Psychopharmacology (2003) 167:211–218 DOI 10.1007/s00213-003-1408-z

ORIGINAL INVESTIGATION

Kirsten E. Culver · Henry Szechtman

Clorgyline-induced switch from locomotion to mouthing in sensitization to the dopamine D2/D3 agonist quinpirole in rats: role of sigma and imidazoline I2 receptors Received: 19 October 2002 / Accepted: 16 January 2003 / Published online: 22 March 2003
Springer-Verlag 2003

Abstract Rationale: The monoamine oxidase inhibitor (MAOI) clorgyline, blocks locomotor sensitization to the D2/D3 dopamine agonist quinpirole and sensitizes selfdirected mouthing behavior in rats by a mechanism independent of MAO inhibition. Clorgyline has a high affinity for imidazoline I2 and sigma receptors, which could account for its effects on quinpirole sensitization. Objectives: To examine whether the effect of clorgyline on quinpirole sensitization is attributed to stimulation of either I2 or sigma receptors. Methods: In one experiment, rats received injections of the I2 receptor agonist 2-BFI (0.2 mg/kg, IP) or vehicle, 90 min prior to each injection of quinpirole (0.5 mg/kg, SC,  8, twice weekly) or saline. A similar protocol was used to examine the effects of the MAOI Ro 41-1049 (10 mg/kg, SC) on quinpirole sensitization. Unlike clorgyline, Ro 41-1049 has no affinity for sigma or I2 sites. An initial experiment demonstrated that intermittent injections of clorgyline (1 mg/kg, SC) are as effective as a continuous clorgyline administration (1 mg/kg per day via osmotic mini-pump) on quinpirole sensitization. Results: Like clorgyline, Ro 41-1049, but not 2-BFI, blocked the development of quinpirole-induced locomotor sensitization and induced instead sensitization of self-directed mouthing. Conclusions: Because Ro 41-1049 produced the same effects as clorgyline, and 2-BFI had no effects on quinpirole sensitization, it is unlikely that clorgyline exerts its effects via an action at sigma or I2 receptors. Our results are consistent with the suggestion that clorgyline and Ro 41-1049 affect the behavioral response to quinpirole via the MAOI-displaceable quinpirole binding (MQB) site, and the hypothesis that the MQB site selects what motor output becomes sensitized to repeated injections of quinpirole. K. E. Culver · H. Szechtman ()) Department of Psychiatry and Behavioural Neurosciences, McMaster University, Health Science Center, Room 4N82, 1200 Main Street West, Hamilton, Ontario, Canada L8N 3Z5 e-mail: [email protected] Tel.: +1-905-5259140 Fax: +1-905-5228804

Keywords Sensitization · Intermittent versus continuous administration · Locomotion · Mouthing · MAOI-displaceable quinpirole binding site (MQB) · Sigma receptor · Imidazoline · I2 receptor

Introduction Chronic administration of the dopamine agonists amphetamine, cocaine, apomorphine, bromocriptine, and quinpirole (QNP) results in a progressive augmentation of the motor response to these drugs (Segal and Schuckit 1983; Mattingly and Gotsick 1989; Hirabayshi et al. 1991; Hoffman and Wise 1992; Szechtman et al. 1994b). This phenomenon, termed behavioral sensitization (Robinson and Becker 1986), has received considerable attention as a potential experimental model for several psychopathological states in humans, including schizophrenia, obsessive-compulsive disorder and addiction (Angrist 1983; Robinson and Berridge 1993; Szechtman et al. 1999). Given these implications, much investigation has been focused on elucidating the mechanisms that underlie behavioral sensitization. In the present report we focus on the mechanisms of sensitization induced by the dopamine D2/D3 receptor agonist quinpirole. Repeated injections of quinpirole induce locomotor sensitization, while mouthing behaviors tolerate (Eilam and Szechtman 1989; Eilam et al. 1992; Szechtman et al. 1994b). Recently, we have shown that continuous administration of the potent monoamine oxidase inhibitor (MAOI) clorgyline (1 mg/kg per day, via osmotic minipump), blocks the development of quinpirole-induced locomotor sensitization and sensitizes self-directed mouthing behaviors in rats (Culver and Szechtman 1997; Szechtman et al. 1999). This effect of clorgyline is unrelated to both the inhibition of MAO activity, and the associated changes in dopamine and serotonin levels in the striatum and prefrontal cortex (Culver et al. 2000, 2002), raising the possibility that the recently identified MAOI-displaceable quinpirole binding site may be

