Suppressive Effects of Bee Venom Acupuncture on Paclitaxel-Induced Neuropathic Pain in Rats: Mediation by Spinal α2-Adrenergic Receptor

Paclitaxel, a chemotherapy drug for solid tumors, induces peripheral painful neuropathy. Bee venom acupuncture (BVA) has been reported to have potent analgesic effects, which are known to be mediated by activation of spinal α-adrenergic receptor. Here, we investigated the effect of BVA on mechanical hyperalgesia and spinal neuronal hyperexcitation induced by paclitaxel. The role of spinal α-adrenergic receptor subtypes in the analgesic effect of BVA was also observed. Administration of paclitaxel (total 8 mg/kg, intraperitoneal) on four alternate days (days 0, 2, 4, and 6) induced significant mechanical hyperalgesic signs, measured using a von Frey filament. BVA (1 mg/kg, ST36) relieved this mechanical hyperalgesia for at least two hours, and suppressed the hyperexcitation in spinal wide dynamic range neurons evoked by press or pinch stimulation. Both melittin (0.5 mg/kg, ST36) and phospholipase A2 (0.12 mg/kg, ST36) were shown to play an important part in this analgesic effect of the BVA, as they significantly attenuated the pain. Intrathecal pretreatment with the α2-adrenergic receptor antagonist (idazoxan, 50 µg), but not α1-adrenergic receptor antagonist (prazosin, 30 µg), blocked the analgesic effect of BVA. These results suggest that BVA has potent suppressive effects against paclitaxel-induced neuropathic pain, which were mediated by spinal α2-adrenergic receptor.


Introduction
Paclitaxel is an important chemotherapeutic agent from the bark of Taxus brevifolia [1], which is widely used to treat various tumors [2][3][4].However, despite its role against the tumors, its usage is often limited, due to the painful peripheral neuropathy occurring after its administration [5].Symptoms commonly reported are sensory neuropathies, which are paresthesia, loss of tendon reflexes, numbness and pain in the upper and lower extremities.Although these neuropathies decrease Toxins 2017, 9, 351 2 of 12 patients' quality of life (QoL), there is still no optimal treatment method or drug to alleviate these neuropathies [5,6].Thus, an effort to explore novel treatments is needed.
Bee venom acupuncture (BVA), a treatment method that injects diluted bee venom into acupoints, is widely used in traditional Korean medicine against various diseases, such as adhesive capsulitis [7], idiopathic Parkinson's disease [8], knee osteoarthritis [9], and musculoskeletal pain diseases [10].Especially, BVA has been reported to have potent analgesic effect in studies conducted using various animal models of pain [11][12][13][14][15], and two case series also reported that BVA treatment may help to reduce the chemotherapy-induced peripheral neuropathy (CIPN), including paclitaxel-induced neuropathy [16,17].Recently, our laboratory has demonstrated that BVA treatment could significantly alleviate mechanical and cold allodynia in a rat model of oxaliplatin-induced neuropathic pain [12,14,18,19].Moreover, although the precise mechanism of BVA analgesic effect is unknown, we have also demonstrated that this analgesic effect was mediated by the descending noradrenergic pain modulation pathway via the activation of spinal α-adrenergic receptor, which was consistent with other previously conducted studies [11][12][13][14]20].
Thus, the aims of this study were, firstly, to examine whether the BVA has suppressive effects against paclitaxel-induced mechanical hyperalgesia and neuronal hyperexcitation in the spinal cord, and secondly, to observe the role of BVA components, such as melittin and phospholipase A2 (PLA2) in their analgesic effect, and finally, to investigate which α-adrenergic receptor subtypes mediate the analgesic effect of BVA in the spinal cord.

