Psychopharmacology

, Volume 212, Issue 2, pp 155–170

Validation and pharmacological characterisation of MK-801-induced locomotor hyperactivity in BALB/C mice as an assay for detection of novel antipsychotics

Authors

  • Andrea M. Bradford
    • Biology Department, Neurosciences Centre of Excellence for Drug DiscoveryGlaxoSmithKline plc
  • Kevin M. Savage
    • Research and Development Information TechnologyGlaxoSmithKline plc
  • Declan N. C. Jones
    • European Centre of Excellence for External Drug DiscoveryGlaxoSmithKline plc
    • CNS Business Unit, Addex Pharmaceuticals, Bâtiment A&CImmeuble Alliance
    • CNS Business Unit, Addex Pharmaceuticals, Bâtiment A&CImmeuble Alliance
original investigation

DOI: 10.1007/s00213-010-1938-0

Cite this article as:
Bradford, A.M., Savage, K.M., Jones, D.N.C. et al. Psychopharmacology (2010) 212: 155. doi:10.1007/s00213-010-1938-0
  • 444 Views

Abstract

Rationale

We evaluated locomotor hyperactivity induced in BALB/C mice by an N-methyl-d-aspartate receptor antagonist MK-801 as an assay for the detection of antipsychotic drugs.

Objectives

We assessed the effects of antipsychotic drugs to validate the assay (study 1), selective dopamine and serotonin ligands for pharmacological characterisation of the model (study 2) and a number of compounds with efficacy in models of schizophrenia to understand the predictive validity of the model (study 3).

Methods

Adult males (n = 9/group) were pretreated with a test compound, habituated to locomotor activity cages before receiving MK-801 (0.32 mg/kg) and activity recorded for a further 75 or 120 min. In study 1, we tested haloperidol, clozapine, olanzapine, risperidone, ziprasidone, aripiprazole, sertindole and quetiapine. In study 2, we tested SCH23390 (D1 antagonist), sulpiride (D2/D3 antagonist), raclopride (D2/D3 antagonist), SB-277011 (D3 antagonist), L-745,870 (D4 antagonist), WAY100635 (5-HT1A antagonist), 8-OH-DPAT (5-HT1A agonist), ketanserin (5-HT2A/5-HT2C antagonist) and SB-242084 (5-HT2C antagonist). In study 3, we tested xanomeline (M1/M4 receptor agonist), LY379268 (mGluR2/3 receptor agonist), diazepam (GABAA modulator) and thioperamide (H3 receptor antagonist).

Results

All antipsychotics suppressed MK-801-induced hyperactivity in a dose-dependent and specific manner. The effects of antipsychotics appear to be mediated via dopamine D1, D2 and 5-HT2 receptors. Xanomeline, LY379268 and diazepam were active in this assay while thioperamide was not.

Conclusions

MK-801-induced hyperactivity in BALB/C mice model of positive symptoms has shown predictive validity with novel compounds acing at M1/M4, mGluR2/3 and GABAA receptors and can be used as a screening assay for detection of novel pharmacotherapies targeting those receptors.

Keywords

GlutamateMK-801SchizophreniaLocomotor activityHyperactivityAntipsychotic

Introduction

Glutamatergic dysfunction has been proposed by various investigators as an underlying cause of the pathophysiology of schizophrenia (Kim et al. 1980; Javitt and Zukin 1991; Olney and Farber 1995; Coyle 1996). In support, non-competitive antagonists at the N-methyl-d-aspartate (NMDA) receptor, MK-801, phencyclidine (PCP) and ketamine cause psychotic-like symptoms in normal humans that are similar to acute episodes of schizophrenia (Snyder 1976; Javitt and Zukin 1991; Steinpreis and Salamone 1993; Steinpreis 1996) and exacerbate existing symptoms in schizophrenic patients (Javitt and Zukin 1991). These observations include positive and negative symptoms as well as cognitive abnormalities. Hence, in a putative pharmacological model of schizophrenia in rodents (Angelucci et al. 1999; Krystal et al. 1999), acute administration of the NMDA receptor antagonists results in behavioural syndrome which includes hyperlocomotion, stereotypic behaviours, disruption of sensorimotor gating and social and cognitive deficits (Verebey et al. 1981; Ford et al. 1989; Jentsch and Roth 1999; Mandillo et al. 2003; Yee et al. 2004; Manahan-Vaughan et al. 2008) that may be analogous to symptoms of schizophrenia in humans. In particular, lower doses of MK-801 (0.05–1.0 mg/kg) induce hyperlocomotion, reduce prepulse inhibition (PPI) and cause cognitive abnormalities, whereas higher doses (>1.0 mg/kg) result in stereotypy and ataxia (Iverson et al. 1988; Keok et al. 1988; Butelman 1989; Tricklebank et al. 1989; Whishaw and Auer 1989; Heale and Harley 1990; Wozniak et al. 1990; Hargreaves and Cain 1992; Mandillo et al. 2003; Yee et al. 2004; Manahan-Vaughan et al. 2008).

Detailed pharmacological evaluation of these behavioural changes in order to understand the predictive validity of this approach has been limited or inconclusive. Differences in experimental variables such as species, gender, strain, as well as choice of antipsychotics, doses and routes of their administration have made it difficult to compare the findings across published reports. For example, in male CD-1 mice, clozapine failed to reverse MK-801-induced PPI deficit at lower doses (0.1 and 0.3 mg/kg; Curzon and Decker 1998), while at higher dose (3.0 mg/kg) it reserved PPI deficit exhibited by dopamine transporter knockout mice (Powell et al. 2008). Also, in female TO mice, haloperidol inhibited MK-801-induced hyperactivity only at a dose (0.1 mg/kg) that also attenuated spontaneous activity (O’Neill and Shaw 1999), whereas in male Swiss mice a higher dose (0.3 mg/kg) reduced MK-801-induced hyperactivity without affecting spontaneous activity (Boulay et al. 2010).

