Suppression of the morphine-induced rewarding effect in the rat with neuropathic pain: implication of the reduction in µ-opioid receptor functions in the ventral tegmental area

Authors


Address correspondence and reprint requests to Dr Tsutomu Suzuki, Department of Toxicology, School of Pharmacy, Hoshi University, 2-4-41 Ebara, Shinagawa-ku, Tokyo 142–8501, Japan. E-mail: suzuki@hoshi.ac.jp

Abstract

The present study was designed to investigate the rewarding effect, G-protein activation and dopamine (DA) release following partial sciatic nerve ligation in the rat. Here we show for the first time that morphine failed to produce a place preference in rats with nerve injury. Various studies provide arguments to support that the mesolimbic dopaminergic system, which projects from the ventral tegmental area (VTA) to the nucleus accumbens (N.Acc), is critical of the motivational effects of opioids. In the present study, there were no significant differences between sham-operated and sciatic nerve-ligated rats in the increases in guanosine-5′-o-(3-[35S]thio)triphosphate ([35S]GTPγS) binding to membranes of the N.Acc stimulated by either DA, the D1 receptor agonist SKF81297, the D2 receptor agonist N-propylnoraporphine or the D3 receptor agonist 7-hydroxy-2-dipropylaminotetralin (7-OH DPAT). In contrast, the increases in [35S]GTPγS binding to membranes of the VTA induced by either morphine or a selective µ-opioid receptor agonist [d-Ala2, NMePhe4, Gly(ol)5]enkephalin were significantly attenuated in nerve-ligated rats as compared with sham- operated rats. Furthermore, the enhancement of DA release in the N.Acc stimulated by morphine was significantly suppressed by sciatic nerve ligation. These findings suggest that attenuation of the morphine-induced place preference under neuropathic pain may result from a decrease in the morphine-induced DA release in the N.Acc with reduction in the µ-opioid receptor-mediated G-protein activation in the VTA.

Abbreviations used
aCSF

artificial cerebrospinal fluid

DA

dopamine

DAMGO

[d-Ala2, NMePhe4, Gly(ol)5]enkephalin

ERK

extracellular signal regulated kinase

GRK2

G protein-coupled receptor kinase 2

GTPγS

guanosine-5′-o-(3-[35S]thio) triphosphate

N.Acc

nucleus accumbens

7-OH DPAT

7-hydroxy-2-dipropylaminotetralin

PLCγ

phospholipase Cγ

TH

tyrosine hydroxylase

VTA

ventral tegmental area.

At the present time, pharmacological and biochemical evidences support the existence of three major classes of opioid receptors (µ, δ and κ). In contrast, dopamine (DA) receptors are divided into two subclasses, i.e. the D1-like receptors (D1 and D5) and the D2-like receptors (D2, D3 and D4). Opioid receptors and DA receptors belong to the superfamily of receptors that regulate G-proteins. The ability of each receptor to regulate different classes of G-proteins causes the multiple G-protein signaling pathways. The stimulation of opioid receptors that couple with Gi/Go-protein types has been shown to inhibit forskolin-stimulated cyclic AMP accumulation, activate an inwardly rectifying K+ channel and close voltage-sensitive Ca2+ channel (Standifer and Pasternak 1997). The D1-like receptors activate adenylyl cyclase through coupling with Gs-protein types, while the D2-like receptors inhibit it through Gi/Go-protein types (Kebabian and Calne 1979). It has been documented that the agonist-induced guanosine-5′-o-(3-[35S]thio)triphosphate ([35S]GTPγS), the non-hydrolyzable analog of GTP, binding to G-proteins constitutes a valuable tool for the study of the functional properties of these receptors (Lazareno 1997; Narita et al. 1999).

The mesolimbic DAergic system, projecting from the ventral tegmental area (VTA) of the midbrain to the nucleus accumbens (N.Acc), has been identified as the critical substrate of the reinforcing effects of morphine (Funada et al. 1995; Narita et al. 2001). We previously reported that the enhancement of DA turnover in the limbic forebrain area containing the N.Acc induced by morphine was associated with the expression of place preference by morphine (Funada et al. 1993). We also demonstrated that intra-VTA administration of a selective µ-opioid receptor agonist [d-Ala2, NMePhe4, Gly(ol)5]enkephalin (DAMGO) caused a dose-related preference for the drug-associated place (Narita et al. 2001). Opioid receptor agonists have been shown to increase DAergic signals in the N.Acc via the activation of DA cells in the VTA, an area that possesses high densities of µ-opioid receptors (Garzon and Pickel 2001). This activation may result mainly from the inhibition of inhibitory GABAergic interneurons in the VTA. This contention can be supported by the finding that morphine and the selective µ-opioid receptor agonist DAMGO inhibit the firing frequency of non-DA cells in the VTA (Johnson and North 1992; Bonci and Williams 1997). Furthermore, we previously found that intra-VTA microinjection of DAMGO reduced the release of GABA in the VTA (Narita et al. 2001). These findings strongly suggest that the activation of the µ-opioid receptor in the VTA may facilitate the mesolimbic DAergic system and increase the extracellular DA levels in the N.Acc through the inhibition of GABAergic neurotransmission in the VTA, resulting in the initiation of a reinforcing effect induced by µ-opioids.

