Psychopharmacology

, Volume 190, Issue 3, pp 373–382

Interaction of the novel antipsychotic aripiprazole with 5-HT1A and 5-HT2A receptors: functional receptor-binding and in vivo electrophysiological studies

Authors

    • Neuroscience Drug DiscoveryBristol-Myers Squibb Pharmaceutical Research Institute
  • Shaun Jordan
    • Otsuka Maryland Research Institute, Inc.
  • Kelly A. Allers
    • Department of PharmacologyOxford University
  • Robert L. Bertekap
    • Neuroscience Drug DiscoveryBristol-Myers Squibb Pharmaceutical Research Institute
  • Ruoyan Chen
    • Otsuka Maryland Research Institute, Inc.
  • Tanaz Mistry Kannan
    • Neuroscience Drug DiscoveryBristol-Myers Squibb Pharmaceutical Research Institute
  • Thaddeus F. Molski
    • Neuroscience Drug DiscoveryBristol-Myers Squibb Pharmaceutical Research Institute
  • Frank D. Yocca
    • Neuroscience Drug DiscoveryBristol-Myers Squibb Pharmaceutical Research Institute
  • Trevor Sharp
    • Department of PharmacologyOxford University
  • Tetsuro Kikuchi
    • Otsuka Pharmaceutical Co. Ltd.
  • Kevin D. Burris
    • Palatin Technologies
Original Investigation

DOI: 10.1007/s00213-006-0621-y

Cite this article as:
Stark, A.D., Jordan, S., Allers, K.A. et al. Psychopharmacology (2007) 190: 373. doi:10.1007/s00213-006-0621-y

Abstract

Background

Aripiprazole (7-{4-[4-(2,3-dichlorophenyl)-1-piperazinyl]butoxy}-3,4-dihydro-2(1H)-quinolinone) is a novel antipsychotic with a mechanism of action that differs from current typical and atypical antipsychotics. Aripiprazole interacts with a range of receptors, including serotonin [5-hydroxytryptamine (5-HT)] and dopamine receptors.

Materials and methods

This study examined aripiprazole’s interactions with 5-HT systems in vitro and in vivo to further clarify its pharmacologic properties.

Results

Aripiprazole produced increases in [35S]GTPγS binding to rat hippocampal membranes. Its potency (pEC50 = 7.2) was similar to that of ziprasidone (7.1) and greater than that of 5-HT (6.7) and buspirone (6.4), a 5-HT1A-receptor partial agonist, whereas its intrinsic activity was similar to that of ziprasidone and buspirone. The stimulatory effect of aripiprazole was blocked by WAY-100635, a 5-HT1A-receptor antagonist. In in vivo electrophysiology studies, aripiprazole produced a dose-related reduction in the firing rate of 5-HT-containing dorsal raphe neurons in rats, which was both prevented and reversed by WAY-100635 administration. Aripiprazole showed a high affinity for human 5-HT1A receptors (Ki = 4.2 nM) using parietal cortex membrane preparations. In membranes from cells expressing human recombinant receptors, aripiprazole bound with high affinity to 5-HT2A receptors (Ki = 3.4 nM), moderate affinity to 5-HT2C (Ki = 15 nM) and 5-HT7 (Ki = 39 nM) receptors, and low affinity to 5-HT6 receptors (Ki = 214 nM) and 5-HT transporter (Ki = 98 nM). In addition, aripiprazole potently blocked 5-HT2A-receptor-mediated increases in intracellular Ca2+ levels in a rat pituitary cell line (IC50 = 11 nM).

Discussion

These results support a partial agonist activity for aripiprazole at 5-HT1A receptors in vitro and in vivo, and suggest important interactions with other 5-HT-receptor subtypes. This receptor activity profile may contribute to the antipsychotic activity of aripiprazole in humans.