212

involved in the mechanism of quinpirole-induced sensitization. Several MAOIs, including clorgyline, pargyline and Ro 41-1049, inhibit the binding of [3H]quinpirole to rat striatal membranes, but exhibit very low potency in competition with either the D2 antagonist [3H]spiperone or the dopamine agonist [3H](-)-N-n-propylnorapomorphine (Levant et al. 1993, 1996). Recently, Ro 41-1049 has also been shown to inhibit [3H]quinpirole binding in MAOA deficient mice (Levant et al. 2001), further suggesting that these MAOI inhibit quinpirole binding by a mechanism independent of MAO. Based on such findings, we postulate that clorgyline affects the response to repeated injections of quinpirole via the MQB site (Culver and Szechtman 1997) and that this site selects which motor pathway (locomotion versus mouthing) becomes sensitized to quinpirole (Culver et al. 2000). Although the results with clorgyline (Culver et al. 2000, 2002) are consistent with a role for the MQB site in quinpirole sensitization, clorgyline has a high affinity for sigma (Itzhak and Kassim 1990; Itzhak et al. 1991) and imidazoline I2 receptors (Olmos et al. 1993; Alemany et al. 1995), raising the possibility that clorgyline may affect quinpirole-induced sensitization via its action at either of these sites. To investigate whether clorgyline exerts its effects on quinpirole sensitization via an action on I2 and/or sigma receptors, two separate experiments were performed. First, the effect of the imidazoline I2 receptor agonist 2-BFI (Lione et al. 1998), on the development of quinpirole-induced sensitization was investigated. Second, the effect of Ro 41-1049 on quinpirole-induced sensitization was investigated. Unlike clorgyline, Ro 411049 has no affinity for sigma (Itzhak et al. 1991) or imidazoline I2 receptors (Olmos et al. 1993). In previous studies examining the effects of clorgyline on quinpirole-induced sensitization (Culver and Szechtman 1997; Culver et al. 2000), clorgyline was administered continuously via osmotic mini-pump. Due to the practical constraints associated with Ro 41-1049 and 2-BFI administration (high dosage and intra-peritoneal injections), an intermittent rather than continuous regimen of drug injections was chosen. Consequently, it was necessary to determine whether intermittent injections of clorgyline were as effective as continuous administration in switching the quinpirole sensitized response from locomotion to mouthing. Therefore, before the 2-BFI and Ro 41-1049 experiments were performed, we first compared the effects of continuous and intermittent clorgyline administration on quinpirole sensitization.

(8:00 a.m. to 8:00 p.m.), with free access to food and water. Rats were allowed to acclimatize to the colony room for 1 week following arrival and were handled 2 min daily for 7 days before the start of experiments. All treatments and testing were conducted during the light hours. Animals were housed and tested in compliance with the guidelines described in the Guide to the Care and Use of Experimental Animals (Canadian Council on Animal Care 1993). Surgery For experiment 1, ten animals were anaesthetized with the inhalant anesthetic Halothane (MTC Pharmaceuticals, Cambridge, Ontario, Canada) and, under aseptic conditions, implanted with 2ML4 Alzet (Alza Corp., Palo Alto, Calif., USA) osmotic mini-pumps (mean pumping rate 2.42 ml/h, 28 days) in the intra-scapular region for subcutaneous drug delivery. Animals were allowed three days of recovery before the start of testing. Drugs Clorgyline hydrochloride (Sigma, St Louis, Mo., USA) was dissolved in saline and administered continuously (1 mg/kg per day) via osmotic mini-pump, or intermittently (1 mg/kg) via subcutaneous injection. Mini-pump filling concentrations were calculated according to the method described by Greenshaw (1986). 2-BFI hydrochloride (Tocris, Ballwin, Mo., USA) was dissolved in saline (0.2 mg/kg, IP). Ro 41-1049 hydrochloride (RBI, Natick, Mass., USA) was dissolved in saline (10 mg/kg, SC). Quinpirole hydrochloride (RBI) was dissolved in saline (0.5 mg/kg, SC) and administered twice weekly for a total of eight injections. All doses refer to the salt form of the drug. The particular dose and injection regimen of quinpirole was chosen because it is representative of the behavioral effects induced by quinpirole doses ranging from 0.25 to 2.5 mg/kg, and because the locomotor effects of chronic treatment reach a plateau after eight to ten drug injections administered 2–8 days apart (Szechtman et al. 1994a, 1994b). The dose of Ro 41-1049 was chosen to produce equivalent levels of MAO inhibition and MQB binding produced by 1 mg/kg clorgyline (Levant et al. 1996; Heeringa et al. 1997). The dose of 2-BFI was chosen to yield binding at I2 receptors that is equivalent to the I2 binding produced by 1 mg/kg dose of clorgyline used in all our studies (Culver and Szechtman 1997; Culver et al. 2000, 2002). The 0.2 mg/kg dose of 2-BFI was calculated using the Ki values of clorgyline (11.7 nM) and 2-BFI (1.71 nM) for I2 receptors (Lione et al. 1998). Clorgyline, 2-BFI, Ro 41-1049 and vehicle were all administered 90 min prior to each injection of quinpirole, based the findings that all three drugs exert their maximal effect on MAO and dopamine levels 90 min postinjection (Clement et al. 1990; Colzi et al. 1990; Hudson et al. 1999). Apparatus The testing environment consisted of a Plexiglas locomotor activity chamber (404035 cm) located in a non-colony room. Six activity chambers were interfaced to a Digiscan 16 monitor that provided an automated recording of the locomotor distance traveled by each rat (AccuScan Instruments, Columbus, Ohio, USA).