Development and Maintanance of Paclitaxel-Induced Mechanical Hyperalgesia
In order to see the time-elapsed change of paclitaxel-induced mechanical hyperalgesia, we evaluated the withdrawal responses of hind paws to mechanical stimulation using a von Frey filament with 15 g bending force.In the paclitaxel group, significant increase in paw withdrawal frequency (PWF) was shown from 10 to 21 days after the first injection (p < 0.01, day 10 and 14; p < 0.05, day 21) (Figure 1).Therefore, we performed the following experiments on day 10 through to 21.
Toxins 2017, 9, 351 2 of 12 decrease patients' quality of life (QoL), there is still no optimal treatment method or drug to alleviate these neuropathies [5,6].Thus, an effort to explore novel treatments is needed.Bee venom acupuncture (BVA), a treatment method that injects diluted bee venom into acupoints, is widely used in traditional Korean medicine against various diseases, such as adhesive capsulitis [7], idiopathic Parkinson's disease [8], knee osteoarthritis [9], and musculoskeletal pain diseases [10].Especially, BVA has been reported to have potent analgesic effect in studies conducted using various animal models of pain [11][12][13][14][15], and two case series also reported that BVA treatment may help to reduce the chemotherapy-induced peripheral neuropathy (CIPN), including paclitaxelinduced neuropathy [16,17].Recently, our laboratory has demonstrated that BVA treatment could significantly alleviate mechanical and cold allodynia in a rat model of oxaliplatin-induced neuropathic pain [12,14,18,19].Moreover, although the precise mechanism of BVA analgesic effect is unknown, we have also demonstrated that this analgesic effect was mediated by the descending noradrenergic pain modulation pathway via the activation of spinal α-adrenergic receptor, which was consistent with other previously conducted studies [11][12][13][14]20].
Thus, the aims of this study were, firstly, to examine whether the BVA has suppressive effects against paclitaxel-induced mechanical hyperalgesia and neuronal hyperexcitation in the spinal cord, and secondly, to observe the role of BVA components, such as melittin and phospholipase A2 (PLA2) in their analgesic effect, and finally, to investigate which α-adrenergic receptor subtypes mediate the analgesic effect of BVA in the spinal cord.

Development and Maintanance of Paclitaxel-Induced Mechanical Hyperalgesia
In order to see the time-elapsed change of paclitaxel-induced mechanical hyperalgesia, we evaluated the withdrawal responses of hind paws to mechanical stimulation using a von Frey filament with 15 g bending force.In the paclitaxel group, significant increase in paw withdrawal frequency (PWF) was shown from 10 to 21 days after the first injection (P < 0.01, day 10 and 14; P < 0.05, day 21) (Figure 1).Therefore, we performed the following experiments on day 10 through to 21.

Effects of BVA on Paclitaxel-Induced Mechanical Hyperalgesia
Since the BVA treatment at Zusanli (ST36) acupoint, but not at Quchi (L11), showed significant anti-hyperalgesic effect (Figure 2), BVA was used at ST36 in the following experiments.Figure 3 Figure 1.Time course of paclitaxel-induced mechanical hyperalgesia.Rats were divided into two groups; paclitaxel (n = 7), vehicle (n = 7).Paclitaxel (2.0 mg/kg per injection) or vehicle was injected to rats four times (arrows; days 0, 2, 4 and 6).Significant differences between two groups were observed from the day 10 to day 21.Data are presented as mean ± SEM (* p < 0.05, ** p < 0.01; two-way ANOVA followed by Bonferroni's multiple comparison test).

Effects of BVA on Paclitaxel-Induced Mechanical Hyperalgesia
Since the BVA treatment at Zusanli (ST36) acupoint, but not at Quchi (L11), showed significant anti-hyperalgesic effect (Figure 2), BVA was used at ST36 in the following experiments.Figure 3 Toxins 2017, 9, 351 3 of 12 shows the analgesic effects of BVA on paclitaxel-induced mechanical hyperalgesia with the time course.BVA treated group (paclitaxel + BVA) showed significant reduction in PWF compared to control group (paclitaxel + PBS (phosphate buffered saline)) at one and two hours after BVA (47% reduction, p < 0.05 and 66% reduction, p < 0.01, respectively).No significant difference between the two groups was shown from four hours after BVA.These results indicate that the treatment of BVA has a potent analgesic effect on paclitaxel-induced neuropathic pain, lasting at least two hours.shows the analgesic effects of BVA on paclitaxel-induced mechanical hyperalgesia with the time course.BVA treated group (paclitaxel + BVA) showed significant reduction in PWF compared to control group (paclitaxel + PBS (phosphate buffered saline)) at one and two hours after BVA (47% reduction, P < 0.05 and 66% reduction, P < 0.01, respectively).No significant difference between the two groups was shown from four hours after BVA.These results indicate that the treatment of BVA has a potent analgesic effect on paclitaxel-induced neuropathic pain, lasting at least two hours.Rats were dispensed arbitrarily into two groups; paclitaxel + BVA (n = 7), paclitaxel + phosphate buffered saline (PBS) (n = 7).BVA (1.0 mg/kg) and PBS were treated at ST36.Significant reduction of PWF was observed from one to two hours after BVA.Data are presented as mean ± SEM (* P < 0.05, ** P < 0.01; two-way ANOVA followed by Bonferroni's multiple comparison test).

