Pharmacological Research 51 (2005) 183188

Antinociceptive mechanisms associated with diluted bee venom

acupuncture (apipuncture) in the rat formalin test: involvement
of descending adrenergic and serotonergic pathways
Hyun Woo Kima,1 , Young Bae Kwona,1 , Ho Jae Hanb , Il Suk Yanga ,
Alvin J. Beitzc , Jang Hern Leea,
a

Department of Veterinary Physiology, College of Veterinary Medicine and School of Agricultural Biotechnology, Seoul National University, San 56-1,
Shilim-dong, Kwanak-gu, Seoul 151-742, South Korea
b Hormone Research Center, College of Veterinary Medicine, Chonnam National University, Kwang-ju, South Korea
c Department of Veterinary and Biomedical Science, University of Minnesota, St. Paul, MN, USA
Accepted 21 July 2004

Abstract
In a previous report, subcutaneous injection of diluted bee venom (dBV) into a specific acupuncture point (Zusanli, ST36), a procedure
termed apipuncture, was shown to produce an antinociceptive effect in the rat formalin pain model. However, the central antinociceptive
mechanisms responsible for this effect have not been established. Traditional acupuncture-induced antinociception is considered to be mediated
by activation of the descending pain inhibitory system (DPIS) including initiation of its opioidergic, adrenergic and serotonergic components.
The purpose of the present study was to investigate whether the antinociceptive effect of apipuncture is also mediated by the DPIS. Behavioral
experiments verified that apipuncture significantly reduces licking behavior in the late phase of formalin test in rats. This antinociceptive
effect of apipuncture was not modified by intrathecal pretreatment with naltrexone (a non-selective opioid receptor antagonist), prazosin (a
1 adrenoceptor antagonist) or propranolol (an adrenoceptor antagonist). In contrast, intrathecally injected idazoxan (an 2 adrenoceptor
antagonist) or intrathecal methysergide (a serotonin receptor antagonist) significantly reversed apipuncture-induced antinociception. These
results suggest that apipuncture-induced antinociception is produced by activation of 2 adrenergic and serotonergic components of the DPIS.
2004 Elsevier Ltd. All rights reserved.
Keywords: Bee venom; Acupuncture; Antinociception; 2 Adrenergic; Serotonergic

1. Introduction
Although the precise antinociceptive mechanisms underlying acupuncture remain to be elucidated, this procedure has
been widely used to relieve various types of acute and chronic
pain [1]. Electrical or mechanical stimulation of an acupuncture point (acupoint) is the most popular form of acupunture
and this type of stimulation produces significant antinociceptive effects [2]. Recently, we have demonstrated that chem
1

Corresponding author. Tel.: +82 2 880 1272; fax: +82 2 885 2732.
E-mail address: jhl1101@snu.ac.kr (J.H. Lee).
These authors contributed equally to this study.

1043-6618/$ see front matter 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phrs.2004.07.011

ical stimulation of an acupoint with subcutaneous injection

of diluted bee venom (dBV), a procedure termed apipuncture, produces a very potent and long-lasting antinociceptive
effect in both acute and chronic rodent pain models [4,5]. Collectively, these data imply that apipuncture may be a more
effective acupoint stimulant than other forms of acupuncture. While apipuncture appears to be a more effective form
of acupuncture, there is very little known regarding the neuronal mechanisms underlying apipuncture-induced antinociception as compared to other types of acupuncture.
Both spinal and supraspinal mechanisms contribute to
acupuncture-induced analgesia [3]. We have recently demonstrated that dBV injection into the Zusanli acupoint (ST36)

184

H.W. Kim et al. / Pharmacological Research 51 (2005) 183188

dose dependently decreased paw licking time and spinal Fos

expression in the formalin inflammatory pain model. While
the antinociception produced by apipuncture was clearly
demonstrated in this study, the central mechanisms responsible for this analgesic effect are currently unknown. In an
earlier study, apipuncture was shown to produce visceral
antinociception by activation of the DPIS involving spinal
adrenoceptors. In the present study, we further investigated
the involvement of opioidergic, adrenergic and serotonergic
components of the DPIS in the apipuncture-induced antinociceptive effect observed in the rat formalin test.

