Introduction

Abstract Introduction Materials & Methods Results Discussion References

Remoxipride is a substituted benzamide (Florvall and Ögren, 1982) with antipsychotic efficacy (den Boer and Westenberg, 1990; Keks et al., 1994; Lambert et al., 1995; see Lewander et al., 1990). In laboratory studies, remoxipride has antagonized DA receptor agonist-induced behaviors in rats (Ögren et al., 1984) and produced an increased DA synthesis and turnover upon systemic administration (Magnusson et al., 1987a, b). In receptor binding studies, remoxipride displays a selective, albeit weak, affinity for DA D₂ receptors, in comparison with other DA receptor subtypes or other neurotransmitter receptors, such as alpha adrenergic, histaminergic or cholinergic (Hall et al., 1986; Mohell et al., 1993; see Jackson et al., 1993). Altogether, this provides the picture of a selective DA D₂ receptor antagonist.

In the initial pharmacological characterization, the interesting observation was made that the difference in doses needed to antagonize apomorphine-induced hyperlocomotion and stereotypy was much greater for remoxipride than for a classical antipsychotic such as haloperidol. In further contrast to haloperidol, remoxipride had very low propensity to produce catalepsy in rats (Ögren et al, 1984). These observations suggested an atypical extrapyramidal side-effect profile for remoxipride as an antipsychotic, and this was confirmed in controlled clinical studies (Keks et al., 1994; Klieser et al., 1994; Lambert et al., 1995) in the few years before remoxipride was withdrawn from the market because of case reports of aplastic anemia. Also, regarding endocrine side-effects, remoxipride may have certain advantages over existing pharmacotherapies (Chouinard, 1987; Awad et al., 1990; Lahdelma et al., 1991; von Bahr et al., 1991).

The possibility that remoxipride is a prototype, nonclozapine, atypical antipsychotic warrants close scrutiny of its mechanism(s) of action, which presently appears related to its high selective affinity for the DA D₂ receptor (see Jackson et al., 1993). It should also be noted, however, that remoxipride has several metabolites, some of which also display high affinity for DA D₂ receptors (Högberg et al., 1987; Mohell et al., 1993). This applies to metabolic reactions in the aromatic, but not in the pyrrolidine, moiety of remoxipride. There is also a species difference here in that the former pathway is relatively more important in rodents, as compared with dogs and humans (see Widman et al., 1993). A possible role of one or more of remoxipride metabolites for its in vivo pharmacological profile was highlighted in this laboratory by the accidental observation that the local intrastriatal application of remoxipride (40 µg) in the rat brain did not affect neostriatal DA turnover. The aromatic ring metabolites FLA-797 and FLA-908, by the same route and dose, produced an increase by 150 and 250%, respectively (unpublished observations). It should also be noted that remoxipride has few, if any, effects in in vitro tests for DA D₂ receptor antagonist properties (Westlind-Danielsson et al., 1994; Nilsson and Eriksson, 1995). These observations received further support from the recent finding that remoxipride does not affect DA-induced hyperpolarization of lactotrophs in a rat pituitary preparation (J. Luthman, personal communication). This evidence, taken together, is a strong indication that remoxipride, by itself, is a poor antagonist at brain DA D₂ receptors.

In the present report, a possible contribution of the remoxipride metabolites FLA-797, FLA-908, NCQ-436, NCQ-469 and NCQ-344 (see Widman et al., 1993; Nilsson, 1997) (see fig. 1) for in vivo effects of remoxipride was critically evaluated. Three different approaches were used: (1) the time course of action for behavioral effects of remoxipride in comparison with the related substituted benzamide raclopride (Köhler et al., 1985), both of which have a similar short half-time of about 35 min in the rat (Widman et al., 1993; Ahlenius et al., 1991); (2) the effect of intracerebroventricular administration of remoxipride on brain DA synthesis and on behavior, normally and after d-amphetamine or quinpirole treatment, in comparison with effects of ()-sulpiride (Trabucchi et al., 1975). This latter substituted benzamide has poor ability to pass the blood-brain barrier and has considerably less potency than remoxipride or haloperidol on systemic administration regarding behavioral and biochemical effects in the rat (e.g., Magnusson et al., 1986), whereas the reverse is true for the intracerebroventricular route of administration (Nishibe et al., 1982); (3) the dose-effect comparison between remoxipride and the above-mentioned metabolites regarding brain DA synthesis, as estimated by the accumulation of DOPA after decarboxylase inhibition (see Carlsson et al., 1972), and effects on spontaneous open-field locomotor activity (see Ericson et al., 1991).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Aromatic ring metabolites of remoxipride and suggested relations to the parent compound.

