European Journal of Pharmacology 594 (2008) 125–131

Contents lists available at ScienceDirect

European Journal of Pharmacology

j ourna l homepage: www.e lsev ie r.com/ locate /e jphar

Melatonin reverses the expression of morphine-induced conditioned placepreference through its receptors within central nervous system in mice

Jing Han 1, Ying Xu 1, Chang-Xi Yu ⁎, Jie Shen, Yi-Ming WeiDepartment of Pharmacology, Fujian Medical University, Fuzhou 350004, Fujian, People's Republic of China

⁎ Corresponding author. Tel.: +86 591 83569311; fax:E-mail address: changxiyu@mail.fjmu.edu.cn (C.-X. Y

1 The first two authors contributed equally to this wo

0014-2999/$ – see front matter © 2008 Elsevier B.V. Aldoi:10.1016/j.ejphar.2008.07.049

a b s t r a c t

a r t i c l e i n f o

Article history:

A growing body of eviden Received 23 February 2008Received in revised form 10 July 2008Accepted 23 July 2008Available online 31 July 2008

Keywords:Conditioned place preferenceMelatoninMorphineCentral nervous systemMelatonin receptor subtypeReward(Mouse)

ce indicates the prominent actions of melatonin on the opioidergic system.Nevertheless, effect of melatonin on rewarding properties of morphine is still obscure. In particular, effect ofmelatonin on the expression of morphine reward is unknown. We evaluated the effect of exogenousadministration of melatonin on the expression of morphine reward in mice using a conditioned placepreference (CPP) paradigm. The conditioned place preference was induced by morphine (s.c., 3 mg/kg, onceeach day for 5 consecutive days) in mice. Our data showed that the intraperitoneal (i.p.) administration ofmelatonin (12.5–50 mg/kg) reversed the expression of morphine-induced conditioned place preference in adose-dependent manner. Furthermore, the intracerebroventricular (i.c.v.) administration of melatonin (0.125–0.5mg/kg) also resulted in dose-dependent reversal effect on the expression ofmorphine-induced conditionedplace preference. We further investigated which of melatonin receptor subtypes within the central nervoussystem was mediating this reversal action in mice using luzindole (2-benzyl-N-acetyltryptamine, a non-selective antagonist for melatonin MT1 and MT2 receptors) and K185 {N-butanoyl-2-(5,6,7-trihydro-11-methoxybenzo[3,4]cyclohept[2,1-α]indol-13-yl)ethanamine, a selective antagonist for melatonin MT2receptor}. It was shown that the i.c.v. administration of either K185 (5, 20 μg) or luzindole (6.25, 12.5 μg)significantly antagonized the reversal effect of melatonin (50 mg/kg, i.p) on the expression of morphine-induced conditioned place preference, while the i.c.v. administration of 20 μg of K185 or 12.5 μg of luzindole byitself did not alter the expression of morphine-induced conditioned place preference. These results suggestthat melatonin reverses the expression of morphine-induced rewarding effect, and this action is mediated bythe activation of melatonin MT2 receptor subtype within the central nervous system.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

Melatonin (N-acetyl-5-methoxytryptamine), an indoleamine neu-rohormone, which is synthesized and secreted primarily by the pinealgland in all mammalian species, has many effects on a wide range ofbiological functions (Pandi-Perumal et al., 2006). Besides its wellestablished function of regulation of circadian rhythms and seasonalresponses, melatonin has been involved in some neuropsychophar-macological actions, such as the sedative/hypnotic, anticonvulsant andantinociceptive activity (Munoz-Hoyos et al., 1998; van den Heuvelet al., 2005; Yu et al., 2000b).Melatoninproduces effects via avariety ofmechanisms in a receptor independent and dependent manner(Boutin et al., 2005; Rodriguez et al., 2004; Tomas-Zapico and Coto-Montes, 2005; Witt-Enderby et al., 2006). Melatonin receptors arefound in a number of tissues, including the central nervous system andperipheral tissues. Relatively abundant melatonin receptors are

+86 591 83569311.u).rk.

l rights reserved.

present in several brain regions, particularly in brain nuclei involvedin the control of behavior (Ekmekcioglu, 2006). Acting throughreceptors is an important mechanism of melatonin to exert its effects(Witt-Enderby et al., 2006). Two subtypes of mammalian melatoninreceptors have been cloned and characterized, the melatonin MT1(Mel1a) and MT2 (Mel1b) receptor subtypes (Alexander et al., 2007;Boutin et al., 2005). To date, quite a few of selective agonists andantagonists on melatonin receptor subtypes have been developed(Alexander et al., 2007; Boutin et al., 2005; Delagrange et al., 2003). Forexample, luzindole (2-benzyl-N-acetyltryptamine) is a non-selectiveantagonist formelatoninMT1 andMT2 receptors,with some selectivityfor melatonin MT2 receptor (Dubocovich, 1988); K185 {N-butanoyl-2-(5,6,7-trihydro-11-methoxybenzo[3,4]cyclohept[2,1-α]indol-13-yl)ethanamine} is the selective antagonist for melatonin MT2 receptor(Sugden et al., 1999). They have been widely used to analyze thereceptor subtypes underlyingmelatonin's actions, to provide evidencefor designing subtype-selected specific ligands for treatment.

Addiction caused by drug abuse has been recognized increasingly asa leading cause of death, morbidity and lost productivity. Recently, anincreasing body of evidence suggests that melatonin may have somereversal effects on different drugs of abuse. For instance, melatonin is

126 J. Han et al. / European Journal of Pharmacology 594 (2008) 125–131

shown to reverse the development of tolerance and physical depen-dence to morphine and to inhibit the morphine withdrawal syndromesin mice (Qiu et al., 1998; Raghavendra and Kulkarni, 1999; 2000; Zhouet al., 2002). Our recent preliminary work has shown that melatoninattenuated the withdrawal contraction induced by naloxone in anisolated guinea pig ileum (Wei et al., 2004). In addition, Sircar has shownthat cocaine-induced behavioral sensitization is inhibited by melatoninin mice (Sircar, 2000). Nevertheless, effect of melatonin on rewardingproperties of morphine is still obscure.