Keywords

AripiprazoleDopamineSerotoninSchizophreniaPartial agonistAntagonist

Introduction

Aripiprazole (7-{4-[4-(2,3-dichlorophenyl)-1-piperazinyl]butoxy}-3,4-dihydro-2(1H)-quinolinone) is a novel antipsychotic that demonstrates improvement of positive and negative symptoms of schizophrenia (Kane et al. 2002; Potkin et al. 2003), has a low propensity for extrapyramidal symptoms (EPS), causes minimal weight gain and sedation, and produces no elevation in serum prolactin levels (Kane et al. 2002; Marder et al. 2003; Potkin et al. 2003). Several studies have also indicated that aripiprazole may possess an adjunctive antidepressant activity when used together with selective serotonin (5-HT) reuptake inhibitors for the treatment of major depressive disorder (Papakostas et al. 2005; Simon and Nemeroff 2005).

The mechanism of action of aripiprazole may differ from that of current typical and atypical antipsychotics. Aripiprazole and other atypical antipsychotics interact with a range of G-protein-coupled receptors, including both 5-HT and dopamine receptor subtypes (Davies et al. 2004). Preclinical studies have provided evidence that aripiprazole exhibits potent partial agonist activity at dopamine D2 receptors (Burris et al. 2002; Inoue et al. 1996; Kikuchi et al. 1995), whereas antipsychotics are believed to exert their effects, in part, through antagonist activity at D2 receptors (Miyamoto et al. 2000).

Subtypes of receptors for 5-HT, most notably 5-HT1A and 5-HT2A receptors, were also implicated in antipsychotic drug action (Millan 2000; Roth and Meltzer 1995). In particular, partial agonist activity at 5-HT1A receptors is thought to contribute to improvements in anxiety; in depressive, negative and cognitive symptoms; and to decreased EPS liability. The results of a recent study (Bardin et al. 2005) showed that aripiprazole-induced catalepsy was partially reversed by pretreatment with WAY-100635, a 5-HT1A-receptor antagonist, supporting the role for 5-HT1A in the favorable EPS profile of aripiprazole. Antagonist activity at 5-HT2A receptors was also linked to positive effects on negative symptoms and cognition, and low liability for EPS.

To date, preclinical studies examining the action of aripiprazole at 5-HT1A receptors showed partial agonist activity at cloned human receptors (Jordan et al. 2002; Shapiro et al. 2003), but have not characterized its functional activity at 5-HT1A receptors in vivo. In addition, pharmacological analysis of the actions of aripiprazole at 5-HT2A and 5-HT2C receptors are incomplete, although a number of studies showed partial agonist actions of aripiprazole at 5-HT2C receptors (Shapiro et al. 2003; Zhang et al. 2006). The purpose of this study was to define the scope of interaction of aripiprazole with 5-HT1A and 5-HT2A receptors, using functional in vitro binding assays and in vivo electrophysiological studies, and to provide further data on the binding affinities of aripiprazole at other 5-HT-receptor subtypes.

Materials and methods

All studies were carried out in accordance with the Guide for the Care and Use of Laboratory Animals, as adopted by the US National Institutes of Health.

Materials

8-hydroxy-dipropylaminotetralin (8-OH-DPAT), clozapine, haloperidol, risperidone, (+)8-OH-DPAT, buspirone, 5-HT, WAY-100635, Tris–HCl, ethylenediaminetetra acetic acid (EDTA), bovine serum albumin (BSA), oxaloacetate/pyruvate/insulin, and polyethylenimine were purchased from Sigma-Aldrich, Chemical (St. Louis, MO, USA). Olanzapine and ziprasidone were synthesized by Otsuka Pharmaceutical Company (Tokushima, Japan). [3H]-8-OH-DPAT, [3H]-lysergic acid diethylamide (LSD), [125I]-iodo-LSD, [3H]-5-CT, [35S]-GTPγS, and [3H]-citalopram were purchased from NEN (Boston, MA, USA). [3H]-myo-inositol was purchased from Amersham Pharmacia Biotech (Piscataway, NJ, USA).

Tissue culture

P11 cells were cultured in a humidified atmosphere of 10% CO2 at 37°C in high glucose Dulbecco’s modified Eagle’s medium supplemented with 2 mM l-glutamine, 150 μg/ml oxaloacetate, 50 μg/ml pyruvate, 0.2 U/ml insulin, and 10% charcoal-treated fetal bovine serum, as described by (Ivins and Molinoff 1990).