Materials and methods

Procedure

Subjects

To examine the effect of continuous versus intermittent clorgyline (CLG) treatment on quinpirole-induced sensitization, 30 rats were injected with quinpirole (QNP), while ten rats were injected with saline (SAL). Quinpirole-treated rats were assigned at random into three groups (n=10/group): a continuous clorgyline group (CLGmini-pump+QNP), an intermittent clorgyline group

One hundred and thirty-six experimentally naive male Long-Evans rats (Charles-River, Canada) weighing 200–230 g at the start of treatment were used. Rats were individually housed in a temperature controlled colony room (22C) under a 12-h light-dark cycle

213 (CLGinjection+QNP), and a quinpirole control group (VEH+QNP). All quinpirole treated rats received clorgyline or vehicle (VEH) 90 min prior to each injection of quinpirole; while saline treated rats (n=10) were similarly pretreated with a subcutaneous injection of vehicle (VEH+SAL). To determine whether continuous or intermittent clorgyline administration blocks the induction of locomotor sensitization to quinpirole, or merely blocks its expression, clorgyline treatment was discontinued after the eighth quinpirole injection, and all groups received a test injection of quinpirole (0.5 mg/kg, SC) 1 week later. Immediately following each quinpirole or saline injection, rats were placed in activity chambers and their locomotor activity was recorded for 90 min. Since clorgyline sensitizes self-directed mouthing behaviors in quinpirole treated rats, mouthing activity was also measured during the first and last 15 min of the 90-min testing period, according to the protocol described in Culver et al. (2000). Mouthing activity was defined as any contact between the rat’s mouth and/or tongue with either an external object (e.g. licking or gnawing walls) or parts of its own body (e.g. nibbling of paws or tail, grooming). Because clorgyline pretreated rats primarily engage in self-directed mouthing behaviors, externaland self-directed mouthing activity was scored separately. To examine the effect of 2-BFI pretreatment on quinpiroleinduced sensitization, 48 rats were assigned at random, into two groups for repeated injections of 2-BFI or vehicle. Half of the rats in each group received an injection of quinpirole (QNP) 90 min after 2-BFI or vehicle pretreatment, while the other half received saline (SAL). Thus, four groups were tested: a chronic 2-BFI group (2-BFI+QNP, n=10), a chronic quinpirole control group (VEH+QNP, n=10) and two saline injected control groups (2-BFI+SAL, n=10; VEH+SAL, n=18). Locomotor and mouthing behavior was scored throughout as described for experiment 1. 2-BFI treatment was discontinued after the eighth quinpirole injection, and all rats received a test injection of quinpirole (0.5 mg/ kg) 1 week later, except for ten rats from the VEH+SAL group which received saline. The VEH+SAL rats that received quinpirole constituted an acute quinpirole group (Acute QNP, n=8). The purpose of the third experiment was to examine the effects of Ro 41-1049 pretreatment on the development of quinpirole sensitization. The design of this experiment was identical to the 2-BFI experiment described above and involved the following groups: a chronic Ro 41-1049 group (Ro 41-1049+QNP, n=10), a chronic quinpirole control group (VEH+QNP, n=10) and two saline injected control groups (Ro 41-1049+SAL, n=10; VEH+SAL, n=18), as well as an Acute QNP group (n=10) after discontinuation of Ro 41-1049 treatment. All injections were administered in the testing environment, 3–4 days apart (i.e. twice weekly) and at the same time of the day. For the pretreatment injections, animals were brought in their home cage from the colony room into the testing environment, administered the appropriate pretreatment, and then returned to their colony room. Ninety min later they were brought back to the testing room and injected with quinpirole or vehicle, and placed into activity monitors. Statistics For experiment 1, a GroupInjections analysis of variance (ANOVA) with repeated measures on the second factor was employed. Duncan multiple range test was used for posthoc between group comparisons. To assess the statistical significance of the sensitization regimen, performance on Injection 8 was compared to performance on Injection 1, using a paired t-test. To assess the statistical significance of differences in the level of sensitization reached by the groups, one-way ANOVA and Duncan multiple range test were used to evaluate performance at Injection 8 and at the quinpirole test injection (Injection 9). Similar analyses were performed for experiment 2 and experiment 3 data. Separate analyses were performed for locomotor distance and duration of mouthing data. In all three experiments, the data was normally distributed and the variances were homogenous. The chosen level

of significance was P<0.05. Calculations were performed using SPSS 10.0 for Windows.

