Attenuation of mecamylamine-precipitated nicotine-withdrawal aversion by the 5-HT3 receptor antagonist ondansetron

Life Sciences, VoI. 61, No. 16, pp. PL 24%254,1997 CopyrIght0 l!w7 EIsmier science Inc. Printed in the USA. All rights reserved elm-32as/57 $17.00 t .oo

PII soo24-3205(97)00745-5 PfURM4COLOGY LEilERs Accelerated Gmmunicadon

ATTENUATION WITHDRAWAL

OF MECAMYLAMINE-PRECIPITATED AVERSION BY THE S-HT, RECEPTOR ONDANSETRON

NICOTINEANTAGONIST

Tsutomu Suzuki, Yuya Ise, Tomohisa Mori and Miwa Misawa

Department of Pharmacology, School of Pharmacy, Hoshi University, Shinagawa-ku, Tokyo 142, Japan (Submitted May X3,1997, accepted June 19, lm, received in final form July 15,1997)

Abstract: The effect of ondansetron (0.01-0.1 mg/kg, s.c.), a selective 5HT, receptor antagonist, on mecamylamine-precipitated nicotine-withdrawal aversion was examined in the conditioned place preference paradigm. Male Sprague-Dawley rats were chronically treated subcutaneously with 9 mg/kg/day (-)-nicotine tartrate using an osmotic minipump. After nicotine treatment for 7 days, mecamylamine (1 mg/kg, s.c.), a nicotinic receptor antagonist, produced place aversion in nicotine-dependent rats. This aversive effect was dose-dependently antagonized by pretreatment with ondansetron 30 min prior to the conditioning. These results suggest that ondansetron may attenuate the place aversion associated with nicotine withdrawal, and may be useful for the treatment of nicotine dependence. 8 1997Elsevier Science ~nc.

Key W&is: nicotine, physical dependence, ondansetron, place aversion, conditioned place preference paradigm

Introduction Nicotine addiction, which has been likened to the addiction produced by cocaine or heroin, has become widely accepted as the mechanism which supports chronic tobacco use (US Depment of Health and Human Services), and nicotine-withdrawal produces a withdrawal syndrome, which includes irritability, anxiety, inhibition of concentration, insomnia and craving in nicotine-dependent subjects (1). Although animal models of physical dependence on nicotine are potentially useful for investigating nicotine dependence, there are currently few such models. There are several reports that chronic administration of nicotine can produce physical dependence in animals, which show various withdrawal signs when nicotine is withdrawn. For example, Fung et al. (2) reported that nicotine-dependent rats showed a decrease in locomotor activity after nicotine is withdrawn. ln addition, nicotine-withdrawal leads to an increase in food consumption in nicotine-dependent rats (3). Moreover, Malin et al. (4) reported that the adminis&ation of mecamylamine to rats, which had been chronically treated with nicotine using an osmotic minipump, induced various withdrawal signs, such as gasping/writhing, teeth chatter/chewing, shakes/tremors and ptosis.

All correspondence

to Tsutomu Suzuki, Ph. D.

PGW)

Nicotine Dependence and Ondansetron

Vol. 61, No. 16, 1997

The conditioned place preference paradigm has been demonstrated to be an effective method for assessing the rewarding effects of several abused drugs, such as cocaine, amphetamine, morphine and nicotine (5). The aversive effects of drugs can also be evaluated using this paradigm. For example, the kappa opioid receptor agonist U50,488H produces conditioned place aversion (6). Moreover, antagonist-precipitated withdrawal aversion can be assessed using this method, e.g., naloxone-precipitated morphine-withdrawal aversion (7,8). In fact, we have reported that mecamylamine-precipitated nicotine-withdrawal aversion can be evaluated using the conditioned place preference paradigm, which may reflect physical dependence on nicotine (9). There have been several reports that 5-HT, receptor antagonists attenuate the withdrawal signs of abused drugs in animals. For example, Oakley et al. (10) reported that the 5-HT, receptor antagonist GR38032F (ondansetron) suppresses aggressive behaviors when alcohol is withdrawn in marmosets. In addition, ondansetron may also attenuate the weight loss associated with benzodiazepine withdrawal (11). Moreover, Costall et al. (12) reported that ondansetron attenuates the aversive behavior following withdrawal from chronic benzodiazepine, ethanol, cocaine and nicotine treatment using the light/dark exploration test in mice. In the present study, the effect of the 5-I-H, receptor antagonist ondansetron on mecamylamine-precipitated nicotine-withdrawal aversion in rats that had been chronically treated with nicotine was examined using the conditioned place. preference paradigm.

