking resemblance between temporal patterns of spinal IAAs
decrease and primary hyperalgesia maintenance prompts us to pre-
dict that a probable loss of tonic inhibition [56,57] due to reduced
presynaptic transmitter release, causing disruption of the balance
between excitatory and inhibitory inﬂuences, may serve as a can-
didate mechanism underlying formation of central sensitization
under BV-produced inﬂammatory pain state, which, at least in part,
contributes to the behavioral expression of thermal and mechanical
Consistent with our previous demonstrations of the peripheral-
dependence of BV-induced behavioral and electrophysiological
responses [27,31,32,35], in this study, local pre-administration of
bupivacaine could cause a complete abolition of PSN and spinal
amino acid release. The possibility that this inhibitory effect might
be due to a systemic effect of the locally injected anesthetic agent
seems unlikely because peripheral injection of the same dose of
bupivacaine into the non-injured hindpaw caused no signiﬁcantly
pharmacological inﬂuence on BV-evoked persistent spontaneous
paw ﬂinching reﬂex (Fig. 1). Of primary interest is the ﬁnding that,
in some cases, local anesthesia of the hindpaw resulted in not only
an entire abolition of BV-induced EAA and IAAs release, but also an
even lower level than the baseline (Figs. 2 and 3). This result is in
general agreement with one electrophysiological study, in which
blockade of the sciatic nerve with 4% lidocaine suppressed not only
formalin-induced spike discharges but also background activities
of spinal nociceptive neurons . Based upon the above results,
we propose that EAAs–IAAs imbalance at the spinal level, requir-
ing the intact primary afferent inputs from the periphery, serves as
another major mechanism underlying occurrence/maintenance of
central sensitization and chronicity of pain state. And this concern
will be important to explain why pain treatment (such as local nerve
blockade) should be given as early as possible before the balance
between nociception and anti-nociception is disrupted.
A growing numbe r of earlier reports have provided compelling
evidence for key roles of NMDA and non-NMDA receptors in numer-
ous behavioral manifestations of pathological pain by using various
pain models [65–67]. According to previous studies, NMDA but not
non-NMDA receptor antagonists could block the persistent noci-
ceptive responses induced by s.c. injection of formalin , while
both NMDA and non-NMDA receptor antagonists were effective
in inhibiting the nociceptive behaviors produced by intraplantar
injection of capsaicin into the mouse hindpaw . There are also
a large number of previous reports indicating complicated roles
of iGluRs in multiple kinds of inﬂammatory or neuropathic pain
[13–16,68–71]. Taken together with those ﬁndings, it might be log-
ically supposed that specialized roles of NMDA and non-NMDA
receptors in development and maintenance of pain and hyperal-
gesia may be different from case to case and need to be further
studied by using a pathological pain model being able to reﬂect
multiple “phenotypes” of pain-related behaviors.
The unique behavioral “phenotypes” of nociception and hyper-
algesia identiﬁe d in the rodent BV test are believed to reﬂect a
complex pathological state of inﬂammatory pain and might be
appropriate to the study of phenotype-based mechanisms of pain
and hyperalgesia [26,28,30,31,37,39]. Using this unique model of
pain following chemical tissue injury, we performed series of phar-
macological experiments to investigate potential roles of peripheral
and central NMDA or non-NMDA receptors in BV-induced multifar-
ious pain-related behaviors and electrophysiological alterations in
spinal nociceptive neurons [12,17,32,35,72]. In the present study,
we further contributed the new ﬁndings that activation of both
NMDA and non-NMDA receptors in the spinal cord is involved in
the induction and maintenance of BV-induced PSN. However, they
were not likely to contribute to the maintenance of primary hyper-
algesia produced by intraplantar BV injection. The failure of NMDA
and non-NMDA receptor antagonisms to reverse established ther-
mal and mechanical hyperalgesia by i.t. route was not likely due to
the dose insufﬁciency, because the same dose of both drugs could
effectively repress BV-initiated PSN. Moreover, our previous report
demonstrated that systemic administration of MK801 (0.01 mg/kg,
i.p.) was not effective either . Collectively, these results, on one
hand, suggest that sufﬁcient amounts of EAAs released from spinal
cord (as mentioned above) are likely acting on iGluRs to mediate
the development and persistence of BV-evoked PSN, while on the
other hand, further conﬁrm our previous hypothesis that different
neurochemical components (or intracellular messenger-mediated
signal transduction pathways) are likely to be involved in medi-
ating different “phenotypes” of nociception and hypersensitivity
In another series of experiments, we found that i.t. antagonism
of different subgroups of mGluRs could effectively block estab-
lished thermal and mechanical hyperalgesia in BV-inﬂamed rats.