Effects of BVA on Paclitaxel-Induced Hyperexcitation in the Spinal Wide Dynamic Range (WDR) Neurons
In order to see whether paclitaxel induces hyperexcitation in WDR neurons and BVA treatment reduces paclitaxel-induced hyperexcitation in WDR neurons, we conducted extracellular recording  shows the analgesic effects of BVA on paclitaxel-induced mechanical hyperalgesia with the time course.BVA treated group (paclitaxel + BVA) showed significant reduction in PWF compared to control group (paclitaxel + PBS (phosphate buffered saline)) at one and two hours after BVA (47% reduction, P < 0.05 and 66% reduction, P < 0.01, respectively).No significant difference between the two groups was shown from four hours after BVA.These results indicate that the treatment of BVA has a potent analgesic effect on paclitaxel-induced neuropathic pain, lasting at least two hours.

Effects of BVA on Paclitaxel-Induced Hyperexcitation in the Spinal Wide Dynamic Range (WDR) Neurons
In order to see whether paclitaxel induces hyperexcitation in WDR neurons and BVA treatment reduces paclitaxel-induced hyperexcitation in WDR neurons, we conducted extracellular recording Rats were dispensed arbitrarily into two groups; paclitaxel + BVA (n = 7), paclitaxel + phosphate buffered saline (PBS) (n = 7).BVA (1.0 mg/kg) and PBS were treated at ST36.Significant reduction of PWF was observed from one to two hours after BVA.Data are presented as mean ± SEM (* p < 0.05, ** p < 0.01; two-way ANOVA followed by Bonferroni's multiple comparison test).

Effects of BVA on Paclitaxel-Induced Hyperexcitation in the Spinal Wide Dynamic Range (WDR) Neurons
In order to see whether paclitaxel induces hyperexcitation in WDR neurons and BVA treatment reduces paclitaxel-induced hyperexcitation in WDR neurons, we conducted extracellular recording in vivo (Figure 4a-d).The number of spike responses of WDR neurons to mechanical stimulation (brush, press, and pinch) was significantly increased in paclitaxel group (p < 0.05; brush, p < 0.001; press and pinch, vs. vehicle, Figure 4e).In the BVA treatment group (1 mg/kg, ST36), significant reduction of paclitaxel-induced hyperexcitation in WDR neurons was observed (p < 0.01; press, p < 0.001; pinch, vs. before BVA, Figure 4f). in vivo (Figure 4a-d).The number of spike responses of WDR neurons to mechanical stimulation (brush, press, and pinch) was significantly increased in paclitaxel group (P < 0.05; brush, P < 0.001; press and pinch, vs. vehicle, Figure 4e).In the BVA treatment group (1 mg/kg, ST36), significant reduction of paclitaxel-induced hyperexcitation in WDR neurons was observed (P < 0.01; press, P < 0.001; pinch, vs. before BVA, Figure 4f).