Recent clinical studies have demonstrated that when morphine is used to control pain, psychological dependence is not a major concern. We previously reported that morphine failed to induce rewarding effects in inflammatory rats with formalin or carrageenan in the hind paw (Suzuki et al. 1996; Suzuki et al. 1999). These findings suggest the possibility that being in a state of pain could lead to physiological changes in neurotransmission at supraspinal levels, which could be responsible for the decrease in opioid dependence. In fact, inflammatory nociception have been proposed to activate endogeous κ-opioidergic systems in the brain, resulting in the suppression of the morphine-induced rewarding effect (Suzuki et al. 1999).

Neuropathic pain can elicit abnormal pain characterized in part by hyperalgesia, such that noxious stimuli are perceived as more painful, and allodynia, such that normally innocuous stimuli elicit pain. Neuropathic pain is particularly difficult to treat in the clinic, as it is often only partially relieved by high doses of opioids. Many studies have focused on the long-term changes in functions of the spinal cord dorsal horn neurons, containing some receptors, protein kinases and peptides following nerve injury (Nichols et al. 1995; Mayer et al. 1999; Narita et al. 2000). However, little information is available regarding changes in function at the supraspinal level after nerve injury. The aim of present study was to assess whether morphine could produce rewarding effects in the sciatic nerve-ligated rat, and whether sciatic nerve ligation could affect the morphine-induced DA release in the N.Acc. We also determined whether nerve injury could change the DA and opioid receptor-mediated G-protein activation in brain membranes.

Materials and methods

The present study was conducted in accordance with the Guiding Principles for the Care and Use of Laboratory Animals, Hoshi University, as adopted by the Committee on Animal Research of Hoshi University, which is accredited by the Ministry of Education, Science, Sports and Culture of Japan.

Animals

Male Sprague–Dawley rats (250–300 g) were obtained from Tokyo Animal Laboratories (Tokyo, Japan). The animals were housed at a room temperature of 22°C ± 1°C with a 12-h light–dark cycle (light on 08 : 00 h to 20 : 00 h). Food and water were available ad libitum.

Nerve injury pain models

The rats were anesthetized with sodium pentobarbital (50 mg/kg, intraperietoneally) in diethyl ether. We produced a partial nerve injury model that approximately one-third to one-half of the sciatic nerve located on the right side (ipsilateral side) was ligated by tying a tight ligature with a 7–0 silk suture as previously described (Seltzer et al. 1990). In sham-operated rats, the nerve was exposed, but not ligated.

Measurement of thermal hyperalgesia

Prior to the testing of the behavioral responses to thermal stimuli, rats were habituated to the test environment for 30 min. To assess heat thermal sensitivity, each of the hind paws of the rats was measured individually using a thermal stimulus apparatus (Biological Research Apparatus Type 7370, UGO BASILE, Varese, Italy). Paw withdrawal latency was determined as the average of two measurements per paw. Only quick hind paw movements (with or without licking of the hind paw) away from the stimulus were considered as a withdrawal response. Paw movements associated with locomotion or weight shifting were not counted as a response, and the rats were re-tested for that trial. Just before sham operation or nerve ligation, and 1, 2, 3, 4, 7 and 10 days after the surgery, withdrawal latencies of the operated (ipsilateral) and unoperated (contralateral) paws from the thermal stimulus were measured.