Previous behavioral and electrophysiological evidence has been
accumulating showing involvement of mGluRs in spinal nocicep-
tive processing [1,20–24,74,75]. An important ﬁnding added by the
present study is that blockade of spinal group I mGluRs activation
by AIDA suppressed BV-induced primary mechanical hyperalgesia,
while inhibition of group II and III mGluRs activity by EGLU and
MSOP markedly reduced the extent of primary thermal hypersensi-
tivity. This result suggests that the BV-evoked primary hyperalgesia
is maintained by a central sensitized state caused by not only spinal
disinhibition but also sustained activation of mGluR subtypes and
the sequential signaling cascades. The exact sources of Glu that is
recruited and responsible for such prolonged activation of mGluRs
are still not clear but likely ascribed to evoked release by periph-
eral heat or mechanical stimuli when testing being performed. In
addition, the present pharmacological blockade observations also
suggest that activation of iGluRs or mGluRs is state- or stimulus
modality-dependent, and spinal antagonism of any speciﬁc gluta-
mate receptor subtype fails to block all behavioral “phenotypes” of
pathological pain, implicating possible involvement of inhibitory
inﬂuences during the whole story.
In summary, it is likely that there is an EAAs–IAAs balance at the
spinal level under normal state, however, the balance would be dis-
rupted by the arrival of ongoing activity from peripheral persistent
nociception (prolonged pathological state), leading to abnormal
changes in spinal synaptic transmission and modulation (central
sensitization) which in turn contribute to persistence or chronicity
of pain. Based upon this hypothesis, the new therapeutic strategy
of analgesia should focus not only on the EAAs receptor-mediated
signaling targets but also on the restoration of EAAs–IAAs balance
at the spinal level to prevent persistence or chronicity of pain at the
early stage of pathological processing.
This work was supported by the National Basic Research (973)
Program of China 2006CB500800, National Innovation Team Pro-
gram of Ministry of Education IRT0560, National Natural Science
Foundation of China grants 30670692 and 30770668, and Natural
Science Foundation of Beijing grant KZ200510025016 to J.C.
 Fundytus ME. Glutamate receptors and nociception: implications for the drug
treatment of pain. CNS Drugs 2001;15:29–58.
 Lawand NB, Willis WD, Westlund KN. Excitatory amino acid receptor
involvement in peripheral nociceptive transmission in rats. Eur J Pharmacol
298 L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299
 Jeftinija S, Jeftinija K, Liu F, Skilling SR, Smullin DH, Larson AA. Excitatory amino
acids are released from rat primary afferent neurons in vitro. Neurosci Lett
 Kangra I, Randi
c M. Outﬂow of endogenous aspartate and glutamate f rom the
rat spinal dorsal horn in vitro by activation of low- and high-threshold primary
afferent ﬁb ers. Modulation by -opioids. Brain Res 1991;553:347–52.
 Stiller CO, Taylor BK, Linderoth B, Gustafsson H, Warsame Afrah A, Brodin E.
Microdialysis in pain research. Adv Drug Deliv Rev 2003;55:1065–79.
 Marsala M, Malmberg AB, Yaksh TL. The spinal loop dialysis catheter: charac-
terization of use in the unanesthetized rat. J Neurosci Methods 1995;62:43–53.
 Malmberg AB, Yaksh TL. The effect of morphine on formalin-evoked b ehav-
ior and spinal release of excitatory amino acids and prostaglandin E2 using
microdialysis in conscious rats. Br J Pharmacol 1995;114:1069–75.
 Vetter G, Geisslinger G, Tegeder I. Release of glutamate, nitric oxide and
and metabolic activity in the spinal cord of rats following
peripheral nociceptive stimulation. Pain 2001;92:213–8.
 Sluka KA, Willis WD. Increased spinal release of excitatory amino acids follow-
ing intradermal injection of capsaicin is reduced by a protein kinase G inhibitor.
Brain Res 1998;798:281–6.
 Sorkin LS, McAdoo DJ. Amino acids and serotonin are released into the lumbar
spinal cord of the anesthetized cat following intradermal capsaicin injections.
Brain Res 1993;607:89–98.
 Zahn PK, Sluka KA, Brennan TJ. Excitatory amino acid release in the spinal cord
caused by plantar incision in the rat. Pain 2002;100:65–76.