Effect of BVA, Melittin, or PLA2 on Paclitaxel-Induced Mechanical Hyperalgesia
To observe the role of different BV components in the analgesic effect of the BVA, BVA (1 mg/kg), melittin (0.5 mg/kg), or PLA2 (0.12 mg/kg) were injected at ST36.The two major protein components of the honey bee are melittin and PLA2, which occupies 50 and 12% of its dry weight, respectively [21].Behavioral assessments were conducted one hour after the injection of BVA, melittin, or PLA2, as BVA showed its strongest analgesic effect one hour after the injection (Figure 3).This result showed that melittin had a stronger analgesic effect against paclitaxel-induced mechanical hyperalgesia than BVA or PLA2 (Figure 5).

Effect of BVA, Melittin, or PLA2 on Paclitaxel-Induced Mechanical Hyperalgesia
To observe the role of different BV components in the analgesic effect of the BVA, BVA (1 mg/kg), melittin (0.5 mg/kg), or PLA2 (0.12 mg/kg) were injected at ST36.The two major protein components of the honey bee are melittin and PLA2, which occupies 50 and 12% of its dry weight, respectively [21].Behavioral assessments were conducted one hour after the injection of BVA, melittin, or PLA2, as BVA showed its strongest analgesic effect one hour after the injection (Figure 3).This result showed that melittin had a stronger analgesic effect against paclitaxel-induced mechanical hyperalgesia than BVA or PLA2 (Figure 5).