Conditioned place preference (CPP) test has been widely used toassess the rewarding properties of drugs with addiction liability(Tzschentke, 2007). This test draws on Pavlovian learning in order toidentify the hedonic properties of drugs and other stimuli. In thisparadigm, animals are trained to associate a specific environmentwith experimenter-administered drugs. Following several such con-ditioning sessions, animals are presented a choice of returning to thedrug-paired or a placebo-paired environment. If the drug under studyis rewarding, then animals will favor the drug-associated environ-ment. Accordingly, the increased time spent in the drug-pairedenvironment is termed “conditioned place preference” (Bardo andBevins, 2000). Morphine produces a significant dose-dependent effecton the magnitude of place preference (Barr et al., 1985). Absorbing,Yahyavi-Firouz-Abadi et al. have shown using a conditioned placepreference paradigm that the combination of melatonin and sub-effective dose of morphine (0.5 mg/kg) led to rewarding effect, whileintraperitoneal administration of melatonin alone did not induceconditioned place preference (Yahyavi-Firouz-Abadi et al., 2007).However, the effect of melatonin on the expression of already acquiredconditioned place preference induced by morphine is unknown.Moreover, whether melatonin activates its receptors within centralnervous system to produce this effect remains to be clarified.

Given this background knowledge, the present study was designedto evaluate the effect of melatonin on the expression of already ac-quired conditioned place preference induced by morphine and inves-tigate the involvement of melatonin receptors within central nervoussystem in this effect inmice.We show for the first time that exogenousadministration of melatonin reverses the expression of morphine-induced conditioned place preference, and that this effect may bemediated through the melatonin MT2 receptor subtype within thecentral nervous system in mice.

2. Materials and methods

2.1. Animals

Male KM mice (Experimental Animal Center, Fujian MedicalUniversity, Fuzhou, China), weighing 18–22 g at the beginning ofexperiments, were used throughout the study. The animals werehoused in groups of six and had free access to standardmouse diet andwater except during behavioral experiments. They were kept in atemperature-controlled room (25±2 °C) under conditions of 12-hlight/12-h dark cycle (lights on from 08:00 to 20:00 h). All mice wereacclimatized for at least 5 days before experiments. Experiments wereperformed during the period from 09:00 to 17:00 h. Each mouse wasused in one experiment only. The study was conducted in compliancewith the international laws on animal experimentation and approvedby the Committee of Ethics of Fujian Medical University, China. Allprocedures were carried out in accordance with the guidelines foranimal care and use in Fujian Medical University, China.

2.2. Drugs

Melatonin was purchased from Sigma-Aldrich Co., St Louis, USA. Itwas dissolved in 5% ethanol saline (v/v) immediately before use.Luzindole and K185 were obtained from Sigma-Aldrich Co., St Louis,USA, and were dissolved in 2.5% Dimethyl Sulfoxide (DMSO) artificial

cerebrospinal fluid (aCSF, v/v) before use. Morphine hydrochloride waspurchased from the First Pharmaceutical Factory of Shenyang, China,and was dissolved in sterile physiological saline (0.9% NaCl). The re-quisite doses of drugs were injected subcutaneously (s.c.) or intraper-itoneally (i.p.) in a volume of 10 ml/kg of the mice body weight.Appropriate vehicle controls were performed for each experiment.

2.3. Apparatus

The place preference apparatus was an acrylonitrile-butadine-styrene (ABS) box divided into two compartments (15 cm×15 cm×40 cm each) of equal size by a wall with a door. To distinguish the twocompartments, visual and sensory texture cues were used. Onecompartment was painted white with smooth floor and the otherwas black with wire mesh floor. A video camera was placed over theboxes and linked to a computer system. Themovement and location ofanimals during the pre-conditioning and test phases were recordedand analyzed by a behavioral analyze software provided by the JiliangSoftware Technology Co Ltd, Shanghai, China. The whole experimentwas conducted under dim illumination and stable noise.

2.4. Surgery and intracerebroventricular injection

Under pentobarbital anesthesia (50 mg/kg, i.p.), the mouse wasimmobilized on a stereotaxic apparatus (Stoelting Co, American). Theskinwas cut along the midline of the skull. A cotton swab soaked with3% H2O2 was used to abrade the tissue until the skull was exposed. Astainless steel guide cannula of 0.6 mm outer diameter was placedstereotaxically and directed to one side of the lateral ventricleaccording to the methods of Zhang et al. (Zhang and Yu, 2002)(1.0 mm posterior to the bregma, 2.0 mm lateral to the midline and2.5 mm from the surface of the skull). The cannula was fixed on theskull with dental resin. To prevent clogging, a stainless steel stylet of0.3 mm outer diameter was put in the guide cannula until the mousewas given the intracerebroventricular (i.c.v.) injection. All animalswere allowed at least 5 days of recovery from surgery before theywereused for the experiments.

For i.c.v. injection, the animal was restrained gently by hand. Thestylet was removed from the guide cannula. A stainless steel injectiontube was inserted into the guide cannula with the lower endprotruding 0.5 mm beyond the lower end of the guide cannula toreach the lateral ventricle. Then 5 μl of drug solution was injected intothe lateral ventricle within 1 min. Control group was injected with thevehicle. The injection needle was left in place for an additional 60 s toallow diffusion. At the end of each experiment, a stainless steel tubingof the same size as the injection tube was inserted into the guidecannula, 5 μl of a solution of pontamine sky blue was injected. Tenminutes later, the mouse was killed and its brain was removed to becut for checking the validity of the i.c.v. injection. Data from animals inwhich the cannula had been incorrectly placed were discarded.