Results Experiment 1: effect of continuous versus intermittent clorgyline administration on quinpirole-induced sensitization Repeated injections of quinpirole induced locomotor sensitization, as evidenced by a six fold increase in the distance that quinpirole control rats traveled at Injection 8 compared to the distance they traveled at Injection 1 [t(9)=4.24, P<0.01] (Fig. 1A). Continuous and intermittent clorgyline pretreatment blocked the induction of locomotor sensitization: clorgyline pretreated rats showed no significant difference in the distance traveled at Injection 8 compared to Injection 1 [t(9)=0.106, P>0.05; t(9)=1.022, P>0.05, CLGmini-pump+QNP and CLGinjection+ QNP groups, respectively] (Fig. 1A). No significant difference in locomotor performance was observed between continuous and intermittent clorgyline groups at Injection 8 (P>0.05, Duncan multiple range test). Continuous and intermittent clorgyline sensitized the mouthing response to repeated injections of quinpirole: the duration of time spent mouthing at quinpirole Injection 8 was significantly greater than at Injection 1 [t(9)=3.470, P<0.01; t(9)=4.65, P<0.01, CLGmini-pump+ QNP and CLGinjection+QNP groups, respectively] (Fig. 1A). The effect of continuous and intermittent clorgyline pretreatment involved not only an increase in the duration of time spent mouthing, but also a change from external to self-directed mouthing (Fig. 1A): VEH+SAL and VEH+QNP treated rats displayed mostly externally-directed mouthing behavior (coprophagia, licking of cage walls), while clorgyline pretreated rats engaged predominantly in the mouthing of their own body, which included paw chewing and tail biting (Fig. 1A). One week after clorgyline treatment was discontinued (Fig. 1B), clorgyline pre-treated rats continued to display a level of locomotor activity that was significantly less than that of sensitized VEH+QNP treated rats (P<0.05, Duncan multiple range test). Similarly, despite the absence of clorgyline, sensitized mouthing behaviors persisted in continuous and intermittent clorgyline pretreated rats compared to all other treatment groups (P<0.05, Duncan multiple range test). Together, these results show that both continuous and intermittent clorgyline administration are equally effective in blocking the development of quinpirole-induced locomotor sensitization and in sensitizing mouthing behavior.

214

Fig. 1 A Locomotor and mouthing responses to quinpirole (QNP; 0.5 mg/kg, SC, 8, twice weekly) in rats co-administered clorgyline (CLG) via osmotic mini-pump (CLGmp; 1 mg/kg per day) or subcutaneous injection (CLGinj; 1 mg/kg). Gray filled curve indicates mouthing that was directed at the rat’s own body and black filled curve shows mouthing of external stimuli. Duration of mouthing is expressed as the percentage of the observation period that the rats spent mouthing. Values are mean€SEM. A 48 ANOVA with repeated measures revealed significant main and interaction effects for the locomotor [for group, F(3,36)=48.58,

P<0.001; for injections, F(7,252)=12.24, P<0.01; for groupinjections, F(21,252)=8.83, P<0.01] and mouthing [for group, F(3,36)=24.18, P<0.001; for injections, F(7, 252)=5.31, P<0.01; for groupinjections, F(21,252)=3.41, P<0.01] data. *P<0.05 compared to performance at Injection 1, paired t-test. B Locomotor and mouthing responses to a test injection of quinpirole (0.5 mg/kg) administered 1 week after the discontinuation of chronic treatment with clorgyline. +P<0.05 compared to Acute QNP group, Duncan multiple range test

Experiment 2: effect of 2-BFI pretreatment on quinpirole-induced sensitization

One week after the discontinuation of 2-BFI treatment (Fig. 2B), VEH+QNP and 2-BFI+QNP pretreated rats displayed a locomotor response to the test injection of quinpirole that was greater than the Acute QNP group, but no different from each other (Duncan multiple range test), suggesting that 2-BFI does not block the development or expression of quinpirole locomotor sensitization. It is also unlikely that 2-BFI pretreatment affects the acute response to quinpirole, since 2-BFI+SAL pretreated rats displayed a locomotor response to the test injection of quinpirole no different from the Acute QNP group (Fig. 2B; P>0.05, Duncan multiple range test). Together, these results show that 2-BFI does not block the development of quinpirole-induced locomotor sensitization or sensitize the mouthing response to repeated injections of quinpirole.

