Differential agonist and inverse agonist profile of antipsychotics at D2L receptors coupled to GIRK potassium channels

Neuropharmacology 52 (2007) 1106e1113 www.elsevier.com/locate/neuropharm

Differential agonist and inverse agonist profile of antipsychotics at D2L receptors coupled to GIRK potassium channels Peter Heusler a,*, Adrian Newman-Tancredi b, Annabelle Castro-Fernandez a, Didier Cussac a a

Cellular and Molecular Biology Department, Pierre Fabre Research Center, 17 Avenue Jean Moulin, 81106 Castres Cedex, France b Neurobiology 2 Division, Pierre Fabre Research Center, 17 Avenue Jean Moulin, 81106 Castres Cedex, France Received 8 September 2006; received in revised form 24 October 2006; accepted 27 November 2006

Abstract The D2 dopaminergic receptor represents a major target of antipsychotic drugs. Using the coupling of the human D2long (hD2L) receptor to G protein-coupled inward rectifier potassium (GIRK) channels in Xenopus laevis oocytes, we examined the activity of antipsychotic agents of different classes e typical, atypical, and a ‘‘new generation’’ of compounds, exhibiting a preferential D2 and 5-HT1A receptor profile. When the hD2L receptor was coexpressed with GIRK channels, a series of reference compounds exhibited full agonist (dopamine, and quinpirole), partial agonist (apomorphine, ()3-PPP, and (þ)-UH232) or inverse agonist (raclopride, and L741626) properties. Sarizotan exhibited only very weak partial agonist action. At higher levels of receptor cRNA injected per oocyte, both partial agonist activity and inverse agonist properties were generally more pronounced. The inverse agonist action of L741626 was reversed by interaction with sarizotan, thus confirming the constitutive activity of wild-type hD2L receptors in the oocyte expression system. When antipsychotic agents were tested for their actions at the hD2L receptor, typical (haloperidol) as well as atypical (nemonapride, ziprasidone, and clozapine) compounds acted as inverse agonists. In contrast, among D2/5-HT1A antipsychotics, only SLV313 and F15063 behaved as inverse agonists, whilst the other members of this group (bifeprunox, SSR181507 and the recently marketed antipsychotic, aripiprazole) exhibited partial agonist properties. Thus, the X. laevis oocyte expression system highlights markedly different activity of antipsychotics at the hD2L receptor. These differential properties may translate to distinct therapeutic potential of these compounds. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Antipsychotic; Dopamine D2L receptor; Partial agonist; Inverse agonist; Xenopus oocytes; G protein-coupled inward rectifier potassium channels

1. Introduction Classical or ‘‘typical’’ antipsychotic drugs have been successfully employed for treatment of positive symptoms in schizophrenia. These agents, however, exhibit a considerable disposition to induce adverse side effects, the most prominent of them being extrapyramidal motor side effects (EPS). Moreover, treatment with these compounds is less efficacious against negative symptoms (characterized by lowered motivation and interest, anhedonia and social withdrawal) and cognitive impairments associated with schizophrenia, such as working and reference memory dysfunction, decreased * Corresponding author. Tel.: þ33 56371 6704; fax: þ33 56371 4363. E-mail address: [email protected] (P. Heusler). 0028-3908/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropharm.2006.11.008

vigilance, and impaired executive function. Considerable progress has been made with the introduction of the second generation ‘‘atypical’’ antipsychotic drugs, exhibiting reduced side effects and improved efficacy against negative symptoms (Kapur and Remington, 2001; Meltzer et al., 2003; Farah, 2005). The common basis of efficacy of typical and atypical antipsychotics is their antagonism at dopaminergic receptors of the D2 type (Remington, 2003), while the particular properties of atypical drugs are attributed to the amount and/or kinetics of D2 receptor occupancy (Kapur and Mamo, 2003) as well as to a broader spectrum of action at several receptors, notably at serotonergic 5-HT2A receptors (Meltzer et al., 2003). In addition, the serotonin 5-HT1A receptor has become an important target for antipsychotic medication. Associating

