Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Glutamatergic and monoaminergic networks in corticolimbic structures and the basal ganglia play an important role in the control of motor function. Correspondingly, an understanding of the mechanisms via which they exert their actions may lead to novel therapies for the improved management of neurological disorders, such as Parkinson's disease, and psychiatric diseases, such as schizophrenia (Carlsson and Carlsson, 1990; Lange et al., 1997; Schmidt and Kretschmer, 1997). To this end, it is important to identify functional models that allow the exploration of interrelationships among glutamatergic and monoaminergic pathways. In previous studies, we demonstrated that the interruption of transmission at N-methyl-D-aspartate (NMDA) receptors by open channel blockers, such as dizocilpine, and antagonists of the NMDA receptor recognition site, such as 3-(2-carboxypiperazin-4-yl)-propyl-1-phosphonic acid (CPP), but not by glycine_B antagonists, such as L701,324, elicits STFs in rats (Millan et al., 1991). The population of NMDA receptors involved is localized in the nucleus accumbens (Millan et al., 1999a). This structure receives a pronounced glutamatergic input from frontal cortex, hippocampus, thalamus, and amygdala, and in interaction with monoaminergic pathways, glutamatergic mechanisms in the accumbens modulate motor function and mood (Carlsson and Carlsson, 1990; Meltzer et al., 1997; Schmidt and Kretschmer, 1997; Morari et al., 1998; Millan et al., 1999b).

The ventral tegmental area and the substantia nigra, pars compacta, the origin of ascending mesocorticolimbic and nigrostriatal dopaminergic projections, respectively, also possess a pronounced glutamatergic innervation from the frontal cortex, the subthalamic nucleus, and other regions (Meltzer et al., 1997). Indeed, reciprocal interactions of NDMA receptors with mesolimbic, mesocortical, and nigrostriatal dopaminergic projections are well documented, although the precise involvement of accumbens-integrated dopaminergic mechanisms in the actions of NMDA receptor antagonists remains to be clarified (Ouagazzal et al., 1993, l994; Narayanan et al., 1996; Meltzer et al., 1997; Millan et al., 1999b). Interestingly, the functional interrelationship of NMDA receptors with D₁ compared with D₂ receptors may differ. Indeed, in rats, NMDA receptor blockade and D₁ receptor stimulation exert a mutual, facilitatory influence on motor behavior (Morelli et al., 1992; Starr and Starr, 1993; Svensson et al., 1994; Morari et al., 1998; Snyder et al., 1998).

Regarding adrenergic mechanisms, NMDA receptors exert a phasic, facilitatory influence on the release of norepinephrine (NE) in the hippocampus and frontal cortex, both directly and indirectly, via actions at the level of adrenergic terminals, as well as in the locus coeruleus itself (Jodo and Aston-Jones, 1997; Yoshida et al., 1997). Contrariwise, in these structures, ₂-adrenergic receptors (ARs) exert an inhibitory influence on the release of glutamate (Kamisaki et al., 1991; Pralong and Magistretti, 1995). Interestingly, although activation of _2A-autoreceptors elicits sedation (Millan et al., 1994a,b), clonidine and dizocilpine exert a synergistic and excitatory influence on motor behavior in reserpine-treated mice. This suggests that on blockade of activity at NMDA receptors, postsynaptic ₂-ARs may facilitate motor behavior (Carlsson and Svensson, 1990; Nutt, 1994; Niittykoski et al., 1997).

Serotonergic mechanisms are of particular interest regarding the induction of spontaneous tail-flicks (STFs) by open channel blockers at NMDA receptors and NMDA receptor antagonists. First, only one other drug class is known to elicit STFs: high-efficacy agonists at 5-hydroxytryptamine (serotonin; 5-HT)_1A receptors, as well as 5-HT itself, after the administration of the 5-HT releasers methylene-dioxy-methamphetamine ("ecstasy") and p-chloro-amphetamine (Millan et al., 1991; Bervoets et al., 1993). Second, there is evidence for functional interactions among NMDA and 5-HT_1A receptors, which exert opposing excitatory and inhibitory influences on neuronal activity (Aston-Jones et al., 1991; Strosznajder et al., 1996). Third, while NMDA antagonists modify certain behaviors elicited by 5-HT_1A agonists (Ross et al., 1992), 5-HT_1A antagonists attenuate stereotyped motor responses provoked by the open channel blocker dizocilpine (Löscher and Hönack, 1993). Fourth, 5-HT_1A receptors modulate glutamate release in the central nervous system (CNS; Matsuyama et al., 1996), whereas open channel blockers at NMDA receptors enhance the activity of serotonergic neurons originating in the dorsal raphe nucleus (Whitton et al., 1992; Lejeune et al., 1994). Finally, selective 5-HT_2A receptor antagonists attenuate the hyperlocomotion elicited by the open channel blocker phencyclidine and, in certain studies, dizocilpine (Carlsson, 1995; Maurel-Rémy et al., 1995; Svensson et al., 1995; Millan et al., 1999b).