2.5. Conditioned place preference procedure

A biased procedure was used to conduct the conditioned placepreference test, which consisted of three phases: pre-conditioning,conditioning and post-conditioning. The conditioned place preferenceparadigm took place on nine consecutive days. For the pre-condition-ing phase (Days 1–3), the mouse was placed in the white compart-ment and the door between the two compartments was left open toallow free access to the entire box for 15 min each day. On day 2 andday 3, the amount of time spent in each compartment was monitoredand averaged to use as the pre-conditioning time of each animal. Ourpreliminary work showed that there was significant difference be-tween the time spent in the black compartment (622.79±11.74 s,n=55) and the time spent in the white compartment (277.21±11.74 s,n=55) before drug conditioning [F(1, 108)=433.54, Pb0.01], which

Fig.1. Effects of i.p. administration of melatonin on the expression of morphine-inducedconditioned place preference in mice. The saline-paired mice or morphine-paired micewere conditioned as described in the Materials and methods section. On the post-conditioning day, the animals received an i.p. injection ofmelatoninvehicle ormelatonin12.5, 25 and 50 mg/kg, respectively, 20 min before placement of animals in thecompartment. Place preference was tested for 15 min and the CPP score was calculated.Mel: melatonin; veh: melatonin vehicle; NS: physiological saline. Bars represent meanfrom 15 to 18 mice and vertical lines S.E.M. ⁎⁎Pb0.01 compared with saline-paired plusmelatonin vehicle. △△Pb0.01 compared with morphine-paired plus melatonin vehicle.

127J. Han et al. / European Journal of Pharmacology 594 (2008) 125–131

indicated the conditioned place preference apparatus that we usedwas of a biased design (Tzschentke, 2007). The mice showed placepreference for the white box before conditioning (15 of 298 mice)were excluded from further analysis. During the conditioning phase(Days 4–8), the door was shut so that two compartments wereseparated. The mice were divided into morphine-paired group andsaline-paired group, and underwent two conditioning sessions eachday. The first session was performed in the morning, when the micereceived morphine (3 mg/kg, s.c.) in morphine-paired group or sterilephysiological saline (10 ml/kg, s.c.) in saline-paired group, and wereimmediately confined to the white compartment for 1 h. After aninterval of 6 h, the second session of the day began. All the micereceived saline and were confined to the black compartment imme-diately for 1 h during the second session. 24 h after the last drug-paired conditioning trial, the post-conditioning phase (Day 9) wascarried out, and was exactly the same as the pre-conditioning phase.The time the mouse spent in the compartments was recorded for the15-min trial. Conditioned place preference was evaluated as thedifference (in seconds) in post-conditioning vs. pre-conditioning timespent in the drug-paired compartment, which is reported as the “CPPscore” (Gadd et al., 2003).

2.6. Experimental design

2.6.1. Effects of i.p. administration of melatonin on the expression ofmorphine-induced conditioned place preference

In order to assess the effects of systemic administration of mela-tonin on the expression of morphine-induced conditioned place pre-ference, the morphine-paired mice conditioned as described above toestablish conditioned place preference to morphine were divided intofour groups randomly. They were given an i.p. injection of vehicle10 ml/kg (5% ethanol saline, v/v), melatonin 12.5, 25 and 50 mg/kg,respectively, 20 min before placement of animals in the compartmenton the post-conditioning day. Place preference was tested for 15 minand the CPP score was calculated. One additional group of saline-paired mice was injected with the vehicle (10 ml/kg, i.p.) prior toplacement in the apparatus under the same schedule and served as asaline-paired control.

2.6.2. Effects of i.c.v. administration of melatonin on the expression ofmorphine-induced conditioned place preference

Mice after recovery from surgery were conditioned with morphineas described in the conditioned place preference paradigm. In order toevaluate the effects of central administration of melatonin on theexpression of morphine-induced conditioned place preference, themorphine-paired mice were divided into four groups randomly. Onthe post-conditioning day, they received an i.c.v. injection of 5-μlvehicle (5% ethanol aCSF, v/v), melatonin 0.125, 0.25 and 0.5 mg/kg,respectively. Ten minutes later, the place preference was tested for15 min and the CPP score was calculated. One additional group ofsaline-paired mice received an i.c.v. injection of the vehicle and servedas a saline-paired control.

2.6.3. Effect of luzindole on the reversal effect of melatonin on theexpression of morphine-induced conditioned place preference

Mice after recovery from surgery were conditioned with morphineas mentioned above. To investigate the effect of luzindole on thereversal effect of melatonin on the expression of morphine-inducedconditioned place preference, themorphine-pairedmicewere dividedinto four groups randomly. On the post-conditioning day, three groupsof mice were treated with an i.c.v. injection of 5-μl vehicle (2.5% DMSOaCSF, v/v), luzindole 6.25, 12.5 μg, respectively, 10 min before treat-ment with an i.p. injection of melatonin 50 mg/kg, whereas the fourthgroup received an i.c.v. injection of the vehicle (5 μl) and an i.p.injectionwith 5% ethanol saline (10 ml/kg) and served as a morphine-paired control. One additional group of saline-paired mice received an

i.c.v. injection of the vehicle (5 μl) and an i.p. injectionwith 5% ethanolsaline (10ml/kg) and served as a saline-paired control.15min after thei.p. injection, the place preference was tested for 15 min and the CPPscore was calculated.