Place conditioning

Place conditioning studies were conducted using the apparatus consisted of a shuttle box (width 30 cm × ∼ length 60 cm ×∼ height 30 cm) which was made of an acrylic resin board and divided into two equal-sized compartments. One compartment was white with a textured floor and the other was black with a smooth floor to create equally inviting compartments. Conditioning sessions (three for morphine: three for saline) were started at 4 days after operation (sham or ligation) and conducted once daily for 6 days. Immediately after subcutaneous injection of morphine (4 or 8 mg/kg), animals were placed in one compartment for 1 h. On alternate days, animals receiving with vehicle were placed in the other compartment for 1 h. The order of injection (morphine or saline) and compartment (white or black) was counter-balanced (unbiased) across the subjects. On day 7, tests of conditioning were performed as follows: the partition separating the two compartments was raised to 12 cm above the floor, and a neutral platform was inserted along the seam separating the compartments. The rats were not treated with either morphine or saline, and then placed on the platform. The time spent in each compartment during a 900-s session was then recorded automatically using an infrared beam sensor (KN-80, Natsume Seisakusyo Co., Tokyo, Japan). All sessions were conducted under conditions of dim illumination (28 lux lamp) and white masking noise.

[35S]GTPγS binding assay

For the membrane preparation, the VTA and N.Acc regions were removed from rats at 4 days after the surgery, and rapidly transferred to a tube filled with an ice-cold buffer. According to the atlas of Paxions and Watson, the N.Acc and VTA samples were obtained with dissecting spatula from coronal brain slice from 0.0 to 2.0 mm anterior and from 5.0 to 7.0 mm posterior to bregma, respectively. The tissue was homogenized using a Potter-Elvehjem tissue grinder with a Teflon pestle in 20 volumes (w/v) of an ice-cold Tris-Mg2+ buffer containing 50 mm Tris-HCl (pH 7.4) for 5 mm MgCl2 and 1 mm EGTA. The homogenate was centrifuged at 4°C for 10 min at 48 000 g. The pellet was resuspended in [35S]GTPγS binding assay buffer containing 50 mm Tris-HCl (pH 7.4), 5 mm MgCl2, 1 mm EGTA and 100 mm NaCl and centrifuged at 4°C for 10 min at 48 000 g. The resultant pellet was resuspended in [35S]GTPγS binding assay buffer and stored at − 70°C until use. The membrane homogenate (3–8 µg protein/assay) was incubated at 25°C for 2 h in 1 mL of assay buffer with various concentrations of the agonists, 30 µm guanosine-5′-diphosphate (GDP) and 50 pm[35S]GTPγS (specific activity, 1000 Ci/mmol; Amersham, Arlington Heights, IL, USA). The reaction was terminated by the filtration using Whatman GF/B glass filters. The filters were washed three times and then transferred to scintillation-counting vials containing tissue solubilizer (Soluene-350, Packard Instrument Company, Meriden, CT) and scintillation cocktail (Hionic Fluor, Packard Instrument Company). The radioactivity in the samples was determined with a liquid scintillation analyser. Non-specific binding was measured in the presence of 10 µm unlabeled GTPγS. Comparable results were obtained from at least three independent sets of experiments.

In vivo microdialysis

The rats were mounted in a stereotaxic frame after sham operation or sciatic nerve ligation. The skull was exposed and a small hole was made using a dental drill. Guide cannula (AG-8, EICOM CORP., Japan) was implanted into N.Acc (A/p + 1.5 mm, L−1.5 mm, V−7.0 mm) according to the atlas of Paxions and Watson 1998). The guide cannula was fixed to the skull with cranioplastic cement.

Four days after the surgery, the microdialysis probe (AI-8–2; 2 mm membrane length, EICOM CORP., Japan) was slowly inserted into the N.Acc through the guide cannula under anesthesia with diethyl ether. After rats had been woken, they were then settled in the experimental cages (width 30 cm × ∼ depth 30 cm × ∼ height 30 cm). The rats were habituated for 4–5 h after the probe insert. The probe was perfused continuously at a flow rate of 2 µL/min with artificial cerebrospinal fluid (aCSF) containing 147.0 mm NaCl, 4.0 mm KCl and 3.0 mm CaCl2. The outflow fractions were collected every 30 min. Following the subsequent collection of three baseline fractions, rats were treated with subcutaneous injection of morphine (8 mg/kg) or saline (1 mL/kg). For these experiments, dialysis samples were collected for 210 min after morphine or saline treatment. Dialysis fractions were then analyzed using high-performance liquid chromatography (HPLC) (EICOM CORP., Japan) with an electrochemical detection (ECD) system. The animals were killed by decapitation under pentobarbital anesthesia at the end of the experiments. The brains were removed and the localization of the probes was confirmed by staining with cresyl violet.