In light of the above, the purpose of the present study was to evaluate the role of multiple monoaminergic receptors in the induction of STFs through the interruption of transmission at NMDA receptors. First, we evaluated the influence of dizocilpine and CPP compared with the glycine_B antagonist L701,324 on dialysate levels of dopamine (DA), NE, and 5-HT in single dialysate samples of the accumbens of freely moving rats. Second, we examined the influence of selective ligands at multiple dopaminergic, adrenergic, and serotonergic receptors on STFs elicited by the open channel blocker dizocilpine and, for certain drugs, the NMDA receptor antagonist CPP. Furthermore, we evaluated the specificity of antagonistic effects against dizocilpine-induced STFs by determining their comparative ability to block the induction of ataxia by dizocilpine. Third, we examined the actions of the neuroleptic haloperidol, the atypical antipsychotic clozapine, and a diversity of novel antipsychotic agents (Brunello et al., 1995; Meltzer, 1995; Millan et al., 1998a) on the induction of STFs by dizocilpine. The choice of drug doses used here was based on our extensive in vivo studies of models reflecting their actions at specific receptor types (for selective agents) and paradigms of potential therapeutic activity (for antipsychotic agents; Millan et al., 1991, 1994b, 1998b, 1999b; Schreiber et al., 1995; Gobert et al., 1998).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Measurement and Definition of STFs. Male Wistar rats of 220 to 230 g (Iffa Credo, L'Arbresle, France) were housed in sawdust-lined cages with unrestricted access to rat chow and water. There was a 12-h light/dark cycle with lights on at 7:00 AM and off at 7:00 PM. All experiments were undertaken during the light phase. STFs were determined exactly as detailed previously (Millan et al., 1991) in rats loosely restrained in horizontal, opaque, plastic cylinders with the tail emerging from the back to hang over the edge of the bench. One STF was defined as the elevation of the tail to a level higher than that of the body axis. The number of STFs emitted was recorded over 5 min. There was a 5-min adaptation period to the cylinder before the recording of STFs.

Drug Treatment for Inhibition and Induction of STFs. For interaction studies, dizocilpine (0.08 mg/kg s.c.) was administered 30 min before evaluation of STFs. This dose elicits a maximal STF response (Millan, 1991; Millan et al., 1999a), and this time corresponds to its peak effect. Drugs were injected 10 min before dizocilpine (i.e., 40 min before testing). For interaction studies with CPP (20.0 mg/kg s.c.), this NMDA receptor antagonist was administered 60 min before the evaluation of STFs: this corresponds to the maximally effective dose and its time of peak effect (Millan, 1991; Millan et al., 1999a). Drugs were injected 10 min before CPP (i.e., 70 min before testing). For evaluation of the ability of agonists at D₁ and ₂-ARs to elicit STFs, drugs were administered 30 min before testing. In the combination studies, they were administered via two injections given simultaneously 30 min before the evaluation of STFs. These doses and times correspond to those at which they maximally exert their effects at ₂-AR and D₁ receptors, respectively (Millan et al., 1994b).

5,7-Dihydroxytryptamine (5,7-DHT) Lesions of Serotonergic Pathways. The procedure was described previously (Bervoets et al., 1993). Briefly, rats were pretreated with desipramine (25 mg/kg i.p.) and anesthetized with pentobarbital (40.0 mg/kg i.p.), and 5,7-DHT (100 µg/10 µl) or vehicle (ascorbic acid) was injected over 1 min into the lateral ventricle at coordinates of AP = 0.0, L = 1.7, and DV = 3.1. The dose-response relationship for induction of STFs by dizocilpine was evaluated 1 week after the administration of 5,7-DHT. For confirmation of the neurochemical effects of 5,7-DHT, levels of 5-HT, DA, and NE were determined, as described previously (Bervoets et al., 1993), through HPLC and coulometric detection in several CNS regions.

Dialysis Studies. The procedure used was described in detail previously (Gobert et al., 1998). Briefly, male Wistar rats of 200 to 220 g were anesthetized with pentobarbital (60.0 mg/kg i.p.), and a guide cannula was implanted into the core of the nucleus accumbens (AP = +1.4, L = ±2.0, and DV = 5.8). After allowing 5 days for recovery, a concentric dialysis probe (CMA 11; 2 mm length, 0.24 mm o.d.) was introduced; 2 h later, three basal samples (each of 20 min) were collected. Thereafter, vehicle, dizocilpine, CPP, or L701,324 was injected, and 20-min samples were taken for an additional 3 h. DA, NE, and 5-HT levels were quantified in individual dialysis samples through HPLC/coulometric detection as described previously (Gobert et al., 1998).

Loss of Righting Reflex in Rats and Induction of Ataxia in Mice. The loss of righting reflex provoked by dizocilpine was determined 30 min after its s.c. administration using a scoring system described previously (Millan et al., 1994a). Briefly, male Wistar rats of 200 to 240 g were placed on their backs on a bench covered with tissue paper, and their ability to right themselves was assessed according to the following score: 0, immediate and complete righting reflex; 1, attempted righting reflex, with a turn of at least 90 degrees; 2, attempted righting reflex, with a turn of less than 90 degrees; and 3, complete loss of righting reflex, no attempt to turn. The ED₅₀ for dizocilpine was calculated, in line with previous studies (Millan et al., 1994a), on the basis of the percentage of rats displaying a score of at least 1. Dizocilpine was administered 30 min before testing, and in the antagonist study, SCH23390 (0.01 and 0.63) or idazoxan (2.5) (mg/kg, s.c. throughout unless otherwise specified) was injected 10 min before dizocilpine (2.5) or vehicle. The induction of ataxia in male, NMRI mice (22-25 g) was measured as the latency to fall from an accelerating Rotarod located over plates connected to an automatic counter (Ugo Basile, Varese, Italy): the rod accelerated from 4 to 40 rpm over 300 s (Millan et al., 1994a). There was a cutoff of 360 s. Dizocilpine was administered 30 min before testing, and in the antagonist studies, SCH23390 (0.01 and 0.63) or idazoxan (2.5) was administered 10 min before dizocilpine (0.63) or vehicle.