2.6.4. Effect of K185 on the reversal effect of melatonin on the expressionof morphine-induced conditioned place preference

To assess the effect of K185 on the reversal effect of melatonin onthe expression of morphine-induced conditioned place preference, weperformed the experiment which design similarly to that in the aboveexperiment except that K185 (5, 20 μg) was administered i.c.v. insteadof luzindole (6.25, 12.5 μg).

2.6.5. Effect of luzindole or K185 on the expression of morphine-inducedconditioned place preference

To investigatewhether luzindole or K185 itself affects the expressionof morphine-induced conditioned place preference, the morphine-paired mice conditioned as described above to establish conditionedplace preference tomorphinewere divided into three groups randomly.On the post-conditioning day, they were given an i.c.v. injection of 5-μlvehicle (2.5% DMSO aCSF, v/v), 12.5 μg of luzindole and 20 μg of K185,respectively,10min before placement of animals in the apparatus. Placepreference was tested for 15 min and the CPP score was calculated. Oneadditional group of saline-paired mice received an i.c.v. injection of thevehicle and served as a saline-paired control.

2.6.6. Effect of melatonin on conditioned place preference inductionAn additional experimentwasdesigned to seewhethermelatonin by

itself induces conditioned place preference. The mice were divided intothree groups randomly: vehicle-paired, morphine-paired and melato-nin-paired. The conditioned place preference paradigm underwent asdescribed above. The mice of melatonin-paired group receivedmelatonin (50 mg/kg, i.p.) instead of morphine (3 mg/kg, s.c.) duringthe conditioning phase.

Fig. 3.Effect of luzindoleon the reversal effect ofmelatoninon theexpression ofmorphine-induced conditioned place preference inmice. The saline-pairedmice ormorphine-pairedmice were conditioned as described in the Materials and methods section. On the post-conditioning day, the animals received an i.c.v. injection of luzindole vehicle or luzindole12.5, 6.25 μg, respectively, 10 min before treatment with an i.p. injection of melatoninvehicle ormelatonin 50mg/kg. Fifteenminutes after the i.p. injection, the place preferencewas tested for 15 min and the CPP score was calculated. Luz: luzindole; Mel: melatonin;veh1: luzindole vehicle; veh2: melatonin vehicle; NS: physiological saline. Bars representmean from8 to 9mice and vertical lines S.E.M. ⁎⁎Pb0.01 comparedwith saline-pairedplusluzindole vehicle and melatonin vehicle. △△Pb0.01 compared with morphine-paired plusluzindole vehicle and melatonin vehicle. ##Pb0.01 compared with morphine-paired plusluzindole vehicle and melatonin 50 mg/kg.

128 J. Han et al. / European Journal of Pharmacology 594 (2008) 125–131

2.7. Statistical analysis

All results are presented as the mean±S.E.M. Multiple groupcomparisons of the CPP scores were assessed by one-way analysis ofvariance (ANOVA), followed by LSD-t test. P values less than 0.05 wereconsidered as statistically significant. All analyses were performedusing the SPSS statistical program (version 10.0).

3. Results

3.1. Effects of i.p. administration of melatonin on the expression ofmorphine-induced conditioned place preference

Fig. 1 demonstrates the effects of i.p. administration of melatoninon the expression of morphine-induced place preference in mice. Itwas shown that morphine was able to induce significant placepreference [F (4, 78)=8.104, Pb0.01; t (78)=4.37, Pb0.01] (Fig. 1). Theresults also showed that the i.p. administration of melatoninattenuated the expression of morphine-induced conditioned placepreference in a dose-dependent manner. The CPP score in the grouptreated with melatonin either 25 mg/kg or 50 mg/kg was significantlyless than that in the morphine-paired control [F (4, 78)=8.104,Pb0.01; t (78)=3.07, Pb0.01 and t (78)=3.86, Pb0.01] (Fig. 1). Thesedata indicated that melatonin was likely to reverse the expression ofmorphine-induced place preference in mice.

3.2. Effects of i.c.v. administration of melatonin on the expression ofmorphine-induced conditioned place preference

The effect of i.c.v. administration of melatonin on the expression ofmorphine-induced place preference in mice was shown in Fig. 2. TheCPP score in the morphine-paired control exhibited a significantincrease [F (4, 46)=5.148, Pb0.01; t (46)=3.73, Pb0.01] (Fig. 2),indicating that significant conditioned place preference produced bymorphine had taken place in the mice after recovery from surgery fori.c.v. injection. The i.c.v. administration of melatonin resulted in dose-dependent reduction of CPP scores. The CPP score in the group treatedwithmelatonin either 0.25mg/kg or 0.5mg/kg significantly decreased

Fig. 2. Effects of i.c.v. administration of melatonin on the expression of morphine-induced conditioned place preference in mice. The saline-paired mice or morphine-pairedmicewere conditioned as described in theMaterials andmethods section. On thepost-conditioning day, the animals received an i.c.v. injection of melatonin vehicle ormelatonin 0.125, 0.25 and 0.5 mg/kg, respectively. Ten minutes later, the placepreferencewas tested for 15min and the CPP scorewas calculated.Mel: melatonin; veh:melatonin vehicle; NS: physiological saline. Bars represent mean from 8 to 12 mice andvertical lines S.E.M. ⁎⁎Pb0.01 compared with saline-paired plus melatonin vehicle.△Pb0.05 and △△Pb0.01 compared with morphine-paired plus melatonin vehicle.