The DA was separated by a column with a mobile phase containing sodium acetate (4.05 g/L), citric acid monohydrate (7.35 g/L), sodium 1-octane sulfonate (170 mg/L), EDTA(2Na) (10 mg/L) and 15% methanol. The mobile phase was delivered at a flow rate of 0.23 mL/min. Identification of DA was determined according to the retention times of a DA standard, and the amounts of DA were quantified by calculating peak area.

The baseline microdialysis data were calculated as concentrations in the dialysates. Other microdialysis data were expressed as percentages of the corresponding baseline level.

Statistical analysis

The data are expressed as the mean ± SEM. The statistical significance of differences between the groups was assessed with a one-way anova, followed by Dunnett's test.

Results

The hyperalgesic response after nerve injury is shown in Fig. 1. There was no difference in the paw withdrawal latencies between sham-operated and sciatic nerve-ligated rats before surgery (0 day). The withdrawal latencies of the ipsilateral paw to the heat thermal stimulus were maximally decreased at 4 days after sciatic nerve ligation, and this reduction in paw-withdrawal latencies lasted for at least 10 days. In contrast, paw withdrawal latencies in the ipsilateral side of sham-operated mice were not changed. No differences in the response thresholds were noted in the contralateral side of the two groups.

Figure 1.

Effect of sciatic nerve ligation on withdrawal responses to thermal stimulation. There was no difference in the basal response between sham-operated and sciatic nerve-ligated rats before surgery (day 0). Thermal hyperalgesia was only observed in the ipsilateral side of sciatic nerve-ligated rats. The data are presented as the mean time ± SEM of six rats.

Under these conditions, we investigated whether sciatic nerve ligation could affect the place conditioning induced by morphine. The subcutaneous injection of morphine produced a dose-related preference for the drug-associated place in sham-operated rats (4 mg/kg: p < 0.05 versus saline groups; 8 mg/kg: p < 0.001 versus saline groups) (Fig. 2). In contrast, the morphine-induced place preference was significantly suppressed in sciatic nerve-ligated group (p < 0.05 versus sham groups).

Figure 2.

The place preference produced by subcutaneous administration of morphine (4 or 8 mg/kg) in sham-operated and sciatic nerve-ligated rats using conditioned place preference assay. Ordinate: mean difference(s) between times spent in the morphine- and saline-paired sides of the test box. Immediately after subcutaneous injection of morphine or saline, the rat was placed and conditioned in either compartment for 1 h. The data represent the mean ± SEM of 12–16 rats. *p < 0.05, ***< 0.001 versus saline groups. #< 0.05 versus sham groups.

Since it has been demonstrated that the activation of the mesolimbic DA system is critically linked to the expression of the rewarding effects of morphine, we examined whether the suppression of the rewarding effect induced by morphine after nerve injury could result from changes in the DA receptor functions which activate G-proteins in the N.Acc (Fig. 3). The ability of DA to activate G-proteins in the N.Acc of sham-operated and sciatic nerve-ligated rats was examined by monitoring the binding of [35S]GTPγS to N.Acc membranes. DA (0.1–10 µm) produced a concentration-dependent increase in [35S]GTPγS binding to N.Acc membranes from both sham-operated and sciatic nerve-ligated rats with the same degree (Fig. 3a). No differences in the increase of [35S]GTPγS binding stimulated by either the selective D1 receptor agonist SKF81297, the D2 receptor agonist N-propylnoraporphine or the D3 receptor agonist 7-hydroxy-2-dipropylaminotetralin (7-OH DPAT) were noted between two groups (Fig. 3b).

Figure 3.

(a) Effect of dopamine (DA) on [35S]GTPγS binding to membranes of the nucleus accumbens obtained from sham-operated and sciatic nerve-ligated rats. (b) Effect of the selective D1 receptor agonist SKF81297 (SKF), the D2 receptor agonist N-propylnoraporphine (NPA) and the D3 receptor agonist 7-OH DPAT (DPAT) on [35S]GTPγS binding to nucleus accumbens membranes of sham-operated and sciatic nerve-ligated rats. Membranes were incubated with 50 pm[35S]GTPγS and 30 µm GDP with and without different concentrations (0.1–10 µm) of DA, 10 µm SKF, 10 µm NPA or 10 µm DPAT for 2 h at 25°C. The data are expressed as the percentage of basal [35S]GTPγS binding measured in the presence of GDP and absence of agonist. The data represent the mean ± SEM from at least three independent experiments.