Statistics. All data for dose-response curves and involving multiple groups and comparisons were analyzed by ANOVA. Post hoc Dunnett's tests and Newman-Keuls test were then applied as appropriate. Data from two groups were analyzed by Student's two-tailed t tests. Log-log correlation analyses were performed by generation of Pearson product-moment correlation coefficients. Correlation analyses for ₂-AR antagonists were based on data published by Millan et al. (1994a) and Renouard et al. (1994).

Drugs. For systemic administration, all drugs were dissolved in sterile water, plus a few drops of lactic acid if necessary, and the pH adjusted to as close to neutrality as possible (pH > 5.0) with sodium hydroxide. Drugs were injected s.c. in a volume of 1 ml/kg b.wt., unless otherwise indicated. For intracerebral microinjection, 5,7-DHT was dissolved in sterile saline and administered in a volume of 1 µl. Doses given refer to the base. Chlorpromazine HCl, clozapine base, (+)-dizocilpine maleate, 5,7-DHT sulfate, haloperidol base, ()-pindolol base, prazosin HCl, ()-(S)-raclopride tartrate, ritanserin base, RX821,002 [2-(2,3-dihydro-2-methoxy-1,4-benzodioxin-2-yl)-4,5-dihydro-1H-imidazole HCl], (+)-SCH23390 [(R)-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine HCl], (+)-SKF81297 [(R)-(+)-6-chloro-7,8-dihydroxy-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine HBr], and WB4101 [2-(2,6-dimethoxyphenoxyethyl)aminomethyl-1,4-benzodioxane HCl] were obtained from Sigma (St. Quentin-Fallavier, France). (±)-CPP, dihydrexidine HCl, L741,626 [3-[[4-(4-chlorophenyl)-4-hydroxypiperidin-L-yl]methyl]-1H-indole], and MDL29,551 [2-carboxy-4,6-dichloro-(1H)-indole-3-propanoic acid] were obtained from Tocris-Cookson (Bristol, UK). Amperozide chlorhydrate, atipamezole HCl, BMY7378 [8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione HCl], citalopram HBr, desfluparoxan HCl, DUP734 [1-(cyclopropylmethyl)-4-[2-(4-fluorophenyl)-2-oxoethyl)-piperidine base], fluoxetine HCl, GR218,231 [2(R,S)-(dipropylamino)-6-(4-methoxyphenylsulfonylmethyl)-1,2,3,4-tetrahydronaphtalene) base], (±)-idazoxan HCl, L745,870 [3-{[4-(4-chlorophenyl)piperazin-1-yl]methyl}-1H-pyrrolo[2,3-b]pyridine base], (+)-MDL100,907 [[(R)-(+)--(2,3-dimethoxyphenyl)-1-[2-(4-fluorophenylethyl)]-4-piperidine-methanol] base], NAN190 [1-(2-methoxyphenyl)-4-[4-(2-phthalimido)butyl] piperazine HCl], ondansetron HCl, quetiapine hemifumarate, risperidone base, UK14304 [5-bromo-N-(4,5-dihydro-1H-imidazol-2-yl)-6-quinoxalinamine tartrate], ziprasidonechlorhydrate, (+)-S14297 [(N,N-dipropylamino)-7-tetrahydro-5,6,7,8 naphto-[2,3b]dihydro-2,3 furane dibenzoyltartrate], (+)-S16924 [1-benzodioxanne-5y-3-[3-(4-fluorophenacyl)pyrrolidine]-1-oxapropane HCl], and S18126 [{2-4-(2,3-dihydrobenzo[1,4]dioxin-6-yl)piperazin-1-yl methyl]indan-2-yl} 2HCl] were synthesized by Servier chemists. ARC239 [2-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-4,4-dimethyl-(2H,4H)-isoquinoline-1,3-dione HCl] was obtained from Boehringer (Ingelheim, France). L657,743 [1,3,4,5',6,6',7,12b-octahydro-1',3'-dimethyl- spiro[2H-benzofuro[2,3-a)quinolizine-2,4'(1')-pyrimidin]-2'(3'H)-one base] was obtained from Merck & Co. (Rahway, NJ). Ocaperidone base was obtained from Janssen (Beerse, Belgium). Olanzapine base was obtained from Eli Lilly (Indianapolis, IN). NNC756 [(5S)-8-chloro-5-(2,3-dihydro-7-benzofuranyl)-2,3,4,5-tetrahydro-3-methyl-1H-3-benzazepin-7-ol hemisuccinate] was obtained from Novo (Copenhagen, Denmark). ORG5222 [[trans-5-chloro-2-methyl-2,3,3a,12b-tetrahydro-1H-dibenz[2,3:6,7]oxepino[4,5-c]pyrrole] fumarate]] was obtained from Organon (Oss, Netherlands). Paroxetine HCl was obtained from Beecham (Brentford, England). SCH39166 [[()-trans-6,7,7a,8,9,13b-hexahydro-3-chloro-2-hydroxy-N-methyl-5H-benzo[d]-naphto-[2,1-b azepine] HCl] was obtained from Schering-Plow (Kenilworth, NJ). Sertindole base was obtained form Lundbeck (Copenhagen, Denmark) and zotepine base was obtained from Fujisawa (Osaka, Japan).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Influence of Dizocilpine and CPP Compared with L701,324 on Dialysate Levels of DA, NE, and 5-HT in Nucleus Accumbens. Administered to freely moving rats, dizocilpine elicited a dose-dependent, marked, and sustained elevation in extracellular levels of NE in the nucleus accumbens (Fig. 1). Indeed, at a dose (0.08) that elicits a maximal STF response (Millan, 1991; Millan et al., 1999a), dialysate levels of NE were increased about 2-fold relative to basal values. At this dose, there was, in contrast, no alteration in levels of either 5-HT or DA in the same dialysate samples (Fig. 1). At the highest dose of dizocilpine examined (0.63), there was, in fact, a significant elevation in extracellular levels of 5-HT. However, this response was transient, and its magnitude was substantially less pronounced than that for NE at the equivalent dose (Fig. 1). Furthermore, there was only a minor, variable, and nonsignificant increase in DA levels even at the highest dose (0.63; Fig. 1). In line with these observations, at a dose eliciting a maximal STF response (Millan, 1991; Millan et al., 1999a), CPP (20.0) provoked a significant elevation in levels of NE in nucleus accumbens without influencing those of 5-HT or DA (Fig. 2). The selective glycine_B antagonist L701,324 (40.0), which does not evoke STFs (Millan et al., 1999a), did not influence accumbens levels of NE, DA, or 5-HT (not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Dizocilpine preferentially elevates dialysate levels of NE (NA) versus DA and 5-HT simultaneously quantified in single samples of the nucleus accumbens of freely moving rats. Data are means ± S.E. (n = 5 per value). Absolute basal levels of DA, NE, and 5-HT were as follows: 4.6 ± 0.7; 0.45 ± 0.13, and 0.62 ± 0.06 pg/20 µl dialysate, respectively. ANOVA results for the effect of dizocilpine (0.01 mg/kg s.c.) were 5-HT: F1,12 = 0.3, P > .05; DA: F1,11 = 1.3, P > .05; and NE: F1,9 = 0.1, P > .05. ANOVA results for the effect of dizocilpine (0.08 mg/kg s.c.) were 5-HT: F1,10 = 0.2, P > .05; DA: F1,10 = 3.2, P > .05; and NE: F1,10 = 29.0, P < .05. ANOVA results for the effect of dizocilpine (0.63 mg/kg s.c.) were 5-HT: F1,12 = 9.2, P < .05; DA: F1,11 = 2.4, P > .05; and NE: F1,9 = 10.3, P < .05. P < .05, significance of drug versus vehicle.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   CPP elevates dialysate levels of NE (NA) but not DA and 5-HT simultaneously quantified in single samples of the nucleus accumbens of freely moving rats. Data are means ± S.E. (n = 5 per value). ANOVA results were DA: F1,12 = 0.3, P > .05; NE: F1,11 = 19.6, P < .05; and 5-HT: F1,12 = 0.1, P > .05. P < .05, significance of drug versus vehicle.