	Materials and Methods

Abstract Introduction Materials & Methods Results Discussion References

Animals. Adult male Sprague-Dawley rats (B&K Universal AB, Sollentuna, Sweden), 280 to 320 g body weight, were used. The animals arrived in the laboratory at least 10 days before the experiments started and were housed, 5 per cage (Makrolon IV), under controlled conditions of light-dark cycle (12:12 h, lights off 6:00 A.M.), relative humidity (55-65%) and temperature (18-20°C). Food (R36, Ewos, Södertälje, Sweden) and tap water were freely available in the home cage.

Drugs. The drugs used in the present series of experiments are listed in table 1. Reserpine, sulpiride, FLA-908 and FLA-797 were dissolved in a minimal quantity of glacial acetic acid and made up to volume with isotonic glucose, whereas the other drugs were dissolved in physiological saline only. The injection route was subcutaneous, intraperitoneal or intracerebroventricular, as given in the text, tables and figures. For systemic injections the injected volume was kept constant at 2 ml kg¹.

Surgery for intracerebroventricular cannulation. The animals were deeply anesthetized with a pentobarbital formulation (Mebumal Vet., Nordvacc, Huddinge, Sweden), 60 mg kg¹ i.p (1 ml kg¹) Thereafter, the rats were mounted in a sterotaxic frame and provided with guide cannulas (21-gauge) reaching the dura mater above the lateral ventricles at the following coordinates relative bregma (1.0 mm) and the midline (±1.3 mm). The injection needles (31-gauge) reached the ventricles 4 mm below the brain surface. Coordinates were adopted from the stereotaxic atlas of Paxinos and Watson (1986). The animals were allowed 1 week of postoperative recovery before being included in experiments. Immediately after the operation and henceforth, the animals were housed individually in Makrolon III cages. As the animals recovered from surgery, no gross behavioral abnormalities were noted, and the animals gained weight rapidly. Thus, the mean weight ± S.D. at the time of surgery was 299 ± 8 g and had increased to 367 ± 23 g at the end of behavioral experiments, 2 to 3 weeks later. The animals were given bilateral intracerebroventricular injections of ()-sulpiride or remoxipride (2-5 µl, injected at a rate of 1.33-3.33 µl min¹).

Locomotor activity observations. The spontaneous motor activity was observed in a square, open-field arena (approximately 0.5 m²), equipped with photocells sensitive to infrared light. The photocells were spaced 90 mm apart, and the last photocell in a row was spaced 25 mm from the wall. The open-field was enclosed in a ventilated, sound-attenuating box with a perspex top. Locomotor activity was defined as the square root of the number of photocell beam interruptions. Measurements were made in the dark and performed between 9:00 A.M. and 4:00 P.M. For further details about the apparatus used, see Ericson et al. (1991).

Treadmill locomotion. The animals were trained to walk on a drum (Ø = 166 mm), rotating at a speed of 8 rpm, resulting in a speed of approximately 4 m min¹. The rats were trained to walk for 3 min twice a day for 2 consecutive days. On the day of experimentation, a pretest was performed, and only those rats that were able to walk continuously for 3 min were included in experiments. Further tests were performed for a maximal time of 2.25 min. Treadmill performance was scored from 0 to 5 according to the time (SQR transformation) the animals walked on the drum: 0 = 0 to 0.08, 1 = 0.09 to 0.35, 2 = 0.36 to 0.80, 3 = 0.81 to 1.42, 4 = 1.43 to 2.24, 5 = >2.25 min.

Catalepsy. Animals were placed on an inclined (60°) grid, and the time the rat remained immobile in the same position was measured for a maximum of 2.25 min. The catalepsy was scored as described for the treadmill performance above, i.e., if the rat remained in the same position for >2.25 min it was scored as 5, etc. For further details on procedures and equipment used for the treadmill and catalepsy observations, see Ahlenius and Hillegaart (1986).

Core temperature. Core temperature measurements were made in a temperature-controlled room (ambient temperature, 21.0 ± 0.4°C). Recordings were made by means of a commercially available telethermometer (YSI-2100, Yellow Springs Instruments Co., Yellow Springs, OH) and an accompanying probe (YSI-402). The probe, lubricated with mucilago etalosi AF-68 (ACO Läkemedel AB, Stockholm, Sweden), was inserted rectally (about 90 mm) in the rat, which was gently restrained by hand. The telethermometer was connected to an automatic printer device that was activated when the temperature reading had stabilized (±0.1°C) for 10 s (see Salmi et al., 1994).

