Twenty-two neuroleptic drugs were studied for interaction with the behavior induced by intravenous injection of apomorphine in rats. All compounds dose-dependently shortened the duration of the apomorphine-induced agitation and—with the exception of clozapine—shortened the onset of the de-arousal grooming that typically occurs immediately after the agitation phase has been terminated. Progressively higher doses were required to antagonize higher levels of apomorphine at earlier time intervals after the intravenous injection. The compounds also decreased palpebral opening, and most of them suppressed grooming behavior at higher doses. Compounds differed considerably in dose increments required for: 1) suppression of grooming, from 0.33 for clozapine to >600 for remoxipride, raclopride, and droperidol; 2) blockade of agitation at 5 minutes after apomorphine, from 2.6 for pimozide to 165 for chlorprothixene and 254 for remoxipride; 3) mild decrease of palpebral opening, from 0.21 for sertindole to 191 for remoxipride; and 4) pronounced decrease of palpebral opening, from 10 for melperone to >820 for raclopride. Only four compounds were able to advance grooming to 15 minutes postapomorphine, but again dose increments varied considerably: droperidol (3.4), pimozide (9.1), raclopride (42), and remoxipride (383). Based on these results obtained in a single animal model, compounds were differentiated in terms of behavioral specificity, incisiveness (the power to counteract the effects of progressively higher apomorphine concentrations), and sedative side-effect liability. Possible explanations for the observed differences and clinical relevance are discussed.
All marketed antipsychotic drugs act by blocking dopamine D2 receptors (Seeman, 2006). They can be classified based on their profile in behavioral and receptor binding tests (Janssen et al., 1965; Niemegeers and Janssen, 1979; Leysen and Niemegeers, 1985; Leysen, 2000). The present study attempts to differentiate between D2 receptor blockers based on a single in vivo test, viz., apomorphine-induced behavior in rats. Apomorphine is a direct dopamine-receptor agonist that mimics the action of dopamine in the brain (Colpaert et al., 1976). Inhibition of apomorphine-induced behavior has been used extensively for screening and profiling neuroleptic drugs in our laboratories (Niemegeers and Janssen, 1979; Megens et al., 1992; Megens and Kennis, 1996; Kapur et al., 2002; Langlois et al., 2012), which has resulted in the development of a large series of antipsychotic drugs (Niemegeers, 1993).
The apomorphine-induced behavior is evaluated by scoring agitation (stereotyped chewing, sniffing, and licking), palpebral opening, and grooming behavior every 5 minutes during the first 60 minutes after intravenous injection of apomorphine. The present study evaluates for the first time the antagonism of the apomorphine-induced behavior as a function of time after the intravenous injection of apomorphine. Peak levels of apomorphine occur within 5 minutes after intravenous injection in rats (Smith et al., 1979; Paalzow and Paalzow, 1986), presumably corresponding to maximum levels of D2 receptor stimulation. Apomorphine levels subsequently rapidly decline to reach the critical level below which the behavior is terminated (~45 minutes after injection). D2 receptor blockers readily block the terminal phase of the behavior when apomorphine levels (and the corresponding degree of D2 receptor stimulation) are close to the critical level below which the behavior is terminated. Progressively higher doses are required to counteract the higher levels of apomorphine at earlier time intervals after apomorphine and thereby reduce the duration of the apomorphine-induced agitation. Based on the required dose increments, we differentiate compounds based on “incisiveness,” i.e., the power to counteract the effects of progressively higher apomorphine concentrations.
Self-grooming is often seen in rats after a variety of activities (e.g., social contact, sexual behavior, exploratory behavior) and usually precedes sleeping. In addition, novelty and other stressors can elicit self-grooming in rats. Grooming reflects the process of de-arousal due to the termination of or habituation to a stressful situation (Spruijt et al., 1992). Such de-arousal grooming is also typically observed in the apomorphine test immediately after the evoked stereotyped behavior has been terminated. D2 receptor blockers dose-dependently shorten the duration of the apomorphine-induced agitation and thereby advance the occurrence of de-arousal grooming to earlier time intervals. However, most D2 receptor blockers are not specific and have associated receptor interactions that suppress grooming at higher doses. Suppression of grooming relative to inhibition of apomorphine-induced agitation can therefore be used as a “behavioral specificity” index. The dose increments required for shortening of the grooming latencies provide another way to evaluate incisiveness. In contrast to blockade of agitation, shortening of grooming latency is not favored by behavioral depressant side effects and therefore is more specifically mediated via D2 receptor blockade.