2. Materials and methods

2.1. Animals
Experiments were performed on male SpragueDawley
rats weighing 200250 g. All experimental animals were obtained from the Laboratory Animal Center of Seoul National
University. They were housed in colony cages with free access to food and water and were maintained in temperature
and light controlled rooms (23 0.5 C, 12/12 h light/dark
cycle with lights on at 07:00 h) for at least 1 week prior to
the study. All of the methods used in the present study were
approved by the Animal Care and Use Committee at SNU
and conform to NIH guidelines (NIH Publication No. 86-23,
revised 1985). In addition, the ethical guidelines for investigating experimental pain in conscious animals recommended
by the International Association for the Study of Pain [6] were
followed.

2.2. Surgery for intrathecal catheterization

Rats were anaesthetized by intraperitoneal injection of
chloral hydrate (400 mg kg1 , Merck, Germany). A piece of
PE10 tubing (0.28 mm i.d., 0.61 mm o.d., Becton and Dickinson, MD, USA) was inserted via the atlanto-occipital membrane using a stereotaxic instrument as previously described
[7]. In order for the catheter to reach the lumbar enlargement,
an intraspinal length of 8 cm was used. The dead space of the
catheter was about 5 l. To verify proper catheter placement
in the lumbar enlargement, an intrathecal injection of lidocaine was performed. When the intraspinal location of the
catheter was correct, then motor paralysis of the animals hind
limb lasting for a period of 2030 min was observed immediately after lidocaine injection. Catheterized animals were not
used for intrathecal studies until 7 days post-implantation.
To determine if catheterized animals showed normal motor
function, a spontaneous activity chamber (MED Associates
Inc., USA, model # SG-506) was used to measure freely traveled distance as previously described [8]. Any animals that
displayed a dysfunction of the limbs while walking were not
used in this study.

2.3. Drugs
Diluted bee venom of Apis mellifera, idazoxan, prazosin,
propranolol were purchased from Sigma (St. Louis, MO,
USA). Naltrexone was purchased from Research Biochemicals (Natick, MA, USA) and methysergide was purchased
from Tocris (UK). All drugs were dissolved in physiologic
saline (0.9% (w/v) NaCl) just before use.
2.4. Experimental protocol
Anesthesia was induced in catheterized rats by inhalation of 5% isoflurane (Baxter, USA) in a mixed N2 O/O2 gas
for 30 s and then maintained with 3% isoflurane. Naltrexone
(NTX, 10 g per rat), idazoxan (IDZ, 50 g per rat), prazosin (PRA, 50 g per rat), propranolol (PRO, 50 g per rat),
or methysergide (MET, 30 g per rat) in a volume of 10 l
saline (Sal) was intrathecally injected. The dose of each receptor antagonist was determined from the literature and was
based on previous studies in which intrathecal administration of the antagonist effectively blocked its receptor and a
spinal cord associated antinociceptive effect [915]. Control
animals received an equivalent volume of physiologic saline
through the same route of administration. Animal groups
names are abbreviated throughout the text and in the figures based on their specific treatment. Thus, the animal group
that received intrathecal injections of saline (Sal) followed by
subcutaneous injection of bee venom (BV) and then formalin injection is designated Sal-BV-F, while the group that
received intrathecal saline, subcutaneous saline and formalin is designated Sal-Sal-F. Similarly animals receiving intrathecal injection of the drugs, naltrexone (NTX), idazoxan
(IDZ), prazosin (PRA), propranolol (PRO), or methysergide
(MET) plus subcutaneous BV and then formalin were designated NTX-BV-F, IDZ-BV-F, PRA-BV-F, PRO-BVF, and MET-BV-F, respectively.
Ten minutes after intrathecal drug treatment, dBV
(0.08 mg kg1 in a volume of 20 l saline) was subcutaneously administered into the Zusanli (ST36) acupoint
located 5 mm below and lateral to the anterior tubercle of the
tibia. Thirty minutes post-dBV administration, 1% formalin
(20 l) was subcutaneouly injected under the plantar surface
of the right hindpaw. The formalin test is a commonly used
model of persistent pain and formalin-induced behavior is
characterized by two phases: an early phase and a late phase.
Pain behaviors during the early phase are thought to be due
to direct chemical stimulation of nociceptors, while pain
behaviors associated with the late phase are attributed to
inflammatory pain induced by formalin-released inflammatory mediators including histamine and prostaglandin [16].
Following intraplantar injection of formalin, the animals
were immediately placed on an acrylic observation chamber
(40 cm high, 20 cm diameter), and behaviors were recorded
using a video camera for 30 min. Following the video-taping,
paw licking time (in seconds per each 5 min increment) was
calculated by two experienced investigators, blinded to the