There is a growing awareness of the differences among mouse strains in behavioural reactivity and in sensitivity to drugs, especially in the context of modelling aspects of schizophrenia (Ralph et al. 2001; Varty et al. 2001; Connolly et al. 2004; Spielewoy and Markou 2004; Kalinichev et al. 2008). For example, inbred BALB/C mice are more sensitive to locomotor-stimulant effects of MK-801 when compared to CD-1 or MF-1 mice (Kalinichev et al. 2008). BALB/C mice are also more sensitive to MK-801-mediated antagonism of electrically precipitated seizures and elicitation of intense jumping behaviour (‘popping’) when compared to NIH Swiss mice (Deutsch et al. 1997, 1998). Furthermore, in a novel environment or open-field, BALB/C mice exhibit signs of elevated ‘neophobia-’ and ‘anxiety-like’ reactivity and are slower to habituate than mice of several other strains (Tang et al. 2002; Kalinichev et al. 2008). Based on these findings, it has been suggested that BALB/C could be a useful mouse strain for studying ‘psychosis-proneness’ (Perera et al. 2008) and for use in a model based on the NMDA receptor inhibition.

Therefore, the goal of this study was to achieve a comprehensive evaluation of MK-801-induced hyperactivity model of psychosis in BALB/C mice, with a question of whether the assay has predictive validity and can be used as a screening assay for detection of novel compounds with antipsychotic efficacy. To achieve this goal we (1) validated the assay with the clinically most commonly prescribed antipsychotics, (2) provided pharmacological characterisation of this model by utilising previously tested hypothesis investigating specific receptor profiles and (3) tested several compounds directed against non-dopamine or serotonin molecular targets hypothesised to be relevant for the treatment of schizophrenia and active in other models for antipsychotic activity in order to assess predictive validity of the model. These novel targets involve pharmacological systems beyond the known profiles of commonly used antipsychotics and include muscarinic M1/M4, metabotropic glutamate type II (mGluR2/3), GABAA and histamine H3 receptors. Accumulating cellular, preclinical and clinical evidence over the last decade provides strong evidence that these receptors are viable targets for the treatment of schizophrenia and can lead to the development of novel pharmacotherapies for this disease.

Materials and methods

Subjects

Adult male BALB/C mice (17–24 g) were purchased from Harlan UK Ltd. (Bicester, UK). Upon arrival at our animal facility, they were housed in groups of five under standard conditions (temperature 20 ± 1°C and humidity 50–58%) with a 12-h light–dark cycle (lights on 06:00 GMT) with food (Harlan Teklad 2014; Harlan, UK Ltd., Bicester, UK) and water available ad libitum. Animals were allowed a minimum of 5 days to acclimatise prior to placement in the experimental room. Experiments were conducted in the light phase between 09:00 and 16:00. Studies were conducted in full compliance with the Home Office Guidance on the operation of the UK Animals (Scientific Procedures) Act 1986 and were approved by the GlaxoSmithKline Procedures Review Panel.

Drugs

MK-801 was purchased from Tocris (Bristol, UK). Sulpiride and raclopride were purchased from Sigma-Aldrich (Dorset, UK). Diazepam has been purchased from Courtin and Warner (Lewis, UK). Haloperidol, clozapine, olanzapine, risperidone, sertindole, aripiprazole, ziprasidone, quetiapine, WAY100635, 8-OH-DPAT, ketanserin, SB-242084, SB-277011, SCH23390, L-745,870, xanomeline, thioperamide and LY379268 were synthesised by Medicinal Chemistry, GlaxoSmithKline (Harlow, UK).

MK-801, sulpiride, ketanserin, WAY100635, 8-OH-DPAT, SB-242084, L-745,870, SCH23390, quetiapine, xanomeline, diazepam and thioperamide were dissolved in saline. Raclopride and LY379268 were dissolved in water. SB-277011, sertindole, clozapine, ziprasidone, haloperidol, risperidone, aripiprazole and olanzapine were dissolved or suspended in 1% methylcellulose. All drugs were administered at 10 ml/kg volume.

Locomotor activity

Apparatus

Locomotor activity (horizontal counts) was monitored using 24 Perspex boxes (\( {42} \times {21} \times 21\;{\hbox{cm}} \)) in conjunction with “AM Logger” AM1052 activity monitors (Linton Instruments, Diss, UK) as described previously (Kalinichev et al. 2008). Activity monitors were equipped with infrared photobeam interruption sensors and a computerised analysis system. The number of horizontal photobeam interruptions was recorded in 5-min bins.

Procedure

Adult male BALB/C mice (n = 9/group) were used in all studies. In study 1, animals were pretreated with either a vehicle or a compound of interest and habituated to locomotor activity (LMA) cages for 60 min. Following habituation, animals were challenged with saline or MK-801 (0.32 mg/kg, i.p.) and activity was recorded for a further 120 min. In our previous study (Kalinichev et al. 2008), 0.32 mg/kg stimulated robust hyperactivity in BALB/C mice with minimal expression of ataxia. Based on the pattern of habituation and hyperactivity obtained in study 1 (see “Results”), this procedure was modified in studies 2 and 3. Based on practical reasons, recording of hyperactivity was shortened to 75 min (except xanomeline), whereas habituation period was aligned with the compounds pretreatment time.