Methods Animals: Male Sprague-Dawley rats (Tokyo Experimental Animals Co. Ltd., Tokyo, Japan), weighing 270-320 g, were housed in groups of 2-3 in a temperature-controlled room (22 2 1 “C). The animals were maintained on a 12-h light/dark cycle (lights on 8:OO a.m. to 8:00 p.m.) with laboratory rat chow and tap water available ad libitum. Apparatus: Place conditioning was conducted according to the method of Suzuki et al. (13). The apparatus consisted of a shuttlebox (30 x 60 x 30 cm: w x 1 x h) which was divided into two compartments of equal size. One compartment was white with a textured floor and the other was black with a smooth floor. Procedure: On day 1, an osmotic minipump (Alzet 2001, Alza Corporation, CA, USA) with a flow rate of 1.03 c~l/h filled with (-)-nicotine tartrate solution in saline was subcutaneously implanted in rats that had been anesthetized with diethylether. The concentration of nicotine was adjusted for differences in body weight, but was approximately 116 mg/ml, resulting in continuous subcutaneous infusion of nicotine tartrate at the late of 9 mg/kg/day according to the method of Suzuki et al. (9). Non-treated rats received sham operations; they were subjected to the same anesthesia and surgical procedure as the implanted animals except for the implantation of an osmotic minipump. In the morning (9:00 a.m.) on day 7 of nicotine infusion, rats were subcutaneously injected with mecamylamine (0.1-1.0 mg/kg, s.c.) or saline (1.0 ml/kg, s.c.), and immediately confined to one compartment of the test apparatus for 60 min. In the evening (7:00 p.m.) on the same day, rats were then treated with saline or mecamylamine, respectively, and confined to the other compartment for 60 min. The pairings of injection (drug or saline) and compartment (white or black) were counterbalanced across all of the subjects. The control rats in the non-treated and nicotine-treated rats were injected with saline (1.0 ml/kg, s.c.) instead of drug in the conditioning session. After the saline injections, the rats were confined to one compartment in the morning and to the other compartment in the evening. Either black or white place of saline control was regarded

Vol. 61, No. 16,1997

as a substitution experiments.

Nicotine Dependence and Onbtron

for drug-associated

place at random.

It was decided before

PLZl

the start of

Ondansetron (0.01-0.1 mg&g, s.c.) or saline (1.0 ml/kg, s.c.) was administered 30 min prior to the mecamylamine (1 mg/kg, s.c.) injection. Moreover, place conditioning by ondansetron (0.1 ml/kg, s.c.) was determined as well as mecamylamine-induced place conditioning. On the morning of day 8, tests of conditioning were performed as follows: the partition which separated the two compartments was raised to 12 cm above the floor, and a neutral platform was inserted along the seam separating the compartments. The time spent in each compartment during a 900-s session was measured automatically in a blind fashion by an infrared beam sensor (KN-80, Natsume Seisakusho, Tokyo, Japan). The position of the rat was defined by the position of its body. All sessions were conducted under conditions of dim illumination (40 lux) and masking white noise. Drugs: The drugs used in the present study were (-)-nicotine hydrogen tartrate (Sigma Chemical Co., St. Louis, USA), mecamylamine hydrochloride (Sigma Chemical Co.) and ondansetron hydrochloride (Nisshin Flour Milling Co., Saitama, Japan). All drugs were dissolved in saline. Mecamylamine and ondansetron were injected in a volume of 1 .O mlikg. Dutcr analysis: Conditioning scores represent the time spent in the drug-paired place minus the time spent in the vehicle-paired place and are expressed as the mean * S.E.M. Behavioral data were statistically evaluated with a one-way ANOVA followed by Dunnett’s multiple comparison test, which was used to determine whether an individual dose produced significant place conditioning, and with a two-way ANOVA, which was used to determine the effects of the treatments on mecamylamine-induced place conditioning. The Wilcoxon test was used to determine whether ondansetron produced significant place conditioning.