Influence of Antagonists at Multiple Dopaminergic Receptors on Induction of STFs by Dizocilpine. The selective antagonists at DA D₁ receptors, SCH23390, SCH39166, and NNC756, all potently, dose dependently, and completely blocked the induction of STFs by dizocilpine (Fig. 3). In contrast, the selective DA D₂ receptor antagonist L741,626 (10.0 mg/kg s.c.) and the selective antagonists at DA D₃ receptors, GR218,231 (2.5 mg/kg s.c.) and S14297 (2.5 mg/kg s.c.), as well as the selective antagonists at DA D₄ receptors, L745,870 (0.16 mg/kg s.c.) and S18126 (0.16 mg/kg s.c.), all failed to significantly modify the induction of STFs by dizocilpine at doses corresponding to those selectively occupying their respective targets (Audinot et al., 1998; Millan et al., 1998b; STFs/5 min: vehicle/dizocilpine, 52.7 ± 8.9, L741,626/dizocilpine, 48.4 ± 5.0, P > .05; vehicle/dizocilpine, 43.2 ± 7.2, GR218,231/dizocilpine, 45.0 ± 8.0, P > .05; vehicle/dizocilpine, 45.3 ± 6.3, S14297/dizocilpine, 45.0 ± 8.0, P > .05, and L745,870/dizocilpine, 45.8 ± 14.1, P > .05; vehicle/dizocilpine, 54.0 ± 6.2, S18126/dizocilpine, 47.8 ± 10.6, P > .05). None of these antagonists elicited STFs on administration alone (not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   The selective D1 receptor antagonists SCH23390, SCH 39166, and NNC756 block induction of STFs by dizocilpine (0.08 mg/kg s.c.). Data are means ± S.E. (n = 5 per value). ANOVA results were SCH23390: F3,42 = 12.5, P < .001; SCH39166: F3,26 = 9.8, P < .001; and NNC756: F3,21 = 29.7, P < .001. P < .05, significance of drug versus vehicle (VEH). ID50 values (95% confidence limits) were NNC756, 0.02 (0.001-0.03); SCH23390, 0.04 (0.02-0.07); and SCH39166, 0.13 (0.04-0.23).