H.W. Kim et al. / Pharmacological Research 51 (2005) 183188

185

experimental conditions, during both the early phase (phase

1; 010 min post-formalin injection) and the late phase (phase
2; 1030 min post-formalin injection) of the formalin test.
Subsequently, the mean licking time was calculated from the
data obtained by these two experimenters and the mean licking time values were subsequently analyzed statistically to
determine significant differences among the various groups.
2.4.1. Data and statistical analysis
The paw licking time at early and late phase was expressed
as the mean S.E.M. Data was analyzed by either an independent t-test or an analysis of variance (ANOVA) followed
by post-hoc comparisons using Fishers least significant difference test. Statistical significance was determined at p <
0.05.

3. Results
3.1. Antinociceptive effect of dBV on formalin-induced
pain behavior
Pretreatment with dBV (0.08 mg kg1 , s.c.) strongly suppressed the formalin-induced paw licking time in the late
phase (Fig. 1). Total paw licking time in the late phase was
92.83 8.35 s in the Sal-Sal-F group and 23.17 9.04 s in
the Sal-dBV-F group ( p < 0.001). In addition, spontaneous
activity of i.t. catheterized and nave animals was not statistically different during 10 min (freely traveled distance: nave
= 346.10 79.92 cm; catheterized = 325.28 55.52 cm).

Fig. 2. The effect of the non-selective opioidergic antagonist naltrexone on

dBV-induced antinociception is illustrated in this graph. Pretreatment (i.t.)
with NTX (10 g 10 l1 ) did not alter the dBV-induced antinociception
observed during the late phase of the formalin test as compared with the
NTXsaline treated (NTX-Sal-F) group ( p < 0.001, two-tailed t-test).
Administration sites and routes were: i.t. injectionZusanli (s.c.)paw (s.c.).
Abbreviations: dBV, diluted bee venom; Sal, saline; F, formalin; NTX, naltrexone.

3.2. Opioidergic involvement in dBV-induced

antinociception
Pretreatment (i.t.) with the non-selective opioid receptor
antagonist, naltrexone (NTX, 10 g 10 l1 ) did not reverse
the dBV-induced antinociception in the late phase of the formalin test (Fig. 2). Paw licking time in the NTX-Sal-F group
was 87.20 6.69 s and in the NTX-dBV-F group was 25.00
6.87 s ( p < 0.001).
3.3. 1 and adrenergic involvement in dBV-induced
antinociception
Both i.t. pretreatment with the 1 adrenoceptor antagonist prazosin (PRA, 50 g 10 l1 ) and the adrenoceptor
antagonist propranolol (PRO, 50 g 10 l1 ) did not inhibit
the dBV-induced antinociceptive effect on formalin-evoked
pain behavior in the late phase (Fig. 3A and B). In the late
phase, paw licking times were 90.60 20.58 s (PRA-Sal-F),
8.11 2.46 s (PRA-dBV-F), 90.40 11.95 s (PRO-Sal-F)
and 27.33 6.90 s (PRO-dBV-F).
3.4. 2 adrenergic and serotonergic involvement in
dBV-induced antinociception

Fig. 1. This graph shows the antinociceptive effect of dBV on formalininduced pain behavior 10 and 30 min post-formalin injection. Treatment with
dBV remarkably reduced formalin-induced pain behavior in the late phase
( p < 0.001, two-tailed t-test). Administration sites and routes were: i.t.
injectionZusanli (s.c.)paw (s.c.). Abbreviations: dBV, diluted bee venom;
Sal, saline; F, formalin.