Radioligand binding assays

For central 5-HT1A receptors, membrane homogenates prepared from human parietal cortex (Analytical Biological Services, Wilmington, DE, USA) were centrifuged at 20,000×g for 20 min. Binding was performed in buffer [50 mM Tris–HCl (pH 7.7), 5 mM MgSO4, 2 mM ethylene glycol bis(2-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 10 μM pargyline, 0.01% ascorbic acid, and 1% dimethyl sulfoxide] and membrane homogenates (50 μg) were incubated with [3H]-8-OH-DPAT (2 nM for competition assays and 0.1–10 nM for saturation assays) for 10 min at 37°C. Nonspecific binding was defined with 1 μM 5-HT. Maximum binding (Bmax) and KD values were determined by unweighted linear regression analysis of transformed saturation binding data (Scatchard 1949). Protein concentrations were determined by the method of protein–dye binding with BSA as a standard (Bradford 1976). We assumed that all curves were of normal steepness; thus, values for Ki were determined using IC50 values, determined by competition for the binding of [3H]-8-OH-DPAT (Cheng and Prusoff 1973).

For 5-HT2A receptors, P11 cells were homogenized and centrifuged at 32,000×g for 30 min. Pellets were resuspended in buffer [50 mM Tris (pH 7.5) at 26°C and 1 mM EDTA], homogenized, and centrifuged again at 32,000×g for 30 min. Binding of [125I]-iodo-LSD was performed in buffer [50 mM Tris–HCl (pH 7.5), 1 mM MgCl2, and 10 μM pargyline]. Homogenates (15–20 μg protein) were incubated with [125I]-iodo-LSD (0.5 nM) for 1 h at 37°C. Nonspecific binding was defined with 100 μM 5-HT.

For 5-HT6 receptors, transfected HeLa-E6-1 cells were transferred from plates, homogenized, and centrifuged at 30,000×g for 20 min. Binding of [3H]-LSD was performed in buffer [50 mM Tris (pH 7.4), 2 mM MgCl2, 200 μM ascorbic acid, and 0.04% BSA] and homogenates (5 μg protein) were incubated with [3H]-LSD for 60 min at 37°C. Nonspecific binding was defined with 10 μM clozapine.

For 5-HT7 receptors, transfected Chinese Hamster Ovary (CHO) cells were homogenized and centrifuged at 30,000×g for 20 min. Binding of [3H]-5-CT was performed in buffer [50 mM N-(2-hydroxyethyl)-piperazine-N1-2-ethanesulforic acid, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) (pH 7.4), 2.5 mM MgCl2, and 2 mM EGTA] and homogenates (15 μg protein) were incubated with [3H]-5-CT for 60 min at 25°C. Nonspecific binding was defined with 10 μM 5-HT.

For 5-HT transporter binding assays, transfected CHO cells were homogenized and centrifuged at 18,000×g for 10 min. Binding of [3H]-citalopram was performed in buffer [50 mM Tris (pH 7.4), 120 mM NaCl, and 5 mM KCl] and homogenates (20 μg protein) were incubated with [3H]-citalopram for 60 min at 25°C. Nonspecific binding was defined with 10 μM fluoxetine.

Competition binding assays with human recombinant 5-HT2A and 5-HT2C receptors were performed by Cerep (Celle L’Evescault, France) using [3H]-ketanserin and [3H]-mesulergine, respectively, and membranes from transfected CHO cells. Values for Ki were determined using IC50 values that had been determined by the competition binding assays (Cheng and Prusoff 1973).

Measurement of levels of intracellular Ca2+ using FLIPR

Measurement of changes in the levels of intracellular Ca2+ in P11 cells induced by 5-HT were performed as described previously (Wohlpart and Molinoff 1998). Changes in levels of intracellular Ca2+ were measured by monitoring the relative change in fluorescence at 500–560 nm after excitation at 488 nm using a fluorescent imaging plate reader (FLIPR) device (Molecular Devices, La Jolla, CA, USA). After obtaining baseline recordings for 10 s, 5-HT was added and data were collected at 1.6-s intervals for 96 s, followed by readings every 2 s for an additional 120 s. Antagonists were added 60 min before addition of 5-HT.