As apomorphine stimulates both D1 and D2 receptors, the grooming observed at the end of the apomorphine-induced behavior, rather than being an expression of de-arousal grooming, might reflect D1 receptor-mediated stereotyped grooming (Molloy and Waddington, 1987; Perreault et al., 2010), masked by the evoked agitation at earlier time intervals after apomorphine. To evaluate the involvement of D1 receptors in the grooming behavior, available binding affinities for human (h)-D1 and h-D2 receptors are compared with the ability of the compounds to suppress grooming.
Palpebral opening is a measure of general arousal, the eyes being wide open (exophthalmos) during arousal but closed (ptosis) during sedation. Most D2 receptor blockers decrease palpebral opening, probably related to behavioral depression; the decrease of palpebral opening can be promoted by associated interactions, in particular by blockade of α1-adrenoceptors (Janssen et al., 1965; Niemegeers, 1974; Niemegeers and Janssen, 1979; Leysen and Niemegeers, 1985). The effect of apomorphine per se on palpebral opening is relatively neutral, its behavioral stimulant effect just precluding reduction of palpebral opening. This contrasts with the increase of palpebral opening observed after, e.g., amphetamine, which—in addition to dopamine—also releases norepinephrine. Therefore, the decrease of palpebral opening was used to differentiate compounds in terms of sedative side-effect liability. Note that the terms “incisiveness” and “sedative side-effect liability” are pharmacological definitions used to differentiate between compounds and do not necessarily correspond to similar, previously reported clinical classifications (Lambert, 1971). To evaluate the contribution of α1-adrenoceptor blockade, the compounds were also tested for protection against norepinephrine-induced mortality.
The following D2 receptor blockers were included in the analysis: amisulpride, aripiprazole, azaperone, chlorpromazine, chlorprothixene, clozapine, droperidol, haloperidol, melperone, olanzapine, paliperidone, perphenazine, pimozide, quetiapine, raclopride, remoxipride, risperidone, sertindole, ziprasidone, zotepine, and two experimental, fast dissociating D2 receptor blockers, N-[1-(3,4-difluorobenzyl)piperidin-4-yl]-6-(trifluoromethyl)pyridazin-3-amine (JNJ-37822681) (Langlois et al., 2012) and its analog, 4-phenyl-6-piperazin-1-yl-3-(trifluoromethyl)pyridazine (JNJ-39269646). All compounds were studied 1 hour after subcutaneous injection.
Materials and Methods
Test Article and Controls.
All compounds were synthesized in our own laboratories, with the exception of quetiapine, remoxipride, and sertindole (all from Sequoia Research Products, Pangbourne, UK); apomorphine, norepinephrine, and raclopride (Sigma-Aldrich, St. Louis, MO); and zotepine (Knoll AG, Ludwigshafen, Germany). [3H]SCH23390 and [3H]spiperone were obtained from GE Healthcare (Little Chalfont, Buckinghamshire, UK). The compounds used to determine nonspecific binding of these radioligands [piflutixol and (+)butaclamol for the h-D1 and h-D2 receptors, respectively] were synthesized internally. For in vitro studies, the test compounds were dissolved in dimethylsulfoxide (final concentration in the assays 1%). For in vivo studies, the test compounds were dissolved in distilled water (apomorphine, chlorpromazine, chlorprothixene, melperone, raclopride, remoxipride) and, if required, acidified with tartaric acid (amisulpride, azaperone, clozapine, droperidol, haloperidol, olanzapine, risperidone, paliperidone), or in 10% hydroxypropyl-β-cyclodextrin acidified with tartaric acid (aripiprazole, JNJ-37822681, JNJ-39269646, pimozide, quetiapine, sertindole, ziprasidone, zotepine). The solutions were stored at room temperature in closed containers protected from light. The preparations were subcutaneously administered in a single dose using a standard volume of 10 ml/kg. Solvents were also tested, and a large collection of data obtained on the effects of apomorphine in solvent-pretreated rats has been generated.
Animals (Species, Weight, and Sex).
Male Wistar rats (Crl:WI; Charles River Germany; 220 ± 40 g) were housed under standard laboratory conditions (21 ± 2°C; 50–65% relative humidity; light-dark cycle set at 12 hours; lights on at 6:00 AM) and fasted overnight before the start of the experiments (tap water remained available ad libitum). Not taking into account solvent-pretreated control rats, 990 rats were used in the present study for testing 22 compounds over a range of 7–12 doses (n = 5 per dose; see Fig. 4 for the doses tested per compound). During the test period, they were housed in individual cages. The local Ethics Committee approved all studies in compliance with the Declaration of Helsinki. See individual test descriptions for information on body weight and sex.
In Vitro Binding Affinity for h-D2L and h-D1 Receptors.