We next investigated whether sciatic nerve ligation could affect the G-protein activation by the stimulation of opioid receptors in membranes of the rat VTA (Fig. 4). Morphine (0.1–10 µm) produced a concentration-dependent increase in [35S]GTPγS binding to VTA membranes obtained from sham-operated rats. In VTA membranes obtained from sciatic nerve-ligated rats, the significant lower level of [35S]GTPγS binding stimulated by morphine was observed as compared to that in sham-operated rats (10−7m: p < 0.05 versus sham groups; 10−6m, 10−5m: p < 0.001 versus sham groups) (Fig. 4a). Since morphine has the affinity for not only µ- but also δ- and κ-opioid receptors, we tried to confirm whether sciatic nerve ligation could result in the specific decrease in the µ-opioid receptor function. Like morphine, the significant lower level of [35S]GTPγS binding by a highly selective µ-opioid receptor agonist DAMGO (10 µm) in sciatic nerve-ligated rats was detected as compared in sham-operated rats (p < 0.001 versus sham group) (Fig. 3b). Under these conditions, there were no significant differences in [35S]GTPγS binding stimulated by either the selective δ-opioid receptor agonist SNC80 or κ-opioid receptor agonist U-50,488H between two groups (Fig. 4b).

Figure 4.

(a) Effect of morphine on [35S]GTPγS binding to membranes of the ventral tegmental area (VTA) obtained from sham-operated and sciatic nerve-ligated rats. (b) Effect of the selective µ-opioid receptor agonist DAMGO, δ-opioid receptor agonist SNC80 (SNC) and κ-opioid receptor agonist U-50,488H (U50) on [35S]GTPγS binding to VTA membranes of sham-operated and sciatic nerve-ligated rats. Membranes were incubated with 50 pm[35S]GTPγS and 30 µm GDP with and without different concentrations (0.1–10 µm) of morphine, 10 µm DAMGO, 10 µm SNC or 10 µm U50 for 2 h at 25°C. The data are expressed as the percentage of basal [35S]GTPγS binding measured in the presence of GDP and absence of agonist. The data represent the mean ± SEM from at least three independent experiments. *p < 0.05,***p < 0.001 versus sham groups.

In the microdialysis study, basal levels of DA in the N.Acc were not differed among all groups (sham-saline, 0.86 ± 0.47 nm; sham-morphine, 0.82 ± 0.27 nm; ligation-saline, 0.90 ± 0.22 nm; ligation-morphine, 0.79 ± 0.28 nm). The effects of subcutaneous administration of morphine on the extracellular levels of DA in the rat N.Acc are shown in Fig. 5. The DA levels were markedly increased by subcutaneous injection of morphine at 8 mg/kg as compared with that by saline treatment in sham-operated rats. However, sciatic nerve ligation significantly suppressed the increased levels of extracellular DA in the N.Acc stimulated by morphine (p < 0.05 versus sham group).

Figure 5.

Effects of treatment with morphine on the extracellular level of dopamine (DA) in the nucleus accumbens of sham-operated and sciatic nerve ligated rats. Morphine (8 mg/kg, subcutaneous) or saline (1 mL/kg, subcutaneous) was injected at time 0. The data are expressed as percentages of the corresponding baseline levels with SEM of six rats. *< 0.05 versus sham-morphine group.

Discussion

In the present study, we demonstrated that the subcutaneously administered morphine-induced place preference was attenuated under neuropathic pain following partial sciatic nerve ligation in the rat. Furthermore, sciatic nerve ligation resulted in the diminishment of µ-, but not δ- and κ-opioid receptor functions to activate G-protein in the VTA. On the other hand, no significant changes in DA receptor functions to activate G-protein in the N.Acc were observed in sciatic nerve-ligated rats. These data suggest the possibility that the reduction in µ-opioid receptor-mediated G-protein activation in the VTA may be responsible for the inhibition of rewarding effects induced by morphine.

The mesolimbic DAergic pathway projecting from the VTA to the N.Acc is thought to play a major role in mediating the rewarding effects of many stimuli, such as electrical brain stimulation and drugs of abuse (Funada et al. 1995; Koob et al. 1998; Narita et al. 2001). Since in vivo microdialysis and electrophysiological studies have provided evidence for an indirect activation of mesolimbic DAergic neurotransmissions by morphine, the stimulation of DA receptors in the N.Acc by the increased release of DA induced by morphine may be an essential process of the expression of the morphine-induced rewarding effect (Matthews et al. 1984; Narita et al. 2001). In fact, treatment with a selective D1 receptor antagonist SCH23390 has been shown to block the morphine-induced place preference (Shippenberg and Herz 1988; Suzuki et al. 1993). A recent study using mice with a genetic disruption of the D2 receptors has demonstrated the involvement of D2 receptors in the rewarding effect produced by morphine (Maldonado et al. 1997). Like the D2 receptor, a growing body of evidences indicates that the D3 receptor is implicated in some behavior elicited by drugs of abuse including morphine (Suzuki et al. 1995; Carta et al. 2000). Considering these backgrounds, we had postulated that changes in the DA receptor function may contribute to the suppression of rewarding effect induced by morphine following nerve injury. However, we clearly demonstrated in this study that animals with sciatic nerve ligation failed to produce such changes in the DA receptor function in the rat N.Acc.