Binding of [35S]-GTPγS in rat hippocampus

Male Wistar rats (body weight, 275–330 g) were decapitated and their hippocampi were dissected and homogenized in 10 volumes of ice-cold buffer (50 mM Tris–HCl, 1 mM dithiothreitol, and 1 mM EGTA, pH 7.4). The homogenate was centrifuged (1,000×g for 5 min at 4°C) and the resultant supernatant (S1) was removed. The remaining pellet was resuspended in buffer, homogenized, and centrifuged as before. This supernatant was mixed with S1, centrifuged (11,000×g for 20 min at 4°C), and the final membrane pellet was resuspended in buffer as reported by Bradford (1976) and stored at −80°C.

Drugs were incubated for 30 min at 30°C with hippocampal membranes (∼20 μg protein) and 445 μl buffer (50 mM Tris–HCl, 100 mM NaCl, 1 mM MgCl2, 1 mM EGTA, 1 mM dithiothreitol, and 100 μM GDP, pH 7.4). Reactions were started by the addition of 50 μl [35S]-GTPγS (final concentration = 0.2 nM), followed by a 30-min incubation period at 30°C. Reactions were terminated by rapid filtration. Filters were placed in counting vials containing liquid scintillation cocktail and radioactivity bound to the filter paper was measured 15 h later by liquid scintillation spectroscopy.

Estimates of potency, relative efficacy, and inhibitory potency were determined by nonlinear regression analysis using GraphPad Prism version 3.00 for Windows (GraphPad Software, San Diego, CA, USA).

Electrophysiology studies

Experiments were carried out on male Sprague–Dawley rats (265–300 g, Harlan Olac, Bicester, UK) using a 12:12 light ratio, with food and water available ad libitum. All experiments were carried out in accordance with the UK Animals (Scientific Procedures) Act (1986) and associated UK Home Office guidelines.

Animals were anesthetized with chloral hydrate (400 mg/kg, i.p.) and mounted in a stereotaxic frame. A lateral tail vein was cannulated for administration of additional anesthetic (if necessary) and drug. Body temperature of the animals was maintained at 36°C throughout the experiment by means of a homeothermic heating pad connected to a rectal probe.

Glass electrodes were lowered below the brain surface. Neurons were usually recorded during electrode descents down in the midline, with the second or third descents being 0.1 mm anterior or posterior to the first descent.

5-HT neurons were discriminated on the baseline of their characteristic electrophysiological properties. One neuron was recorded per animal.

Spontaneous single-unit activity was followed for a minimum period of 5 min before vehicle (water, captisol, and tartaric acid) or drug administration. Firing rates were determined during the 30-s period immediately before vehicle/drug injection and then during the last 30-s period of the 2 min immediately after vehicle/drug administration. No significant difference was observed between vehicle and baseline firing rates.

At the end of each experiment, the final position of the electrode tip was marked by iontophoretic dye ejection for verification of electrode placement. In experiments that were designed to assess the dose–response relationship, after baseline measurements, aripiprazole was administered in increasing doses (0.01, 0.1, and 1 mg/kg, i.v.). In other experiments, WAY-100635 (0.1 mg/kg, i.v.) was administered to block the effect of a bolus injection of aripiprazole (5 mg/kg, i.v.).

Results

Binding of aripiprazole to human 5-HT receptors

The affinity of aripiprazole for 5-HT1A receptors was examined in tissue from human brain. In membranes from parietal cortex, [3H]-8-OH-DPAT labeled an apparent single population of receptors with a KD value of 1.2 nM and a density of 95 fmol/mg. The affinity of agonists (8-OH-DPAT and 5-HT), partial agonists (clozapine), and antagonist (WAY-100635) for binding was consistent with labeling of 5-HT1A receptors (Fig. 1, inset). Aripiprazole bound with high affinity (Ki = 4.2 nM) to 5-HT1A receptors in human parietal cortex.
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-006-0621-y/MediaObjects/213_2006_621_Fig1_HTML.gif
Fig. 1