Results As shown in Fig. 1, the saline-control rats exhibited no preference for either compartment. The mean conditioning scores in non-treated and nicotine-treated rats were -8.8 2 50.7 s (n=12) and -42.0 * 45.3 s (n=12), respectively. Mecamylamine (0.1, 0.3 and 1.0 mgikg) produced neither significant place preference nor place aversion in non-treated rats. The mean conditioning scores associated with 0.1, 0.3 and 1.0 mg/kg mecamyhunine were -8.4 * 90.2 s (n=8), -3.8 * 47.4 s (n=12) and -28.0 * 59.0 s (n=12), respectively. On the other hand, mecamylamine -4.40, PcO.01) in nicotine-treated rats. produced dose-dependently place aversion (F, .,,+Mecamylamine (0.1 and 0.3 mg/kg) induced slight place aversion, but the effect was not significant, with mean conditioning scores of -101.1 * 65.3 s (n=8) and -183.7 * 53.0 s (n=16), respectively. Significant place aversion was observed at 1.0 mg/kg mecamylamine, with a mean conditioning score of -304.0 * 49.4 s (n=12, PcO.01). As shown in Fig. 2-A, mecamylamine (1 .O mg/kg) produced neither significant place preference nor place aversion in non-treated rats which had been pretreated with saline or ondansetron (0.01, 0.03 and 0.1 mg/kg). The mean conditioning scores were 2.5 + 70.5 s (n=8), 15.6 * 24.4 s (n=8), -41.0 * 99.7 s (n=8) and 0.8 * 64.4 s (n=8), respectively. Ondansetron (0.1 mgikg, s.c.) itself did not induce either significant place preference or place aversion in nontreated rats (Fig. 2-A). The mean conditioning score was -40.5 1: 65.4 s (n=8). As shown in Fig. 2-B, mecamylamine (1.0 mg/kg) produced place aversion in nicotinetreated rats. The mean conditioning score was -323.4 * 57.8 s (n=8). This place aversion .was significantly attenuated by pretreatment with ondansetron (0.01, 0.03 and 0.1 mgkg) 30 min prior

PL-252

Nicotine Dependence and Ondansetron

Vol. 61, No. 16, 1997

to the mecamylamine injection (F1,54-5.29, P.zO.05). The mean conditioning scores were -156.4 * 72.0 s (n=S), -82.0 * 86.5 s (n=S) and -10.8 * 71.2 s (n=S), respectively. Significant inhibition of mecamylamine-induced place aversion was observed at 0.1 mg/kg of ondansetron (P~0.05). There was no significant effect of dose (F3,54-0.99, ~~90.05) or treatment x dose interaction (F3.54=1.49, P>O.O5).

-200 r

-

S c

300 -

tg

-400

?!

--o_

Non-treated group

+

Nle-treated group 1 0

I a1 Mecamylamlne

Fig.

I

I

(x3

1

(mg/kg)

1.