Membranes expressing the different receptors of interest were prepared as follows. Cells were transfected with cloned human receptor cDNA, collected by scraping, and homogenized in 50 mM Tris-HCl, pH 7.4, using an Ultra Turrax homogenizer (IKA-Werke GmbH & Co. KG, Staufen, Germany). The homogenate was centrifuged for 10 minutes at 23,500g in a Sorvall RC-5B centrifuge (4°C). The cells were then suspended in 5 mM Tris-HCl, pH 7.4, and after recentrifugation for 20 minutes at 30,000g at 4°C, the pellet was homogenized in 50 mM Tris-HCl, pH 7.4, divided into aliquots, and stored at –80°C. Binding assays were carried out under incubation conditions as summarized in Table 1.
Apomorphine-Induced Agitation in Rats.
Apomorphine-induced (1.0 mg/kg i.v.) agitation (compulsive sniffing, licking, chewing), grooming behavior, and palpebral opening were scored every 5 minutes over the first hour after injection of apomorphine. The score system was as follows: (A) for agitation and grooming: (3) pronounced, (2) moderate, (1) mild, and (0) absent; (B) for palpebral opening: (5) exophthalmos, (4) wide open, (3) open for three-quarters, (2) half open, (1) open for one-quarter, (0) closed. The onset of grooming behavior (a de-arousal phenomenon, marking the end of the apomorphine-induced behavior) was noted as the first 5-minute block at which the score for grooming was >0.
Norepinephrine-Induced Lethality in Rats.
Survival time after norepinephrine (0.63 mg/kg i.v.) was recorded up to 1 hour after challenge in rats pretreated 1 hour earlier with test compound or solvent. In control rats, this dose of norepinephrine is lethal within 5 minutes. Survival times greater than 60 minutes were considered to reflect significant norepinephrine antagonism (0% false positives in controls; n = 241). Protection against norepinephrine lethality evaluates the ability of test compounds to block peripheral α1-adrenoceptors.
All experiments were performed by unbiased trained technicians using coded solutions. Doses were selected from the geometrical series 0.0025–0.005–0.01...............40.0–80.0–160 mg/kg. Each dose group consisted of five animals that were tested in separate daily experimental sessions to account for day-to-day variability and to minimize systematic errors. Control injections of solvent were included in each experimental session.
The percentage of binding was plotted against the log concentration of the test compound and the sigmoidal log concentration-effect curve of best fit was calculated by nonlinear regression analysis using the Lexis software interface (developed at Janssen Research & Development, Beerse, Belgium). From these concentration-response curves, the IC50 values (concentration causing 50% inhibition of binding) were calculated. IC50 values were converted to Ki values assuming competitive inhibition using the Cheng-Prusoff equation. The pKi values from several experiments were averaged and presented as Ki values with corresponding 95% CL.
All-or-none criteria for drug-induced effects were defined by analyzing a frequency distribution of a large series of historical control data obtained in solvent-pretreated rats (see Results), aiming for <5.0% false positives. Based on the criteria obtained in this way, ED50s (the dose at which 50% of the tested animals respond to the criterion adopted for drug-induced effect) and corresponding 95% CL were determined according to the modified Spearman-Kaerber estimate, using theoretical probabilities instead of empirical ones (Tsutakawa, 1982). This modification allows determination of ED50 and its confidence interval as a function of the slope of the log dose-response curve. An internal report on the method is available upon request (Lewi et al., 1977). At least three doses, covering the 0–100% effect range of the dose-response curve, were used for ED50 determinations.
Linear Regression and Correlation.
The inter-relationship between ED50s for different effects was studied by performing and graphing linear regression and Spearman rank correlation using GraphPad Prism version 5.01 for Windows (GraphPad Software, San Diego, CA, www.graphpad.com).
Results and Conclusions
Historical Solvent-Pretreated Control Rats
Figure 1 (upper panel) shows the summation of all scores for agitation, grooming, and palpebral opening given per 5-minute time interval after the apomorphine challenge over a large historical series of solvent-pretreated rats (n = 6180). Agitation was maximal from 5 to 30 minutes postapomorphine and then declined to 0 from 35 minutes to 50 minutes. De-arousal grooming was observed during the period from 40 to 50 minutes after apomorphine, marking the recovery from the apomorphine-induced agitation. Palpebral opening slightly increased during the 35 minutes of agitation, decreased during the period of grooming (when the rats close their eyes while grooming), and then increased again.