The 2 adrenoceptor antagonist, idazoxan (IDZ,

50 g 10 l1 ) significantly blocked the antinociceptive
effect of dBV on the formalin-induced pain behavior in the
late phase (Fig. 4A). During this phase, the total sum of
the paw licking time was increased from 81.80 5.00 s in

186

H.W. Kim et al. / Pharmacological Research 51 (2005) 183188

Fig. 3. A graph summarizing the effect of intrathecal administration of an 1 and a adrenergic antagonist on dBV-induced antinociception. In the late phase
of the formalin test, dBV treatment still had a significant antinociceptive effect on formalin-induced pain behavior in animals pretreated intrathecally with PRA
(50 g 10 l1 ; A) or PRO (50 g 10 l1 ; B) ( p < 0.001, two-tailed t-test). Administration sites and routes were: i.t. injectionZusanli (s.c.)paw (s.c.).
Abbreviations: dBV, diluted bee venom; Sal, saline; F, formalin; PRA, prazosin; PRO, propranolol.

the IDZ-dBV-F group to 88.80 9.63 s in the IDZ-Sal-F

group. The antinociceptive effect of dBV was also blocked
by i.t. pretreatment with the serotonin receptor antagonist
methysergide (MET, 30 g 10 l1 ) (Fig. 4B). In the late
phase, the total sums of paw licking time were 98.58
8.66 and 92.75 12.01 s in the MET-Sal-F group and the
MET-dBV-F groups, respectively.

4. Discussion
A variety of stimulation techniques including electroacupuncture (EA), moxibustion and acupressure have been
used to stimulate acupoints in order to produce antinociceptive effects that are selectively mediated by the activation of
descending modulatory systems [1721]. Two major components of this endogenous descending antinociceptive system
have been implicated in inhibition of nociceptive input at the
level of the spinal cord. One is the intrinsic opioidergic system
and the other is a descending monoaminergic (i.e., serotonin
and adrenaline) system in the brainstem [2]. It has been proposed that acupunctures antinociceptive effect is mediated
by different neuronal mechanisms depending on the type of
stimulation that is applied to an acupoint [22,23]. For instance, low frequency electroacupuncture-induced analgesia
appears to be mediated by the endogenous opioidergic system, while the analgesic effect of high frequency EA is mediated by a non-opioidergic system [24]. In the present study,
we observed that the antinociceptive effect of dBV induced
by acupoint stimulation (apipuncture) was totally reversed by
intrathecal pretreatment with the 2 adrenoceptor antagonist
idazoxan or the non-selective serotonin receptor antagonist
methysergide. In contrast, apipuncture-induced antinociception was not affected by intrathecal injection of antagonists

of other adrenoceptor subtypes or by i.t. injection of a nonselective opioid receptor antagonist. These results imply that
the antinociceptive effect of apipuncture is mediated by specific descending monoaminergic pathways rather than by the
intrinsic opioidergic system. A similar phenomenon has also
been observed in an acetic acid-induced visceral pain model
[5]. Based on these findings, it is suggested that apipunctureinduced antinociception is mediated by the spinal release of
norepinephrine and/or serotonin and the subsequent activation of specific adrenoceptors and/or 5-HT receptors in the
spinal cord. It is well documented that the adrenergic and
serotonergic components of the descending pain modulation system arise principally from the nucleus raphe magnus (RMg) and the locus coeruleus (LC), respectively [19].
Therefore, it is likely that the activation of these nuclei by
acupoint stimulation with dBV produces an antinociceptive
effect via activation of spinal 2 and/or serotonergic receptors. Support for this hypothesis comes from recent work
in our laboratories showing that BV acupoint stimulation
increases neuronal activity in brainstem catecholaminergic
neurons [25]. While this hypothesis seems likely for noradrenergic neurons, it remains to be determined if the 5-HT
specific antinociceptive mechanism of dBV is mediated via
higher brain centers including the RMg.
We observed that the 2 adrenoceptor antagonist idazoxan
antagonized apipuncture-induced antinociception. It is well
known that intrathecal administration of the non-selective
adrenergic antagonist phentolamine or the selective 2 adrenergic antagonist yohimbine attenuates descending inhibition
of nociceptive reflexes produced by electrical and/or chemical stimulation in the PAG, nucleus raphe magnus, nucleus
reticularis paragigantocellularis, and locus coeruleus [26]. In
addition, intrathecal administration of noradrenergic receptor agonists including clonidine is antinociceptive/analgesic

H.W. Kim et al. / Pharmacological Research 51 (2005) 183188

187

Fig. 4. This graph illustrates the effect of i.t. pretreatment with the 2 and serotonergic receptor antagonist, IDZ (50 g 10 l1 ) and MET (30 g 10 l1 ) on
dBV-induced antinociception. Both IDZ (A) and MET (B) pretreatment significantly reversed the antinociceptive effect of dBV on the formalin-evoked pain
behaviors during the late phase of the formalin test. Administration sites and routes were: i.t. injectionZusanli (s.c.)paw (s.c.). Abbreviations: dBV, diluted
bee venom; Sal, saline; F, formalin; IDZ, idazoxan; MET, methysergide.