Discussion
Multiple injection of paclitaxel can occur peripheral neuropathy, which can limit its usage and decreases patients' QoL.Although the treatments such as gabapentin, pregabalin, and morphine have been used to alleviate the neuropathic pain, these treatments have, themselves, various side effects, such as nausea, vomiting, somnolence, dizziness, suicidal thought, and drug dependence [22][23][24][25].Therefore, an effort to search for effective treatment options is critically needed.In traditional Korean medicine, BVA has been used to treat musculoskeletal pain and arthritis from the past [10,26].In addition, these days, BVA has also been founded to be effective in treating patients with CIPN [16,17].Thus, in this study, we experimented to find out whether BVA can alleviate the paclitaxel-induced neuropathy and to clarify the mechanism that lies behind it.
Our data showed that BVA treatment at ST36, not LI11, had a significant analgesic effect.It should be noted that ST36 acupoint is closer to the hind paw, where mechanical test was performed, than LI11 acupoint.It is consistent with the previous study in which BVA had more potent analgesic effect when treated closer to the tested area [14].Then, we examined the time course of the analgesic effect of the BVA at ST36.The result showed that the analgesic effect was significant until two hours after BVA treatment.Our previous study also showed that the analgesic effect of BVA was effective until two hours after BVA treatment in oxaliplatin-induced cold allodynia [14].Considering that moderate concentration of morphine without side effects was no longer effective in oxaliplatin-induced cold allodynia at two hours after administration [12,27], this result would be clinically significant.
The spinal wide dynamic range (WDR) neuron receives non-nociceptive and nociceptive inputs via A-and C-fibers, and descending pain modulatory systems synapse at the WDR neuron [28].Therefore, the spinal WDR neuron is suitable for assessing the degree of pain.In addition, the hyperexcitation of spinal WDR neuron was observed previously in a rat model of paclitaxel-induced hyperalgesia [29].In our study, electrophysiological data confirmed that hyperexcitation of WDR neurons is induced by paclitaxel.We further demonstrated that BVA treatment could significantly inhibit this paclitaxel-induced hyperexcitation in the spinal WDR cells.
In subsequent experiments, we administered BVA, melittin, or PLA2 at ST 36, to observe the role of different BV components in the analgesic effect of BVA against paclitaxel-induced mechanical hyperalgesia.Melittin is a major component of the BV, occupying 50% of its total dry weight.PLA2 occupies 12%.Our results showed that 0.5 mg/kg of melittin was more powerful than 1 mg/kg of BVA or 0.12 mg/kg of PLA2.In our previous study, we showed that intraperitoneal injection of PLA2 could significantly decrease the cold and mechanical allodynia induced by single oxaliplatin injection in mice [30].Moreover, although not on chemotherapy induced pain model, other lab has reported that melittin injected at ST36 had a powerful analgesic effect against complete Freund's adjuvant-induced rheumatoid arthritis, showing a similar effect to BVA [31].In this study, the analgesic effect of BVA or melittin was blocked by spinal α 2 -adrenergic receptor antagonist (idazoxan), showing that BVA and melittin act on similar spinal adrenergic receptors to inhibit mechanical hyperalgesia induced by paclitaxel.
EA (electro-acupuncture) is a modified acupuncture which utilizes electrical current to treat pain.BVA is another form of acupuncture which uses chemical compounds; bee venom.The two different forms of acupuncture have similarities and differences.One of the similarities is that the endogenous analgesic systems are involved in both of their analgesic mechanisms, and the difference is that the analgesic effects of EA are mainly mediated by the opioidergic system [32], whereas those of the BVA are mostly mediated by the noradrenergic system [33].However, despite this difference, EA and BVA were both reported to be effective in different types of allodynia assessed using thermal [34] and chemical [12] stimulations.These results show that other inhibitory systems, such as serotonergic, GABA, and/or cholinergic systems, may also play an important role, along with opioidergic and adrenergic system, in the action of EA and BVA.Furthermore, interaction of periaqueductal gray (PAG) and locus coeruleus (LC) in the brain should also play an important part in their analgesic effect, as both the EA and BVA were reported to activate PAG [35] and LC [36], which are important opioid and noradrenaline producing site in the CNS, respectively.
BVA induced analgesia was shown to be mediated by spinal α 2 -adrenergic receptor [11][12][13][14][15], and it increased c-Fos expression in LC and A5 cell group (A5) [37,38].Moreover, BVA reduced c-Fos expression in the spinal dorsal horn of rats with formalin or acetic acid-induced pain [13,39].Considering that both the LC and A5 are part of the descending noradrenergic pathway [40], it is suggested that BVA suppresses conduction of afferent nociceptive signals in the spinal dorsal horn affecting descending noradrenergic pathway.Our data are consistent with previous studies showing that spinal α 2 -adrenergic receptor mediates BVA-induced analgesia.Furthermore, the pain attenuating effect of melittin was also blocked by spinal α 2 -adrenergic receptor antagonist (idazoxan) showing that melittin, the richest component of the BV, also acts on spinal α 2 -adrenergic receptor to inhibit mechanical hyperalgesia induced by paclitaxel.
Drug combination is widely used to treat dreadful diseases, such as AIDS and cancer.The main aim of drug combination is to reduce dose and toxicity, and to delay the induction of drug resistance.Our previous study showed the combined effect of BVA and morphine on oxaliplatin-induced neuropathic pain [12].BVA treated with morphine showed prolonged analgesic effects compared to the BVA or morphine alone.Moreover, another article showed that BVA could enhance the analgesic effect of intrathecal injection of clonidine in chronic constriction injury-induced neuropathic pain model [41].Because such combined effect on paclitaxel-induced neuropathic pain has yet to be studied, further studies are needed to examine the combined effect of BVA with other drugs, like morphine, clonidine, SSRI, SNRI, gabapentin, and cannabinoid.Furthermore, in the future studies, it will be interesting to investigate the effect of various components of the BVA on paclitaxel-induced neuropathic pain model, as several active components exist in the BV, such as melittin [42] and PLA2 [30], which have been reported to be effective in other pain models.

Conclusions
In conclusion, BVA (1 mg/kg) at ST36 significantly attenuated mechanical hyperalgesia induced by paclitaxel.The significant analgesic effect lasted two hours, which was long enough compared to the effect of morphine.Suppressive action was verified by conducting extracellular recording in the spinal WDR neurons.Moreover, both melittin (0.5 mg/kg) and PLA2 (0.12 mg/kg), which are major components of the BV, significantly attenuated the paclitaxel-induced mechanical hyperalgesia.This analgesic effect of BVA or melittin was significantly blocked by intrathecal injection of idazoxan, but not by prazosin, demonstrating that the action of spinal α 2 -adrenergic receptor, but not α 1 -adrenergic receptor, is involved in the mechanism of analgesic effect.