Figure 1 (bottom panel) shows the percentage of rats showing specific scores for agitation, grooming, and palpebral opening as a function of time after apomorphine. Agitation started to decline from 35 minutes onwards for score 3, from 40 minutes onwards for scores ≥2, and from 45 minutes for scores ≥1. As agitation was almost always present (at least score ≥1) during the first 40 minutes after apomorphine, dose-dependent blockade of agitation (to score 0) could be evaluated for all 5-minute time intervals during these first 40 minutes after apomorphine. Grooming was observed during the period of 40–50 minutes after apomorphine; the graphs for scores >0, >1, and >2 coincided because grooming was almost always scored 3. Low scores for palpebral opening (≤2) were almost never observed during the period of apomorphine-induced behavior but often occurred during the period of grooming behavior.
Figure 2 shows the frequency distributions of the number of specific scores given for agitation and palpebral opening and for grooming latency. Regarding agitation (Fig. 2A), fewer than 7 scores of ≥1, fewer than 6 scores of ≥2, or fewer than 6 scores of 3 never or extremely rarely occurred (0.00, 0.00, and 0.13% false positives, respectively). In our standard procedure, this criterion was adopted as the all-or-none criterion to determine ED50s of the test compounds for inhibition of agitation. Fewer than 2 scores of ≥1 or 0 scores of ≥2 never occurred, and this criterion was adopted as the all-or-none criterion to determine ED50s of the test compounds for blockade of agitation. Regarding palpebral opening (Fig. 2B), more than 0 scores of 0, more than 2 scores of ≤1, or more than 8 scores of ≤2 occurred very rarely (0.05, 0.48, and 0.47% false positives, respectively), which was adopted as the criterion for a mild decrease of palpebral opening. As many as 12 scores of ≤1 never occurred in control rats, which was adopted as the criterion for pronounced decrease of palpebral opening after test compounds. Regarding grooming behavior (Fig. 2C), the onset of grooming was almost never within the first 40 minutes after apomorphine (0.44% false positives); dose-dependent occurrence of grooming at earlier time intervals was used to determine ED50s for advancement of grooming to earlier time intervals. Finally, a grooming latency >50 minutes (3.4% false positives) was adopted as a criterion indicating suppression of grooming.
Pretreatment with Test Compounds
Table 2 lists the ED50s (95% CL; mg/kg s.c.; –1 hour) for inhibition of agitation and decrease of palpebral opening in the apomorphine test according to our standard procedure. Many D2 receptor blockers have associated α1-adrenoceptor blocking activity. Apart from central D2 receptor blockade, this activity is considered to be an important factor in the effects of the compounds on palpebral opening, agitation, and grooming behavior. To evaluate this possible interference, ED50s for protection against norepinephrine-induced lethality (reflecting blockade of peripheral α1-adrenoceptors) have also been listed for comparison.
Antagonism of Apomorphine-Induced Behavior as a Function of Time after Apomorphine.
Blockade of Apomorphine-Induced Agitation.
Table 3 lists the ED50s of the test compounds for blockade of the apomorphine-induced agitation to score 0 as a function of time after apomorphine. Agitation was most readily blocked at the 40-minute time interval. The ED50s obtained at 40 minutes (Table 3, first column) were very comparable to those obtained for inhibition of apomorphine-induced agitation according to the standard procedure (Table 2, first column). The ratio of both ED50 values was close to 1.0 (mean ± S.D.: 1.2 ± 0.3), ranging from 0.33 for aripiprazole to 2.3 for perphenazine. The ED50 at the 40-minute time interval after apomorphine was used as a reference to calculate dose increments for all other effects discussed below. These ED50s are graphically illustrated in Fig. 3, sorted from low to high. Progressively higher doses were required to block agitation at earlier time intervals after apomorphine. The required dose increments (relative to the basal ED50 at 40 minutes) as function of time after apomorphine have been listed in Table 4 and are graphically illustrated in Fig. 4 (red circles). The results will be further discussed below (see “Blockade of Agitation at 5 Minutes after Apomorphine”).
Advancement of De-arousal Grooming.
Table 5 lists the ED50s of the test compounds for advancing the onset of grooming behavior to progressively earlier time intervals after apomorphine. Table 6 lists the same ED50s but expressed as dose increments relative to the basal ED50 of the compound for blockade of agitation at 40 minutes after apomorphine (listed in Table 3, first column). These dose increments are graphically illustrated in Fig. 4 (green symbols), together with the dose-increments required to block agitation (red circles), as function of time after apomorphine. With increase of dose of the D2 receptor blockers, grooming is advanced to earlier time intervals, corresponding to the blockade of agitation at progressively earlier time intervals after apomorphine and reflecting the shortening of the duration of the apomorphine-induced behavior. Up to the dose at which grooming is suppressed, the curves for blockade of agitation and for advancement of grooming almost coincided. A close linear relation between the ED50s for both effects for the 40-minute interval after apomorphine is shown in Fig. 5; the linear regression line almost coincides with the bisector (Fig. 5; r2 = 0.87, P <0.001), with a slope close to unity (0.87 ± 0.03). For most compounds, however, advancement of grooming to earlier time intervals was limited by suppression of grooming behavior at higher doses (Tables 5 and 6, final columns (>50 minutes); see also below). The results will be further discussed below.