2-BFI pretreatment did not block the induction of locomotor sensitization to repeated injections of quinpirole: the distance traveled by 2-BFI co-treated rats at Injection 8 was significantly greater than that at Injection 1 [t(9)=4.64, P<0.01]. Moreover, 2-BFI pretreated rats showed a level of locomotor activity no different from that of the quinpirole control group at Injection 8 (P>0.05, Duncan multiple range test), further suggesting the lack of any effect of 2-BFI on locomotor sensitization to quinpirole (Fig. 2A). 2-BFI pretreatment was ineffective in sensitizing the mouthing response to repeated injections of quinpirole: the percentage of time 2-BFI pretreated rats spent mouthing at Injection 8 was significantly less than that at Injection 1 [t(9)=2.10, P>0.05], and was no different than the percentage of time VEH+QNP treated rats spent mouthing at Injection 8 (P>0.05 Duncan multiple range test) (Fig. 2A).

215

Fig. 2 A Locomotor and mouthing responses to quinpirole (QNP; 0.5 mg/kg, SC, 8, twice weekly) in rats co-administered 2-BFI (0.2 mg/kg, IP). Values are mean€SEM. A 48 ANOVA with repeated measures revealed significant main and interaction effects for the locomotor [for group, F(3,44)=15.76, P<0.001; for injections, F(7,308)=27.51, P<0.001; for groupinjections, F(21,308)= 11.85, P<0.001] and mouthing [for group, F(3,44)=6.55, P<0.01; for injections, F(7,308)=9.36, P<0.001; for groupinjections,

F(21,308)=2.74, P<0.001] data. *P<0.05 compared to performance at Injection 1, paired t-test. B Locomotor and mouthing responses to a test injection of quinpirole (0.5 mg/kg) administered one week after the discontinuation of chronic treatment with 2-BFI. All groups received an injection of quinpirole except for the VEH+SAL group, which was injected with saline. +P<0.05 compared to Acute QNP group, Duncan multiple range test

Experiment 3: effect of Ro 41-1049 pretreatment on quinpirole-induced sensitization

mouthing, but also induced a change from external to selfdirected mouthing behaviors. One week after Ro 41-1049 treatment was discontinued (Fig. 3B), Ro 41-1049 pretreated rats displayed a locomotor response to the test injection of quinpirole no different from the Acute QNP group (P>0.05, Duncan multiple range test), and significantly less than that of sensitized VEH+QNP treated rats (P<0.05, Duncan multiple range test). Ro 41-1049+QNP treated rats also displayed a mouthing response to the test injection of quinpirole that was significantly greater than the percentage of time spent mouthing by any other group (P<0.05, Duncan multiple range test). Together, these findings show that Ro 41-1049 pretreatment blocks the development of locomotor sensitization and instead sensitizes the mouthing response to repeated injections of quinpirole.

In contrast to 2-BFI, Ro 41-1049 pretreatment did block the induction of locomotor sensitization: rats pretreated with Ro 41-1049 showed no significant difference in the distance traveled at Injection 8 compared to the distance traveled at Injection 1 [t(9)=2.10, P>0.05] (Fig. 3A). Moreover, at Injection 8, the locomotor performance of Ro 41-1049+QNP treated rats was significantly less than that of the quinpirole control group (P<0.05, Duncan multiple range test), further suggesting that Ro 41-1049 administration blocks locomotor sensitization to quinpirole. Ro 41-1049 pretreatment sensitized the mouthing response to repeated injections of quinpirole: the duration of mouthing observed in Ro 41-1049 pretreated rats at Injection 8 was significantly greater than at Injection 1 [t(9)=3.40, P<0.01], and significantly greater than the percentage of time spent mouthing by the VEH+QNP group at Injection 8 (P<0.05, Duncan multiple range test) (Fig. 3A). As observed in clorgyline pretreated rats, Ro 41-1049 not only increased the duration of time spent

Discussion Consistent with previous studies investigating the effects of continuous clorgyline treatment on quinpirole-induced