[26]. Currently, it is presumed that norepinephrine inhibits A

and C fiber-mediated nociceptive transmission to spinal cord
through the activation of the 2 adrenoceptors (probably the
2A subtype) that are present on primary afferent terminals
[27].
While serotonin was initially proposed as a purely
antinociceptive transmitter, the influence of 5-HT upon the
activity of nocisponsive neurones in the dorsal horn is heterogeneous, with observations of both inhibition and less frequently excitation. In this regard, a variety of supraspinal
structures can inhibit and/or facilitate spinal nociceptive reflexes via activation of serotonergic receptors [26]. In the
present study, we have shown that intrathecal administration
of the non-selective serotonin receptor blocker methysergide
antagonized apipuncture-induced antinociception suggesting
that a serotoninergic spinal cord component is involved in
apipuncture analgesia.
It is interesting that the effects of BV were principally
on the late phase of the formalin response. As mentioned
earlier, pain behaviors during the early phase of the formalin test are thought to be due to direct chemical stimulation of nociceptors, while pain behaviors associated with
the late phase are attributed to inflammatory pain induced
by formalin-released inflammatory mediators including histamine and prostaglandin [16]. This concept is supported
by a recent study showing that a novel non-peptide antagonist of the bradykinin (BK) B1 receptor also inhibits only
the late phase of the formalin response suggesting that this
phase is related to formalin-released inflammatory mediators
[28]. These results taken together with our own data suggest that apipuncture is affective against the inflammatory
state induced by formalin injection that occurs 1055 min
post-formalin injection, while it is ineffective in blocking the
initial direct formalin stimulation of nociceptors.

In summary, we conclude that the antinociceptive effect of

apipuncture is mediated by the selective activation of spinal
2 adrenergic and serotonergic receptors. Conversely, neither
spinal opioidergic nor 1 or adrenergic components of the
descending pain inhibitory system are involved with dBVinduced antinociception in the rat formalin test.

Acknowledgements
This research was supported by a grant
(M103KV01000903K220100940) from the Brain Research Center of the 21st Century Frontier Research
Program funded by the Ministry of Science and Technology
of the Republic of Korea. The publication of this manuscript
was also supported by a Research Fund from the Research
Institute for Veterinary Science (RIVS) in the College of
Veterinary Medicine, Seoul National University, as well as
the Brain Korea 21 project.

References
[1] Cao X. Scientific bases of acupuncture analgesia. Acupunct Electrother Res 2002;27:114.
[2] Ernst E, White A. Acupuncutre: a scientific appraisal. Oxford:
ButterworthHeinemann; 1999. p. 1.
[3] Carlsson C. Acupuncture mechanisms for clinically relevant longterm effectsreconsideration and a hypothesis. Acupunct Med
2002;20:8299.
[4] Kim HW, Kwon YB, Ham TW, Roh DH, Yoon SY, Lee HJ, et al.
Acupoint stimulation using bee venom attenuates formalin-induced
pain behavior and spinal cord fos expression in rats. J Vet Med Sci
2003;65:34955.
[5] Kwon YB, Kang MS, Han HJ, Beitz AJ, Lee JH. Visceral antinociception produced by bee venom stimulation of the Zhongwan