 Chen HS, Chen J. Secondary heat, but not mechanical, hyperalgesia induced
by subcutaneous injection of bee venom in the conscious rat: effect of
systemic MK-801, a non-competitive NMDA receptor antagonist. Eur J Pain
 Ren K, Dubner R. NMDA receptor antagonists attenuate mechanical hyper-
algesia in rats with unilateral inﬂammation of the hindpaw. Neurosci Lett
 Laird JM, Mason GS, Webb J, Hill RG, Hargreaves RJ. Effects of a partial ago-
nist and a full antagonist acting at the glycine site of the NMDA receptor
on inﬂammation-induced mechanical hyperalgesia in rats. Br J Pharmacol
 Mao J, Price DD, Hayes RL, Lu J, Mayer DJ. Differential roles of NMDA
and non-NMDA receptor activation in induction and maintenance of ther-
mal hyperalgesia in rats with painful peripheral mononeuropathy. Brain Res
 Munglani R, Hudspith MJ, Fleming B, Harrisson S, Smith G, Bountra C, et al. Effect
of pre-emptive NMDA antagonist treatment on long-term Fos expression and
hyperalgesia in a model of chronic neuropathic pain. Brain Res 1999;822:210–
 Chen HS, Chen J, Sun YY. Contralateral heat hyperalgesia induced by unilaterally
intraplantar bee venom injection is produced by central changes: a behavioral
study in the conscious rat. Neurosci Lett 2000;284:45–8.
 Coderre TJ, Melzack R. The contribution of excitatory amino acids to central
sensitization and persistent nociception after formalin-induced tissue injury. J
 Sakurada T, Wako K, Sugiyama A, Sakurada C, Tan-No K, Kisara K. Involvement of
spinal NMDA receptors in capsaicin-induced nociception. Pharmacol Biochem
 Karim F, Wang CC, Gereau IV RW. Metabotropic glutamate receptor subtypes 1
and 5 are activators of extracellular signal-regulated kinase signaling required
for inﬂammatory pain in mice. J Neurosci 2001;21:3771–9.
 Neugebauer V. Metabotropic glutamate receptors-important modulators of
nociception and pain behavior. Pain 2002;98:1–8.
 Neugebauer V, Chen PS, Willis WD. Role of metabotropic glutamate receptor
subtype mGluR1 in brief nociception and central sensitization of primate STT
cells. J Neurophysiol 1999;82:272–82.
 Fisher K, Fundytus ME, Cahill CM, Coderre TJ. Intrathecal administration
of the mGluR compound, (S)-4CPG, attenuates hyperalgesia and allodynia
associated with sciatic nerve constriction injury in rats. Pain 1998;77:59–
 Dolan S, Nolan AM. Behavioural evidence supporting a differentialrole for group
I and II metabotropic glutamate receptors in spinal nociceptive transmission.
 Lariviere WR, Melzack R. The bee venom test: a new tonic-pain test. Pain
 Chen J, Luo C, Li HL, Chen HS. Primary hyperalgesia to mechanical and heat
stimuli following subcutaneous bee venom injection into the plantar surface of
hindpaw in the conscious rat: a comparative study with the formalin test. Pain
 Chen J, Chen HS. Pivotal role of capsaicin-sensitive primary afferents in devel-
opment of both heat and mechanical hyperalgesia induced by intraplantar bee
venom injection. Pain 2001;91:367–76.
 Lariviere WR, Wilson SG, Laughlin TM, Kokayeff A, West EE, Adhikari SM, et
al. Heritability of nociception. III. Genetic relationships among commonly used
assays of nociception and hypersensitivity. Pain 2002;97:75–86.
 Chen J. Processing of different ‘phenotypes’ of pain by different spinal signaling
pathways. In: Kumamoto K, editor. Cellular and molecular mechanisms for the
modulation of nociceptive transmission in the peripheral and central nervous
systems. Kerala: Recent Research Development Series, Research SignPost; 2007.
 Chen J. Spinal processing of bee venom-induced pain and hyperalgesia. Acta
Physiologica Sinica 2008;60:645–52.
 Chen J, Luo C, Li HL. The contribution of spinal neuronal changes to development
of prolonged, tonic nociceptive responses of the cat induced by subcutaneous
bee venom injection. Eur J Pain 1998;2:359–76.
 Chen J, Li HL, Luo C, Li Z, Zheng JH. Involvement of peripheral NMDA and non-
NMDA receptors in development of persistent ﬁring of spinal wide-dynamic-
range neurons induced by subcutaneous bee venom injection in the cat. Brain
 You HJ, Arendt-Nielsen L. Unilateral subcutaneous bee venom but not
formalin injection causes contralateral hypersensitized wind-up and after-
discharge of the spinal withdraw reﬂex in anesthetized spinal rats. Exp Neurol
 You HJ, Chen J. Differential effects of subcutaneous injection of formalin and
bee venom on responses of wide-dynamic range neurons in spinal dorsal horn
of the rat. Eur J Pain 1999;3:177–80.