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113

activity at D2 and 5-HT1A receptors represents a promising approach to combine antipsychotic activity with reduced EPS liability as well as with improved properties in the treatment of cognitive deficits and depressive states associated with schizophrenia (Prinssen et al., 1999; Millan, 2000; Meltzer et al., 2003). Thus, a new generation of antipsychotic drugs has been developed, preferentially targeting these two receptor subtypes (Newman-Tancredi et al., 2005; Bruins Slot et al., 2006; Bardin et al., 2006). These drugs, including bifeprunox (Van Vliet et al., 2000), SLV313 (McCreary et al., 2007), SSR181507 (Claustre et al., 2003) and F15063 (Newman-Tancredi et al., 2006; Depoortere et al., 2006) differ in their 5-HT1A/D2 affinity ratio as well as in their intrinsic efficacy at both receptor subtypes. All of them possess agonist activity at the 5-HT1A receptor, albeit with different efficacies and potencies (Newman-Tancredi et al., 2005; Bruins Slot et al., 2006). At D2 receptors, the efficacy of the compounds shows diverse properties. When tested in vitro in [35S]GTPgS binding and MAP kinase phosphorylation assays SLV313 and F15063 behave as antagonists, whereas bifeprunox and SSR181507 exhibit partial agonist properties (Cosi et al., 2006; Bruins Slot et al., 2006; Newman-Tancredi et al., 2006). Indeed, these latter compounds share some properties with aripiprazole, a recently marketed antipsychotic. Aripiprazole is a partial agonist at both 5-HT1A and D2 receptors, in addition to its multiple other receptor interactions (Lawler et al., 1999; Jordan et al., 2002; Shapiro et al., 2003; compare also Cosi et al., 2006; Bruins Slot et al., 2006). In order to further characterize and compare the actions of antipsychotics, especially those with D2/5-HT1A receptor properties, we now report their efficacies at hD2L receptors in an assay based on coupling to GIRK (G protein-coupled inward rectifier potassium) channels in Xenopus oocytes. This experimental system represents a well-established electrophysiological approach for the study of G protein-coupled receptors (GPCRs) that couple to Gi/o, but it has not yet been employed to characterize the actions of antipsychotic compounds at D2 receptors. Hyperpolarisation of neurons via the opening of GIRK channels represents a major functional response to D2 receptor activation in the CNS (Neve et al., 2004), and the D2 receptor has been shown to exist in a close functional association with GIRK channels in a cellular context (Levine et al., 2002). In addition, the activation of GIRK channels can be considered as prototypical for an effect mediated by the bg subunits of heterotrimeric G proteins and it is possible that differences exist in the receptor-dependent regulation of pathways mediated by a and bg subunits. Our results reveal the differential activities of antipsychotic compounds at the hD2L receptor. In particular, among the new generation of antipsychotics, aripiprazole, bifeprunox and SSR181507 partially activated GIRK currents via the hD2L receptor, while F15063 and SLV313 were devoid of agonist activity and behaved as inverse agonists, similar to typical and atypical antipsychotics. Parts of the results presented here have been published in abstract form (Newman-Tancredi et al., 2006).

1107

2. Materials and methods 2.1. RNA preparation The plasmid containing the coding sequence for the human D2L receptor was subcloned into the Xenopus high-expression vector pGEMHE (Liman et al., 1992) and designated pGEMHE/hD2L. Plasmids pSP/GIRK1 and pBScMXT/GIRK2 encoding the GIRK1 and GIRK2 potassium channel subunits were prepared as described (Heusler et al., 2005). Plasmids were linearised with NheI (pGEMHE/hD2L), EcoRI (pSP/GIRK1) or SalI (pBScMXT/ GIRK2) and in vitro transcription of RNA was performed using the T7 (pGEMHE/hD2L), SP6 (pSP/GIRK1) or T3 (pBScMXT/GIRK2) mMessage mMachine transcription kit (Ambion, Austin, TX). cRNA was purified using the RNeasy RNA cleanup kit (Qiagen, Courtaboeuf, France), quantified spectrometrically, diluted to the appropriate concentration in RNAse-free water and stored at 80  C prior to use.

2.2. X. laevis oocyte expression Isolation and separation of oocytes were performed as previously described (Heusler et al., 2005). Defolliculated oocytes were injected with a roughly 50 nl volume of cRNA solution containing the cRNAs coding for the GIRK1 and GIRK2 channel subunits at a concentration of 10e50 pg cRNA/ oocyte each with or without addition of hD2L receptor cRNA at 10e 1000 pg/oocyte (no measurable dopamine-induced currents were observed in oocytes injected with 1000 pg hD2L receptor RNA alone, not shown). After injection, oocytes were kept at 17  C in ND96 solution (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, 5 mM HEPES, pH 7.5 with NaOH) supplemented with 50 ng/ml gentamicin for at least 2 days before experiments. Animals were handled and cared for in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996) and the European Directive 86/609/EEC, and the protocols were carried out in compliance with French Regulations and with local Ethical Committee Guidelines for Animal Research.

2.3. Electrophysiological recordings Whole-cell oocyte currents were recorded using the two-electrode voltage clamp technique and a Geneclamp 500 amplifier (Axon Instruments, Union City, CA), as described previously (Heusler et al., 2005). Briefly, oocytes were placed in a recording chamber where they were continuously superfused with ND96 solution (2.5e3.5 ml/min). GIRK currents were recorded in highpotassium solution (hKþ, similar to ND96, but containing 96 mM KCl, 2 mM NaCl) at a holding potential of 70 mV. Drugs were dissolved in hKþ and applied by superfusion. If not specified otherwise, saturating concentrations of pharmacological compounds were utilised. At the end of each experiment, BaCl2 (1 mM, dissolved in hKþ) was applied to quantify the GIRK-independent currents in high-potassium solution. For drugs exhibiting receptorindependent GIRK current inhibitions, ligand efficacy values in the assay on receptor activation were corrected for the respective value (see Heusler et al., 2005).

2.4. Data analysis The pClamp 8 software (Axon Instruments) was used for data acquisition. Data are presented as mean  S.E.M. Isotherms were analysed by non-linear regression, using GraphPad Prism (GraphPad Software Inc., San Diego, CA) to yield EC50 values.