Influence of ₁- and ₂-AR Antagonists on Induction of STFs by Dizocilpine. Figure 4 illustrates the influence of drugs interacting with ₁- and ₂-ARs on the induction of STFs by dizocilpine. Several structurally diverse and preferential antagonists at ₂- versus ₁-ARs, RX821,002 (a benzodioxane), L657,743 (a benzofuroquinolizine), atipamezole and idazoxan (imidazolines), and desfluparoxan (a benzopyrrolidine), all potently, dose dependently, and completely blocked the induction of STFs by dizocilpine. In distinction, the preferential ₁- versus ₂-AR antagonists, prazosin (a quinazolinylpiperazine), WB4101 (a benzodioxane) and ARC239 (an isoquinolinephenylpiperazine), only weakly inhibited the action of dizocilpine. ID₅₀ values (in mg/kg s.c., with 95% confidence limits) were 3.7 (1.6-8.1) for prazosin, 4.0 (2.2-7.6) for WB4101, and 5.0 (3.3-7.4) for ARC239. None of these antagonists elicited STFs on administration alone (not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   The 2-AR antagonists RX821,002, L657,743, desfluparoxan, idazoxan, and atipamezole block induction of STFs by dizocilpine (0.08 mg/kg s.c.). Data are means ± S.E. (n = 5 per value). ANOVA results were RX821,002: F6,51 = 9.8, P < .001; L657,743: F4,34 = 8.6, P < .001; desfluparoxan: F3,23 = 8.0, P < .001; idazoxan: F3,36 = 7.8, P < .001; atipamezole: F4,26 = 6.8, P < .001. P < .05, significance of drug versus vehicle (VEH). ID50 values (95% confidence limits) were RX821,002, 0.02 (0.01-0.04); L657,743, 0.03 (0.01-0.06); atipamezole, 0.05 (0.02-0.10); idazoxan, 0.12 (0.06-0.30); and desfluparoxan, 0.3 (0.2-0.5).

Correlation Analysis Relative to Affinity and In Vivo Activity at ₂-ARs. Although they are more potent ₁-AR than ₂-AR ligands, prazosin, WB4101, and ARC239 all have modest affinity for ₂-ARs. Thus, it was possible to perform a correlation analysis for all eight ₁/₂-AR antagonists between potency for inhibition of dizocilpine-induced STFs and affinity at ₂-ARs (Fig. 5). It may be seen that potency for blockade of dizocilpine-induced STFs correlated significantly (r = 0.90, P < .005) and markedly with affinity at rat _2A-ARs (Renouard et al., 1994). Furthermore, antagonist potency for reduction of STFs also correlated strongly and significantly (P < .001) with several functional parameters in which a role of rat _2A-ARs has previously been demonstrated (Millan et al., 1994a; Hunter et al., 1997): that is, inhibition of the loss of righting reflex (0.95), ataxia (0.97), and antinociception (0.97) provoked by ₂-AR agonists. These data strongly support a specific role of ₂-AR blockade in the inhibition of dizocilpine-induced STFs by these antagonists.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   The potency of 1/2-AR antagonists in blocking dizocilpine-induced STFs is significantly correlated with their affinity for 2-ARs and with their activity in functional paradigms of 2-AR-mediated activity. A, affinity at 2-ARs. B, blockade of xylazine-induced loss of righting reflex in the rat; C and D, blockade of antinociception elicited in the mouse by the 2-AR agonist UK14304. AC, abdominal constriction test (C); HP, hot-plate test (D). Affinity values are from Renouard et al., 1994, and in vivo data are from Millan et al., 1994.

Lack of Influence of D₁ and ₂-AR Antagonists on Induction of STFs by a Low Dose of Dizocilpine. Inasmuch as the dose-response for induction of STFs by dizocilpine is biphasic (Millan, 1991), it might be argued that a loss of dizocilpine-induced STFs in the presence of D₁ or ₂-AR antagonists may reflect a potentiation rather than attenuation of its actions. To this end, we examined the influence of SCH23390 and RX821,002 on the potential induction of STFs by a low dose of dizocilpine (0.02). However, there was no enhancement [vehicle/dizocilpine, 1.2 ± 0.4 STFs/5 min, SCH23390 (0.63)/dizocilpine, 0.0 ± 0.0, RX821,002 (0.16)/dizocilpine, 1.7 ± 0.7, no significant differences (P > .05)].

Lack of Influence of D₁ and ₂-AR Antagonists on Induction of Ataxia by Dizocilpine. Dizocilpine dose dependently (0.16-2.5 mg/kg s.c.) elicited a loss of righting reflex in rats, with a peak effect at a dose of 2.5 mg/kg s.c. (Table 1). Its ED₅₀ (95% confidence limits) value was 0.6 (0.3-0.9). Even at doses sufficient to abolish the induction of STFs by dizocilpine, the D₁ antagonist SCH23390 and the ₂-AR antagonist idazoxan failed to modify the loss of righting reflex provoked by dizocilpine (Table 1). In the mouse, dizocilpine dose dependently decreased the latency to fall in an accelerating Rotarod test with an ID₅₀ (95% confidence limits) value of 0.08 (0.06-0.11) mg/kg s.c). Idazoxan and SCH23390 also failed to block the action of dizocilpine in this model (Table 1).

Influence of D₁ and ₂-AR Antagonists on Induction of STFs by CPP. In analogy to the inhibition of STFs elicited by the open channel blocker dizocilpine, STFs evoked by the NMDA receptor antagonist CPP were potently and dose dependently inhibited by the D₁ antagonists SCH23390, SCH39166, and NNC756 (Table 2). Likewise, the ₂-AR antagonists atipamezole, RX821,002, and idazoxan all blocked the induction of STFs by CPP (Table 2).