The density of 5-HT1A receptors was determined by saturation binding of [3H]-8-OH-DPAT to membranes that had been prepared from human parietal cortex. KD and Bmax values were determined by fitting data to a one-site binding isotherm. Inset: Ki values were determined by the method of Cheng and Prusoff (1973) using IC50 values determined by competition for the binding of the [3H]-8-OH-DPAT. Data are the mean±SEM, n = 3 experiments

The affinity of aripiprazole for other subtypes of human receptors for 5-HT was determined. Aripiprazole bound with high affinity to 5-HT2A receptors, with moderate affinity to 5-HT2C and 5-HT7 receptors, and with low affinity to 5-HT6 receptors (Table 1). In addition, aripiprazole exhibited moderate affinity for the human 5-HT transporter (Table 1).
Table 1

Affinity of aripiprazole for cloned human 5-HT receptors and the cloned human 5-HT transporter

Receptor

Ki (nM)

Reference (Ki, nM)

Radioligand

5-HT1Aa

1.7

(+)8-OH-DPAT (2.2)

[3H]-8-OH-DPAT

5-HT2A

3.4

Ketanserin (0.8)

[3H]-ketanserin

5-HT2C

15

Mesulergine (0.7)

[3H]-mesulergine

5-HT6

214 ± 20

Clozapine (10.6 ± 1.2)

[3H]-LSD

5-HT7

39 ± 15

Clozapine (44 ± 6)

[3H]-5-CT

5-HT transporter

98 ± 7

Fluoxetine (1.3)

[3H]-citalopram

Ki values were determined by the method of Cheng and Prusoff (1973) using IC50 values that had been determined by radioligand competition binding assays.

aJordan et al. (2002)

Functional activity of aripiprazole at 5-HT receptors

The effects of aripiprazole and reference drugs on binding of [35S]GTPγS to rat hippocampal membranes were examined. Aripiprazole produced increases in binding of [35S]GTPγS, with a potency similar to (+)8-OH-DPAT and ziprasidone and greater than that of 5-HT and buspirone (Table 2). Aripiprazole, buspirone, and ziprasidone behaved as partial agonists with similar intrinsic activity, whereas (+)8-OH-DPAT and 5-HT were full agonists. Risperidone, olanzapine, clozapine, and WAY-100635 did not stimulate an increase in binding of [35S]GTPγS over basal levels (Table 2). However, WAY-100635 potently blocked, in a concentration-dependent manner, the stimulatory effects of aripiprazole, (+)8-OH-DPAT, ziprasidone, and buspirone (Fig. 2). WAY-100635 potently inhibited the activity of 5-HT in this assay, although WAY-100635 did not fully block the stimulatory effect of 5-HT.
Table 2

Functional parameter estimates for aripiprazole and reference drugs in a [35S]GTPγS binding assay using rat hippocampal membranes

Agonist

Potency

Intrinsic activity

R2

pEC50 ± SEM

Emax ± SEM (%)

5-HT

6.7 ± 0.1

138 ± 3

0.996

(+)8-OH-DPAT

7.1 ± 0.1

92 ± 2

0.993

Buspirone

6.4± 0.2

28 ± 2

0.932

Aripiprazole

7.2 ± 0.2

37 ± 2

0.935

Ziprasidone

7.1 ± 0.2

32 ± 2

0.963

Clozapine

No effect

No effect

No effect

Risperidone

No effect

No effect

No effect

Olanzapine

No effect

No effect

No effect

WAY-100635

No effect

No effect

No effect

Agonist potency (pEC50) and relative efficacy (Emax, maximal drug effect on basal [35S]GTPγS binding expressed as a percentage of that produced by 10 μM (+)8-OH-DPAT) were estimated by nonlinear regression analysis. Nonlinear regression was also used to estimate the inhibitory potency (pIC50) of WAY-100635, tested at 0.01, 0.1, 1, 10, 100, 1,000, 10,000, and 50,000 nM against 1 μM concentrations each of 5-HT, (+)8-OH-DPAT, aripiprazole, ziprasidone, and buspirone. R2 represents the goodness of fit between observed concentration effect data points and nonlinear functions that were derived for each drug or drug combination studied.