Study 1—validation of the assay with antipsychotic drugs

We tested the following antipsychotic drugs (all given intraperitoneally): haloperidol (0.1, 0.32 and 1.0 mg/kg), clozapine (0.1, 0.32 and 1.0 mg/kg), olanzapine (0.1, 0.32 and 1.0 mg/kg), risperidone (0.01, 0.032 and 0.1 mg/kg), ziprasidone (0.1, 0.32 and 1.0 mg/kg), aripiprazole (0.32, 1.0 and 3.2 mg/kg), sertindole (0.32, 1.0 and 3.2 mg/kg) and quetiapine (3.2, 10 and 15.6 mg/kg). The doses were determined based on published reports (O’Neill and Shaw 1999; Su et al. 2007) and our in-house data on the effects of these drugs in in vivo assays.

Study 2—pharmacological characterisation of the assay

We tested dopamine D1 receptor antagonist SCH23390 (0.01, 0.032 and 0.1 mg/kg, s.c.), dopamine D2/D3 receptor antagonists sulpiride (10, 32 and 60 mg/kg, s.c.) and raclopride (0.01, 0.032 and 0.1 mg/kg, s.c.), selective dopamine D3 receptor antagonist SB-277011 (10, 30 and 45 mg/kg, p.o.) and selective dopamine D4 antagonist L-745,870 (3.2, 5.6 and 10 mg/kg, i.p.).

The role of serotonin 5-HT1A receptor in the MK-801-induced hyperactivity was evaluated using its selective antagonist WAY100635 (0.032, 0.1 and 0.32 mg/kg, s.c.) and agonist 8-OH-DPAT (1.0, 3.2 and 10 mg/kg, s.c.). We also evaluated serotonin 5-HT2A/5-HT2C receptor antagonist ketanserin (0.32, 0.56 and 1.0 mg/kg, i.p.) and selective serotonin 5-HT2C receptor antagonist SB-242084 (1.0, 3.2 and 10 mg/kg, s.c.).

Study 3—testing compounds with potential efficacy in psychosis

The following compounds with potential antipsychotic efficacy were tested: a muscarinic M1/M4 receptor agonist xanomeline (0.1, 0.32 and 1.0 mg/kg, i.p.), a GABAA receptor modulator diazepam (0.32, 1.0 and 3.2 mg/kg, i.p.), an mGluR2/3 receptor agonist LY379268 (0.32, 1.0 and 3.2 mg/kg, s.c.) and a histamine H3 receptor antagonist thioperamide (3.2, 10 and 15.6 mg/kg, i.p.).

Data analysis

In all studies, horizontal activity counts were assessed. We analysed total counts over the habituation phase and total counts over the first 60 min of hyperactivity phase excluding the initial 15 min following injection. Data were analysed using one-way ANOVA followed by planned comparisons. The alpha level chosen was p < 0.05 and the data were log-transformed, where appropriate in order to stabilise the variance. The statistical package used was Statistica 6.1 (StatSoft Inc., Tulsa, OK, USA).

Results

In all experiments, mice exposed to novel LMA cages exhibited progressive decreases in activity indicative of habituation. Decreases in activity were rapid during the first 5–10 min followed by a more gradual reduction and subsequent levelling of activity during the last 30 min of habituation. Following 0.32 mg/kg MK-801 challenge, vehicle-pretreated animals (i.e., the reference group) exhibited robust hyperactivity, which was rapid in onset (10–20 min) and lasted 90 to 110 min, confirming our previous findings in BALB/C mice (Kalinichev et al. 2008). The peak activity counts were four- to sixfold higher than those observed at baseline and were reliably reproduced in every experiment.

Study 1

Haloperidol (0.1, 0.32 and 1.0 mg/kg, i.p.) had a significant effect on habituation [F(4, 40) = 13.70, p < 0.001], suppressing it markedly (65%, p < 0.001) at the highest dose (1.0 mg/kg; Fig. 1a, b; Table 1). Haloperidol dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 36.09, p < 0.001] causing 25% (p < 0.01), 56% (p < 0.001) and 86% (p < 0.001) reduction in total counts when administered at 0.1, 0.32 and 1.0 mg/kg, respectively (Fig. 1a, b; Table 1). Clozapine (0.1, 0.32 and 1.0 mg/kg, i.p.) had no effect on habituation, but attenuated MK-801-induced hyperactivity [F(4, 40) = 31.14, p < 0.001] causing a 21% (p < 0.05) reduction in total counts when dosed at the lowest dose (0.1 mg/kg). However, at higher doses (0.32 and 1.0 mg/kg), the effect did not reach statistical significance (Fig. 1c, d; Table 1). Olanzapine (0.1, 0.32 and 1.0 mg/kg, i.p.) had no effect on habituation, but dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 16.76, p < 0.001] causing 28% (p < 0.05), 53% (p < 0.001) and 48% (p < 0.001) reduction in total counts when administered at 0.1, 0.32 and 1.0 mg/kg, respectively (Fig. 1e, f; Table 1). Risperidone (0.01, 0.032 and 0.1 mg/kg, i.p.) had no effect on habituation, but dose-dependently attenuated MK-801-induced hyperactivity [F(4, 39) = 40.38, p < 0.001], with a 45% (p < 0.001) reduction in total counts seen at the highest dose (0.1 mg/kg; Fig. 1g, h; Table 1). Ziprasidone (0.1, 0.32 and 1.0 mg/kg, i.p.) had no effect on habituation, but dose-dependently attenuated MK-801-induced hyperactivity [F(4, 39) = 21.80, p < 0.001] with a 50% (p < 0.001) reduction in total counts seen at the highest dose (1.0 mg/kg; Fig. 2a, b; Table 1). Aripiprazole (0.32, 1.0 and 3.2 mg/kg, i.p.) had no effect on habituation, but dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 5.75, p < 0.001] with a 28% (p < 0.05) reduction in total counts seen at the highest dose (3.2 mg/kg; Fig. 2c, d; Table 1). Sertindole (0.32, 1.0 and 3.2 mg/kg, i.p.) impaired habituation activity [F(4, 40) = 14.71, p < 0.001] causing 29% and 52% (p < 0.001) reduction in total counts at 1.0 and 3.2 mg/kg respectively (Fig. 2e, f; Table 1). Sertindole also dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 50.31, p < 0.001] reducing it by 48%, 72% and 94% (all p < 0.001) at 0.32, 1.0 and 3.2 mg/kg, respectively (Fig. 2e, f; Table 1). Quetiapine (3.2, 10 and 15.6 mg/kg, i.p.) impaired habituation activity [F(4, 40) = 37.34, p < 0.001] with 49% and 82% (p < 0.001) reduction seen at 10 and 15.6 mg/kg, respectively (Fig. 2g, h; Table 1). Quetiapine also dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 26.50, p < 0.001) causing 22% (p < 0.05), 35% (p < 0.01) and 86% (p < 0.001) reduction in total hyperactivity counts when administered at 3.2, 10 and 15.6 mg/kg, respectively (Fig. 2g, h; Table 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-010-1938-0/MediaObjects/213_2010_1938_Fig1_HTML.gif
Fig. 1