216

Fig. 3 A Locomotor and mouthing responses to quinpirole (QNP 0.5 mg/kg, SC 8, twice weekly) in rats co-administered Ro 41-1049 (10 mg/kg, SC). Values are mean€SEM. A 48 ANOVA with repeated measures revealed significant main and interaction effects for the locomotor [for group, F(3,44)=17.86, P<0.001; for injections, F(7,308)=26.38, P<0.001; for groupinjections, F(21,308)= 15.99, P<0.001] and mouthing [for group, F(3,44)=56.12, P<0.001; for injections, F(7,308)=2.30, P<0.05; for

groupinjections, F(21,308)=4.71, P<0.001] data. *P<0.05 compared to performance at Injection 1, paired t-test. B Locomotor and mouthing responses to a test injection of quinpirole (0.5 mg/kg) administered 1 week after the discontinuation of chronic treatment with Ro 41-1049. All groups received an injection of quinpirole except for the VEH+SAL group, which was injected with saline. + P<0.05 compared to Acute QNP group, Duncan multiple range test

sensitization (Culver and Szechtman 1997; Szechtman et al. 1999; Culver et al. 2000), the results of the present study show that intermittent clorgyline administration is equally effective in blocking the development of quinpirole-induced locomotor sensitization and in sensitizing the mouthing response to repeated injections of quinpirole. Thus, an intermittent regimen of clorgyline administration is sufficient, and a continuous regimen is not required, to switch the sensitized response to quinpirole from locomotion to mouthing. In that context, the present results showed that Ro 41-1049 pretreatment, like clorgyline, blocked locomotor sensitization to quinpirole and sensitized self-directed mouthing behavior, but 2-BFI pretreatment had no effects on quinpirole sensitization. Below we discuss the implications of these findings regarding the role of sigma and I2 receptors in the effects of clorgyline on quinpirole sensitization, and compare the role of these receptors in sensitization to quinpirole and other psychostimulant drugs.

A non-sigma, non-I2 effect of clorgyline on quinpirole sensitization In addition to its inhibitory effects on the MAO enzyme (Culver et al. 2000), clorgyline has an affinity for several other sites, including imidazoline I2 receptors, sigma receptors and the MQB site (Itzhak and Kassim 1990; Itzhak et al. 1991; Olmos et al. 1993; Alemany et al. 1995; Levant et al. 1996). The results of the present study strongly suggest, however, that the effects of clorgyline on the development of quinpirole sensitization are not mediated by an action of clorgyline at either imidazoline I2 or sigma receptors. With respect to the role of I2 receptors in the effects of clorgyline on quinpirole sensitization, the evidence is especially convincing because Ro 41-1049, which has no affinity for I2 receptors (Olmos et al. 1993), switched the sensitized quinpirole response from locomotion to mouthing, while the potent imidazoline I2 receptor agonist 2-BFI (Lione et al. 1998) had no such effects. Thus, it is unlikely that clorgyline exerts its effects on quinpirole sensitization via an action at I2 receptors. Considering that Ro 41-1049, like

217

clorgyline, switched the sensitized quinpirole response from locomotion to mouthing, it is also unlikely that the effects of clorgyline on quinpirole sensitization involve sigma receptors, since Ro 41-1049 has no affinity for sigma receptors (Itzhak et al. 1991; Levant et al. 1993). The finding that clorgyline and Ro 41-1049 both switched the sensitized response to quinpirole from locomotion to mouthing suggests that the switching effect of these drugs may be exerted through a common mechanism of action. Clorgyline and Ro 41-1049 (but not 2-BFI) possess a high affinity for the MQB site (Levant et al. 1993, 1996). As such, the present results strengthen, but do not prove the hypothesis that clorgyline produces its effects on quinpirole sensitization via an action at the MQB site (Culver and Szechtman 1997; Culver et al. 2000). Sigma and I2 receptors in development of sensitization Although the present findings show that clorgyline acts via a non-sigma receptor mechanism to alter quinpirole sensitization, the results leave open the possibility that the development of sensitization to chronic injections of quinpirole may involve sigma receptors. To ascertain the role of sigma receptors in sensitization to quinpirole, the effects of sigma ligands on sensitization should be investigated. However, quinpirole has a very low affinity for sigma receptors (Levant et al. 1993), suggesting that a direct action of quinpirole on sigma receptors in sensitization seems unlikely. In contrast to quinpirole, psychostimulant drugs such as amphetamine and cocaine do possess an affinity for sigma receptors (Ujike et al. 1996), and it has been proposed that the development of sensitization to these psychostimulant drugs requires stimulation of sigma receptors (Ujike et al. 1992b, 1996). Indeed, rats sensitized to methamphetamine show behavioral cross-sensitization to sigma receptor agonists (Ujike et al. 1992b), while sigma receptor antagonists block the development of cocaine and methamphetamine-induced sensitization (Menkel et al. 1991; Ujike et al. 1992a, 1996; Takahashi et al. 2000). Given these findings, it would not be surprising if sigma receptors were involved in sensitization to quinpirole as well. However, as indicated above, the apparent lack of affinity of quinpirole to sigma receptors would suggest that their involvement (if any) is more indirect, and therefore different from that described for methamphetamine and cocaine-induced sensitization. Unlike sigma, the imidazoline receptor has yet to be implicated in the development of sensitization to any psychostimulant drug. Several studies have shown however, that I2 receptor ligands such as 2-BFI significantly modulate monoamine levels in rat brain (Nutt et al., 1995; Sastre-Coll et al. 2001). Nevertheless, the present results suggest that I2 receptors are unlikely to play a role in quinpirole sensitization since 2-BFI did not alter quinpirole-induced sensitization, nor did it induce sensitization when administered alone.