https://static-content.springer.com/image/art%3A10.1007%2Fs00213-006-0621-y/MediaObjects/213_2006_621_Fig2_HTML.gif
Fig. 2

WAY-100635 inhibition of 1 μM concentrations of a 5-HT- and buspirone-, and b (+)8-OH-DPAT-, aripiprazole- and ziprasidone-induced [35S]GTPγS binding to rat hippocampal membranes. All data points are means±SEM of triplicate determinations of a single representative experiment, and are expressed as a percentage of the stimulatory effect of 10 μM (+)8-OH-DPAT on basal [35S]GTPγS binding

Rat P11 cells express 5-HT2A receptors that are linked to the phosphoinositide signaling pathway and subsequent elevation of intracellular levels of Ca2+ (Wohlpart and Molinoff 1998). Exposure of cells to increasing concentrations of 5-HT resulted in a threefold increase in the level of intracellular Ca2+ (Fig. 3). The effect of 5-HT (1 μM) was potently blocked by the antipsychotics risperidone, clozapine, aripiprazole, and haloperidol, with IC50 values being consistent with binding to rat 5-HT2A receptors (Table 3 and Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-006-0621-y/MediaObjects/213_2006_621_Fig3_HTML.gif
Fig. 3

Antagonism of 5-HT2A-receptor-mediated increases in levels of intracellular Ca2+. P11 cells that had been loaded with the calcium-sensitive dye Fluo-4 were exposed to 1 μM 5-HT in the absence and presence of increasing concentrations of antipsychotics. Changes in fluorescent intensity were measured and data were recorded using FLIPR. The maximum response to 5-HT at each concentration tested was expressed as fold basal. Data are the mean±SEM, n = 3 experiments

Table 3

Affinity and potency of aripiprazole and selected antipsychotics at 5-HT2A receptors in rat P11 cells

 

Binding

Function

Ki, nM

IC50, nM

Risperidone

0.13 ± 0.02

0.14

Clozapine

4.2 ± 0.4

1.1

Aripiprazole

12 ± 3

11

Haloperidol

31 ± 1

54

Ki values were determined by the method of Cheng and Prusoff (1973) using IC50 values determined by competition for the binding of [125I]-iodo-LSD to membranes prepared from P11 cells. IC50 values were obtained by fitting data to a one-site competitive binding curve using GraphPad Prism version 3.0 (GraphPad Software). IC50 values were determined from the data in Fig. 3 and are the average of three experiments.

Electrophysiology studies

In one group of animals, aripiprazole (0.01–1.0 mg/kg, i.v.) caused a dose-related inhibition of 5-HT cell firing in the dorsal raphe nucleus (Fig. 4). At the highest dose tested in these animals (1 mg/kg, i.v.), the maximum mean inhibition of firing was reduced to 13% of predrug levels. Statistical analysis of the raw data using one-way analysis of variance revealed a significant effect of treatment (p < 0.002). Post hoc analysis (Dunnett’s test) revealed a statistically significant effect of 1 mg/kg (i.v.) vs predrug values.
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-006-0621-y/MediaObjects/213_2006_621_Fig4_HTML.gif
Fig. 4

Dose--response effect of aripiprazole on 5-HT cell firing in the dorsal raphe nucleus. Each point is a mean±SEM value obtained from the number of neurons shown in brackets. *p < 0.05 vs baseline values (Dunnett’s test)

A bolus injection of aripiprazole (5 mg/kg, i.v.) caused a marked inhibition of 5-HT cell firing, and this effect was completely reversed by injection of WAY-100635 (0.1 mg/kg, i.v.; n = 2). In addition, the effect of aripiprazole was totally abolished in animals that were treated 10 min previously with WAY-100635 (Fig. 5).
https://static-content.springer.com/image/art%3A10.1007%2Fs00213-006-0621-y/MediaObjects/213_2006_621_Fig5_HTML.gif
Fig. 5

Summary of the data showing the blockade and reversal of aripiprazole-induced inhibition of 5-HT cell firing by the 5-HT1A antagonist, WAY-100635 (0.1 mg/kg, i.v.). Each column is a mean±SEM value obtained from five neurons. *p < 0.05 vs baseline (Dunnett’s test)