 You HJ, Chen J, Morch CD, Arendt-Nielsen L. Differential effect of peripheral
glutamate (NMDA, non-NMDA) receptor antagonists on bee venom-induced
spontaneous nociception and sensitization. Brain Res Bull 2002;58:561–7.
 Chen HS, Lei J, He X, Wang Y, Wen WW, Wei XZ, et al. Pivotal involvement
of neurogenic mechanism in subcutaneous bee venom-induced inﬂammation
and allodynia in unanesthetized conscious rats. Exp Neurol 2006;200:386–
 Cao FL, Liu MG, Hao J, Li Z, Lu ZM, Chen J. Different roles of spinal p38 and c-
Jun N-terminal kinase pathways in bee venom-induced multiple pain-related
behaviors. Neurosci Lett 2007;427:50–4.
 Li KC, Chen J. Differential roles of spinal protein kinases C and A in development
of primary heat and mechanical hypersensitivity induced by subcutaneous bee
venom chemical injury in the rat. Neurosignals 2003;12:292–301.
 Yu HY, Liu MG, Liu DN, Shang GW, Wang Y, Qi C, et al. Antinociceptive effects of
systemic paeoniﬂorin on bee venom-induced various ‘phenotypes’ of nocicep-
tion and hypersensitivity. Pharmacol Biochem Behav 2007;88:131–40.
 Zheng JH, Chen J. Differential roles of spinal neurokinin 1/2 receptors in
development of persistent spontaneous nociception and hyperalgesia induced
by subcutaneous bee venom injection in the conscious rat. Neuropeptides
 Zimmermann M. Ethical guidelines for investigations of experimental pain in
conscious animals. Pain 1983;16:109–10.
 Shi L, Smolders I, Umbrain V, Lauwers MH, Sarre S, Michotte Y, et al. Peripheral
inﬂammation modiﬁes the effect of intrathecal IL-1␤ on spinal PGE
mainly through cyclooxygenase-2 activity. A spinal microdialysis study in freely
moving rats. Pain 2006;120:307–14.
 Sorkin LS, Steinman JL, Hughes MG, Willis WD, McAdoo DJ. Microdialysis
recovery of serotonin released in spinal cord dorsal horn. J Neurosci Methods
 Parent M, Bush D, Rauw G, Master S, Vaccarino F, Baker G. Analysis of amino
acids and catecholamines, 5-hydroxytryptamine and their metabolites in brain
areas in the rat using in vivo microdialysis. Methods 2001;23:11–20.
 Hertz L, Zielke HR. Astrocytic control of glutamatergic activity: astrocytes as
stars of the show. Trends Neurosci 2004;27:735–43.
 Hinoi E, Takarada T, Tsuchihashi Y, Yoneda Y. Glutamate transporters as drug
targets. Curr Drug Targets CNS Neurol Disord 2005;4:211–20.
 Gibbs ME, Hutchinson D, Hertz L. Astrocytic involvement in learning and mem-
ory consolidation. Neurosci Biobehav Rev 2008;32:927–44.
 Dickenson AH, Chapman V, Green GM. The pharmacology of excitatory and
inhibitory amino acid-mediated events in the transmission and modulation of
pain in the spinal cord. Gen Pharmacol 1997;28:633–8.
 Pellicciari R, Luneia R, Costantino G, Marinozzi M, Natalini B, Jakobsen P, et
al. 1-Aminoindan-1,5-dicarboxylic acid: a novel antagonist at phospholipase
C-linked metabotropic glutamate receptors. J Med Chem 1995;38:3717–9.
 Moroni F, Lombardi G, Thomsen C, Leonardi P, Attucci S, Peruginelli F, et al. Phar-
macological characterization of 1-aminoindan-1,5-dicarboxylic acid, a potent
mGluR1 antagonist. J Pharmacol Exp Ther 1997;281:721–9.
 Schoepp DD, Jane DE, Monn JA. Pharmacological agents acting at subtypes
of metabotropic glutamate receptors. Neuropharmacology 1999;38:1431–
 Pellicciari R, Costantino G. Metabotropic G-protein-coupled glutamate recep-
tors as therapeutic targets. Curr Opin Chem Biol 1999;3:433–40.
 Sluka KA, Westlund KN. An experimental arthritis in rats: dorsal horn aspartate
and glutamate increases. Neurosci Lett 1992;145:141–4.