Differentiation between Compounds.
The test compounds can be differentiated based on potency (i.e., the ED50 for inhibition of apomorphine-induced agitation) and on the dose increments required to suppress grooming behavior, to block agitation shortly (5 minutes) after the intravenous injection, to advance grooming to early time intervals, and to decrease palpebral opening. For all of these effects, which will be further discussed below, the dose increments are relative to the basal ED50 for apomorphine antagonism, i.e., the ED50 for blockade of agitation 40 minutes after apomorphine. These ED50s have been graphically depicted in Fig. 3 to differentiate between compounds based on potency: A) high potency (ED50 <0.05 mg/kg): droperidol, raclopride, perphenazine, and haloperidol; B) intermediate potency (ED50 between 0.05 and 0.5 mg/kg): pimozide, remoxipride, risperidone, ziprasidone, olanzapine, paliperidone, zotepine, aripiprazole, chlorprothixene, JNJ-37822681, azaperone, chlorpromazine, JNJ-39269646, melperone; C) low potency (ED50: >0.5 mg/kg): sertindole, quetiapine, amisulpride, clozapine.
Suppression of Grooming Behavior.
Suppression of the de-arousal grooming can be used as a measure of the behavioral specificity of the compounds. As long as grooming behavior is not suppressed, interference of secondary behavioral depressant or stimulant actions of the compounds is not likely. Figure 6 shows the dose increments (relative to blockade of agitation at 40 minutes) required for suppression of de-arousal grooming. Clozapine suppressed grooming already at a lower dose than that required for inhibition of agitation (factor 0.33). On the other hand, relatively specific D2 receptor blockers such as droperidol, raclopride, and remoxipride did not suppress grooming behavior up to >600 times the basal ED50 for blockade of agitation at 40 minutes. Thus, suppression of de-arousal grooming after the apomorphine challenge does not seem to depend on D2 receptor blockade but rather to be mediated via secondary, nondopaminergic behavioral side effects of the compounds.
Table 7 compares the dose increments required for suppression of grooming behavior (shown in Fig. 6) with the corresponding ratios of D1 over D2 binding affinity. Compounds such as olanzapine and quetiapine that display similar affinities for both types of receptor suppress grooming behavior only after increasing dose with a factor 48 and 14, respectively, making it unlikely that the grooming behavior is an expression of D1 receptor-mediated stereotypy.
Blockade of Agitation at 5 Minutes after Apomorphine.
Peak plasma and brain levels of apomorphine and corresponding maximum levels of D2 receptor stimulation may be expected to occur shortly after the intravenous injection of apomorphine. Therefore, it is of interest to compare the dose increments of the compounds that are required to block agitation at 5 versus 40 minutes after apomorphine, corresponding to high and low levels, respectively, of D2 receptor stimulation. As shown by the data in Fig. 7, the dose increments varied considerably among the tested compounds.
Advancement of De-arousal Grooming to Earlier Time Intervals.
As the levels of apomorphine and the corresponding degree of D2 receptor stimulation are higher at progressively earlier time intervals after apomorphine, higher doses of D2 receptor blockers are required to reduce the excessive D2 receptor stimulation to normal and thereby trigger de-arousal grooming. However, the advancement of grooming behavior to earlier time intervals is limited by associated nondopaminergic behavioral depressant effects that suppress grooming at higher doses of the test compounds. Compounds can thus be classified based on the maximal obtained advancement of grooming and the corresponding dose increments to achieve that effect (Fig. 8).
No advancement of grooming was observed after clozapine. This sedative compound suppressed grooming already below the dose required for apomorphine antagonism, precluding advancement of grooming to earlier time intervals.
Maximum advancement of grooming to 15 minutes was achieved with only four compounds but at very different dose increments (in parentheses): pimozide (3.4), droperidol (9.1), raclopride (42), remoxipride (383) (see Fig. 8). All four compounds are potent and relatively specific D2 receptor blockers. Therefore, it is rather remarkable that the required dose increment varied considerably.