Discussion

Alterations in serotonergic neurotransmission are thought to contribute to the pathophysiology of schizophrenia (Lieberman et al. 1998). Interaction of 5-HT at its’ multiple receptor sites was hypothesized to regulate neuronal network activity (Andrade 1998). The 5-HT1A receptor was recently implicated as a molecular target for the development of improved antipsychotic drugs, as partial agonist activity at this receptor was shown to improve anxiety, depression, cognitive and negative symptoms, and to decrease EPS liability (Millan 2000). Several lines of evidence supporting a role for the 5-HT2A receptor in the mechanism of action of antipsychotic drugs include the restorative effect of 5-HT2A-receptor antagonists on N-methyl d-aspartate (NMDA) antagonist disruption of prepulse inhibition, a model of sensorimotor gating deficit in schizophrenia (Varty et al. 1999). In addition, most second-generation antipsychotics were shown to be potent antagonists of 5-HT2 receptors (Stockmeier et al. 1993). Furthermore, indications suggest that second-generation antipsychotics exhibiting a robust 5-HT2A-receptor antagonist affinity vs D2-receptor affinity, possess the ability to reduce D2-receptor antagonist adverse effects, such as EPS (Roth and Meltzer 1995).

Aripiprazole is a potent partial agonist at human recombinant 5-HT1A receptors (Jordan et al. 2002; Shapiro et al. 2003) and may possess antidepressant activity (Papakostas et al. 2005; Simon and Nemeroff 2005); therefore, a functional characterization of the compound at serotonergic receptor subtypes in vivo and in vitro was deemed necessary. Results from the present study are in line with previous studies confirming both the binding affinity and receptor density of the selective 5-HT1A-receptor agonist [3H]-8-OH-DPAT at the human parietal cortex (O’Neill et al. 1991). The affinity values of agonists and antagonists for the site labeled by [3H]-8-OH-DPAT were consistent with binding to 5-HT1A receptors; potent binding was observed with aripiprazole.

The present study provides further evidence for the partial agonist activity of aripiprazole at 5-HT1A receptors. Stimulation of [35S]-GTPγS binding in rat hippocampal membranes by aripiprazole, as well as by ziprasidone, buspirone, and (+)8-OH-DPAT, was completely blocked by the 5HT1A-receptor antagonist WAY-100635. Clozapine did not show partial agonist activity at 5-HT1A receptors, although this activity was demonstrated in previous [35S]-GTPγS binding studies (Jordan et al. 2002; Newman-Tancredi et al. 1998). The lack of effect observed with clozapine in the current study is thought to reflect the reduced sensitivity of the rat hippocampal membrane assay compared with that involving cloned human receptors.

Studies performed in vivo also revealed the partial agonist activity of aripiprazole at 5-HT1A receptors. 5-HT1A partial agonists, such as buspirone and gepirone, inhibit dorsal raphe activity via an interaction with somatodendritic autoreceptors (Blier and de Montigny 1987; Van der Maelen et al. 1986). These compounds display full intrinsic activity at these receptors because of the existence of a large receptor reserve for agonists at somatodendritic 5-HT1A receptors in the dorsal raphe nuclei (Meller et al. 1990). In rats, aripiprazole produced a dose-dependent reduction in the firing rate of 5-HT-containing dorsal raphe neurons. In contrast to the partial agonist activity of aripiprazole at hippocampal 5-HT1A receptors, aripiprazole appeared to be a full agonist at 5-HT1A receptors in the dorsal raphe. This may reflect the lack of 5-HT1A receptor reserve in the hippocampus compared to the raphe (Meller et al. 1990; Yocca et al. 1987). Aripiprazole-induced reductions in serotonergic neuronal activity were both prevented and reversed by the administration of WAY-100635. These data indicate that aripiprazole-induced inhibition of 5-HT neuronal activity is mediated by an agonist interaction with 5-HT1A somatodendritic autoreceptors on dorsal raphe serotonergic neurons. However, it is also possible that some of the inhibitory effects that are produced by aripiprazole may be mediated through neurons in the medial prefrontal cortex. Studies indicate that agonist activity at postsynaptic 5-HT1A receptors in the medial prefrontal cortex can inhibit the neuronal activity of serotonergic dorsal raphe neurons via a long feedback loop (Ceci et al. 1994; Hajos et al. 1999). A partial agonist action of aripiprazole at 5-HT2C receptors was also reported (Shapiro et al. 2003; Zhang et al. 2006).