Biochemical measurements. After decapitation by means of a guillotine, the whole brain, including the olfactory bulb rostrally and the medulla oblongata caudally, was removed quickly and placed in a mould where it could be sliced into 2.5-mm sections by a thin stainless steel wire (Ø = 70 µm). The ventral neostriatum (including the nucleus accumbens, the olfactory tubercle, the diagonal band of Broca and the bed nucleus of the stria terminalis), the dorsolateral (d-l) neostriatum and the overlying neocortex were dissected on ice from one of these slices. The rostral edge of this slice was approximately +2.1 mm in relation to bregma. The brain was cut at an inclination of 7°, such that ventrally the sections extended slightly rostrally, according to the horizontal plane in the atlas of Paxinos and Watson (1986). This report is based on results from 14 separate experiments, and the grand mean weight (mg) ± S.D. of the samples thus obtained were: 26.0 ± 3.6 (ventral neostriatum); 21.7 ± 2.3 (d-l neostriatum); 104.0 ± 5.5 (neocortex). In a few experiments a larger sample of the neostriatum was dissected, including also its ventromedial aspect. The corresponding figures for this sample were 45.4 ± 2.4 mg. The brain samples were immediately frozen on dry ice and stored at 70°C until processing. DOPA and 5-HTP were determined in the brain samples by means of coupled-column liquid chromatography with electrochemical detection. The preparations of the samples and further details are given in Magnusson et al. (1980, 1988) and in Mohringe et al. (1986).

Statistics. Parametric description and analysis was used for biochemical and locomotor activity data (see Winer, 1971), whereas corresponding nonparametric procedures were used for results obtained in experiments on treadmill locomotion and on catalepsy (see Siegel, 1956). The specific procedures are indicated in the tables and figure legends.

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Effects of NCQ-436 and NCQ-469 on DOPA accumulation after decarboxylase inhibition in the ventral striatum and on locomotor activity in normal and reserpine-treated rats. For schedule of drug injections and doses of reserpine and NSD-1015, see figure 7. The results are presented as means ± S.D. based on three values per group. Statistical evaluation by means of a two-way ANOVA, followed by t tests for comparisons with appropriate controls as indicated in the figure. NCQ-436. DOPA: F1,23 = 0.30, n.s. (pretreatment); F3,23 = 11.49, P < .001 (dose); F3,23 = 24.35, P < .001 (pretreatment × dose). Locomotion: F3,16 = 172.38, P < .001 (pretreatment); F3,16 = 6.87, P < .01 (dose); F3,16 = 15.23, P < .001 (pretreatment × dose). NCQ-469. DOPA: F1,16 = 7.08, P < .05 (pretreatment); F3,16 = 4.29, P < .05 (dose); F3,16 = 2 .30, n.s. (pretreatment × dose). Locomotion: F1,16 = 603.92, P < .001 (pretreatment); F3,16 = 2.32, n.s. (dose); F3,16 = 2.22, n.s. (pretreatment × dose). nsP > .05; *P < .05; **P < .01.

View larger version (34K): [in this window] [in a new window]   Fig. 9.   Effects of remoxipride, NCQ-344, FLA-797 and FLA-908 on neostriatal DOPA accumulation in NSD-1015-treated rats. The schedule of drug injections followed the protocol for "normal rats" in figure 7 with the exception that the test compounds (or vehicle) were administered at 70 min and the locomotor activity was monitored for 20 min beginning at 50 min. The results are presented as means ± S.D. based on four to five animals per group. Statistical evaluation was performed by means of one-way ANOVAs for each experiment separately, followed by the Dunnett's t test for comparisons with the pooled controls shown by the hatched area (mean ± S.D., n = 19). NCQ-344, F3,30 = 145.31, P < .001; FLA-797, F3,30 =34.85, P < .001; FLA-908, F3,30 = 28.91, P < .001; FLA-731, F4,30 = 7.15, P < .001. nsP > .05; *P < .05; **P < .01.

	Results

Abstract Introduction Materials & Methods Results Discussion References

Time Course of Action and Importance of Route of Administration for Effects of Remoxipride on Behavior and on Neostriatal DA Synthesis

Effects of remoxipride administration on neostriatal DOPA accumulation, as a function of route of administration. There was a dose-related, statistically significant increase in neostriatal DOPA accumulation 50 min after an i.p. remoxipride injection (0.9-60.0 µmol kg¹). In this dose range, remoxipride had no statistically significant effects by the s.c. route at this time interval (fig. 2), whereas the reference compound raclopride (4 µmol kg¹) produced a maximal effect within 20 min of s.c. administration (data not shown). It should be noted, however, that prolongation of the time interval from 50 to 70 min, between remoxipride administration (0.9-60.0 µmol kg¹) and decapitation, produced a statistically significant effect also by the s.c. route of administration. Thus, a comparison of the dose-effect curves at these two time intervals by means of a two-way ANOVA showed a marked increase in DOPA accumulation with time (F_1,29 = 6.77, P < .025 and F_1,29 = 22.73, P < .01, for the ventral and dorsal neostriatum, respectively.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effects of remoxipride on DOPA accumulation in the ventral striatum of NSD-1015-treated rats. Twenty minutes after the administration of remoxipride, or the saline vehicle, all animals received the decarboxylase inhibitor NSD-1015, 100 mg kg1 i.p., and were decapitated 30 min later for brain dissections. Data are presented as means ± S.D., based on observations of four animals per group. The statistical analysis was performed by means of a one-way ANOVA, followed by the Dunnett's t test for comparisons with saline-treated controls, as indicated in the figure. The mean control value ± S.D. (pooled for i.p. and s.c. administration) is shown by the hatched areas. Intraperitoneal route: F4,15 = 10.82, P < .001; subcutaneous route: F4,13 = 1.43, n.s. nsP > .05; *P < .05; **P < .01.