2.5. Drugs The following compounds were obtained commercially: dopamine hydrochloride, quinpirole, raclopride, L741626 (3-[4-(4-chlorophenyl)-4-hydroxypiperidin-L-yl]-methyl-1H-indole), ()3-PPP (S()-3-(3-hydroxyphenyl)-N-npropylpiperidine), haloperidol (Sigma RBI; St. Quentin Fallavier, France), and clozapine (Tocris; Illkirch, France). F15063 (2-(2,2-dimethyl-2,

1108

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113

3-dihydro-benzofuran-7-yloxy)-ethyl)-(3-cyclopenten-1-yl-benzyl)-amine, fumarate salt), ziprasidone, nemonapride, aripiprazole, bifeprunox (DU127090), sarizotan hydrochloride, SLV313 (piperazine, 1-(2,3-dihydro-1,4-benzodioxin5-yl)-4-[[5-(4-fluorophenyl)-3-pyridinyl]methyl), and SSR181507 ((3-exo)-8benzoyl-N-[[(2S )7-chloro-2,3-dihydro-1,4-benzodioxin-1-yl]methyl]-8-azabicyclo-[3.2.1]octane-3-methanamine monohydrochloride) were synthesized by the Chemistry Department, Pierre Fabre Research Center.

hK+

ND96

BaCl2

ND96

dopamine basal

agonist

3. Results 3.1. GIRK current activation via the hD2L receptor hD2L 10 pg cRNA / oocyte

As a control, all ligands were tested on oocytes expressing GIRK channels alone. At the concentrations employed in the pharmacological studies, it was found that only sarizotan (3 mM), clozapine (10 mM), SLV313 (10 mM) and F15063 (1 mM) had modest receptor-independent blocking effects on basal GIRK currents recorded in these oocytes (blockade by 6.5%, 11.8%, 4.2%, and 8.4%, respectively). As a consequence, quantitative assessments of receptor-mediated GIRK current induction had to be corrected by the respective values (see Section 2). 3.3. Actions of reference agonists Quinpirole (1 mM) behaved as an agonist with an efficacy equivalent to dopamine (not shown) and was therefore utilised as reference ‘‘full agonist’’. At low hD2L receptor expression level (10 pg cRNA per oocyte), apomorphine and ()3-PPP behaved as partial agonists (Fig. 2). (þ)-UH232 and sarizotan (an anti-dyskinetic drug with D2/5-HT1A activity) were virtually devoid of agonist activity. When tested on oocytes injected with higher amounts of hD2L receptor cRNA (100 pg each; Fig. 2), the relative efficacy of apomorphine was enhanced to near ‘‘full agonist’’ properties and (þ)-UH232 now showed marked partial agonist activity. In contrast, the efficacy value of ()3-PPP was only slightly increased under these conditions. In addition, weak partial agonist activity could be detected for sarizotan. At the highest amounts of

125 100 75 50 25 0 -10

-9

-8

-7

-6

-5

log [dopamine] 150

GIRK current activation (% of dopamine at 10 M)

3.2. Test for hD2L receptor-independent effects

GIRK current activation (% of dopamine at 10 M)

150

The hD2L receptor was expressed in Xenopus oocytes together with the GIRK1 and GIRK2 potassium channel subunits, which are considered to form the prototypical heteromultimeric GIRK channel in the CNS (Liao et al., 1996; Koyrakh et al., 2005). Exposure of the cells to dopamine (10 mM) induced an increase in the Ba2þ-dependent potassium current (Fig. 1), an effect that was absent in oocytes not injected with D2 receptor cRNA (not shown, see also Werner et al., 1996; Wiens et al., 1998). The pEC50 for dopamine in oocytes injected with 10 pg hD2L receptor cRNA each was determined as 7.94  0.11, a value similar to that reported previously (w8.6; Werner et al., 1996). When the amount of hD2L receptor cRNA was augmented to 100 pg, the pEC50 was increased to 8.78  0.28 (Fig. 1), indicating that receptor protein expression is increased with higher amounts of receptor cRNA injected.

hD2L 100 pg cRNA / oocyte

125 100 75 50 25 0 -10

-9

-8

-7

-6

-5

log [dopamine] Fig. 1. hD2L receptor-mediated effects on GIRK current activation in Xenopus oocytes. Upper panel: activation of GIRK currents by dopamine (10 mM). Current trace recorded from an oocyte injected with 100 pg hD2L receptor cRNA and GIRK1/2 channel subunit cRNAs (20 pg each). Oocytes were clamped at 70 mV, and dopamine was applied by superfusion for 90 s as indicated by the bar. Basal GIRK current was activated by application of high-potassium solution (hKþ, containing 96 mM Kþ), and the level of GIRK-independent potassium current was determined by superfusion with hKþ solution supplemented with 1 mM BaCl2. At the beginning and at the end of each experiment, oocytes were kept in ND96 solution. Middle and lower panel: concentrationeresponse curve of GIRK current activation by dopamine. Middle: oocytes injected with 10 pg hD2L receptor cRNA per oocyte. Bottom: oocytes injected with 100 pg hD2L receptor cRNA per oocyte. Experiments were performed as described above, and agonist-induced GIRK currents were quantified and normalized to the mean current evoked by the maximal dose of dopamine (10 mM) in oocytes of the respective batch.

cRNA injected (1 ng per oocyte; Fig. 2), the relative efficacies of sarizotan and (þ)-UH232 were further enhanced, but that of ()3-PPP was not affected. 3.4. Inverse agonism The D2 receptor compounds L741626 and raclopride both decreased basal GIRK currents in oocytes injected with hD2L receptor cRNA (Fig. 2). These drug-induced decreases in GIRK