[F (4, 46)=5.148, Pb0.01; t (46)=2.49, Pb0.05 and t (46)=3.017,Pb0.01] (Fig. 2). Compared with Fig. 1, the effective dose for melatoninadministered i.c.v. is roughly about 1/100 of that for melatoninadministered i.p.. These data further indicated that melatonin may

Fig. 4. Effect of K185 on the reversal effect of melatonin on the expression of morphine-induced conditioned place preference in mice. The saline-paired mice or morphine-pairedmice were conditioned as described in the Materials and methods section. On the post-conditioning day, the animals received an i.c.v. injection of K185 vehicle or K185 20, 5 μg,respectively, 10 min before treatment with an i.p. injection of melatonin vehicle ormelatonin 50mg/kg. Fifteenminutes after the i.p. injection, the place preferencewas testedfor 15 min and the CPP score was calculated. Mel: melatonin; veh1: K185 vehicle; veh2:melatonin vehicle; NS: physiological saline. Bars represent mean from 8 to 12 mice andvertical lines S.E.M. ⁎Pb0.05 comparedwith saline-paired plus K185 vehicle andmelatoninvehicle.△Pb0.05 comparedwithmorphine-pairedplusK185vehicle andmelatoninvehicle.##Pb0.01 compared with morphine-paired plus K185 vehicle and melatonin 50 mg/kg.

Fig. 6. Effect of melatonin on conditioned place preference induction in mice. Themice were divided into three groups randomly: vehicle-paired, morphine-paired andmelatonin-paired. The conditioned place preference paradigm underwent as describedin the Materials and methods section. The mice of melatonin-paired group receivedmelatonin (50 mg/kg, i.p.) instead of morphine (3 mg/kg, s.c.) during the conditioningphase. Mor: morphine; Mel: melatonin; Veh: melatonin vehicle. Bars represent meanfrom 9 to 10 mice and vertical lines S.E.M. ⁎⁎Pb0.01 compared with vehicle-pairedgroup. Therewas no significant difference betweenmelatonin-paired group and vehicle-paired group.

129J. Han et al. / European Journal of Pharmacology 594 (2008) 125–131

reverse the expression of morphine-induced place preference andimplied that the central nervous system may be the primary site formelatonin to exert the effect.

3.3. Effect of luzindole on the reversal effect of melatonin on theexpression of morphine-induced conditioned place preference

Fig. 3 illustrates the effect of luzindole on the reversal effect ofmelatonin on the expression of morphine-induced conditioned placepreference in mice. Consistent with the result described above,melatonin (50 mg/kg, i.p.) significantly reversed the expression ofmorphine-induced conditioned place preference [F (4, 37)=14.60,Pb0.01; t (37)=3.67, Pb0.01] (Fig. 3). Either 6.25 μg or 12.5 μg ofluzindole administered i.c.v. significantly antagonized the reversaleffect of melatonin [F (4, 37)=14.60, Pb0.01; t (37)=2.91, Pb0.01 and t(37)=4.22, Pb0.01] (Fig. 3). Moreover, i.c.v. administration of 12.5 μg ofluzindole by itself did not alter the expression of morphine-inducedconditioned place preference as shown in Fig. 5. These results suggestthat the reversal effect of melatonin on the expression of morphine-induced place preference may be mediated by the activation ofmelatonin MT1 and MT2 receptors within the central nervous system.

3.4. Effect of K185 on the reversal effect of melatonin on the expression ofmorphine-induced conditioned place preference

Fig. 4 summarizes the effect of K185 on the reversal effect ofmelatonin on the expression of morphine-induced place preference inmice. K185 is a selective antagonist on melatonin MT2 receptor. Either5 μg or 20 μg of K185 administered i.c.v. completely abolished thereversal effect of melatonin on the expression of morphine-inducedconditioned place preference [F (4, 47)=5.707, Pb0.01; t (47)=3.21,Pb0.01 and t (47)=3.36, Pb0.01] (Fig. 4). Furthermore, i.c.v. adminis-tration of 20 μg of K185 alone did not affect the expression ofmorphine-induced conditioned place preference as shown in Fig. 5.The results further suggest that the reversal effect of melatonin on theexpression ofmorphine-induced place preferencemay bemediated by

Fig. 5. Effects of luzindole or K185 on the expression of morphine-induced conditionedplace preference in mice. The saline-paired mice or morphine-paired mice wereconditioned as described in the Materials and methods section. On the post-conditioning day, the animals received an i.c.v. injection of vehicle, luzindole 12.5 μgand K185 20 μg, respectively,10min before placement of animals in the apparatus. Placepreference was tested for 15 min and the CPP score was calculated. Luz: luzindole; veh:vehicle; NS: physiological saline. Bars represent mean from 6 to 7mice and vertical linesS.E.M. ⁎⁎Pb0.01 compared with saline-paired plus vehicle. Neither luzindole nor K185altered the expression of morphine-induced conditioned place preference.

the activation of melatonin MT2 receptor subtype within the centralnervous system.

3.5. Effect of luzindole or K185 on the expression of morphine-inducedconditioned place preference

Fig. 5 shows that neither 12.5 μg of luzindole nor 20 μg of K185administered i.c.v. had any effect on the expression of morphine-induced conditioned place preference in mice.

3.6. Effect of melatonin on conditioned place preference induction

The effect of melatonin on the conditioned place preference in-duction in mice was illustrated in Fig. 6. Statistical analysis indicatedthat the CPP score in the morphine-paired group was significantlyhigher than that in the vehicle-paired group [F (2, 26)=11.546,Pb0.01; t (26)=4.75, Pb0.01] (Fig. 6). Although melatonin seemed toincrease the CPP score, statistical analysis revealed that there was nosignificant difference between the melatonin-paired group and thevehicle-paired group.