The effect of remoxipride and raclopride on exploratory locomotor activity as a function of route of administration. Results from an initial series of experiments showed the ED₅₀ values for suppression of locomotor activity by remoxipride and raclopride to be approximately 30.0 and 2.0 µmol kg¹ i.p., respectively. In the experiments presented in figure 3, the time course of action for these doses was monitored after both s.c. and i.p. administration. In general agreement with results from catalepsy and treadmill experiments (see below), the onset was faster for raclopride (<30 min) than for remoxipride (>1 h). Furthermore, there was a marked difference in the effects obtained by the different routes of administration for the two compounds. Thus, the s.c. route of administration was significantly more efficacious than the i.p. route for raclopride, whereas the opposite was true for remoxipride. In fact, the locomotor activity showed no statistically significant effects at all from s.c. administered remoxipride, in comparison with saline-treated controls (cf. fig. 2).

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effects of remoxipride and raclopride on open-field locomotor activity in rats. Remoxipride and raclopride were administered i.p. or s.c. 30 and 20 min, respectively, before the start of a 15-min session in the open-field arena. Shown are the means ± S.D., based on observations of four animals per group. Statistical evaluation was performed by means of a two-way ANOVA, followed by appropriate t tests for comparisons between the two routes of administration at the different time intervals, as indicated in the figure. Raclopride: F1,36 = 19.51 (route), P < .01; F5,36 = 16.69 (time), P < .01; F5,36 = 3.93, P < .01 (route × time); Remoxipride: F1,36 = 33.71 (route), P < .01; F5,36 = 1.97 (time), P > .05; F5,36 = 2.14, P > .05 (route × time). *P < .05.

Effects of remoxipride and raclopride on cataleptic rigidity and on treadmill performance: time course of action. The lowest dose of remoxipride, or raclopride, at which maximal effects were obtained in the respective test situation, were estimated from careful dose-effect studies in preliminary experiments and were found to be 37.5 and 30.0 µmol kg¹ i.p. (catalepsy) and 18.8 and 8.0 µmol kg¹ i.p. (treadmill), respectively. The time course of action for the two compounds, at the doses thus determined, are shown in figure 4. As shown in the figure, the peak effect of raclopride occurred 30 to 60 min after administration in both test situations, and the duration was from 4 to 8 h (catalepsy) to 2 to 4 h (treadmill). In comparison, remoxipride had its peak effect at 4 h after administration, and the duration was 4 to 8 h, similar for both test situations.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Time course of action for remoxipride and raclopride on catalepsy and on treadmill locomotion in the rat. Remoxipride and raclopride were administered at time 0 h, at the doses indicated in the figure. Shown are the medians ± semi-interquartile range, based on repeated observations of 5 to 12 animals per group. Statistical analysis was performed by means of the Friedman two-way ANOVA. Raclopride: 2(5) = 28.98, P < .01 (catalepsy); 2(3) = 10.50, P < .02 (treadmill). Remoxipride: 2(5) = 10.81, P > .05 (catalepsy); 2(6) = 12.81, P < .05 (treadmill).

Effects of Intracerebroventricular Remoxipride Injections

Effects on DOPA accumulation. As shown in figure 5, there were no statistically significant effects by the remoxipride treatment (12.5-200 nmol i.c.v., bilaterally) on the DOPA accumulation in the ventral striatum, or in any of the other brain areas investigated (unpublished observations). In stark contrast, ()-sulpiride (12.5-200 nmol i.c.v., bilaterally) produced a dose-dependent and statistically significant increase in the DOPA accumulation in the same area of the neostriatum.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Effects of intracerebroventricular administration of ()-sulpiride and remoxipride on DOPA accumulation in the ventral striatum of NSD-1015 treated rats. ()-Sulpiride or remoxipride were administered 40 min, and NSD-1015 (100 mg kg1 s.c.) 30 min, before the animals were sacrificed for biochemical experiments. For details on the intracerebroventricular administration procedure, see "Materials and Methods." The figure shows means ± S.D., based on four determinations per group. The results were analyzed by means of a two-way ANOVA, followed by t tests for comparisons with the respective saline control group, as indicated in the figure. F1,23 = 106.02, P < .001 (treatment); F3,23 = 21.17, P < .001 (dose); F3,23 = 18.99, P < .001 (treatment × dose). nsP > .05; **P < .01.