10 pg hD2L cRNA / oocyte

1109

hK+

ND96

100

BaCl2

60

n=10

n=18

40

n=7

6

e cl op

62

rid

an ra

sa

(+ )U

H

riz ot

23

2

3P PP

ph

(-)

200 nA

L741626 1 µM

20 sec n=6

100 pg hD2L cRNA / oocyte

hK+

ND96

BaCl2

100

ND96

80 n=9

60

n=8

40 20

n=21

n=12

n=9

cl

6 L7

41

op

62

rid

an ra

sa

(+

(-)

)U

riz

H

ot

23

2

PP 3P

in ph or

sarizotan3 µM

ap

om

e

0 e

GIRK current activation (% of full agonist)

ap

om or

n=7

n=5

41

n=2

0

L7

20

100 nA 30 sec

L741626 1 µM 1000 pg hD2L cRNA / oocyte 100

n=5

80 60

n=6

40 n=11

20

n=7

n.d.

6 62

sarizotan + L741626 -5

-10

41

op cl ra

L7

rid

an ot riz sa

(+

)U

H

23

2

PP 3P (-)

in ph or

0

-15

L741626 alone

ap

om

e

0

GIRK current activation (% of full agonist)

n=6

e

GIRK current activation (% of full agonist)

ND96

80

in e

GIRK current activation (% of full agonist)

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113

Fig. 2. hD2L receptor-mediated effects on GIRK currents by reference compounds. hD2L receptor cRNA was injected at 10 pg per oocyte (upper panel), 100 pg per oocyte (middle), or 1000 pg per oocyte (lower panel), as indicated, while equal amounts of GIRK1/2 channel subunit cRNAs were coinjected for each condition. Experiments were performed as described in Section 2 and Fig. 1. Drug concentrations at all conditions were 1 mM for ()3-PPP and raclopride, 3 mM for (þ)-UH232, and 10 mM for apomorphine. Effects on GIRK current were quantified and normalized to the mean current evoked by a full agonist in oocytes of the respective batch. n.d.: not determined.

current were not observed in oocytes not injected with hD2L receptor cRNA and thus represented inverse agonist activity, i.e. an inhibition of a constitutive activity of the hD2L receptor coupled to GIRK channels in the Xenopus oocyte expression system. This was confirmed by examining the effect of L741626 (1 mM) after pre-occupation of the receptor with sarizotan (3 mM; Fig. 3). Due to the lack of a silent antagonist among the compounds tested, sarizotan was chosen as the competing compound. Sarizotan exhibits high potency at D2 receptors and can be considered as a functional hD2L receptor antagonist in experiments at oocytes injected with 100 pg hD2L receptor

Fig. 3. Inhibition of inverse agonist activity of L741626 by sarizotan preincubation. L741626 (1 mM) was administered for 1 min at oocytes not pretreated or in the presence of sarizotan (3 mM; 1 min pre-occupation without L741626). Effects of L741626 on GIRK current were determined for these two conditions and normalized to the mean current evoked by a full agonist in oocytes of the respective batch. Compare Fig. 1 for further details. Note that the GIRK current level in the presence of sarizotan before the delivery of L741626 is considered as the GIRK baseline level in the respective experiments.

cRNA (see Fig. 2). Using this approach, the inverse agonist effect of L741626 was indeed found to be receptor-mediated: while L741626 consistently decreased the basal GIRK current (8.58  3.76% of full agonist effect; n ¼ 7) in non-treated oocytes, no current decrease was observed after occupation of the hD2L receptor with sarizotan (1.0  0.48%; n ¼ 10; Fig. 3). 3.5. Pharmacology of antipsychotics To test for the activity of antipsychotics, we adopted the conditions of intermediate receptor expression (100 pg

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113

1110

cRNA per oocyte) as a standard approach. When tested in these conditions, the typical antipsychotic compound, haloperidol, and the atypicals, clozapine, nemonapride, and ziprasidone, exhibited inverse agonist activity similar to raclopride and L741626 (Fig. 4A). Among the ‘‘new generation’’ of compounds with mixed D2/5-HT1A receptor activity, F15063 and SLV313 acted as inverse agonists, exhibiting similar efficacy to the other drugs showing this behaviour (Fig. 4A), the pIC50 for inverse agonism of F15063 being 6.99  0.27 (Fig. 4B). In contrast, bifeprunox, SSR181507 and aripiprazole acted as partial agonists with relative efficacies comprised between 30% and 44% (Fig. 5). To further characterize the actions of the ‘‘new generation’’ compounds, they were tested at lower as well as at higher levels of receptor expression (Fig. 5). SLV313 and F15063 behaved as inverse agonists at higher levels of receptor cRNA injected, while inverse agonism was barely detectable at 10 pg cRNA per oocyte; thus, the results were similar to the effects of the reference compounds L741626 and raclopride. For the partial agonists, the relative efficacy values of bifeprunox, SSR181507 and aripiprazole were generally found to be higher with increasing receptor density. However, this tendency was less readily detectable in the case of aripiprazole, mimicking in part the effect observed with ()3-PPP.

antipsychotics: inverse agonists (100 pg hD2L cRNA / oocyte) n=6

0

n=9

n=15

n=8

n=7

13

63 F1 50

e

V3

rid

SL

ne

m

on

as i pr zi

5

ap

do

ne pi za

cl o

pe rid lo

ne

-10

ha

GIRK current activation (% of full agonist)