4. Discussion

As described above, melatonin possesses some important neurop-sychopharmacological effects. Interestingly, it has been reported thatalteration in light/dark cycle leads to impairment of morphine-in-duced rewarding effect in mice and this effect is reversed by exo-genous administration of melatonin (Tahsili-Fahadan et al., 2005). Agrowing body of evidence indicates the prominent actions of mela-tonin on the opioidergic system. For example, our previous study hasshown that melatonin promotes the release of β-endorphin from theperiaqueductal gray in rats (Yu et al., 2000a). Moreover, melatonin isshown to reverse the development of tolerance and physical depen-dence to morphine and to attenuate the morphine withdrawal syn-dromes. In addition, it produces the analgesic effect in a dose-

130 J. Han et al. / European Journal of Pharmacology 594 (2008) 125–131

dependent manner that is blocked by naloxone (Lakin et al., 1981; Yuet al., 2000b). Therefore, one may surmise that melatonin may havesome effects on rewarding properties of morphine. The aim of thisstudy is to evaluate the effect of melatonin on the expression ofmorphine-induced rewarding effect and investigate the involvementof melatonin receptors within central nervous system in this action.

It is well known that opiate addiction is a major social and medicalproblem that results in immense harm at both individual and societallevel. Despite the size and scope of this problem, there are feweffective treatments for opiate addiction. In the present study, weshow for the first time that the exogenous administration of mela-tonin reversed the expression of morphine-induced conditioned placepreference in a dose-dependent manner. Moreover, the results pre-sented show that melatonin by itself did not induce conditioned placepreference. In addition, Sugden has reported that melatonin possessedvery low acute toxicity when administered tomice and rats by variousroutes (i.p., p.o., s.c., i.v.) (Sugden, 1983). These findings thus appear tosuggest that melatonin without being addictive itself and its analo-gues may have therapeutic potential in the treatment of opiateaddiction. Indeed, this hypothesis needs to be tested in more animalspecies and in more animal models as well as in human beings withappropriate trials. Noticeably, it was reported that i.p. administrationof melatonin (1–40 mg/kg) alone did not induce conditioned placepreference, while the combination of melatonin (5–20 mg/kg) andsub-effective dose of morphine (0.5 mg/kg) resulted in conditionedplace preference (Yahyavi-Firouz-Abadi et al., 2007). However, thestudy didn't show the effect of melatoninwhen co-administratedwiththe effective dose of morphine. Our preliminarywork revealed that i.p.administration of melatonin (i.p., 25–50 mg/kg, once each day for5 days) during the development of conditioned place preference bymorphine (s.c., 3 mg/kg, once each day for 5 days) did not alteracquisition of morphine-induced conditioned place preference inmice (data not shown). Moreover, Huang et al. have reported apreliminary study that melatonin (i.p., 10–20mg/kg, once each day for5 days) may attenuate the development of morphine (s.c., 10 mg/kg,once each day for 5 days) induced conditioned place preference inmice (Huang et al., 1999). The controversy may be due to differentdoses of morphine. Therefore, the effect of melatonin on thedevelopment of morphine-induced rewarding effect remains to beinvestigated.

Melatonin is well known to have sedative effect and to regulatecircadian rhythms. Thus, one may raise one possibility that melatoninchanged the conditioned place preference due to its sedative effectand effect on circadian rhythms. To exclude the possibility, we havedesigned an experiment to assess the effect of melatonin on theexpression of place preference in saline-paired mice. It has beenshown that melatonin at the doses used in the present study (25 mg/kg and 50 mg/kg) did not change the expression of place preference insaline-paired mice (data not shown), suggesting melatonin may nothave significant non-specific influence on the conditioned placepreference. In fact, we observed that the locomotor activity of micetreatedwithmelatoninwas similar to that of mice of morphine-pairedcontrol during test. In nocturnal animals such as mice, night-timemelatonin secretion coincides with the active period, and in thesespecies, only very high pharmacological doses can produce transientsedation. Thus, the doses of melatonin used in the present studychange the conditioned place preference impossibly due to itssedative effect and effect on circadian rhythms.

Relatively abundant melatonin receptors and endogenous mela-tonin are found in several brain regions (Ekmekcioglu, 2006; Pang andBrown, 1983). Furthermore, melatonin can penetrate the blood-brainbarrier. Therefore, it is logical to assume that central nervous systemmay be one of the most important sites for melatonin to exert thereversal effect of expression of morphine-induced conditioned placepreference. The present study revealed that the i.c.v. administration ofmelatonin reversed the expression of morphine-induced conditioned

place preference in a dose-dependent manner. The effective dose formelatonin administered i.c.v. is roughly about 1/100 of that formelatonin administered i.p.. Moreover, our data also indicated that thei.c.v. administration of either luzindole or K185 antagonized signifi-cantly the reversal of expression of morphine-induced conditionedplace preference by melatonin. These results provide a stronger sup-port to the viewpoint that the central nervous system is the primarysite for melatonin to exert the effect.

Melatonin MT1 and MT2 receptor subtypes have been cloned andcharacterized. The functional significances of these subtypes are stillobscure. Recently, an increasing body of reports indicate that the mela-tonin MT1 and/or MT2 receptors may be involved in some importantneurophysiopathology processes (Ekmekcioglu, 2006; Savaskan et al.,2005; Stewart and Leung, 2005; Uz et al., 2005; Yu et al., 2000c). Thestudy presented here demonstrated for the first time that i.c.v.administration of 5 μg of K185 (1.33×10−8 mol) completely antagonizedthe reversal of expression of morphine-induced conditioned placepreference by melatonin. Also, 6.25 μg of luzindole (2.1×10−8 mol)administered i.c.v. significantly blocked the melatonin reversal effect.These results suggest that the reversal action of expressionofmorphine-induced rewarding effect by melatonin may be mediated by theactivation of melatonin MT2 receptor within central nervous system. Itappears to infer that melatonin MT2 receptor would be a novel the-rapeutic target for the development of MT2 receptor-selective agonistsfor the treatmentof opiate-inducedaddiction. Importantly, Uz et al. havefound evidence for the expression of melatonin MT2 receptor mRNA inthe caudate-putamen and nucleus accumbens of the mouse striatum(Uz et al., 2003). As indicated above, there exists the interaction betweenthe opiate system and melatonin. However, we did not find a study onthe co-localisation of opiate system and melatonin MT2 receptor in theregions relevant to the reward system. Indeed, further research isneeded to elucidate the exact relationship between melatonin MT2receptor and opiate addiction.