Animals
Adult Sprague-Dawley rats (male, 180-210 g, 6 weeks old) (Daehan Biolink, Chungbuk, Korea) were housed in cages with free access to food and water, and were sustained at 23 ± 2 • C room temperature with a 12 hour light/dark cycle.Prior to any experiments, all animals were acclimated in their cages (3-4 rats per cage) for a week.All experiments using animals were ratified by the Institutional Animal Care and Use Committee of Kyung Hee University (KHUASP(SE)-16-153), and were performed on the ground of the guidelines of the International Association for the Study of Pain [43].

Administration of Paclitaxel
Paclitaxel (Wako Pure Chemical Industries, Osaka, Japan) was dissolved in cremophor EL polyethoxylated castor oil (Sigma, St. Louis, MO, USA) and 100% ethanol (Merck KGaA, Marmstadt, Germany) (1:1 solution), and 6 mg/mL stocks were made.Then, stocks were diluted by phosphate buffered saline (PBS) at a concentration of 2 mg/ml and administrated at an amount of 2 mg/kg on four alternate days (days 0, 2, 4, and 6).As control, the same volume of vehicle was intraperitoneally injected.The formula of paclitaxel was slightly modified from previous studies [44,45].

Behavior Tests
Twenty to thirty minutes before the behavior test, animals were adapted to the experimental circumstances.The experimenters were blinded to paclitaxel and any other treatments.The animals were placed on a metal mesh, enclosed within a 20 (d) × 20 (w) × 14 (h) cm clear plastic cage.Mechanical hyperalgesia was assessed using von Frey filament (Stoelting Co., Wood Dale, IL, USA).The measurement method of mechanical hyperalgesia was modified from the previous studies [44][45][46].On the mid-plantar area of both hind paws, the von Frey filament (bending force of 15 g) was stimulated for 10 times each, with the applications held for 5 s.The percentage of withdrawal responses to the von Frey filament application was calculated, and then expressed as an overall percentage response.

Experimental Schedule
The time schedule of this experiment is shown in Figure 7.After baseline mechanical sensitivity was measured at day 0, paclitaxel was injected intraperitoneally on four alternate days (days 0, 2, 4, and 6) (Figure 7a).Behavior tests were performed after paclitaxel administration.The time course of BVA effect was measured at 1, 2, 4, and 6 hours after administration of BVA (Figure 7b).Antagonists were treated 20 min before BVA, and then, behavior tests were conducted 1 hour later (Figure 7c).
Twenty to thirty minutes before the behavior test, animals were adapted to the experimental circumstances.The experimenters were blinded to paclitaxel and any other treatments.The animals were placed on a metal mesh, enclosed within a 20 (d) × 20 (w) × 14 (h) cm clear plastic cage.Mechanical hyperalgesia was assessed using von Frey filament (Stoelting Co., Wood Dale, IL, USA).The measurement method of mechanical hyperalgesia was modified from the previous studies [44][45][46].On the mid-plantar area of both hind paws, the von Frey filament (bending force of 15 g) was stimulated for 10 times each, with the applications held for 5 s.The percentage of withdrawal responses to the von Frey filament application was calculated, and then expressed as an overall percentage response.

Experimental Schedule
The time schedule of this experiment is shown in Figure 6.After baseline mechanical sensitivity was measured at day 0, paclitaxel was injected intraperitoneally on four alternate days (days 0, 2, 4, and 6) (Figure 6a).Behavior tests were performed after paclitaxel administration.The time course of BVA effect was measured at 1, 2, 4, and 6 hours after administration of BVA (Figure 6b).Antagonists were treated 20 min before BVA, and then, behavior tests were conducted 1 hour later (Figure 6c).