B

n=12

-5

ol

GIRK current activation (% of full agonist)

A

Log [F15063] (M)

0 -9

-8

-7

-6

-5

-5

-10

Fig. 4. Inverse agonist activity of antipsychotic compounds at hD2L receptors. (A) Oocytes were injected with 100 pg hD2L receptor cRNA each, and experiments were performed as described in Section 2 and Fig. 1. Drug concentrations were 1 mM for haloperidol, nemonapride and F15063, 3 mM for ziprasidone, and 10 mM for clozapine and SLV313. Effects on GIRK current were quantified and normalized to the mean current evoked by a full agonist in oocytes of the respective batch. (B) Concentrationeresponse curve of the effect of F15063.

4. Discussion 4.1. Xenopus oocyte expression system We have examined the activity of antipsychotics and reference products at the hD2L receptor using a Xenopus oocyte expression system. In particular, the system allows characterisation of partial agonist as well as inverse agonist activity at the hD2L receptor. By the use of different quantities of cRNA injected, the amount of receptor expression can be modulated (see e.g. Kovoor et al., 1998; Vorobiov et al., 2000). Even if one cannot assume that an increase in cRNA injected does imply a proportional increase in protein expression (compare e.g. Peleg et al., 2002), a higher receptor density and a concomitant induction of receptor reserve are demonstrated in our system by the increase of potency of dopamine at higher levels of cRNA injected. The system can thus be adapted to record activity of low efficacy ligands. In general, the relative efficacy of partial agonists is enhanced (or boosted towards full agonism; e.g. apomorphine in our assay) under conditions of receptor reserve. This is clearly demonstrated in our model (see below) and has also been shown for the D2 receptor in other expression systems (see e.g. Burris et al., 2002; Tadori et al., 2005). Inverse agonist activity, on the other hand, has been difficult to demonstrate for the D2 receptor in different in vitro systems. Often, G protein subunits are coexpressed (Burstein et al., 2005) or mutant D2 receptors specially engineered to exhibit a high degree of constitutive activity are expressed in order to record a sufficient degree of inverse agonist activity (Wilson et al., 2001; Wurch et al., 2003; but see Hall and Strange, 1997; Wiens et al., 1998, for studies at wild-type D2L and D2S receptors). In the Xenopus oocyte/GIRK system, however, we were able to record inverse agonist activity of several compounds at the wild-type hD2L receptor without G protein coexpression. However, inverse agonism seemed more consistently observed at higher receptor expression levels, a factor that is known to profoundly influence the degree of constitutive activity and inverse agonism (Kenakin, 2004). The inverse agonist nature of the basal current decrease observed was confirmed by the absence of effects in oocytes without hD2L receptor as well as in oocytes in which hD2L receptors were occupied by sarizotan, a compound that acts as a quasineutral antagonist in this system. These data therefore demonstrate that constitutive activity of the hD2L receptor is readily detectable in Xenopus oocytes. The mechanism underlying the detection of constitutive activity in the present system is unclear, but efficacious coupling to endogenous G proteins or the presence of suitable accessory proteins in these cells may be involved. It should be mentioned that the existence of constitutive activity of the D2 receptor in Xenopus oocyte has been suggested before (as indicated by the increase of basal GIRK currents with increasing receptor densities; see Gregerson et al., 2001). However, this is the first time that the system was employed to characterize inverse agonist activity at D2 receptors in a pharmacological approach.

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113

GIRK current activation (% of full agonist)

10 pg cRNA / oocyte

100 pg cRNA / oocyte

1111

120

1000 pg cRNA / oocyte n=6

100

n=10

80

n=8

40 20

n=12

n=7

60 n=6 n=11

n=13

n=9 n=7 n=7

n=8 n=7

n=6 n=7

0

ar

ip ip ra z bi fe ole pr SS uno x R 18 15 0 SL 7 V3 1 F1 3 50 63 ar ip ip ra zo bi le fe pr un SS ox R 18 15 SL 07 V3 13 F1 50 63 ar ip ip ra z bi fe ole pr SS un ox R 18 15 SL 07 V3 13 F1 50 63

-20

Fig. 5. Activity of ‘‘new generation’’ antipsychotic compounds at hD2L receptors. Drug concentrations were 1 mM for all compounds except SLV313 (10 mM). Oocytes were injected with hD2L receptor cRNA as indicated, while equal amounts of GIRK1/2 channel subunit cRNAs were coinjected for each condition. Experiments were performed as described in Fig. 1, and effects on GIRK current were quantified and normalized to the mean current evoked by a full agonist in oocytes of the respective batch.