Effects on spontaneous or d-amphetamine-induced locomotor activity. Remoxipride (6.3-50 nmol i.c.v., bilaterally) administration produced no statistically significant effects on spontaneous locomotor activity, whereas a small, but statistically significant decrease (17%) was noted at the highest dose (100 nmol) (fig. 6). In a separate experiment, plasma remoxipride levels were measured 20 min after administration of this 100-nmol dose and were found to be 32 ± 6 nmol l¹. It should be noted that upon systemic i.p administration remoxipride plasma concentrations in this range also are associated with a corresponding, statistically significant decrease in locomotor activity. In fact, the plasma EC₅₀ value of remoxipride for effects on locomotor activity, on systemic administration, is less than 30 nmol l¹ (unpublished observations). The administration of ()-sulpiride (6.3-100 nmol i.c.v., bilaterally) caused a marked and statistically significant decrease in the locomotor activity (fig. 6).

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effects of intracerebroventricular administration of ()-sulpiride and remoxipride on spontaneous open-field locomotor activity in the rat. ()-Sulpiride or remoxipride were administered 20 min before a 20-min observation period in the open-field arena. Repeated measurements were made on the same animals in a changeover design (Li, 1964) (n = 8 and 9 for the sulpiride and the remoxipride experiments, respectively). The results are presented as means ± S.D. Statistical analysis was performed by means of a two-way ANOVA for repeated measurements, followed by paired t tests for comparisons with the appropriate saline control condition, as indicated in the figure. F1,15 = 15.31, P < .01 (treatment); F3,45 = 64.14, P < .001 (dose); F3,45 = 29.02, P < .001 (treatment × dose). nsP > .05; *P < .05; **P < .01.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Schedule of drug injections for experiments presented in figures 8 to10. lma = locomotor activity.

1

View this table: [in this window] [in a new window]   TABLE 2 Effects of ()-sulpiride and remoxipride on amphetamine-induced locomotor stimulation in rats Immediately upon d-amphetamine (AMPH) administration (1.0 mg kg1 s.c.), the animals were placed in the open-field arena and 20 min later (marked by the arrow in the table), ()-sulpiride (SPR) or remoxipride (RMX) were injected i.c.v. (25 nmol, bilaterally). Controls were given the solvent vehicle at corresponding time points. The table shows mean locomotor activity min1 ± S.D., expressed as per cent of vehicle-treated controls at the respective time interval. Repeated measurements were made on the same animals (n = 10), which served as their own controls with a changeover design (Li, 1964). Statistical analysis for comparisons between the different treatment conditions was performed by means of an appropriate two-way ANOVA (A × B × S design) (Keppel, 1982). F1,5 = 2.56, n.s. (AMPH vs. AMPH + RMX); F1,5 = 11.90, P < .05 (AMPH vs. AMPH + SPR); F1,7 = 8.48, P < .05 (AMPH + RMX vs. AMPH + SPR).

Effects on quinpirole-induced hypothermia. The administration of the DA D_2/3 receptor agonist quinpirole (0.4 mg kg¹ s.c.) produced a marked and statistically significant hypothermia, lasting up to 4 h. The quinpirole-induced hypothermia was fully antagonized by pretreatment with ()-sulpiride (25 nmol i.c.v. bilaterally), whereas the same dose of remoxipride was ineffective (table 3).

View this table: [in this window] [in a new window]   TABLE 3 Effects of ()-sulpiride and remoxipride on quinpirole-induced hypothermia in rats ()-Sulpiride (SPR) or remoxipride (RMX) were injected i.c.v. (25 nmol, bilaterally), and quinpirole (QPR) (0.4 mg kg1 s.c.), 30 and 20 min before the first temperature recording. Controls were given the solvent vehicle at corresponding time points. The table shows mean core temperature ± S.D. in relation to time for i.c.v. treatments. Repeated measurements were made on the same animals (n = 5), which served as their own controls with a changeover design (Li, 1964). Statistical analysis for comparisons between the different treatment conditions was performed by means of an appropriate two-way ANOVA (A × B × S design) (Keppel, 1982). F1,3 = 0.90, n.s. (QPR vs. QPR + RMX); F1,3 = 13.91, P < .05 (QPR vs. QPR + SPR); F1,3 = 42.54, P < .05 (QPR + RMX vs. QPR + SPR).

Effects of Remoxipride and Some Biologically Active Metabolites on Spontaneous Locomotor Activity and on Forebrain Catecholamine Synthesis

Effects of NCQ-436 and NCQ-469 on the neostriatal DOPA accumulation and on Spontaneous Locomotor Activity in Normal and Reserpine-Treated Rats. The administration of NCQ-436 (1.9-30.0 µmol kg¹ s.c.) produced an increase and a decrease of the DOPA accumulation in normal and reserpine-treated animals, respectively (fig. 8, top). There were no effects on the neocortical DOPA or 5-HTP accumulation in normal animals, whereas the neocortical DOPA and 5-HTP accumulation was decreased by the NCQ-436 treatment in reserpine-treated rats (tables 4 and 5). The spontaneous locomotor activity was decreased in a dose-dependent manner in normal animals, whereas no stimulation was found in the reserpine-treated rats (fig. 8, top).