View this table: [in this window] [in a new window]   TABLE 2 Influence of 2-AR and D1 receptor antagonists on induction of STFs by CPP (20.0 mg/kg s.c.)

Lack of Induction of STFs by D₁ and ₂-AR Agonists. The selective agonists at D₁ receptors, SKF38393 (0.63-10.0 mg/kg s.c.), SKF81297 (0.04-2.5), and dihydrexidine (0.04-2.5), did not elicit STFs over a dose-range corresponding to their activity (Deveney and Waddington, 1997) at D₁ receptors in other behavioral models (not shown and Table 3). Similarly, the ₂-AR agonist UK14,304 failed to elicit STFs at doses over which it exerts other actions via ₂-ARs (Millan et al., 1994a; Table 3). Furthermore, the combined administration of UK14304 with SKF81297 or dihydrexidine did not elicit a significant STF response (Table 3).

View this table: [in this window] [in a new window]   TABLE 3 Lack of induction of STFs by the independent or combined administration of 2-AR and D1 receptor agonists Values are means ± S.E. for at least five animals per group.

Influence of 5-HT Depletion and Selective 5-HT Reuptake Inhibitors (SSRIs) on Induction of STFs by Dizocilpine. Animals treated 1 week earlier with an i.c.v. injection of the serotonergic neurotoxin 5,7-DHT showed a pronounced reduction in levels of 5-HT and the 5-HT metabolite 5-hydroxyindoleacetic acid in the CNS. In the lumbar spinal cord, their levels were reduced by 90 and 96%, respectively (for 5-HT: vehicle, 427.7 ± 42.1 pg/mg tissue; 5,7-DHT, 42.7 ± 9.3, P < .001; for 5-hydroxyindoleacetic acid: vehicle, 564.7 ± 74.3 pg/mg tissue; 5,7-DHT, 24.0 ± 7.4, P < .001). In contrast, there was no significant alterations in levels of NE or DA in the spinal cord or other tissues (not shown). Despite the substantial reduction in levels of 5-HT in lesioned rats, the dose-response relationship for induction of STFs by dizocilpine was not significantly modified (Fig. 6). Furthermore, on pretreatment of naive rats with the SSRIs citalopram (2.5 mg/kg s.c.), fluoxetine (2.5 mg/kg s.c.), and paroxetine (2.5 mg/kg s.c.), in no case was the induction of STFs by dizocilpine altered (vehicle/dizocilpine, 48.2 ± 6.1, citalopram/dizocilpine, 39.8 ± 8.9, P > .05; vehicle/dizocilpine, 48.2 ± 6.1, fluoxetine/dizocilpine, 44.8 ± 10.0, P > .05; vehicle/dizocilpine, 46.0 ± 5.7, paroxetine/dizocilpine, 51.0 ± 14.4, P > .05). Together, these data indicate that, in contrast to 5-HT releasers (Millan et al., 1991; Bervoets et al., 1993), dizocilpine elicits STFs independently of serotonergic neurons.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   The dose-response relationship for the induction of STFs by dizocilpine is not altered after 5,7-DHT lesions of serotonergic pathways. Data are means ± S.E. (n = 5 per value). ANOVA results were 5,7-DHT × dizocilpine: F4,41 = 0.2, P > .05; dizocilpine: F4,41 = 11.4, P < .001; and 5,7-DHT: F1,41 = 0.1, P > .05. P < .05, significance of drug versus vehicle (VEH).

Influence of Antagonists at Multiple Serotonergic Receptors on Induction of STFs by Dizocilpine. The 5-HT_1A receptor antagonists BMY7378 (2.5 mg/kg s.c.), NAN 190 (2.5 mg/kg s.c.), and ()-pindolol (10.0 mg/kg s.c.) all failed to modify the induction of STFs by dizocilpine [vehicle/dizocilpine, 48.6 ± 7.4, BMY7378/dizocilpine, 39.4 ± 8.4, P > .05; vehicle/dizocilpine, 45.1 ± 9.9, NAN190/dizocilpine, 36.7 ± 6.4, P > .05; vehicle/dizocilpine, 42.5 ± 9.1, ()-pindolol, dizocilpine 45.8 ± 11.1, P > .05]. These doses are higher than their respective ID₅₀ values of 0.4, 0.03, and 0.9 mg/kg s.c. for inhibition of STFs elicited by the 5-HT_1A agonist 8-OH-DPAT (8-hydroxy-2-dipropylaminotetralin; Millan et al., 1991). The mixed antagonist at 5-HT_2A, 5-HT_2B, and 5-HT_2C receptors, ritanserin (2.5 mg/kg s.c.), as well as the selective 5-HT_2A antagonist MDL100,907 (2.5 mg/kg s.c.), failed to modify the action of dizocilpine (vehicle/dizocilpine, 46.6 ± 7.9, ritanserin/dizocilpine, 31.5 ± 8.8, P > .05; vehicle/dizocilpine, 31.6 ± 4.6, MDL100,907/dizocilpine, 44.3 ± 4.2, P > .05). In addition, the 5-HT₂ agonist (±)-1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (0.04 mg/kg s.c.) did not modify the action of dizocilpine [vehicle/dizocilpine (0.08), 49.8 ± 8.8; (±)-1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane/dizocilpine, 49.0 ± 9.8 STFs/5 min, P > .05). Finally, the selective 5-HT₃ antagonist ondansetron did not influence the induction of STFs by dizocilpine [vehicle/dizocilpine, 47.0 ± 9.3, ondansetron (2.5 mg/kg s.c.)/dizocilpine, 51.2 ± 7.2, P > .05].