Grooming within 10 minutes was not consistently achieved with any of the compounds. Even an intravenous injection of saline causes sufficient stress in most animals to suppress grooming during the first 10 minutes. Consistent advancement of grooming behavior to intervals ≤10 minutes postapomorphine may therefore be impossible, even with compounds that suppress the apomorphine stimulus completely.
Decrease of Palpebral Opening.
Figure 9 shows the dose increments of the compounds required for mild and pronounced decrease of palpebral opening (relative to the basal ED50 for blockade of agitation at 40 minutes). A mild decrease of palpebral opening was generally observed at similar dose levels as inhibition of agitation. The reduction of behavioral stimulation achieved by blockade of central D2 receptors is apparently already sufficient to reduce palpebral opening for most compounds. In line with its predominant α1-adrenoceptor blocking activity (see norepinephrine antagonism in Table 2), sertindole already decreased palpebral opening below the dose affecting agitation (margin: 0.21). Least sensitive toward effect on palpebral opening (mild decrease) were JNJ-39269646 (margin: 21), raclopride (24), JNJ-37822681 (42), aripiprazole (56), and remoxipride (191). Being specific D2 receptor blockers, the latter compounds differentiated in this respect from other rather specific D2 receptor blockers such as pimozide (margin: 2.0) and droperidol (3.4).
Within the tested dose range, pronounced decrease of palpebral opening was obtained only with melperone (margin: 10), paliperidone (18), risperidone (24), pimozide (36), perphenazine (41), droperidol (95), and haloperidol (219).
Although other neurotransmitters may be involved, it is still accepted nowadays that neuroleptic drugs exert antipsychotic activity and produce extrapyramidal side effects (EPS) mainly by blocking D2 receptors. Antagonism of apomorphine-induced behavior has been extensively used in our laboratories for screening neuroleptic drugs (see Introduction) and predicting antipsychotic plasma levels and specificity margins relative to side effects (e.g., catalepsy, palpebral ptosis, interactions with other receptors). By including de-arousal grooming, the present analysis allows characterization of behavioral specificity of compounds in one and the same model, reducing the need for testing in additional models. Moreover, the study of apomorphine antagonism as a function of time reveals remarkable differences in ability of compounds to counteract the effects of the progressively higher apomorphine concentrations at earlier time intervals after the apomorphine injection; these differences may have clinical consequences.
It is unlikely that the grooming observed at the end of the apomorphine-induced behavior reflects D1 receptor-mediated stereotyped grooming (Molloy and Waddington, 1987; Perreault et al., 2010) rather than de-arousal grooming. First, grooming is also observed after nondopaminergic stimulants (e.g., N-methyl-d-aspartate antagonists; unpublished observations), immediately after the stereotyped behavior has been terminated and likely reflecting de-arousal. Moreover, compounds such as olanzapine and quetiapine that display similar D1 and D2 affinities suppress grooming only after relatively high dose increments, making it unlikely that the grooming is a D1 receptor-mediated response. Finally, if the inhibition of D2 receptor-mediated agitation would result in unmasking of D1 receptor-mediated grooming, it would not only shorten the onset but also increase the duration of grooming, this behavior being maintained as long as apomorphine concentrations remain high enough. This was not the case: the grooming was always short-lasting and observed immediately after termination of the agitation and, therefore, very likely reflecting de-arousal.
The dose increments required to suppress grooming highly varied. Whereas the sedative neuroleptic clozapine suppressed grooming already below the dose required for inhibition of agitation, other compounds (aripiprazole, haloperidol, perphenazine, droperidol, remoxipride, raclopride) did not suppress grooming up to >100 times the ED50 for inhibition of agitation. In view of the well-known behavioral depressant effects of D2 antagonists, it was surprising that some of them did not at all suppress grooming or did only at very high doses. This apparent discrepancy can be explained by the mutually antagonistic interactions between the test compounds and apomorphine at the level of the D2 receptor. The occurrence of de-arousal grooming marks the time at which there is a match between both mutually opposing actions. The suppression of grooming observed with other compounds is therefore most likely related to nondopaminergic side effects.
The relatively good correlation with norepinephrine antagonism (Fig. 10) suggests that suppression of grooming is mediated at least partially via blockade of α1-adrenoceptors. Other mechanisms of action may be involved as well. The fast-dissociating D2 antagonist JNJ-37822681 suppressed grooming from 12.4 mg/kg onwards, but a mechanism of action is not evident from its activity profile (Langlois et al., 2012). It is presumed that associated nondopaminergic behavioral depressant effects are involved. Such behavioral depressant effects may be helpful in the treatment of short-term psychosis or exacerbations in subjects with long-term schizophrenia but are not desired in maintenance treatment.