View this table: [in this window] [in a new window]   TABLE 4 Effects of NCQ-436 and NCQ-469 on neocortical DOPA accumulation in NSD-1015-treated rats The table shows mean values ± S.D. (nmol g1). The results presented in the table belong to the experiments presented in figure 8, and for details regarding schedule of drug injections and statistical evaluation see the legend to figure 8. NCQ-436, F1,24 = 14.05, P < .001 (pretreatment); F3,24 = 5.85, P < .01 (dose); F3,24 = 7.30, P < .01 (pretreatment × dose). NCQ-469, F1,16 = 39.91, P < .001 (pretreatment); F3,16 = 0.40, n.s. (dose); F3,16 = 0.08, n.s. (pretreatment × dose).

View this table: [in this window] [in a new window]   TABLE 5 Effects of NCQ-436 and NCQ-469 on neocortical 5-HTP accumulation in NSD-1015-treated rats The table shows mean values ± S.D. (nmol g1). The results presented in the table belong to the experiments presented in figure 8, and for details regarding schedule of drug injections and statistical evaluation see the legend to figure 8. NCQ-436, F1,24 = 6.23, P < .05 (pretreatment); F3,24 = 5.83, P < .01 (dose); F3,24 = 2.01, n.s. (pretreatment × dose). NCQ-469, F1,16 = 7.39, P < .01 (pretreatment); F3,16 = 0.70, n.s. (dose); F3,16 = 0.43, n.s. (pretreatment) × dose).

1

Effects of remoxipride, NCQ-344, FLA-797 and FLA-908 on neostriatal DOPA accumulation and on spontaneous locomotor activity in normal and reserpine-treated rats. The effects of remoxipride (FLA-731) (0.9-60.0 µmol kg¹ s.c.) and some of its identified metabolites on neostriatal DOPA accumulation are shown in figure 9. The three compounds, NCQ-344 (3-200 nmol kg¹ s.c.), FLA-797 (8-500 nmol kg¹ s.c.) and FLA-908 (0.5-30.0 µmol kg¹ s.c.), all produced a dose-dependent increase in the neostriatal DOPA accumulation. Remoxipride, by the s.c. route, had only small (although statistically significant) effects, in contrast to the effects obtained by i.p. administration (fig. 9, cf. fig. 2). All the other compounds were more potent and more efficacious than remoxipride (fig. 9). In contrast to remoxipride, NCQ-344, FLA-797 and FLA-908 also produced an increase in neocortical DOPA accumulation (table 6), whereas no effects were found on neocortical 5-HTP accumulation (table 7).

View this table: [in this window] [in a new window]   TABLE 6 Effects of remoxipride and some of its identified metabolites on neocortical DOPA accumulation in NSD-1015-treated rats The table shows mean values ± S.D. (nmol g1). The same group of animals was used in this experiment as in the experiments presented in figure 9, and for further details see the legend to figure 9. FLA-797, F3,16 = 3.33, P < .05; FLA-908, F3,16 = 7.53, P < .01; FLA-731, F4,15 = 0.72, n.s.

View this table: [in this window] [in a new window]   TABLE 7 Effects of remoxipride and some of its identified metabolites on neocortical 5-HTP accumulation in NSD-1015-treated rats The table shows mean values ± S.D. (nmol g1). The same group of animals was used in this experiment as in the experiments presented in figure 9, and for further details see the legend to figure 9 FLA-797, F3,16 = 0.51, n.s.; FLA-908, F3,16 = 2.30, n.s.; FLA-731, F4,15 = 0.38, n.s.

1

View larger version (30K): [in this window] [in a new window]   Fig. 10.   Effects of remoxipride, NCQ-344, FLA-797 and FLA-908 on locomotor activity in rats. The observations were performed in connection with experiments presented in figure 9. For schedule of drug injections and details on statistical description and analysis, see figures 7 and 9. NCQ-344, F3,30 = 148.29, P < .001; FLA-797, F3,30 = 59.80, P < .001; FLA-908, F3,30 = 45.46, P < .001; FLA-731, F4,30 = 3.90, P < .05. nsP > .05; *P < .05; **P < .01.