Locomotor activity (horizontal counts) of male BALB/C mice expressed as time-course profiles (a, c, e, g) and total counts (b, d, f, h). Animals were pretreated (i.p.) with either haloperidol (0.1, 0.32 and 1.0 mg/kg; a, b), clozapine (0.1, 0.32 and 1.0 mg/kg; c, d), olanzapine (0.1, 0.32 and 1.0 mg/kg; e, f) or risperidone (0.01, 0.032 and 0.1 mg/kg; g, h) and habituated to locomotor activity (LMA) cages for 60 min before being challenged with MK-801 (0.32 mg/kg, i.p.) and returned to LMA cages for another 120 min. Each point represents the observed mean (±SEM). *p < 0.05, **p < 0.01, ***p < 0.001 compared to the Veh/MK-801 group (n = 9/group)

Table 1

A summary of all antipsychotic drugs, selective dopamine and serotonin receptor ligands and compounds with efficacy in models of schizophrenia

Compounds

Receptor

Route

Doses (mg/kg)

MED (mg/kg) suppressing habituation

MED (mg/kg) suppressing MK-801 hyperactivity

Haloperidol

Antipsychotic

i.p.

0.1, 0.32, 1.0

1.0

0.1

Clozapine

Antipsychotic

i.p.

0.1, 0.32, 1.0

0.1a

Olanzapine

Antipsychotic

i.p.

0.1, 0.32, 1.0

0.1

Risperidone

Antipsychotic

i.p.

0.01, 0.032, 0.1

0.1

Ziprasidone

Antipsychotic

i.p.

0.1, 0.32, 1.0

1.0

Aripiprazole

Antipsychotic

i.p.

0.32, 1.0, 3.2

3.2

Sertindole

Antipsychotic

i.p.

0.32, 1.0, 3.2

1.0

0.32

Quetiapine

Antipsychotic

i.p.

3.2, 10, 15.6

10

3.2

SCH23390

D1

s.c.

0.01, 0.032, 0.1

0.01

Sulpiride

D2/D3

s.c.

10, 32, 60

60

Raclopride

D2/D3

s.c.

0.01, 0.032, 0.1

0.1

0.01

SB-277011

D3

p.o.

10, 30, 45

L-745,870

D4

i.p.

3.2, 5.6, 10

10

WAY100635

5-HT1A

s.c.

0.032, 0.1, 0.32

8-OH-DPAT

5-HT1A

s.c.

1.0, 3.2, 10

1.0

10

Ketanserin

5-HT2A/5-HT2C

i.p.

0.32, 0.56, 1.0

1.0

0.32

SB-242084

5-HT2C

s.c.

1.0, 3.2, 10

Xanomeline

M1/M4

i.p.

0.1, 0.32, 1.0

1.0

LY379268

mGluR2/3

s.c.

0.32, 1.0, 3.2

3.2

3.2

Diazepam

GABAA

i.p.

0.32, 1.0, 3.2

3.2

Thioperamide

H3

i.p.

3.2, 10, 15.6

The table includes names of drugs, targeted receptor(s), routes of administration, doses administered (mg/kg) as well as minimal effective doses (MEDs, mg/kg) suppressing habituation and MK-801-induced hyperactivity

aBut inactive at higher doses

https://static-content.springer.com/image/art%3A10.1007%2Fs00213-010-1938-0/MediaObjects/213_2010_1938_Fig2_HTML.gif
Fig. 2

Locomotor activity (horizontal counts) of male BALB/C mice expressed as time-course profiles (a, c, e, g) and total counts (b, d, f, h). Animals were pretreated (i.p.) with either ziprasidone (0.1, 0.32 and 1.0 mg/kg; a, b), aripiprazole (0.32, 1.0 and 3.2 mg/kg; c, d), sertindole (0.32, 1.0 and 3.2 mg/kg; e, f) or quetiapine (3.2, 10, 15.6 mg/kg; g, h) and habituated to locomotor activity (LMA) cages for 60 min before being challenged with MK-801 (0.32 mg/kg, i.p.) and returned to LMA cages for another 120 min. Each point represents the observed mean (±SEM). *p < 0.05, **p < 0.01, ***p < 0.001 compared to the Veh/MK-801 group (n = 9/group)