 Yang LC, Marsala M, Yaksh TL. Characterization of time course of spinal amino
acids, citrulline and PGE2 release after carrageenan/kaolin-induced knee joint
inﬂammation: a chronic microdialysis study. Pain 1996;67:345–54.
 McAdoo DJ, Xu GY, Robak G, Hughes MG. Changes in amino acid concentration
over time and space around an impact injury and their diffusion through the
rat spinal cord. Exp Neurol 1999;159:538–44.
 Duggan AW, Morton CR. Tonic descending inhibition and spinal nociceptive
transmission. Prog Brain Res 1988;77:193–211.
 Kohno T. A role of spinal inhibition in neuropathic pain. In: Kumamoto K, editor.
Cellular and molecular mechanisms for the modulation of nociceptive trans-
mission in the peripheral and central nervous systems. Kerala: Recent Research
Development Series, Research SignPost; 2007. p. 131–45.
 Sivilotti L, Woolf CJ. The contribution of GABAA and glycine receptors to central
sensitization: disinhibition and touch-evoked allodynia in the spinal cord. J
 Baba H, Ji RR, Kohno T, Moore KA, Ataka T, Wakai A, et al. Removal of GABAergic
inhibition facilitates polysynaptic A ﬁbermediated excitatory transmission to
the superﬁcial spinal dorsal horn. Mol Cell Neurosci 2003;24:818–30.
L.-H. Yan et al. / Pharmacological Research 59 (2009) 290–299 299
 Seltzer Z, Cohn S, Ginzburg R, Beilin B. Modulation of neuropathic pain behavior
in rats by spinal disinhibition and NMDA receptor blockade of injury discharge.
 Moore KA, Kohno T, Karchewski LA, Scholz J, Baba H, Woolf CJ. Partial peripheral
nerve injury promotes a selective loss of GABAergic inhibition in the superﬁcial
dorsal horn of the spinal cord. J Neurosci 2002;22:6724–31.
 McMahon SB, Lewin GR, Wall PD. Central hyperexcitability triggered by noxious
inputs. Curr Opin Neurobiol 1993;3:602–10.
 Zimmermann M. Pathobiology of neuropathic pain. Eur J Pharmacol
 Chen J, Koyama N, Yokota T. Effects of subcutaneous formalin on responses of
dorsal horn wide dynamic range neurons and primary afferent neurons in the
cat. Pain Res 1996;11:71–83.
 Zieglgänsberger W, Tölle TR. The pharmacology of pain signalling. Curr Opin
 Dray A, Urban L, Dickenson A. Pharmacology of chronic pain. Trends Pharmacol
 Besson JM. The neurobiology of pain. Lancet 1999;353:1610–5.
 Ren K, Hylden JL, Williams GM, Ruda MA, Dubner R. The effects of a non-
competitive NMDA receptor antagonist, MK-801, on behavioral hyperalgesia
and dorsal horn neuronal activity in rats with unilateral inﬂammation. Pain
 Yamamoto T, Shimoyama N, Mizuguchi T. The effects of morphine, MK-801,
an NMDA antagonist, and CP-96,345, an NK1 antagonist, on the hyperesthesia
evoked by carageenan injection in the rat paw. Anesthesiology 1993;78:124–33.
 Zahn PK, Umali E, Brennan TJ. Intrathecal non-NMDA excitatory amino acid
receptor antagonists inhibit pain behaviors in a rat model of postoperative pain.
 Kim YI, NaHS, YoonYW, Han HC, KoKH, Hong SK. NMDA receptors are important
for both mechanical and thermal allodynia from peripheral nerve injury in rats.
 You HJ, Morch CD, Chen J, Arendt-Nielsen L. Role of central NMDA versus non-
NMDA receptor in spinal withdrawal reﬂex in spinal anesthetized rats under
normal and hyperexcitable conditions. Brain Res 2003;981:12–22.
 Cui XY, Dai Y, Wang SL, Yamanaka H, Kobayashi K, Obata K, et al. Differential acti-
vation of p38 and extracellular signal-regulated kinase in spinal cord in a model
of bee venom-induced inﬂammation and hyperalgesia. Mol Pain 2008;4:17.
 Fisher K, Coderre TJ. The contribution of metabotropic glutamate receptors
(mGluRs) to formalin-induced nociception. Pain 1996;68:255–63.
 Dogrul A, Ossipov MH, Lai Josephine, Malan Jr TP, Porreca F. Peripheral and
spinal antihyperalgesic activity of SIB-1757, a metabotropic glutamate receptor
(mGluR5) antagonist, in experimental neuropathic pain in rats. Neurosci Lett