The compounds also differed remarkably in dose increment required to block agitation 5 minutes after the intravenous injection of apomorphine when levels of apomorphine and corresponding levels of D2 receptor stimulation may be expected to be maximal. Blockade of agitation is likely promoted by other factors, e.g., behavioral depressant effects (e.g., α1-adrenoceptor blockade) or counteracted by other activities (e.g., associated antimuscarinic activity). In case of specific D2 receptor blockers and assuming competitive interaction with apomorphine at the level of the D2 receptor, however, one would expect comparable dose increments to be required for the various compounds to counteract the higher apomorphine levels at earlier time intervals. However, even among specific D2 receptor blockers, the required dose increments varied considerably from a factor 2.6 for pimozide to a factor >60 for JNJ-39269646, JNJ-37822681, aripiprazole, raclopride, chlorprothixene, and remoxipride. Differences between D2 receptor blockers in rate of dissociation from the D2 receptor (Kapur and Seeman, 2001; Langlois et al., 2012) may play an important role. A slow dissociation rate may be responsible for the small dose increment required for pimozide (2.6). This compound is apparently not readily displaced from the D2 receptor by the injection of apomorphine. On the other hand, the large dose increments required for other compounds may result from a fast rate of D2 receptor dissociation (JNJ-39269646, JNJ-37822681, remoxipride, raclopride and possibly chlorprothixene) or other factors, e.g., functional-selective actions at different D2 receptor signaling pathways (aripiprazole) (Urban et al., 2007).
The dose-dependent advancement of de-arousal grooming to earlier time intervals can be measured up to the ED50 for suppression of grooming. Thus, maximum advancement of grooming to 15 minutes after apomorphine is obtained with specific D2 receptor blockers only. It occurred at very different dose increments (in parentheses) with pimozide (3.4), droperidol (9.1), raclopride (42), and remoxipride (383). Again differences in rate of dissociation from the D2 receptor may be involved: slow displacement for pimozide and droperidol and fast displacement for raclopride and remoxipride. Suppression of grooming may have masked the full potential of a fast dissociation rate for compounds such as clozapine and quetiapine and, to a lesser extent, also JNJ-37822681 and JNJ-39269646. As already mentioned, other factors may affect the dose-dependency of the effect, e.g., functional selectivity in the case of aripiprazole (Urban et al., 2007) or associated antimuscarinic activity in the case of clozapine and olanzapine (Langlois et al., 2012).
A mild decrease of palpebral opening was generally already obtained from doses only slightly above the dose affecting agitation, suggesting that antagonism of agitation is already sufficient to decrease palpebral opening; the agitation indeed precludes the closing of the eyes that you would expect to occur in naïve rats. Decrease of palpebral opening is promoted by associated α1-adrenoceptor blockade as indicated by the results obtained with sertindole, quetiapine, paliperidone, and clozapine (Fig. 9). On the other hand, relatively large dose increments were required for JNJ-39269646, raclopride, JNJ-38722681, aripiprazole, and remoxipride. Functional selective actions at different D2 receptor signaling pathways (Urban et al., 2007) may again play a role in the profile observed for aripiprazole, whereas fast dissociation from the D2 receptor may spare palpebral opening over a wide dose range in the case of the other compounds.
In the present analysis, we assume brain levels of the test compounds to remain more or less constant during the critical observation period, i.e., 5–40 minutes after apomorphine and 65–100 minutes after the subcutaneous injection of the test compounds, which seems to be a reasonable assumption. The analysis is based on dose increments. As an increase in dose does not always result in a dose-proportional increase in plasma and/or brain levels, nonlinear kinetics may be another interfering factor in the study. However, nonlinear kinetics may occur less frequently after subcutaneous dosing than after oral dosing.
By protecting against norepinephrine lethality, several compounds displayed α1-adrenoceptor blocking activity. This property undoubtedly affects the interaction profile in the apomorphine test by favoring decrease of palpebral opening, inhibition of agitation, and suppression of grooming. Tachyphylaxis to the effects of α1-adrenoceptor blockade develops rapidly after repeated administration (Graham et al., 1976). For compounds with prominent α1-adrenoceptor blocking activity, the present results obtained after single administration in rats may therefore be less predictive for their activity profile after long-term administration in patients.
Study of the dose-dependent shortening of the apomorphine-induced behavior allows differentiation between D2 receptor blockers based on incisiveness, i.e., the power to counteract the effects of progressively higher apomorphine concentrations. This is particular valid for specific D2 receptor blockers that do not suppress grooming behavior over a large dose range and act primarily by blocking D2 receptors. Specific D2 receptor blockers that readily block agitation at early time intervals while preserving grooming up to high doses levels can be classified as incisive. The largest differentiation in terms of incisiveness was obtained with pimozide, droperidol, raclopride, and remoxipride. To evaluate the clinical relevance of the observed differences in pharmacodynamic incisiveness, a view on the clinical profile of these compounds may be helpful.