Study 2

Selective dopamine D1 receptor antagonist SCH23390 (0.01, 0.032 and 0.1 mg/kg, s.c.) had no effect on habituation, but dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 25.26, p < 0.001] causing 22% (p < 0.05), 50% (p < 0.001) and 61% (p < 0.001) reduction in total hyperactivity counts at 0.01, 0.032 and 0.1 mg/kg, respectively (Fig. 3a, b; Table 1). Dopamine D2/D3 receptor antagonist sulpiride (10, 32 and 60 mg/kg, s.c.) had no effect on habituation, but dose-dependently attenuated MK-801-induced hyperactivity [F(4, 37) = 20.90, p < 0.001] with a 25% (p < 0.01) reduction in total counts seen at the highest dose (60 mg/kg; Fig. 3c, d; Table 1). Dopamine D2/D3 receptor antagonist raclopride (0.01, 0.032 and 0.1 mg/kg, s.c.) caused minor (18%, p < 0.01) reduction of habituation [F(4, 64) = 4.83, p < 0.01] at the highest dose (0.1 mg/kg) while dose-dependently attenuating MK-801-induced hyperactivity [F(4, 64) = 30.86, p < 0.001] with an approximately 20% (p < 0.01) reduction in total counts at 0.01 mg/kg and 45% (p < 0.001) reduction in those counts at 0.1 mg/kg (Fig. 3e, f; Table 1). Selective dopamine D3 receptor antagonist SB-277011 (10, 30 and 45 mg/kg, p.o.) had no effect on locomotor activity neither during habituation nor following MK-801 challenge (Fig. 3g, i; Table 1). Selective dopamine D4 receptor antagonist L-745,870 (3.2, 5.6 and 10 mg/kg, i.p.) transiently reduced (by 30%, p < 0.01) habituation at the highest dose (10 mg/kg) as confirmed by the significant effect of treatment [F(4, 40) = 4.177, p < 0.01], but had no effect upon MK-801-induced hyperactivity (Fig. 3i, j; Table 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-010-1938-0/MediaObjects/213_2010_1938_Fig3a_HTML.gif
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-010-1938-0/MediaObjects/213_2010_1938_Fig3i_HTML.gif
Fig. 3

Locomotor activity (horizontal counts) of male BALB/C mice expressed as time-course profiles (a, c, e, g, i) and total counts (b, d, f, h, j). Animals were pretreated with either SCH23390 (0.01, 0.032 and 0.1 mg/kg, s.c.; a, b), sulpiride (10, 32 and 60 mg/kg, s.c.; c, d), raclopride (0.01, 0.032 and 0.1 mg/kg, s.c.; e, f), SB-277011 (10, 30 and 45 mg/kg, p.o.; g, h) or L-745,870 (3.2, 5.6 and 10 mg/kg, i.p.; i, j) and habituated to locomotor activity (LMA) cages for 30–60 min before being challenged with MK-801 (0.32 mg/kg, i.p.) and returned to LMA cages for another 75 min. Each point represents the observed mean (±SEM). *p < 0.05, **p < 0.01, ***p < 0.001 compared to the Veh/MK-801 group (n = 9/group)

Selective serotonin 5-HT1A receptor antagonist WAY100635 (0.032, 0.1 and 0.32 mg/kg, s.c.) had no effect on locomotor activity neither during habituation nor following MK-801 challenge (Fig. 4a, b; Table 1). Selective serotonin 5HT1A receptor agonist 8-OH-DPAT (1.0, 3.2 and 10 mg/kg, s.c.) dose-dependently attenuated habituation [F(4, 40) = 67.90, p < 0.001] causing 38%, 64% and 73% (all p < 0.001) reduction in total counts at 1.0, 3.2 and 10 m/kg, respectively (Fig. 4c, d; Table 1). 8-OH-DPAT also dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 14.62, p < 0.001], with a 36% (p < 0.001) reduction seen at 10 mg/kg (Fig. 4c, d; Table 1). Serotonin 5-HT2A/5-HT2C receptor antagonist ketanserin (0.32, 0.56 and 1.0 mg/kg, i.p.) caused 30% (p < 0.001) reduction of habituation [F(4, 40) = 8.23, p < 0.001] at the highest dose (1.0 mg/kg). Ketanserin also attenuated MK-801-induced hyperactivity [F(4, 40) = 7.69, p < 0.001] with 30% (p < 0.05) and 44% (p < 0.01) reduction in total counts seen at 0.32 and 1.0 mg/kg, respectively (Fig. 4e, f; Table 1). Selective serotonin 5-HT2C receptor antagonist SB-242084 (1.0, 3.2 and 10 mg/kg s.c.) had no effect on habituation, but potentiated MK-801-stimulated activity [F(4, 40) = 36.30, p < 0.001] by 22% (p < 0.05) in response to the highest dose (10 mg/kg; Fig. 4g, h; Table 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-010-1938-0/MediaObjects/213_2010_1938_Fig4_HTML.gif
Fig. 4

Locomotor activity (horizontal counts) of BALB/C male mice expressed as time-course profiles (a, c, e, g) and total counts (b, d, f, h). Animals were pretreated with either WAY100635 (0.032, 0.1 and 0.32 mg/kg, s.c.; a, b), 8-OH-DPAT (1.0, 3.2 and 10 mg/kg, s.c.; c, d), ketanserin (0.32, 0.56 and 1.0 mg/kg, i.p.; e, f), or SB-242084 (1.0, 3.2 and 10 mg/kg, s.c.; g, h) and habituated to locomotor activity (LMA) cages for 60 min before being challenged with MK-801 (0.32 mg/kg, i.p.) and returned to LMA cages for another 75 min. Each point represents the observed mean (±SEM). *p < 0.05, **p < 0.01, ***p < 0.001 compared to the Veh/MK-801 group (n = 9/group)