Aripiprazole binds with high affinity to human 5-HT2A receptors (McQuade et al. 2002; Shapiro et al. 2003). Consistent with antagonist activity at 5-HT2A receptors, aripiprazole, as well as the antipsychotics clozapine, risperidone, and haloperidol, blocked 5-HT2A-receptor-mediated increases in levels of intracellular Ca2+ in a pituitary cell line in the present study. Antagonist actions of aripiprazole at 5-HT2A receptors in C6 glioma and GF62 cells were also demonstrated (Shapiro et al. 2003). Furthermore, aripiprazole blocks head twitches in mice that are elicited by 5-MeO-N,N-DMT, an agonist at 5-HT2 receptors (Hirose et al. 2004).

Serotonergic neurons modulate the activity of other neurotransmitter systems, including the dopaminergic system. Serotonergic neurons in the dorsal raphe nuclei provide major input to the frontal cortex and the striatum. Agonist activity at 5-HT1A autoreceptors located on these neurons reduces release of 5-HT, resulting in dopaminergic activation of the frontal cortex (Millan 2000). Agonistic functionality of aripiprazole at 5-HT1A autoreceptors may increase cortical dopamine pathway activity, which may improve negative symptoms and cognitive functioning. 5-HT1A-receptor agonists (autoreceptor) increase adrenergic activity in the frontal cortex, which may also play an important role in improving cognition. Furthermore, partial 5-HT1A-receptor agonists may act as functional antagonists at postsynaptic 5-HT1A receptors. Antagonist activity at postsynaptic 5-HT1A receptors was associated with anxiolytic activity and improvements in cognitive functioning (Millan 2000). Of interest, a recent preclinical study showed the neuroprotective effects of 5-HT1A agonism by aripiprazole (Cosi et al. 2005). This neuroprotective action may contribute to the novel mode of action of aripiprazole.

Antagonist activity at 5-HT2A receptors was suggested to increase dopaminergic transmission in neurons that project into the striatum, thereby counteracting atypical (D2-receptor antagonists) antipsychotic decreases (Meltzer 1999). Furthermore, agonist activity at 5-HT1A autoreceptors on the dorsal raphe nuclei increases dopamine release in the striatum (Lieberman et al. 1998). Therefore, partial agonist activity at D2 and 5-HT1A receptors, in addition to antagonist activity at 5-HT2A receptors, may act to maintain dopaminergic activity in the nigrostriatal pathway, thereby contributing to the low risk of EPS that is seen with aripiprazole treatment (Marder et al. 2003).

It should be noted that aripiprazole is the first D2-receptor partial agonist that was shown to have clinical efficacy in both acute and long-term studies. While exhibiting marked potency at D2, 5-HT1A, and 5-HT2A receptors, aripiprazole also displays moderate-to-high affinity for several human cloned biogenic amine G-protein-coupled receptors, most notably H1 (Ki = 25 nM) and α1A-adrenergic (Ki = 26 nM) receptors (see Davies et al. 2004 for review). The novel mechanism of action of aripiprazole supported by the present findings is one that combines D2- and 5-HT1A-receptor partial agonist activity with 5-HT2A-receptor antagonist actions. This mechanism may be predictive of broad therapeutic use beyond schizophrenia to the treatment of affective disorders, such as acute bipolar mania (Keck et al. 2003).

Acknowledgements

This study was supported by Bristol-Myers Squibb Company (Princeton, NJ, USA) and Otsuka Pharmaceutical (Tokyo, Japan). Editorial support for the preparation of this manuscript was provided by Ogilvy Healthworld Medical Education; funding was provided by Bristol-Myers Squibb Company.

Copyright information

© Springer-Verlag 2006