Conclusions The results of the present study suggest that clorgyline does not exert its effects on quinpirole sensitization via an action at either sigma or I2 receptors. Taken together with the results of previous studies which show that clorgyline does not exert its effects via inhibition of MAO, increase in dopamine or serotonin levels (Culver et al. 2000), or inhibition of striatal dopamine reuptake (Culver et al. 2002), the present findings suggest that clorgyline does not exert its effects on quinpirole sensitization via any of the established modes of action of clorgyline. Furthermore, the finding that both clorgyline and Ro 41-1049 (but not 2-BFI) switch the sensitized quinpirole response from locomotion to mouthing, suggests that these two MAOIs share a common site of action in quinpirole sensitization, possibly the MQB site (Levant et al. 1993, 1996). While the present results do not provide proof, they nevertheless are consistent with the hypothesis that clorgyline, Ro 41-1049, and pargyline (Culver et al. 2002) exert their switching effects on quinpirole-induced sensitization by inhibiting quinpirole binding at the MQB site. Acknowledgements We thank Donna Waxman for help in carrying out a portion of the study. This study was supported by the Canadian Institutes of Health Research (MT-12852). H.S. is a Senior Research Fellow of the Ontario Mental Health Foundation.

References Alemany R, Olmos G, Garcia-Sevilla JA (1995) The effects of phenelzine and other monoamine oxidase inhibitor antidepressants on brain and liver I2 imidazoline preferring-receptors. Br J Pharmacol 114:837–845 Angrist B (1983) Psychosis induced by central nervous system stimulants and related drugs. In: Creese I (ed) Stimulants: neurochemical, behavioural and clinical perspectives. Raven Press, New York, pp 1–30 Canadian Council on Animal Care (1993) Guide to the care and use of experimental animals, vol 1, 2nd edn. CCAC, Ottawa, Ontario Clement H, Grote C, Wesemann W (1990) In vivo studies on the effect of monoamine oxidase inhibitors on dopamine and serotonin metabolism in rat brain areas. J Neural Transm Suppl 32:85–88 Colzi A, d’Agostini F, Kettler R, Borroni E, DaPrada M (1990) Effect of selective and reversible MAO inhibitors on dopamine outflow in rat striatum: a microdialysis study. J Neural Transm 32:79–84 Culver KE, Szechtman H (1997) Monoamine oxidase inhibitor sensitive site implicated in sensitization to quinpirole. Eur J Pharmacol 339:109–111 Culver KE, Rosenfeld JM, Szechtman H (2000) A switch mechanism between locomotion and mouthing implicated in sensitization to quinpirole in rats. Psychopharmacology 151:202–210 Culver KE, Rosenfeld JM, Szechtman H (2002) Monoamine oxidase inhibitor-induced blockade of locomotor sensitization to quinpirole: role of striatal dopamine uptake inhibition. Neuropharmacology 43:385–393 Eilam D, Szechtman H (1989) Biphasic effect of D-2 agonist quinpirole on locomotion and movements. Eur J Pharmacol 161:151–157