	Discussion

Abstract Introduction Materials & Methods Results Discussion References

The present results demonstrate that the effects of remoxipride on neostriatal DA synthesis highly depend on route of administration. In normal rats, an increased DA synthesis was obtained 50 min after i.p. administration. No statistically significant effects were obtained when remoxipride was administered by the subcutaneous route, thereby avoiding first-pass hepatic metabolism. Furthermore, also in the locomotor activity tests, remoxipride was only effective by the i.p., and not by the s.c., route of administration, when given at an ED₅₀ dose (i.p.) for suppression of locomotor activity. This was in contrast to raclopride, which was much less effective by the i.p. than the s.c. route. Furthermore, the time course of action was markedly different for remoxipride and raclopride in the catalepsy and treadmill tests. The onset of effect was slower for remoxipride, and the peak effect appeared in both situations about 3 h later in comparison with raclopride, which had a prompt onset of action (<30 min). It should be noted that the half-time for remoxipride and raclopride is approximately the same (30 min), and both drugs display a high first-pass hepatic extraction ratio (see Widman et al., 1993). If anything, remoxipride has a somewhat shorter half-time than raclopride. Thus, the differences found between remoxipride and raclopride regarding route of administration and time course of action do not appear to be related to differences in t_1/2 or other pharmacokinetic parameters (Widman et al., 1993; Ahlenius et al., 1991), which suggests the operation of biologically active metabolites as an explanation for the aberrant behavior of remoxipride.

No, or only minor, effects of remoxipride on neostriatal DA synthesis or on spontaneous locomotor activity occurred when the i.c.v. route of administration was used. This also applied to antagonism of d-amphetamine-induced hyperactivity or quinpirole-induced hypothermia in the rat. In stark contrast, ()-sulpiride in the same dose range as used for remoxipride had marked effects by itself, and also antagonized the d-amphetamine- or quinpirole-induced effects. The suppression of spontaneous locomotor activity, the increase in striatal DA synthesis as well as antagonism of drug-induced hyperactivity or hypothermia by ()-sulpiride are in good agreement with the effects of this compound on systemic administration (see Spano et al., 1978). However, the effects of ()-sulpiride are obtained more readily by the i.c.v. route because ()-sulpiride passes poorly over the blood-brain barrier (see Nishibe et al., 1982; Ahlenius et al., 1990). Thus, the systemic potency relationships between remoxipride and ()-sulpiride are reversed on i.c.v. administration. The weak, but statistically significant, effects on locomotor activity, seen after 100 nmol of remoxipride i.c.v., were obtained at plasma remoxipride levels of 30 nmol l¹. More extensive unpublished studies in this laboratory on correlations between plasma remoxipride levels and suppression of locomotor activity for individual animals disclose a weak, but statistically significant, correlation of r = 0.61 (t₁₃ = 2.77, P < .05). The corresponding value for raclopride is r = 0.89 (t₃₄ = 11.28, P < .001). It was also found that maximal plasma and brain drug levels after an i.p. injection of remoxipride (18.6 and 32.0 µmol kg¹) were reached within 20 min. The calculated EC₅₀ value, based on linear regression analysis, for effects of remoxipride on locomotor activity was <30 nmol l¹. The highest dose of remoxipride used in the i.c.v. studies is thus a maximal dose for the present purpose. Finally, remoxipride by the i.c.v. route of administration neither affected basal DA receptor-mediated functions, nor did it antagonize effects produced by DA receptor stimulation. Together with the above-mentioned comparisons between remoxipride and raclopride regarding the route of parenteral administration and the time course of action, the i.c.v. studies provide strong evidence for hepatic bioactivation of remoxipride.

The concept of remoxipride as a prodrug, for effects as a DA D₂ receptor antagonist in laboratory in vivo studies, receives strong support from recently published in vitro experiments. Thus, remoxipride displays a very low propensity to antagonize DA D₂ receptor-mediated inhibition of cAMP formation in rat neostriatal tissue (Westlind-Danielsson et al., 1994). Furthermore, in a cultured cloned prolactin-producing pituitary tumor cell line, transfected with DA D₂ receptors, raclopride, but not remoxipride, antagonized quinpirole-induced suppression of prolactin release (Nilsson and Eriksson, 1995). Finally, in electrophysiological experiments on lactotrophs from rat pituitary, DA-induced hyperpolarization was effectively blocked by raclopride, but not by remoxipride (J. Luthman, personal communication).

As mentioned above, remoxipride is a drug with a high hepatic extraction ratio. The predominant metabolic reactions in the rat are oxidations and hydroxylations in the aromatic moiety of the compound (Widman et al., 1993). NCQ-436, NCQ-469, FLA-797 and FLA-908 have all been shown to have affinity for brain DA D₂ receptors (Högberg et al., 1987; Gawell et al., 1989; Mohell et al., 1993) and to produce behavioral effects in rats (Ögren et al., 1993). In the present experiments, all these compounds, as well as NCQ-344, displayed biological activity in normal rats, as evidenced by an increased neostriatal DA synthesis and, with the exception of NCQ-469, a suppression of exploratory locomotor activity. With the DOPA accumulation in normal rats as an index of DA receptor-blocking properties, NCQ-436 and NCQ-469 have a low potency, with ED₅₀ values from 25 to 30 µmol kg¹ s.c. (see table 8). The compounds FLA-908, FLA-797 and NCQ-344 are considerably more potent and the corresponding ED₅₀ values were 2.2, 0.1 and 0.01 µmol kg¹ s.c., respectively (see table 8). Thus, all the metabolites range from approximately equipotent to about 2,000 times as potent as remoxipride. Because the remoxipride ED₅₀ values were calculated from i.p. experiments, these ratios are on the conservative side.