Study 3

A muscarinic M1/M4 receptor agonist xanomeline (0.1, 0.32 and 1.0 mg/kg, i.p.) had no effect on habituation, but attenuated MK-801-induced hyperactivity [F(4, 35) = 15.61, p < 0.001] with a 31% (p < 0.01) reduction in total hyperactivity counts when administered at the highest dose (1.0 mg/kg; Fig. 5a, b; Table 1). An mGluR2/3 receptor agonist LY379268 (0.32, 1.0 and 3.2 mg/kg, s.c.) influenced habituation [F(4, 40) = 5.55, p < 0.01] reducing it mildly (∼20%, p < 0.01) at 3.2 mg/kg (Fig. 5c, d; Table 1). LY379268 also dose-dependently attenuated MK-801-induced hyperactivity [F(4, 40) = 18.96, p < 0.01] with a 33% (p < 0.01) reduction seen at 3.2 mg/kg (Fig. 5c, d; Table 1). GABAA receptor modulator diazepam (0.32, 1.0 and 3.2 mg/kg) had no effect on habituation but attenuated MK-801-induced hyperactivity [F(4, 40) = 16.49, p < 0.001] with a 30% (p < 0.05) reduction seen at 3.0 mg/kg (Fig. 5e, f; Table 1). Histamine H3 receptor antagonist thioperamide (3.2, 10 and 15.6 mg/kg, i.p.) had no effect on habituation (Fig. 5g, h; Table 1), but had a biphasic effect on MK-801-induced hyperactivity, reducing it mildly during the initial 30 min post-administration, while stimulating it afterwards.
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-010-1938-0/MediaObjects/213_2010_1938_Fig5_HTML.gif
Fig. 5

Locomotor activity (horizontal counts) of male BALB/C mice expressed as time-course profiles (a, c, e, g) and total counts (b, d, f, h). Animals were pretreated with either xanomeline (0.1, 0.3 and 1.0 mg/kg, i.p.; a, b), LY379268 (0.32, 1.0 and 3.2 mg/kg, s.c.; c, d), diazepam (0.32, 1.0 and 3.2 mg/kg, p.o.; e, f) or thioperamide (3.2, 10 and 15.6 mg/kg, i.p.; g, h) and habituated to locomotor activity (LMA) cages for 30–60 min before being challenged with MK-801 (0.32 mg/kg, i.p.) and returned to LMA cages for another 75 or 120 min. Each point represents the observed mean (±SEM). **p < 0.01 compared to the Veh/MK-801 group (n = 9/group)

Discussion

The overall objective of these studies was to establish a model of positive symptoms of schizophrenia (psychosis) which has predictive validity and therefore can be used as a screening assay in the development of novel pharmacotherapies. With this objective, we provide a comprehensive evaluation of locomotor hyperactivity induced in male BALB/C mice by acute MK-801 challenge. We turned to this model because BALB/C mice are particularly sensitive to MK-801 (Deutsch et al. 1997, 1998; Kalinichev et al. 2008) and have been suggested to be an appropriate strain of mice to study ‘psychosis-proneness’ (Perera et al. 2008). Also, locomotor hyperactivity induced by MK-801 in animals of this strain is robust and technically easy to perform (Kalinichev et al. 2008). The assay is believed to be relevant to positive symptoms of schizophrenia since NMDA antagonists induce psychosis-like symptoms in healthy individuals and exacerbate psychosis in patients with schizophrenia. The NMDA receptor antagonists have been used extensively in rodents to model aspects of schizophrenia (Large 2007; Jones et al. 2008a, b).

The first phase of comprehensive evaluation of the model was to confirm that the hyperactivity induced by MK-801 is sensitive to commonly prescribed antipsychotics. All antipsychotics tested in this study had specific and dose-dependent effects on MK-801-induced hyperactivity, therefore validating the study. Specifically, at the range of doses tested, these drugs either had no effect on spontaneous activity in habituation phase (clozapine, olanzapine, risperidone, ziprasidone and aripiprazole) or suppressed this activity at doses that were higher than those inhibiting MK-801-stimulated activity (haloperidol, sertindole and quetiapine). Thus, we can argue that the suppression of locomotion by antipsychotics is specific to hyperactivity induced by MK-801 and is not a consequence of motor impairment due to ataxia/sedation. Although a comprehensive evaluation of antipsychotics in a single model has been lacking, the most commonly prescribed compounds (haloperidol, clozapine, olanzapine, risperidone, ziprasidone and aripiprazole) have been well characterised in several rodent models of positive symptoms of schizophrenia, including those that are based on locomotor hyperactivity induced by the NMDA receptor antagonists in rats and in mice (Castellani and Adams 1981; Kitaichi et al. 1994; Maurel-Remy 1995; O’Neill and Shaw 1999; Ninan and Kulkarni 1999; Nilsson et al. 1997; Natesan et al. 2008; Leite et al. 2008; Nordquist et al. 2008). It should be noted, that unlike the present findings, effects of antipsychotics on locomotor hyperactivity from earlier reports are often confounded by non-specific reduction of spontaneous locomotor activity (O’Neill and Shaw 1999; Andiné et al. 1999; Zocci et al. 2005; Nordquist et al. 2008).

In comparison with the antipsychotics discussed above, quetiapine and sertindole have been evaluated in fewer studies in models of positive symptoms of schizophrenia. In the present study, significant effects of sertindole and quetiapine in MK-801-induced hyperactivity coincided with a significant suppression of spontaneous locomotor activity at higher doses. The antagonism on histamine H1 receptors can be responsible for the suppression of spontaneous locomotion by quetiapine (Saller and Salama 1993), but not by sertindole. The mechanisms mediating the effect of sertindole and quetiapine on spontaneous activity of BALB/C mice require further studies.