Influence of Antipsychotic Agents on Induction of STFs by Dizocilpine. The influence of antipsychotics on dizocilpine-induced STFs may be compared with doses eliciting catalepsy (Millan et al., 1998a,b). Consistent with the lack of a major role of D₂ (or D₃) receptors in the induction of STFs, the D₂/D₃ antagonist raclopride only weakly inhibited the induction of STFs by dizocilpine (Table 4). Indeed, its ID₅₀ value was 10-fold higher than that for the induction of catalepsy (0.2 mg/kg s.c.; Millan et al., 1998b). Similarly, the neuroleptic and preferential D₂ antagonist haloperidol blocked dizocilpine-induced STFs (Table 4, Fig. 7) only at doses 6-fold higher than those eliciting catalepsy (0.15 mg/kg s.c.; Millan et al., 1998a). ORG5222 and ocaperidone, which possess potent and prominent D₂ antagonist properties, blocked STFs at doses (Table 4) similar to those eliciting catalepsy: 0.3 and 0.2 mg/kg s.c., respectively. However, risperidone, olanzapine, and ziprasidone, which possess less marked activity at D₂ than other monoaminergic receptor types, all reduced STFs at doses lower than those provoking catalepsy: 1.3, 7.5, and 4.0 mg/kg, respectively. Moreover, sertindole, which shows a similar receptorial profile, blocked STFs without evoking catalepsy (>40.0 mg/kg). Similarly, amperozide, clozapine, quetiapine, and S16924, which have modest affinity for D₂ receptors, blocked STFs without eliciting catalepsy at doses up to 40.0 mg/kg s.c.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Antipsychotic agents block the induction of STFs by dizocilpine (0.08 mg/kg s.c.). Data are means ± S.E. (n = 5 per value). ANOVA results were haloperidol: F3,46 = 11.8, P < .001; clozapine: F3,36 = 6.0, P < .01; olanzapine: F3,22 = 10.2, P < .001; quetiapine: F4,24 = 6.6, P < .001; ziprasidone: F3,29 = 3.0, P < .05; risperidone: F4,39 = 6.3, P < .001; sertindole: F4,41 = 7.9, P < .001; and S16924: F3,29 = 7.3, P < .001. P < .05, significance of drug versus vehicle (VEH).

Correlation Analysis for Antipsychotics Relative to Affinity at _2A-ARs. As indicated earlier, ₂-ARs are implicated in the induction of STFs by dizocilpine, and the antagonist potency of antipsychotics against dizocilpine-induced STFs correlated significantly with their affinity for rat _2A-ARs (Millan et al., 1998a; r = 0.65, P < .05). Interestingly, when these data were reanalyzed incorporating the ₂-AR antagonists indicated in Fig. 5, the correlation coefficient was highly significant (r = 0.79, P < .001; Fig. 8).

View larger version (22K): [in this window] [in a new window]   Fig. 8.   The potency of antipsychotics in blocking dizocilpine-induced STFs is significantly correlated to their affinity at 2-ARs [r = 0.65, P < .05 for antipsychotic agents () and r = 0.79, P < .001 for all drugs].

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Modulation of Accumbens Levels of NE, DA, and 5-HT. At doses exerting a maximal STF response, dizocilpine and CPP elevated extracellular levels of NE in the nucleus accumbens, whereas the glycine_B antagonist L701,324, which fails to elicit STFs, was ineffective (Millan, 1991; Millan et al., 1999a). This elevation in nucleus accumbens levels of NE data suggests that NMDA receptors may exert a tonic, inhibitory influence on extracellular levels of NE in the nucleus accumbens. Inasmuch as glutamatergic input onto adrenergic neurons in the locus ceruleus is excitatory (Jodo and Aston-Jones, 1997), this is an unlikely site of action. Thus, in analogy to local NMDA receptors excitatory to NE release in frontal cortex and hippocampus (Yoshida et al., 1997), NMDA receptors inhibitory to NE release, possibly acting via inhibitory GABAergic interneurons, may be localized in the nucleus accumbens itself (Zhang et al., 1993). Consistent with this possibility, local infusion of dizocilpine into the nucleus accumbens increased extracellular levels of NE therein (Yan et al., 1997). This interaction would also be consistent with induction of STFs by intra-accumbens injection of dizocilpine and CPP (Millan et al., 1999a) and with their blockade of ₂-AR antagonists (see later).

Role of ₂-ARs. A major role of ₂-ARs in NMDA receptor-mediated STFs is indicated by their blockade with chemically diverse 2-AR antagonists, the potency of which correlated with their activities in other functional models of ₂-AR-mediated activity (Fig. 5). Interestingly, the _2A-AR subtype was previously implicated in these in vivo paradigms (Millan et al., 1994a; Hunter et al., 1997), and antagonist potency for blockade of dizocilpine-induced STFs correlated markedly with affinity at rat _2A-ARs (Fig. 5). Inasmuch as dizocilpine and CPP augment NE levels in nucleus accumbens (vide supra), direct blockade of ₂-ARs therein may well be involved in the inhibitions by ₂-AR antagonists of STFs. However, adrenergic mechanisms in the frontal cortex also contribute to the control of motor function (Gioanni et al., 1998), and definitive identification of the population of ₂-ARs involved in the induction of STFs requires future study.