Place conditioning produced by mecamylamine in rats that had been chronically treated with nicotine (Nit; 9 mg/kg/day) using an osmotic minipump. Rats were injected with saline (1.0 ml/kg, s.c.) or mecamylamine (0.1-1.0 ml/kg, s.c.) in the conditioning session. Each point represents the mean conditioning score with S.E.M. of 8-16 rats. **P
Discussion The conditioned place preference paradigm can be used to evaluate not only the rewarding effects but also the aversive effects of drugs. Using the conditioned place preference paradigm, naloxone-precipitated withdrawal aversion was observed in morphine-dependent rats (7,8), suggesting that place aversion may be a sign of morphine-withdrawal. Naloxone-precipitated withdrawal aversion was produced in morphine-dependent rats by a low dose of naloxone that does not induce marked withdrawal signs. Therefore, the appearance of naloxone-precipitated withdrawal aversion may be a very sensitive sign of morphine-withdrawal. Since the conditioned place preference paradigm has these properties, antagonist-precipitated withdrawal aversion may be useful for investigating nicotine dependence, which does not exhibit marked withdrawal signs. In fact, we previously reported that mecamylamine-precipitated nicotine-withdrawal aversion in rats can be evaluated using the conditioned place preference paradigm (9). Moreover, in the present study, we found that mecamylamine-precipitated nicotine-withdrawal aversion was antagonized by pretreatment with ondansetron. This result suggests that antagonism of the 5-HT, receptor may participate in the suppression of nicotine-withdrawal aversive behavior.

Nicotine Dependent and Ondansetron

VoL 61, No. 16,1997

A ~~~tmtec~group =P

200’

B

loo -

.% $

B Nle-treatedgroup

MN: 1 mg/kg

MEC 1 mg/kg

DNDMWkg) SAL 0.01 0.03

0.1

PL253

OND(mglkgl 0.1

SAL 0.01 0.03

0.1

0’ in 2 ‘0 5

-100 -

-200 -

g 5 2

-300’

:

400

-400’

Fig.

2.

A: Motivational effects of mecamylamine (MEC) and ondansetron (OND) in rats. B: Effect of ondansetron on mecamylamine-precipitated place aversion in nicotine (NIC)-treated rats. Bats were injected with saline (1 ml/kg, s.c.) or ondansetron (0.01-0.1 mg/kg, s.c.) 30 min before treatment with mecamylamine (1 mg/kg, s.c.). Each column represents the mean conditioning score with S.E.M. of 8 rats. *P
Recent studies have indicated that central serotonergic mechanisms may contribute to the aversive response, which partly results from the anxiogenic effect in laboratory studies. For example, Costa11 et al. (12) reported that diazepam-, cocaine-, ethanol- and nicotine-withdrawal aversive responses can be estimated by the light/dark exploration test, and that these aversive behaviors may result from an anxiogenic response which is mediated by the overstimulation of brain serotonergic neurons from the raphe nucleus to the amygdala. Moreover, Stein et al. (14) reported that aversive stimuli in general can activate 5-HT cell bodies in the raphe nucleus, leading to enhanced serotonergic function in the forebrain. Thus, these findings suggest that 5-HT pathways may be involved in the control of aversive behavior, i.e. mecamylamine-precipitated nicotine-withdrawal aversion. 5-HT, receptors have recently been identified in the limbic and cortical areas of rodent and human brain (15). Costa11 et al. (16) reported that injections of 5-HT, receptor antagonists, including ondansetron as well as the anxiolytic drug diazepam, into the amygdala of the mouse brain attenuate the aversive response using the light/dark exploration test. In addition, the intraamygdaloid injection of ondansetron attenuates the aversive behavior produced by diazepam-, cocaine-, ethanol- and nicotine-withdrawal in mice (12). Moreover, Higgins et al. (8) reported that naloxone-precipitated morphine-withdrawal aversion using the conditioned place preference paradigm was attenuated. Since anxiety is a core feature of withdrawal (17) it could be hypothesized that the suppression of mecamylamine-precipitated nicotine-withdrawal aversion may result from the anxiolytic effect of ondansetron. Ondansetron (0.1 mg/kg) tends to produce place preference (18). In the present study, 0.1 mg/kg of ondansetron did not produce a significant preference for the drug-associated compartment. As a result, the dose of ondansetron used in the present study may not produce a rewarding effect. Moreover, 5-I-IT, receptor antagonists, such as ICS 205-930 and MDL 72222, block the nicotine-induced place preference (19). These results suggest that 5-HT, receptor antagonists including ondansetron may be useful for treating nicotine dependence.