NCQ-436 was found also to antagonize the reserpine-induced increase in the neostriatal DA synthesis, which suggests DA receptor agonist properties. All the other metabolites were ineffective in this regard. It is interesting that similar antagonism of reserpine-induced activation of neostriatal DA synthesis was reported previously for remoxipride (Ahlenius et al., 1993). Considering the evidence presented in this report, those effects in all probability reside in NCQ-436, and not in remoxipride itself. The DA receptor agonism produced by remoxipride is weak, however, and it was not possible to achieve a potentiation of locomotor stimulation in reserpine-treated rats by the addition of the DA D₁ receptor agonist SKF-38,393. In fact, this also applied to NCQ-436 (unpublished observations). Several partial DA D₂ receptor agonists have been tried in schizophrenia (e.g., Naber et al., 1992). It is possible, however, that the intrinsic activity of these compounds, modeled on ()-3-(3-hydroxyphenyl)-N-n-propylpiperidine (3-PPP) (Hjorth et al., 1983), possess too high an efficacy as DA receptor agonists to be effective as antipsychotics, and that weak agonist properties, together with a predominant antagonist profile, is a desirable feature for an atypical antipsychotic (cf. Coward et al., 1989). In addition to its effects on neostriatal DA synthesis, NCQ-436 also produced decreased neocortical DOPA and 5-HTP accumulation in the reserpine-treated animals. In view of the neocortical region sampled being predominantly a noradrenergic area, the DOPA accumulation should primarily reflect changes in brain noradrenaline (NA) synthesis in this case. Together, these findings suggest that the compound NCQ-436 also has at least some intrinsic activity at brain alpha-2 adrenoceptors and at brain 5-HT receptors. The latter finding is of particular interest, because several observations indicate that the side-effect profile of DA receptor-blocking antipsychotics can be improved by concomitant manipulations with brain serotonergic neurotransmission (e.g., Ahlenius, 1989; Ugedo et al., 1989; Wadenberg and Ahlenius, 1991; Wadenberg, 1996).

There are several interesting possibilities to explain the mechanism whereby remoxipride itself should exert its atypical antipsychotic profile in the clinic. Thus, for example, affinity for brain sigma receptor sites (Largent et al., 1988), affinity for a subpopulation of DA D₂ receptors (Malmberg et al., 1993; Ögren et al., 1994) and high selectivity for DA D₂ receptors, albeit with weak affinity (Mohell et al., 1993), have been presented as explanations. It is also noteworthy that other substituted benzamides, like sulpiride, raclopride eticlopride and at least FLA-797 of the remoxipride metabolites, display high selectivity for DA D₂ versus D₁, alpha adrenoceptors, muscarinic and histamine receptors (Högberg et al., 1987). The very high selectivity for DA D₂ versus D₁ receptors has a very interesting implication in view of recent findings which suggest that stimulation of, or intact, DA D₁ receptor-mediated neurotransmission in the prefrontal cortex may be of great importance for atypical antipsychotic efficacy (e.g., Lidow et al., 1997; Okubo et al., 1997). Thus, many compounds in this group of agents could owe their atypical character as antipsychotics to a high selectivity for DA D₂ receptors, leaving DA D₁ receptors intact. In this regard, it is interesting to note that clozapine displays agonist properties in an in vivo model for DA D₁ receptor efficacy (Salmi et al., 1994; Salmi and Ahlenius, 1996).

Conclusion

The present results strongly suggest that remoxipride behaves as a prodrug for effects on brain DA synthesis, induction of catalepsy, antagonism of d-amphetamine-induced hyperlocomotion, quinpirole-induced hypothermia and suppression of treadmill or spontaneous locomotor activity in the rat. This supposition is based on the fact (1) that remoxipride is most effective by the intraperitoneal route of administration, favoring a rapid first-pass hepatic metabolism, (2) has a delayed onset of action and (3) is devoid of effects on intracerebroventricular administration. Formation of metabolites in the aromatic ring moiety of remoxipride, the preferred metabolic route in rodents, such as NCQ-344, FLA-797, FLA-908, NCQ-436, NCQ-489, are possible candidates for in vivo effects of remoxipride in laboratory studies. In view of the documented atypical profile of remoxipride in the laboratory, as well as its clinical efficacy in the treatment of schizophrenia, further studies into the pharmacodynamics of its metabolites as possible new leads for nonclozapine atypical antipsychotics are warranted.