Role of D₁ Receptors. Dizocilpine- and CPP-induced STFs were abolished by the selective D₁ antagonists SCH23390, SCH39166, and NNC756 (Josselyn et al., 1997), demonstrating that D₁ (or closely related D₅) receptors play an essential role in their expression. However, inasmuch as the selective D₁ agonists dihydrexidine and SKF81297 did not elicit STFs, D₁ receptors may, like ₂-ARs, play a permissive role in this response. In contrast to D₁ antagonists, the D₂ antagonist L741,626, the D₃ antagonists GR218,231 and S14297, and the D₄ antagonists S18126 and L745,870 were ineffective (Audinot et al., 1998; Millan et al., 1998b). This implication of D₁ versus D₂ (and D₃/D₄) receptors parallels evidence that: 1) NMDA receptors differentially interact with D₁ versus D₂ sites in the control of motor behavior and 2) activation and blockade of D₁ and NMDA receptors may synergistically facilitate motor function (see Introduction for citations). Very recently, D₁ receptors were shown to enhance NE release in the nucleus accumbens (Vanderschuren et al., 1999), suggesting that their activation might intervene in the elevation of dialysate level of NE elicited by dizocilpine and CPP.

Independence from Serotonergic Mechanisms. Although activation at postsynaptic 5-HT_1A receptors elicits STFs (see Introduction), induction of STFs by dizocilpine and CPP does not reflect serotonergic mechanisms. First, 5-HT_1A receptors mediating STFs are localized in the dorsal horn (Bervoets et al., 1992), yet intrathecal administration of dizocilpine or CPP does not elicit STFs (Millan, 1999a). Second, the activity of serotonergic pathways running to the spinal cord from the raphe magnus is not modified by dizocilpine or CPP (Lejeune et al., 1994). Third, in contrast to 5-HT releasers, induction of STFs by dizocilpine was affected by neither the serotonergic neurotoxin 5,7-DHT nor SSRIs. Fourth, in distinction to 5-HT releasers and 8-OH-DPAT (Millan et al., 1991; Bervoets et al., 1993), 5-HT_1A receptor antagonists did not attenuate induction of STFs by dizocilpine. Furthermore, although 5-HT_2C receptor agonists facilitate induction of STFs by 8-OH-DPAT (Millan et al., 1997), they did not affect the actions of dizocilpine.

Influence of Antipsychotic Agents. In line with the inactivity of L741,626, both the neuroleptic haloperidol and the benzamide raclopride, which are likewise preferential D₂ receptor antagonists (Meltzer, 1995; Millan et al., 1998a), weakly blocked the induction of STFs by dizocilpine, being active only at supracataleptic doses (see Results). In distinction, clozapine and several other potentially "atypical" antipsychotic agents possessing modest affinity for D₂ receptors, S16924, quetiapine, and amperozide, all blocked STFs despite their lack of cataleptogenic potential (Brunello et al., 1995; Meltzer, 1995; Millan et al., 1998a). Similarly, olanzapine and sertindole (Meltzer, 1995) were active at relatively low doses. As mentioned, the selective blockade of ₂-ARs abolished the induction of STFs by dizocilpine and CPP, and there was a significant correlation between antipsychotic potency in blocking STFs elicited by dizocilpine and their affinity for ₂-ARs (Fig. 8). However, correlation coefficients were not significant for D₁, D₂, ₁-AR, 5-HT_2A, or 5-HT_2C receptors (r = 0.14-0.44, P > .05 in each case). This suggests that ₂-AR antagonist properties may, at least partially, be involved in inhibition of STFs by antipsychotic agents. There is increasing interest in the potential significance of the ₂-AR antagonist properties of clozapine and in the management of schizophrenia (Breier et al., 1994; Nutt, 1994; Brunello et al., 1995). However, rather than selective blockade of ₂-ARs per se, it is the association of ₂-AR antagonist actions that may improve clinical profiles of antipsychotic agents (Litman et al., 1996). Similarly, although STFs were abolished by D₁ antagonists, such agents have not, to date, demonstrated antipsychotic efficacy: rather, balanced D₁/D₂ blockade may be a more effective strategy to improve efficacy while limiting extrapyramidal side effects (Brunello et al., 1995; Karlsson et al., 1995).

Conclusions. STFs elicited by open channel blockers and NMDA receptor antagonists are dependent for their expression on ₂-ARs and D₁ receptors. Inhibition of STFs by antipsychotics may involve, at least partially, the blockade of ₂-ARs. Although it would be inappropriate to consider blockade of dizocilpine-induced STFs as predictive of antipsychotic activity per se, this paradigm is of pertinence to schizophrenia in several complementary respects. First, a perturbation of corticolimbic glutamatergic mechanisms may contribute to the pathogenesis of schizophrenia. Second, antagonist properties at D₁ and ₂-ARs are involved in the actions of clozapine and other antipsychotic agents. Third, a further characterization of the circuitry underlying induction of STFs may provide insights into the interrelationships among glutamatergic transmission, monoaminergic networks, and other transmitters implicated in the functional and emotional deficits accompanying psychiatric and neurological disorders.