4.2. Antipsychotics When antipsychotic agents were tested in this assay, essentially two groups of current responses were observed. Aripiprazole, SSR181507 and bifeprunox, on the one hand, behaved as partial agonists, whereas the other compounds behaved as inverse agonists. The members of the first group (partial agonists) belong to a new group of antipsychotics preferentially acting at 5-HT1A and D2 receptors. Comparative studies of these compounds have been published recently, employing [35S]GTPgS binding, measurement of prolactin levels and MAP kinase phosphorylation as functional readouts (Cosi et al., 2006; Bruins Slot et al., 2006; Newman-Tancredi et al., 2006). In agreement with our results, only aripiprazole, SSR181507 and bifeprunox showed partial D2 agonist properties in these assays, while no agonist properties could be detected for SLV313 and F15063 under identical conditions (Cosi et al., 2006; Bruins Slot et al., 2006; Newman-Tancredi et al., 2006). In agreement with expectations on the behaviour of partial agonists, relative efficacies of the antipsychotics aripiprazole, SSR181507 and bifeprunox were increased in our system when higher amounts of receptor were expressed. Thus, when the results are compared with those from the studies cited above, differences in receptor reserve and in signal amplification of the respective model systems have to be taken into account. With respect to our results under conditions of lower amounts of cRNA injected, the efficacy values of the partial agonists are in a similar range as those reported for G protein-activation at the hD2L receptor in vitro (Cosi et al., 2006). At the highest receptor expression tested, the efficacy values reach a level similar to that observed at hD2S receptors in an MAP kinase phosphorylation assay, a system particularly sensitive to partial agonist properties (Bruins Slot et al., 2006). The increase in relative efficacy at higher receptor level seems to be somewhat smaller for aripiprazole when compared

to SSR181507 and bifeprunox. A similar e but more pronounced e phenomenon was observed with ()3-PPP in the group of reference ligands. This suggests that there is a differential sensitivity of some compounds with regard to an increase in receptor density. The reasons for such a behaviour are unclear, but differential coupling to distinct G protein pools (with some of them being more readily activated than others) after receptor activation might be an explanation. Interestingly, there is evidence for ‘‘agonist-directed trafficking of receptor response’’ or ‘‘functional selectivity’’ at the D2 receptor (Nickolls and Strange, 2004; Gay et al., 2004). Importantly, evidence for pathway-specific agonism has been reported for aripiprazole in several studies (Lawler et al., 1999; Shapiro et al., 2003; Urban et al., in press; Jordan et al., 2005). Moreover, these indications include a relatively low sensitivity of aripiprazole, when compared to other partial agonists, with respect to changes in receptor sensitivity (Lawler et al., 1999), in accordance with the effects observed here. It remains to be determined if this behaviour is related to a differential G proteinactivation profile of aripiprazole. Interestingly, ()3-PPP, which shows a somewhat similar pattern to aripiprazole at higher expression level, has been reported to show a selectivity for coupling the D2 receptor to G proteins of the Go type (Gazi et al., 2003a; Cordeaux et al., 2001). The other antipsychotics exhibited inverse agonist properties in the present system with an efficacy similar to the reference compounds, raclopride and L741626. This includes the prototypical antipsychotic, haloperidol, the atypicals, clozapine, nemonapride, and ziprasidone, as well as F15063 and SLV313 as members of the 5-HT1A/D2 group of antipsychotics. It thus seems that antipsychotics that act as antagonists at D2 receptors indeed behave as inverse agonists when tested in an appropriate system, such as the Xenopus oocyte/ GIRK approach applied here or in some other studies as previously reported (Hall and Strange, 1997; Akam and Strange, 2004; Burstein et al., 2005). A possible exception is ziprasidone that has been found to be a neutral antagonist in certain

1112

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113

models (Wurch et al., 2003; Gazi et al., 2003b). However, these results are based on approaches using chimeric D2 receptors; whereas in studies using wild-type receptors, ziprasidone behaved as inverse agonist (Burstein et al., 2005; our study). On the other hand, one could consider pathway-specific effects and/or issues of receptor expression levels, which can profoundly influence constitutive activity. Thus, the compound (þ)-UH232, acting as partial agonist in our system as well as in other studies (e.g. Wurch et al., 2003), has been found to behave as inverse agonist at a D2S:Gao fusion protein (Gazi et al., 2003b). 4.3. Relationship to therapeutical potential and side effects The main findings of the present study concern the group of ‘‘new generation’’ antipsychotic drugs, possessing marked D2 and 5-HT1A properties. Among these, bifeprunox, SSR181507 and aripiprazole exhibit partial agonist activity at the hD2L receptor, while F15063 and SLV313 behave as inverse agonists. With respect to the treatment of schizophrenia, the impact of those differential properties at D2 receptors is still a matter of debate. Whilst partial agonists at this receptor may avoid certain side effects associated with neutral antagonists or inverse agonists, the efficacy of these recent compounds for control of positive symptoms of schizophrenia remains under discussion. On the other hand, all drugs cited above exhibit activities at other receptors than the D2 (see e.g. Cussac et al., 2006), and these will have an important impact on their functional profile. In particular, the drugs aripiprazole, bifeprunox, SSR181507, SLV313 and F15063 exhibit markedly different balances in their 5-HT1A/D2 properties (Newman-Tancredi et al., 2005, 2006). In conclusion, the main determinant of the therapeutic potential of an antipsychotic will be the interplay of its D2 receptor properties with its activity at other receptor subtypes. Acknowledgments We wish to thank Claudie Cathala for her expert support in molecular biology and Sophie Mercure for her kind secretarial assistance. The pGEMHE high-expression vector was kindly donated by Jan Tytgat (University of Leuven, Belgium). References Akam, E., Strange, P.G., 2004. Inverse agonist properties of atypical antipsychotic drugs. Biochem. Pharmacol. 67, 2039e2045. Bardin, L., Kleven, M.S., Barret-Grevoz, C., Depoortere, R., NewmanTancredi, A., 2006. Antipsychotic-like vs cataleptogenic actions in mice of novel antipsychotics having D2antagonist and 5-HT1Aagonist properties. Neuropsychopharmacology 31, 1869e1879. Bruins Slot, L.A., De Vries, L., Newman-Tancredi, A., Cussac, D., 2006. Differential profile of antipsychotics at serotonin 5-HT1A and dopamine D2S receptors coupled to extracellular signal-regulated kinase. Eur. J. Pharmacol. 534, 63e70. Burris, K.D., Molski, T.F., Xu, C., Ryan, E., Tottori, K., Kikuchi, T., Yocca, F.D., Molinoff, P.B., 2002. Aripiprazole, a novel antipsychotic, is