Interestingly, several lines of evidence suggests that circadianrhythms and clock genes are critically involved in behavioral effects ofpsychoactive drugs including cocaine and morphine, and this actionrelies on the intact pineal gland and endogenous melatonin (Abarcaet al., 2002; Baird and Gauvin, 2000; Kurtuncu et al., 2004; Liu et al.,2005; Manev and Uz, 2006; Uz et al., 2003). For instance, Uz et al.showed that endogenous melatonin is critical for circadian mPeriod1expression in the mouse striatum and for circadian cocaine sensitiza-tion. Also, Liu et al. revealed that mPeriod1 is involved in the deve-lopment of morphine-induced conditioned place preference. Weshow that exogenous administration of melatonin reverses the ex-pression of morphine-induced rewarding effect, and this action maybe mediated through melatonin MT2 receptor within the centralnervous system. These studies will help us better understand themechanisms of drug abuse and identify novel targets for the preven-tion and/or treatment of addictions.

In conclusion, the present study shows for the first time thatmelatonin reverses the expression of morphine-induced rewardingeffect, and this action may be mediated by the activation of melatoninMT2 receptor subtype within the central nervous system. Our findingsappear to infer that melatonin without being addictive itself and itsanalogues may have therapeutic potential in the treatment of opiateaddiction, and that melatonin MT2 receptor subtype would be a noveltherapeutic target for the development of MT2-selective agonists forthe treatment of opiate-induced addiction.

Acknowledgements

This work was financially supported by the Project for the Plan ofScience and Technology Exploitation of Fujian Province of China (No.2005D073), the Natural Science Foundation of Fujian Province ofChina (No. C0410018), and the Project of Science and Technology ofEducation Bureau, Fujian Province, China (A Item, No. JA07089).

131J. Han et al. / European Journal of Pharmacology 594 (2008) 125–131

References

Abarca, C., Albrecht, U., Spanagel, R., 2002. Cocaine sensitization and reward are underthe influence of circadian genes and rhythm. Proc. Natl. Acad. Sci. U. S. A. 99,9026–9030.

Alexander, S.P., Mathie, A., Peters, J.A., 2007. Guide to receptors and channels, 2ndedition (2007 revision). Br. J. Pharmacol. 150 (Suppl 1), S53–S54.

Baird, T.J., Gauvin, D., 2000. Characterization of cocaine self-administration andpharmacokinetics as a function of time of day in the rat. Pharmacol. Biochem.Behav. 65, 289–299.

Bardo, M.T., Bevins, R.A., 2000. Conditioned place preference: what does it add to ourpreclinical understanding of drug reward? Psychopharmacology (Berl) 153, 31–43.

Barr, G.A., Paredes, W., Bridger, W.H., 1985. Place conditioning with morphine andphencyclidine: dose dependent effects. Life Sci. 36, 363–368.

Boutin, J.A., Audinot, V., Ferry, G., Delagrange, P., 2005. Molecular tools to studymelatonin pathways and actions. Trends Pharmacol. Sci. 26, 412–419.

Delagrange, P., Atkinson, J., Boutin, J.A., Casteilla, L., Lesieur, D., Misslin, R., Pellissier, S.,Penicaud, L., Renard, P., 2003. Therapeutic perspectives for melatonin agonists andantagonists. J. Neuroendocrinol. 15, 442–448.

Dubocovich, M.L., 1988. Luzindole (N-0774): a novel melatonin receptor antagonist.J. Pharmacol. Exp. Ther. 246, 902–910.

Ekmekcioglu, C., 2006. Melatonin receptors in humans: biological role and clinicalrelevance. Biomed. Pharmacother. 60, 97–108.

Gadd, C.A., Murtra, P., De Felipe, C., Hunt, S.P., 2003. Neurokinin-1 receptor-expressingneurons in the amygdala modulate morphine reward and anxiety behaviors in themouse. J. Neurosci. 23, 8271–8280.

Huang, M., Zhou, L.H., Li, W.H., Li, Z.Y., Zhang, X.D., Fu, Q.S., 1999. Melatonin attenuatedthe dependence to morphine in mice. Acad. J. Sec. Mil. Med. Univ. 20, 749–751.

Kurtuncu, M., Arslan, A.D., Akhisaroglu, M., Manev, H., Uz, T., 2004. Involvement of thepineal gland in diurnal cocaine reward in mice. Eur. J. Pharmacol. 489, 203–205.

Lakin, M.L., Miller, C.H., Stott, M.L., Winters, W.D., 1981. Involvement of the pineal glandand melatonin in murine analgesia. Life Sci. 29, 2543–2551.

Liu, Y., Wang, Y., Wan, C., Zhou, W., Peng, T., Liu, Y., Wang, Z., Li, G., Cornelisson, G.,Halberg, F., 2005. The role of mPer1 inmorphine dependence inmice. Neuroscience130, 383–388.

Manev, H., Uz, T., 2006. Clock genes: influencing and being influenced by psychoactivedrugs. Trends Pharmacol. Sci. 27, 186–189.

Munoz-Hoyos, A., Sanchez-Forte, M., Molina-Carballo, A., Escames, G., Martin-Medina,E., Reiter, R.J., Molina-Font, J.A., Acuna-Castroviejo, D., 1998. Melatonin's role as ananticonvulsant and neuronal protector: experimental and clinical evidence. J. ChildNeurol. 13, 501–509.