188

[6]
[7]
[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

H.W. Kim et al. / Pharmacological Research 51 (2005) 183188

acupuncture point in mice: role of alpha(2) adrenoceptors. Neurosci
Lett 2001;308:1337.
Zimmermann M. Ethical guidelines for investigations of experimental pain in conscious animals. Pain 1983;16:109.
Yaksh TL, Rudy TA. Analgesia mediated by a direct spinal action
of narcotics. Science 1976;192:13578.
Lee JH, Kim HW, Kwon YB, Kang MS, Choi DW, Na JH, et al.
General pharmacology studies on beta-domain deleted recombinant
factor VIII. Arzneimittelforschung 2000;50:8692.
Yu LC, Lu JT, Huang YH, Meuser T, Pietruck C, Gabriel A, et
al. Involvement of endogenous opioid systems in nociceptin-induced
spinal antinociception in rats. Brain Res 2002;945:8896.
Liu NJ, Gintzler AR. Gestational and ovarian sex steroid antinociception: relevance of uterine afferent and spinal alpha(2)-noradrenergic
activity. Pain 1999;83:35968.
Rueter LE, Meyer MD, Decker MW. Spinal mechanisms underlying A-85380-induced effects on acute thermal pain. Brain Res
2000;872:93101.
Chen SR, Eisenach JC, Pan HL. Intrathecal S-nitroso-Nacetylpenicillamine and l-cysteine attenuate nerve injury-induced
allodynia through noradrenergic activation in rats. Neuroscience
2000;101:75965.
Hurley RW, Banfor P, Hammond DL. Spinal pharmacology of
antinociception produced by microinjection of mu or delta opioid receptor agonists in the ventromedial medulla of the rat. Neuroscience
2003;118:78996.
Miller JF, Proudfit HK. Antagonism of stimulation-produced
antinociception from ventrolateral pontine sites by intrathecal administration of alpha-adrenergic antagonists and naloxone. Brain Res
1990;530:2034.
Iwamoto ET, Marion L. Adrenergic, serotonergic and cholinergic
components of nicotinic antinociception in rats. J Pharmacol Exp
Ther 1993;265:77789.
Tjolsen A, Berge OG, Hunskaar S, Rosland JH, Hole K. The formalin test: an evaluation of the method. Pain 1992;51:517.

[17] Han JS. Acupuncture: neuropeptide release produced by electrical

stimulation of different frequencies. Trends Neurosci 2003;26:17
22.
[18] Cheng RS, Pomeranz B. Monoaminergic mechanism of electroacupuncture analgesia. Brain Res 1981;215:7792.
[19] Takeshige C, Sato T, Mera T, Hisamitsu T, Fang J. Descending pain
inhibitory system involved in acupuncture analgesia. Brain Res Bull
1992;29:61734.
[20] Ulett GA, Han S, Han JS. Electroacupuncture: mechanisms and clinical application. Biol Psychiatry 1998;44:12938.
[21] Takagi J, Yonehara N. Serotonin receptor subtypes involved in
modulation of electrical acupuncture. Jpn J Pharmacol 1998;78:
5114.
[22] Altman S. Techniques and instrumentation. Probl Vet Med
1992;4:6687.
[23] Felhendler D, Lisander B. Pressure on acupoints decreases postoperative pain. Clin J Pain 1996;12:3269.
[24] He LF, Dong WQ, Wang MZ. Effects of iontophoretic etorphine
and naloxone, and electroacupuncture on nociceptive responses from
thalamic neurones in rabbits. Pain 1991;44:8995.
[25] Kwon YB, Han HJ, Beitz AJ, Lee JH. Bee venom acupoint stimulation increases fos expression in catecholaminergic neurons in the
rat brain. Mol Cells 2004;17:32933.
[26] Millan MJ. Descending control of pain. Prog Neurobiol
2002;66:355474.
[27] Kawasaki Y, Kumamoto E, Furue H, Yoshimura M. Alpha2
adrenoceptor-mediated presynaptic inhibition of primary afferent glutamatergic transmission in rat substantia gelatinosa neurons. Anesthesiology 2003;98:6829.
[28] Gougat J, Ferrari B, Sarran L, Planchenault C, Poncelet M,
Maruani J, et al. SSR240612 [(2R)-2-[((3R)-3-(1,3-benzodioxol-5yl)-3-[ [(6-methoxy-2-naphthyl)sulfonyl]amino]propanoyl)amino]-3(4-[ [2R,6S]-2,6-dimethylpiperidinyl]methyl]phenyl)-N-isopropyl-Nmethylpropanamide hydrochloride], a new nonpeptide antagonist of
the brad. J Pharmacol Exp Ther 2004;309:6619.