PL2.54

Nicotine

Dependence and Oodansetron

Vol. 61, No. 16, 1997

In conclusion, mecamylamine-precipitated nicotine-withdrawal aversion in chronic nicotine-treated rats, i.e. a withdrawal sign in nicotine dependence, was attenuated by pretreatment with ondansetron, and this effect may result from its anxiolytic effects. Therefore, ondansetron may be useful for the treatment of nicotine dependence. Acknowledgments This work was supported in part by a Research Grant (8A-5) for Nervous and Mental Disorders and a Research Grant from the Ministry of Health and Welfare to T. Suzuki. We wish to thank Ms. Seiko Katsuike and Mr. Tsuyoshi Yoshida for their expert technical assistance. References 1. J. R. HUGHES, S. W. GUST, K. SKOOG, R. M. KEENAN and J. W. FENWICK, Arch. Gen. Psychiatry 48 52-59 (1991) 2. Y. K. FUNG, M. J. SCHMID, T. M. ANDERSON and Y.-S. LAU, Pharmacol. Biochem. Behav. 53 635-640 (1996) 3. N. E. GRUNBERG, D. J. BOWEN and S. E. WINDERS, Psychopharmacology 90 101-105 (1986) 4. D. H. MALIN, J. R. LAKE, V. A. CARTER, J. S. CUNNINGHAM, K. M. HEBERT, D. L. CONRAD and 0. B. WILSON, Psychopharmacology 115 180-184 (1994) 5. D. C. HOFFMAN, Brain Res. Bull. 23 373-387 (1989) 6. R. F. MUCHA and A. HERZ, Psychopharmacology 86 274-280 (1985) 7. R. F. MUCHA, Brain Res. 418 214-220 (1987) 8. G. A. HIGGINS, P. NGUYEN, N. JOHARCHI and E. M. SELLERS, Psychopharmacology 105 322-328 (1991) 9. T. SUZUKI, Y. ISE, M. TSUDA, J. MAEDA and M. MISAWA, Eur. J. Pharmacol. 3 14 281-284 (1996) 10. N. R. OAKLEY, B. J. JONES, M. B. TYERS, B. COSTALL and A. M. DOMENEY, Br. J. Pharmacol. 95 870P (1988) 11. A. J. GOUDIE and M. J. LEATHLEY, Eur. J. Pharmacol. 185 179-186 (1990) 12. B. COSTALL, B. J. JONES, M. E. KELLY, R. J. NAYLOR, E. S. ONAIVI and M. B. TYERS, Pharmacol. Biochem. Behav. 36 97-104 (1990) 13. T. SUZUKI, Y. MASUKAWA and M. MISAWA, Psychopharmacology 102 438-442 (1990) 14. L. STEIN, C. D. WISE and J. D. BELUZZI, h4echanism ofAction ofBenzodiuzepines, E. COSTA and P. GREENGARD (Eds), 29-44, Raven Press, New York (1975) 15. G. J. KILPATRICK, B. J. JONES and M. B. TYERS, Nature 330 746-748 (1987) 16. B. COSTALL, M. E. KELLY, R. J. NAYLOR, E. S. ONAIVI and M. B. TYERS, Br. J. Pharmacol. 96 325-332 (1989) 17. M. W. EMMETT-OGLESBY, D. A. MATHIS, R. T. Y. MOON and H. LAL, Psychopharmacology 10 1 292-309 (1990) 18. G. A. HIGGINS, N. JOHARCHI, P. NGUYEN and E. M. SELLERS, Psychopharmacology 106 315-320 (1992) 19. E. CARBONI, E. ACQUAS, P. LEONE and G. DI CHIARA, Psychopharmacology 97 175-178 (1989)