It is of interest to note that sciatic nerve ligation produced a specific inhibition of µ-opioid receptor-mediated G-protein activation in the VTA which possesses high densities of µ-opioid receptors and plays a critical role in psychological dependence on µ-opioids. Therefore, it is most likely that the reduction in µ-opioid receptor functions in the VTA may be a key factor in inhibition of the morphine-induced place preference. This contention can be strongly supported by the present finding that sciatic nerve ligation caused the inhibition of the morphine-induced DA release in the N.Acc. In addition, no change in the basal level of DA was observed following sciatic nerve ligation. These findings indicate that nerve injury leads to the reduction in the µ-opioidergic system in the VTA, resulting in the suppression of the morphine-induced DA release in the N.Acc.

We have proposed that a correlation exists between µ-opioid receptor-mediated G-protein activity and behavioral responses (Mizoguchi et al. 1999; Narita et al. 1999). In the biochemical and behavioral studies using mice with different gene levels of µ-opioid receptors, both G-protein activity and some pharmacological effects produced by µ-opioid receptor agonists were gene-dose dependent. In the present study, we observed only 30–40% decreases in [35S]GTPγS binding to VTA membranes stimulated by either morphine or DAMGO in sciatic nerve-ligated rats. However, the rewarding effect of morphine was almost eliminated in these rats. Although the specific reason for the present phenomenon remains unclear, a hypothesis might be explained by the µ-opioid receptor subtypes. A considerable number of biochemical and pharmacological studies have provided evidences that µ-opioid receptors consist of, at least, µ1- and µ2-opioid receptor subtypes (Pasternak 2001). In a previous study, we demonstrated that the morphine-induced rewarding effect can be mediated mainly by µ2-opioid receptors, because morphine produced a place preference in µ1-opioid receptor-deficient CXBK mice and in mice pretreated with the µ1-opioid receptor antagonist naloxonazine (Suzuki et al. 1993). Although further study is required, it is possible that the decrease in levels of [35S]GTPγS binding to VTA membranes by morphine following sciatic nerve ligation may result from the reduction in the µ2-, but not µ1-, opioid receptor function, resulting in the suppression of the rewarding effect induced by morphine. This hypothesis of involvement of µ-opioid receptor subtypes can be supported by our recent findings that µ-opioid receptor agonists produced a significantly detectable G-protein activation in CXBK mice, and the remaining activities in this strain were abolished by µ-opioid receptor antagonists (Mizoguchi et al. 2000).

Another possibility is that nerve injury may cause reduction of µ-opioid receptor-mediated signaling downstream. It has been demonstrated that chronic morphine treatment results in the various biochemical adaptations in the VTA (Nestler 1996). The direct changes in the expression and/or functions of some neurofilament proteins, extracellular signal regulated kinase (ERK), phospholipase Cγ (PLCγ), some glutamate receptors and tyrosine hydroxylase (TH) on the VTA DAergic neurons have been observed by repeated morphine exposure (Beitner-Johnson et al. 1992; Berhow et al. 1996; Wolf et al. 1999). These alternations are thought to be implicated in the development of psychological dependence on morphine. If this is the case, we cannot exclude the possibility that the inhibition of the motivational effect induced by morphine after the sciatic nerve ligation may result from the less induction of µ-opioid receptor-associated signaling under a state of neuropathic pain.

In conclusion, the present data suggest that a state of neuropathic pain induced by sciatic nerve ligation leads to the reduction in the µ-opioid receptor function in the VTA. This effect produces a significant decrease in the morphine-induced DA release in the N.Acc, resulting in the inhibition of the rewarding effect of morphine in rats. These findings strongly indicate that treatment of morphine could be highly recommended for the relief of severe chronic pain.

Acknowledgements

This work was supported in part by grants from the Ministry of Health, Labour and Welfare, and the Ministry of Education, Culture, Sports, Science and Technology of Japan.

Ancillary