Remoxipride produced similar effects on positive and negative symptoms as classic neuroleptic drugs but with less extrapyramidal and sedative side effects; amelioration of cognitive, conative, affective, and emotional functions have been reported (Lewander, 1994). Relative to haloperidol, remoxipride maintained alertness (Awad et al., 1997) and caused less impairment of psychomotor performance, cognitive functioning, cortical activation, and subjective well-being (Rammsayer and Gallhofer, 1995). Remoxipride also showed low EPS liability, even after long-term treatment (Mendlewicz et al., 1990; Pflug et al., 1990; Walinder and Holm, 1990). Patients often felt that they were not taking any antipsychotic drugs (Meltzer, 1995). Remoxipride was withdrawn due to incidence of aplastic anemia. Because of less beneficial pharmacokinetic properties (Lewander, 1995), raclopride was studied less extensively clinically than remoxipride; it was an effective antipsychotic drug and well tolerated with low EPS liability (The British Isles Raclopride Study Group, 1992; Lewander, 1995). Pimozide was characterized by incisive effects, notably against autism, delusions, and hallucinations but also by socializing effects (Brugmans, 1968). It has been reported to be particularly effective in treating monosymptomatic hypochondriacal psychosis and delusional jealousy (Opler and Feinberg, 1991). Because of the absence of sedative properties, pimozide was less efficient for treating short-term agitated psychosis (Opler and Feinberg, 1991). Initial claims of low EPS liability could not be maintained (Opler and Feinberg, 1991). In contrast to pimozide, droperidol has α1-adrencoeptor blocking activity (see Table 2), increasing sedative effects. Because of its rapid onset, short duration of action, and capacity to potentiate the morphine-like action of fentanyl, droperidol was mainly used as a preanesthesia sedative and, in combination with fentanyl, for neuroleptanalgesia (Ayd, 1980). Parenteral droperidol appeared particularly useful for treating acutely disturbed behavior in psychotic and nonpsychotic patients; it had EPS liability, in particular regarding dystonic reactions and akathisia (Ayd, 1980).
Based on above clinical results, the pharmacodynamic incisiveness measured in the present study has apparently nothing to do with efficacy against agitation in short-term psychosis but apparently translates into clinical incisiveness, i.e., high efficacy against positive symptoms and high EPS liability (prototype pimozide). By not being readily displaced from the D2 receptor by rising dopamine concentrations after environmental stimuli, these compounds preclude adequate responding. As a consequence, EPS and worsening of negative and cognitive symptoms may be expected. Associated α1-adrenoceptor blocking activity is required to increase sedative effects and thereby enhance efficacy against agitation in short-term schizophrenia (prototype droperidol). Low pharmacodynamic incisiveness (prototype remoxipride) still allows effective antipsychotic therapy. Fast displacement from the D2 receptor by rising concentrations of dopamine results in a more flexible D2 receptor blockade, still allowing adequate responding to environmental stimuli and resulting in less EPS and less impairment of alertness, cognition, and subjective well-being.
The present results obtained in one single animal model allowed differentiation of neuroleptic drugs in terms of potency, behavioral specificity, incisiveness, and sedative side-effect liability. The most interesting differences were observed among the specific D2 antagonists. De-arousal grooming is also observed after other stimulants, e.g., N-methyl-d-aspartate antagonists, allowing similar analyses for nondopaminergic mechanisms.
The authors thank Dr. Willem Talloen for review and discussions and Eva Huybrechts (K15 BVBA, Zoersel, Belgium, email@example.com) for adapting the figures to the journal requirements.
Participated in research design: Megens.
Conducted experiments: Hendrickx.
Performed data analysis: Megens, Lavreysen.
Wrote or contributed to the writing of the manuscript: Megens, Hendrickx, Lavreysen, Langlois.
- Received June 26, 2013.
- Accepted September 26, 2013.
This work was funded by Janssen Research and Development, LLC. The sponsors also provided a formal review of this manuscript. All authors are employees of Janssen Research and Development, a division of Janssen Pharmaceutica NV.
All authors met International Council of Medical Journal Editors criteria. All authors had access to the study data, provided direction and comments on the manuscript, made the final decision about where to publish these data, and approved submission to the journal.
- extrapyramidal side effects
- confidence limits
- Copyright © 2013 by The American Society for Pharmacology and Experimental Therapeutics