In study 2, we further investigated the receptor subtypes in order to pinpoint a specific receptor profile and pharmacological mechanism of antipsychotics-induced suppression of MK-801-mediated hyperactivity. Our findings indicate that the suppression by MK-801-induced hyperactivity may be due to actions at dopamine and serotonin receptors. Specifically, dopamine D1 receptors appear to be necessary for the MK-801-induced hyperactivity in BALB/C mice. SCH23390, a dopamine antagonist with high affinity at D1 receptors dose-dependently attenuated MK-801-induced hyperactivity (minimal effective dose (MED) 0.01 mg/kg) without altering spontaneous locomotion. SCH23390 has been found to have similar effects on MK-801-induced hyperactivity in CD-1 mice (Corbett et al. 1995) and in rats (Ouagazzal et al. 1993). In contrast, in female TO mice, high doses of SCH23390 (0.05–0.25 mg/kg) had only a mild effect on MK-801-induced hyperactivity while markedly suppressing spontaneous locomotor activity (O’Neill and Shaw 1999). Strain and/or sex differences in D1 receptor characteristics may be responsible for discrepancies in the outcome of these studies.

Dopamine D2 receptors also appear to pay a key role in hyperactivity induced by 0.32 mg/kg MK-801, whereas dopamine D3 and D4 receptors do not. Selective dopamine D2/D3 receptor antagonists sulpiride and raclopride were active in the assay, while selective antagonists at dopamine D3 (SB-277011) and D4 (L-745,870) receptors were without any effect. SB-277011 (2–42.3 mg/kg, p.o.) also failed to affect stimulant-induced hyperlocomotion in rats (Reavill et al. 2000). In full accord with our findings, dopamine D3 receptor knockout mice and their wild-type controls were equally sensitive to locomotor-stimulant effects of 0.3 mg/kg MK-801 (Leriche et al. 2003). Thus, hyperactivity induced by NMDA antagonist or psychostimulant does not appear to require D3 activity in rodents.

Among the subtypes of serotonin receptors tested in this study, only 5-HT2A appears to be necessary in MK-801-induced hyperactivity in BALB/C mice. A 5-HT2A/5-HT2C receptor antagonist, ketanserin, resulted in attenuation of MK-801-induced hyperactivity at 0.32 mg/kg while inhibiting spontaneous locomotion at 1.0 mg/kg. According to an earlier study (Ninan and Kulkarni 1999), ketanserin dose-dependently attenuated MK-801-induced hyperactivity (MED 2.5 mg/kg) as well as spontaneous locomotor activity (MED 5.0 mg/kg) in CD-1 mice. BALB/C mice may be more sensitive than CD-1 mice to the effects of 5-HT2A or 5-HT2C antagonists on spontaneous locomotion and on hyperactivity induced by an NMDA antagonist.

WAY100635 did not influence MK-801-induced hyperactivity suggesting a lack of requirement for 5HT1A receptor activation in MK-801-induced hyperactivity. 8-OH-DPAT suppressed spontaneous activity at all doses tested (1.0–10 mg/kg) suggesting underlying sedative effects of this compound, which has been reported previously (Misslin et al. 1990; Söderpalm and Engel 1988; Moser et al. 1990). The 5-HT2 receptor antagonist SB-242084 with the affinity at 2A, 2B and 2C receptor subtypes had no effect on spontaneous activity but caused a slight increase in locomotor activity following MK-801 challenge at the highest dose (10 mg/kg). In comparison, O’Neill et al. (1999) showed that SB-242084 (0.25–1.0 mg/kg) had no effect on MK-801-induced hyperactivity in female TO mice suggesting that 5HT2C receptors are not involved in this strain.

Following the validation and pharmacological characterisation of the assay, we tested several compounds developed for novel targets for the treatment of schizophrenia to assess predictive validity of the model. There is a growing body of preclinical and clinical evidence suggesting that M1 and M4 types of muscarinic cholinergic receptors are viable targets for the treatment of schizophrenia (Jones et al. 2008a, b; Brady et al. 2008; Chan et al. 2008). In a recent clinical study, a selective M1/M4 receptor agonist xanomeline reduced positive and negative symptoms and improved cognition in a cohort of schizophrenic patients (Shekhar et al. 2008). In preclinical studies, xanomeline has a well-documented antipsychotic-like profile. For example, in the rat, xanomeline reversed apomorphine-induced disruption of PPI, attenuated amphetamine-induced hyperactivity (Stanhope et al. 2001) and blocked dopamine agonist-induced turning in unilateral 6-OHDA-lesioned animals (Shannon et al. 2000). In mice, xanomeline blocked apomorphine-induced climbing (Shannon et al. 2000) and attenuated amphetamine-induced hyperactivity (Woolley et al. 2009). The latter effect appears to be mediated largely by M4 and to a lesser extent by M1 receptors since it was fully abolished in M4 receptor knockout mice while being only attenuated in M1 receptor knockout mice (Woolley et al. 2009). Here, xanomeline resulted in a dose-dependent attenuation of MK-801 (MED 1.0 mg/kg) with no effect of spontaneous locomotor behaviour supporting and expanding the evidence on its anti-psychotic-like profile. Thus, the assay has predictive validity in revealing the efficacy of compounds that act through activation of muscarinic M1 or M4 receptors and can aid in the discovery of novel compounds with this mechanism, such as allosteric agonists or potentiators (Jones et al. 2008; Brady et al. 2008; Chan et al. 2008).