218 Eilam D, Talangbayan H, Canaran G, Szechtman H (1992) Dopaminergic control of locomotion, mouthing, snout contact, and grooming: opposing roles of D1 and D2 receptors. Psychopharmacology 106:447–454 Greenshaw AJ (1986) Osmotic mini-pumps: a convenient program for weight adjusted filling concentrations. Brain Res Bull 16:759–761 Heeringa MJ, d’Agostini F, DeBoer P, DaPrada M, Damsma G (1997) Effect of monoamine oxidase A and B and of catecholO-methyltransferase inhibition on L-DOPA-induced circling behaviour. J Neural Transm 104:593–603 Hirabayshi M, Okada S, Tadokoro S (1991) Comparison of sensitization to ambulation-increasing effects of cocaine and methamphetamine after repeated administration in mice. J Pharm Pharmacol 43:827–830 Hoffman DC, Wise RA (1992) Locomotor-activating effects of the D2 agonist bromocriptine show environment-specific sensitization following repeated injections. Psychopharmacology 107:277–284 Hudson HL, Gough R, Tyacke R, Lione L, Lalies M, Lewis J, Husbands S, Knight P, Murray F, Hutson P, Hutt DJ (1999) Novel selective compounds for the investigation of imidazoline receptors. Ann NY Acad Sci 881:81–91 Itzhak Y, Kassim CO (1990) Clorgyline displaces high affinity for sigma binding sites in C57BL/6 mouse brain. Eur J Pharmacol 176:107–108 Itzhak Y, Stein I, Zhang S, Kassim CO, Cristante D (1991) Binding of s-ligands to C57BL/6 mouse brain membranes: effects of monoamine oxidase inhibitors and subcellular distribution studies suggest the existence of s-receptor subtypes. J Pharmacol Exp Ther 257:141–148 Levant B, Grigoradis DE, DeSouza EB (1993) Monoamine oxidase inhibitors inhibit [3H]quinpirole binding in rat striatal membranes. Eur J Pharmacol 246:171–178 Levant B, Moehlenkamp JD, Morgan KA, Leonard NL, Cheng CC (1996) Modulation of [3H]quinpirole binding in brain by monoamine oxidase inhibitors: Evidence for a potential novel binding site. J Pharmacol Exp Ther 278:145–153 Levant B, Morgan KA, Beckendorf-Ahlgren JA, Grandy DK, Chen K, Shih JC, Seif I (2001) Modulation of [3H]quinpirole binding at striatal D2 dopamine receptors by a monoamine oxidaseAlike site: evidence from radioligand binding studies and D2 receptor- and MAOA-deficient mice. Life Sci 70:229–241 Lione LA, Nutt DJ, Hudson AL (1998) Characterization and localization of [3H]2-(2-benzofuranyl)-2-imidazoline binding in rat brain: a selective ligand for imidazoline I2 receptors. Eur J Pharmacol 353:123–135 Mattingly B, Gotsick S (1989) Conditioning and experimental factors affecting the development of sensitization to apomorphine. Behav Neurosci 103:1311–1317 Menkel M, Terry P, Pontecorvo M, Katz J, Witkin J (1991) Selective s ligands block stimulant effects of cocaine. Eur J Pharmacol 201:251–252

Nutt D, French N, Handley S, Hudson A, Husbands S, Jackson S, Jordan S, Lalies M, Lewis J, Lione L, Mallard N, Pratt J (1995) Functional studies of specific imidazoline-2 receptor ligands. Ann NY Acad Sci 763:125–129 Olmos G, Gabilondo A, Miralles A, Escriba P, Garcia-Sevilla E (1993) Chronic treatment with the monoamine oxidase inhibitors clorgyline and pargyline down-regulates non-adrenoceptor [3H]-idazoxan binding sites in the rat brain. Br J Pharmacol 108:597–603 Robinson TE, Becker JB (1986) Enduring change in brain and behavior produced by chronic amphetamine administration: a review and evaluation of animal models of amphetamine psychosis. Brain Res 396:157–198 Robinson TE, Berridge KC (1993) The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Res Rev 18:247–291 Sastre-Coll A, Esteban S, Miralles A, Zanetti R, Garcia-Sevilla J (2001) The imidazoline receptor ligand 2-(2-benzofuranyl)-2imidazoline is a dopamine releasing agent in the rat striatum in vivo. Neurosci Lett 301:29–32 Segal DS, Schuckit MA (1983) Animal models of stimulantinduced psychosis. In: Creese I (ed) Stimulants: neurochemical, behavioural and clinical perspectives. Raven Press, New York, pp 131–167 Szechtman H, Dai H, Mustafa S, Einat H, Sullivan RM (1994a) Effects of dose and interdose interval on locomotor sensitization to the dopamine agonist quinpirole. Pharmacol Biochem Behav 48:921–928 Szechtman H, Talangbayan H, Canaran G, Dai H, Eilam D (1994b) Dynamics of behavioral sensitization induced by the dopamine agonist quinpirole and a proposed central energy mechanism. Psychopharmacology 115:95–104 Szechtman H, Culver KE, Eilam D (1999) Role of dopamine systems in obsessive-compulsive disorder: implications from a novel psychostimulant-induced animal model. Pol J Pharmacol 51:55–61 Takahashi S, Miwa T, Horikomi K (2000) Involvement of s1 receptors in methamphetamine-induced behavioral sensitization in rats. Neurosci Lett 289:21–24 Ujike H, Kanzaki A, Okumura K, Akiyama K, Otsuki S (1992a) Sigma antagonist BMY 14802 prevents methamphetamineinduced sensitization. Life Sci 50:129–134 Ujike H, Okumura K, Zushi Y, Akiyama K, Otsuki S (1992b) Persistent supersensitivity of s receptors develops during repeated methamphetamine treatment. Eur J Pharmacol 211:23–328 Ujike H, Kuroda S, Otsuki S (1996) s receptor antagonists block the development of sensitization to cocaine. Eur J Pharmacol 296:123–128