a high-affinity partial agonist at human dopamine D2 receptors. J. Pharmacol. Exp. Ther. 302, 381e389. Burstein, E.S., Ma, J., Wong, S., Gao, Y., Pham, E., Knapp, A.E., Nash, N.R., Olsson, R., Davis, R.E., Hacksell, U., Weiner, D.M., Brann, M.R., 2005. Intrinsic efficacy of antipsychotics at human D2, D3, and D4 dopamine receptors: identification of the clozapine metabolite N-desmethylclozapine as a D2/D3 partial agonist. J. Pharmacol. Exp. Ther. 315, 1278e1287. Claustre, Y., Peretti, D.D., Brun, P., Gueudet, C., Allouard, N., Alonso, R., Lourdelet, J., Oblin, A., Damoiseau, G., Francon, D., SuaudChagny, M.F., Steinberg, R., Sevrin, M., Schoemaker, H., George, P., Soubrie, P., Scatton, B., 2003. SSR181507, a dopamine D2 receptor antagonist and 5-HT1A receptor agonist. I: neurochemical and electrophysiological profile. Neuropsychopharmacology 28, 2064e2076. Cordeaux, Y., Nickolls, S.A., Flood, L.A., Graber, S.G., Strange, P.G., 2001. Agonist regulation of D2 dopamine receptor/G protein interaction. Evidence for agonist selection of G protein subtype. J. Biol. Chem. 276, 28667e28675. Cosi, C., Carilla-Durand, E., Assie, M.B., Ormiere, A.M., Maraval, M., Leduc, N., Newman-Tancredi, A., 2006. Partial agonist properties of the antipsychotics SSR181507, aripiprazole and bifeprunox at dopamine D2 receptors: G protein activation and prolactin release. Eur. J. Pharmacol. 535, 135e144. Cussac, D., Heusler, P., Martel, J., Newman-Tancredi, A., 2006. The new antipsychotics, bifeprunox, F15063 and SLV313, behave as partial agonists at dopamine D4 receptors: comparison with typical and atypical antipsychotics. Soc. Neurosci. Abstr., 332.4. Depoortere, R., Bardin, L., Auclair, A., Bruins-Slot, L., Kleven, M., NewmanTancredi, A., 2006. F15063, an innovative antipsychotic with D2/D3 antagonist, 5-HT1A agonist and D4 partial agonist properties: II. Behavioural profile in models of positive, negative symptoms and cognitive deficits of schizophrenia. Int. J. Neuropsychopharmacol. 9 (Suppl. 1), P 01.165. Farah, A., 2005. Atypicality of atypical antipsychotics. Prim. Care Companion: J. Clin. Psychiatry 7, 268e274. Gay, E.A., Urban, J.D., Nichols, D.E., Oxford, G.S., Mailman, R.B., 2004. Functional selectivity of D2 receptor ligands in a Chinese hamster ovary hD2L cell line: evidence for induction of ligand-specific receptor states. Mol. Pharmacol. 66, 97e105. Gazi, L., Nickolls, S.A., Strange, P.G., 2003a. Functional coupling of the human dopamine D2 receptor with Gai1, Gai2, Gai3 and Gao G proteins: evidence for agonist regulation of G protein selectivity. Br. J. Pharmacol. 138, 775e786. Gazi, L., Wurch, T., Lopez-Gimenez, J.F., Pauwels, P.J., Strange, P.G., 2003b. Pharmacological analysis of a dopamine D2Short:Gao fusion protein expressed in Sf9 cells. FEBS Lett. 545, 155e160. Gregerson, K.A., Flagg, T.P., O’Neill, T.J., Anderson, M., Lauring, O., Horel, J.S., Welling, P.A., 2001. Identification of G protein-coupled, inward rectifier potassium channel gene products from the rat anterior pituitary gland. Endocrinology 142, 2820e2832. Hall, D.A., Strange, P.G., 1997. Evidence that antipsychotic drugs are inverse agonists at D2 dopamine receptors. Br. J. Pharmacol. 121, 731e736. Heusler, P., Pauwels, P.J., Wurch, T., Newman-Tancredi, A., Tytgat, J., Colpaert, F.C., Cussac, D., 2005. Differential ion current activation by human 5-HT1A receptors in Xenopus oocytes: evidence for agonist-directed trafficking of receptor signalling. Neuropharmacology 49, 963e976. Jordan, S., Chen, R., Johnson, J., Regardie, K., Tadori, Y., Kikuchi, T., 2005. In vitro effects of antipsychotic drugs on second and third messengers of human dopamine D2 receptor signalling. Soc. Neurosci. Abstr. 913, 2. Jordan, S., Koprivica, V., Chen, R., Tottori, K., Kikuchi, T., Altar, C.A., 2002. The antipsychotic aripiprazole is a potent, partial agonist at the human 5HT1A receptor. Eur. J. Pharmacol. 441, 137e140. Kapur, S., Remington, G., 2001. Atypical antipsychotics: new directions and new challenges in the treatment of schizophrenia. Annu. Rev. Med. 52, 503e517. Kapur, S., Mamo, D., 2003. Half a century of antipsychotics and still a central role for dopamine D2 receptors. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 27, 1081e1090. Kenakin, T., 2004. Efficacy as a vector: the relative prevalence and paucity of inverse agonism. Mol. Pharmacol. 65, 2e11.