BVA, Melittin, or PLA2 Treatment
To verify the optimal acupoint for the BVA treatment, paclitaxel administered rats were divided randomly into two groups; Quchi (LI11) and Zusanli (ST36).LI11 is located at the depression medial to the extensor carpi radialis, at the lateral end of the cubital crease.ST36 is located in the anterior tibial muscle, 5 mm lateral and distal from the anterior tubercle of the tibia [47].
BV was manufactured by Jayeonsaeng TJ (Kyeonggi-Do, Korea), and its quality is strictly controlled by regular HPLC analysis (SNU National Instrumentation Center for Environmental Management, Seoul, Korea; see Supplementary Materials Figure S1).BV (1.0 mg/kg), as reported as an effective concentration without side effects from a previous study [12], dissolved in PBS was respectively injected at right side LI11 or ST36 acupoints subcutaneously, after baseline mechanical

BVA, Melittin, or PLA2 Treatment
To verify the optimal acupoint for the BVA treatment, paclitaxel administered rats were divided randomly into two groups; Quchi (LI11) and Zusanli (ST36).LI11 is located at the depression medial to the extensor carpi radialis, at the lateral end of the cubital crease.ST36 is located in the anterior tibial muscle, 5 mm lateral and distal from the anterior tubercle of the tibia [47].
BV was manufactured by Jayeonsaeng TJ (Kyeonggi-Do, Korea), and its quality is strictly controlled by regular HPLC analysis (SNU National Instrumentation Center for Environmental Management, Seoul, Korea; see Supplementary Materials Figure S1).BV (1.0 mg/kg), as reported as an effective concentration without side effects from a previous study [12], dissolved in PBS was respectively injected at right side LI11 or ST36 acupoints subcutaneously, after baseline mechanical sensitivity was measured.The mechanical behavior test was performed following time course schedule (Figure 7b).Melittin (0.5 mg/kg) and PLA2 (0.12 mg/kg) were also injected at ST36.All drugs injected at acupoints were injected subcutaneously.

Extracellular Recording
Extracellular recordings were made from animals 10-21 days following administration of paclitaxel, when rats exhibited significant mechanical hyperalgesia.Extracellular recordings were

Figure 3 .
Figure 3.Time course of the analgesic effect of BVA on paclitaxel-induced mechanical hyperalgesia.Rats were dispensed arbitrarily into two groups; paclitaxel + BVA (n = 7), paclitaxel + phosphate buffered saline (PBS) (n = 7).BVA (1.0 mg/kg) and PBS were treated at ST36.Significant reduction of PWF was observed from one to two hours after BVA.Data are presented as mean ± SEM (* P < 0.05, ** P < 0.01; two-way ANOVA followed by Bonferroni's multiple comparison test).

Figure 3 .
Figure 3.Time course of the analgesic effect of BVA on paclitaxel-induced mechanical hyperalgesia.Rats were dispensed arbitrarily into two groups; paclitaxel + BVA (n = 7), paclitaxel + phosphate buffered saline (PBS) (n = 7).BVA (1.0 mg/kg) and PBS were treated at ST36.Significant reduction of PWF was observed from one to two hours after BVA.Data are presented as mean ± SEM (* P < 0.05, ** P < 0.01; two-way ANOVA followed by Bonferroni's multiple comparison test).

Figure 3 .
Figure 3.Time course of the analgesic effect of BVA on paclitaxel-induced mechanical hyperalgesia.Rats were dispensed arbitrarily into two groups; paclitaxel + BVA (n = 7), paclitaxel + phosphate buffered saline (PBS) (n = 7).BVA (1.0 mg/kg) and PBS were treated at ST36.Significant reduction of PWF was observed from one to two hours after BVA.Data are presented as mean ± SEM (* p < 0.05, ** p < 0.01; two-way ANOVA followed by Bonferroni's multiple comparison test).

Figure 6 .
Figure 6.Time schedule of the experiment.(a) Paclitaxel was administered four alternate days (0, 2, 4, 6 days, i.p.); (b) the time course of BVA effect was conducted at 1, 2, 4, and 6 hours after administration of BVA; (c) antagonists were treated 20 min before administration of BVA or melittin, and behavior tests were conducted one hour after administration of BVA or melittin.

Figure 7 .
Figure 7. Time schedule of the experiment.(a) Paclitaxel was administered four alternate days (0, 2, 4, 6 days, i.p.); (b) the time course of BVA effect was conducted at 1, 2, 4, and 6 hours after administration of BVA; (c) antagonists were treated 20 min before administration of BVA or melittin, and behavior tests were conducted one hour after administration of BVA or melittin.