Pandi-Perumal, S.R., Srinivasan, V., Maestroni, G.J., Cardinali, D.P., Poeggeler, B.,Hardeland, R., 2006. Melatonin: nature's most versatile biological signal? FEBS J.273, 2813–2838.

Pang, S.F., Brown, G.M., 1983. Regional concentrations of melatonin in the rat brain inthe light and dark period. Life Sci. 33, 1199–1204.

Qiu, Y., Kang, L., Qiu, X., 1998. The inhibitory effect of melatonin on morphinewithdrawal syndromes and serum monoamines in morphine dependent mice.Zhonghua Yi Xue Za Zhi 78, 704–706.

Raghavendra, V., Kulkarni, S.K., 1999. Reversal of morphine tolerance and dependenceby melatonin: possible role of central and peripheral benzodiazepine receptors.Brain Res. 834, 178–181.

Raghavendra, V., Kulkarni, S.K., 2000. Possible mechanisms of action in melatoninreversal of morphine tolerance and dependence in mice. Eur. J. Pharmacol. 409,279–289.

Rodriguez, C., Mayo, J.C., Sainz, R.M., Antolin, I., Herrera, F., Martin, V., Reiter, R.J., 2004.Regulation of antioxidant enzymes: a significant role for melatonin. J. Pineal Res. 36,1–9.

Savaskan, E., Ayoub, M.A., Ravid, R., Angeloni, D., Fraschini, F., Meier, F., Eckert, A.,Muller-Spahn, F., Jockers, R., 2005. Reduced hippocampal MT2 melatonin receptorexpression in Alzheimer's disease. J. Pineal Res. 38, 10–16.

Sircar, R., 2000. Effect of melatonin on cocaine-induced behavioral sensitization. BrainRes. 857, 295–299.

Stewart, L.S., Leung, L.S., 2005. Hippocampal melatonin receptors modulate seizurethreshold. Epilepsia 46, 473–480.

Sugden, D., 1983. Psychopharmacological effects of melatonin in mouse and rat.J. Pharmacol. Exp. Ther. 227, 587–591.

Sugden, D., Yeh, L.K., Teh, M.T., 1999. Design of subtype selective melatonin receptoragonists and antagonists. Reprod. Nutr. Dev. 39, 335–344.

Tahsili-Fahadan, P., Yahyavi-Firouz-Abadi, N., Ghahremani, M.H., Dehpour, A.R., 2005.Effect of light/dark cycle alteration on morphine-induced conditioned placepreference. NeuroReport 16, 2051–2056.

Tomas-Zapico, C., Coto-Montes, A., 2005. A proposed mechanism to explain thestimulatory effect of melatonin on antioxidative enzymes. J. Pineal Res. 39, 99–104.

Tzschentke, T.M., 2007. Measuring reward with the conditioned place preference (CPP)paradigm: update of the last decade. Addict. Biol. 12, 227–462.

Uz, T., Akhisaroglu, M., Ahmed, R., Manev, H., 2003. The pineal gland is critical forcircadian Period1 expression in the striatum and for circadian cocaine sensitizationin mice. Neuropsychopharmacology 28, 2117–2123.

Uz, T., Arslan, A.D., Kurtuncu, M., Imbesi, M., Akhisaroglu, M., Dwivedi, Y., Pandey, G.N.,Manev, H., 2005. The regional and cellular expression profile of the melatoninreceptor MT1 in the central dopaminergic system. Brain Res. Mol. Brain Res. 136,45–53.

van den Heuvel, C.J., Ferguson, S.A., Macchi, M.M., Dawson, D., 2005. Melatonin as ahypnotic: con. Sleep Med. Rev. 9, 71–80.

Wei, Y.M., Yu, C.X., Xie, J.M., 2004. Effect of melatonin on morphine dependency in theguinea-pig ileum in vitro. J. Fujian Med. Univ. 38, 8–10.

Witt-Enderby, P.A., Radio, N.M., Doctor, J.S., Davis, V.L., 2006. Therapeutic treatmentspotentially mediated by melatonin receptors: potential clinical uses in theprevention of osteoporosis, cancer and as an adjuvant therapy. J. Pineal Res. 41,297–305.

Yahyavi-Firouz-Abadi, N., Tahsili-Fahadan, P., Ghahremani, M.H., Dehpour, A.R., 2007.Melatonin enhances the rewarding properties of morphine: involvement of thenitric oxidergic pathway. J. Pineal Res. 42, 323–329.

Yu, C.X., Wu, G.C., Xu, S.F., Chen, C.H., 2000a. Melatonin influences the release ofendogenous opioid peptides in rat periaqueductal gray. Sheng Li Xue Bao 52,207–210.

Yu, C.X., Zhu, B., Xu, S.F., Cao, X.D., Wu, G.C., 2000b. The analgesic effects of peripheraland central administration of melatonin in rats. Eur. J. Pharmacol. 403, 49–53.

Yu, C.X., Zhu, C.B., Xu, S.F., Cao, X.D., Wu, G.C., 2000c. SelectiveMT(2) melatonin receptorantagonist blocks melatonin-induced antinociception in rats. Neurosci. Lett. 282,161–164.

Zhang, Z., Yu, C.X., 2002. Effect of melatonin on learning and memory impairmentinduced by aluminum chloride and its mechanism. Yao xue xue bao 37, 682–686.

Zhou, Y.H., Huo, Z.Y., Qiu, X.C., 2002. Inhibitory effect of melatonin on morphinewithdrawal syndromes and the content of NO in plasma and brain tissue inmorphine dependent mice. Yao Xue Xue Bao 37, 175–177.