Another target that can lead to the development of a fundamentally novel approach to the treatment of schizophrenia includes metabotropic group II receptors, mGluR2 and mGluR3 (Conn et al. 2008). In a phase II clinical trial, selective mGluR2/3 agonists markedly improved positive and negative symptoms in patients with schizophrenia (Patil et al. 2007). In preclinical studies, selective mGluR2/3 receptor agonists LY379268 and LY354740 have been well-characterised in rodent models predicting antipsychotic activity. For example, LY379268 dose-dependently attenuated amphetamine- and PCP-induced hyperactivity in rats (Cartmell et al. 1999, 2000), the latter paralleled by attenuation of PCP-mediated increases in prefrontal glutamate (Moghaddam and Adams 1998). Also, LY379268 administered at doses used in the present study (0.3, 1.0 and 3.0 mg/kg) reduced amphetamine- and PCP-induced hyperactivity in C57BL6/J mice without having any effect on spontaneous locomotion in these animals (Galici et al. 2005; Woolley et al. 2008). The effects of LY379268 on hyperactivity appear to be mediated via mGluR2 rather than mGluR3 receptors since they are completely abolished in mGluR2 knockout mice, but preserved in mGluR3 knockout mice (Woolley et al. 2008). Similarly, inhibition of spontaneous and PCP-stimulated activity by an mGluR2/3 agonist LY314582 is completely absent in mGluR2 knockout mice, while preserved in wild-type controls (Spooren et al. 2000). Here, LY379268 resulted in a dose-dependent attenuation in MK-801-induced hyperactivity reaching 33% reduction in total counts at 3.2 mg/kg. The same dose also resulted in mild (20%) reduction in spontaneous locomotion. Similarly, LY379268 at higher doses (>6.0 mg/kg) suppressed spontaneous locomotor activity in C57BL6/J mice (Campo, unpublished data). Thus, there is a possibility that inhibition of MK-801-mediated hyperactivity by LY379268 in the present assay is not specific and is due to the effect on spontaneous locomotion. Some evidence from another report (Woolley et al. 2008) argues against this interpretation. Specifically, LY379268 can reduce PCP-induced hyperactivity at doses (1.0 mg/kg) that have no effect on spontaneous locomotion (Woolley et al. 2008). Also, reductions in spontaneous locomotion in wild-type mice dosed at 3.0 mg/kg LY379268 are also seen in mGluR2 knockout mice despite LY379268 having no effect on amphetamine-induced hyperactivity (Woolley et al. 2008). Thus, the effect of LY379268 may still be specific, despite the effect on spontaneous locomotion.

In a recent clinical trial, an mGluR2/3 agonist LY2140023 failed to improve clinical signs in patients with schizophrenia (Kinon 2009), the results questioning predictive validity of preclinical models in the detection of antipsychotic efficacy of mGluR2/3 agonists. However, the results of the study were considered to be inconclusive due to the robust response seen in placebo-treated patients and the similarity between placebo- and olanzapine-treated groups (Kinon 2009). Thus, additional clinical data are needed in order to fully understand predictive validity of the present model in detection of antipsychotic efficacy of mGluR2/3 agonists.

There is a large body of evidence suggesting that GABAergic dysfunction plays a key role in the aetiology of schizophrenia (Guidotti et al. 2005). Diazepam, a modulator of GABAA receptor, although not considered as an antipsychotic drug, has been frequently used in the treatment of acute psychosis as a monotherapy or as adjunctive medication in patients with schizophrenia. For example, diazepam was shown to be efficacious in preventing the escalation of psychotic symptoms in patients with early signs of schizophrenia (Carpenter et al. 1999). Here, we show the efficacy of diazepam in a rodent model of positive symptoms. Diazepam caused a dose-dependent attenuation of MK-801-induced hyperactivity in BALB/C mice (MED 3.0 mg/kg) without any effect on spontaneous locomotion. In an earlier report, diazepam attenuated hyperactivity induced in female Sprague–Dawley rats by MK-801, but only at the dose (5.0 mg/kg) that caused slowing of the righting reflex indicative of sedation (Andiné et al. 1999). Hence, the present model can be used as a screening assay in any attempt to develop novel drugs that target GABAA receptors for the treatment of schizophrenia (Guidotti et al. 2005).

Recently, there has been accumulating evidence suggesting the role of histaminergic neurotransmission in schizophrenia (Arrang 2007). For example, concentrations of the histamine metabolite tele-methylhistamine in cerebral spinal fluid are elevated in schizophrenic patients and correlate with the severity of the psychotic symptoms (Prell et al. 1995). In preclinical studies, amphetamine-induced hyperactivity is abolished in transgenic mice lacking histamine H3 receptor (Toyota et al. 2002) or following inhibition of H3 receptor in rats (Clapham and Kilpatrick 1994; Morisset et al. 2002). The histamine H3 receptor is an inhibitory presynaptic heteroreceptor which modulates the release of histamine in the central nervous system (CNS). Histamine levels in the CNS are increased by administration of thioperamide, a histamine H3 receptor antagonist (Arrang et al. 1987). It has been shown that 10 mg/kg thioperamide attenuated locomotion activation induced by amphetamine, apomorphine and cocaine (Clapham and Kilpatrick 1994) in CHR mice. Also, another H3 antagonist ciproxifan reduced hyperactivity induced in mice by 0.3 mg/kg dose of MK-801 (Faucard et al. 2006). On the other hand, thioperamide can potentiate cocaine-induced hyperactivity in mice (Brabant et al. 2009), while in Long–Evans rats it can either potentiate or attenuate MK-801-induced hyperactivity depending on the dose of the latter (Bardgett et al. 2009). In the present study, the initial attenuating effect of thioperamide on MK-801-induced hyperactivity was followed by a potentiating effect. This confirms that the relationship between H3 receptors and hyperactivity induced by either psychostimulants or NMDA receptor blockers is complex and requires further studies. Therefore, the present assay may not be suitable for the detection of novel compounds that work through the inhibition of histamine H3 receptor (Southam et al. 2009).

In conclusion, this model showed good sensitivity to all clinically active antipsychotics and could be useful in detecting novel antipsychotic compounds that are developed based on receptor profiles of standard antipsychotics. In addition, the model showed good predictive validity in the detection of compounds that stimulate activity of M1/M4, mGluR2/3 or GABAA receptors and can be used as a screening assay for testing of novel compounds.

Conflict of interest

None.

Copyright information

© Springer-Verlag 2010