P. Heusler et al. / Neuropharmacology 52 (2007) 1106e1113 Kovoor, A., Celver, J.P., Wu, A., Chavkin, C., 1998. Agonist induced homologous desensitization of m-opioid receptors mediated by G protein-coupled receptor kinases is dependent on agonist efficacy. Mol. Pharmacol. 54, 704e711. Koyrakh, L., Lujan, R., Colon, J., Karschin, C., Kurachi, Y., Karschin, A., Wickman, K., 2005. Molecular and cellular diversity of neuronal Gprotein-gated potassium channels. J. Neurosci. 25, 11468e11478. Lawler, C.P., Prioleau, C., Lewis, M.M., Mak, C., Jiang, D., Schetz, J.A., Gonzalez, A.M., Sibley, D.R., Mailman, R.B., 1999. Interactions of the novel antipsychotic aripiprazole (OPC-14597) with dopamine and serotonin receptor subtypes. Neuropsychopharmacology 20, 612e627. Levine, N., Ethier, N., Oak, J.N., Pei, L., Liu, F., Trieu, P., Rebois, R.V., Bouvier, M., Hebert, T.E., Van Tol, H.H., 2002. G protein-coupled receptors form stable complexes with inwardly rectifying potassium channels and adenylyl cyclase. J. Biol. Chem. 277, 46010e46019. Liao, Y.J., Jan, Y.N., Jan, L.Y., 1996. Heteromultimerization of G-proteingated inwardly rectifying Kþ channel proteins GIRK1 and GIRK2 and their altered expression in weaver brain. J. Neurosci. 16, 7137e7150. Liman, E.R., Tytgat, J., Hess, P., 1992. Subunit stoichiometry of a mammalian Kþ channel determined by construction of multimeric cDNAs. Neuron 9, 861e871. McCreary, A.C., Glennon, J.C., Ashby, C.R., Meltzer, H.Y., Li, Z., Reinders, J.H., Hesselink, M.B., Long, S.K., Herremans, A.H., van Stuivenberg, H., Feenstra, R.W., Kruse, C.G., 2007. SLV313 (1-(2,3-dihydro-benzo[1, 4]dioxin-5-yl)-4-[5-(4-fluoro-phenyl)-pyridin-3-ylmethyl]-piperazine monohydrochloride): a novel dopamine D2receptor antagonist and 5-HT1Areceptor agonist potential antipsychotic drug. Neuropsychopharmacology 32, 78e94. Meltzer, H.Y., Li, Z., Kaneda, Y., Ichikawa, J., 2003. Serotonin receptors: their key role in drugs to treat schizophrenia. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 27, 1159e1172. Millan, M.J., 2000. Improving the treatment of schizophrenia: focus on serotonin (5-HT)1A receptors. J. Pharmacol. Exp. Ther. 295, 853e861. Neve, K.A., Seamans, J.K., Trantham-Davidson, H., 2004. Dopamine receptor signaling. J. Recept. Signal Transduct. Res. 24, 165e205. Newman-Tancredi, A., Assie, M.B., Leduc, N., Ormiere, A.M., Danty, N., Cosi, C., 2005. Novel antipsychotics activate recombinant human and native rat serotonin 5-HT1A receptors: affinity, efficacy and potential implications for treatment of schizophrenia. Int. J. Neuropsychopharmacol. 8, 341e356. Newman-Tancredi, A., Assie´, M.-B., Martel, J.-C., Cosi, C., Heusler, P., Bruins Slot, L., Carilla-Durand, E., Cussac, D., 2006. F15063, an innovative antipsychotic with D2/D3 antagonist, 5-HT1A agonist and D4 partial agonist properties: I. In vitro, neurochemical and neuroendocrine profiles. Int. J. Neuropsychopharmacol. 9 (Suppl. 1), P 01.164.

1113

Nickolls, S.A., Strange, P.G., 2004. The influence of G protein subtype on agonist action at D2 dopamine receptors. Neuropharmacology 47, 860e872. Peleg, S., Varon, D., Ivanina, T., Dessauer, C.W., Dascal, N., 2002. Gai controls the gating of the G protein-activated Kþ channel, GIRK. Neuron 33, 87e99. Prinssen, E.P., Kleven, M.S., Koek, W., 1999. Interactions between neuroleptics and 5-HT1A ligands in preclinical behavioral models for antipsychotic and extrapyramidal effects. Psychopharmacology (Berl.) 144, 20e29. Remington, G., 2003. Understanding antipsychotic ‘‘atypicality’’: a clinical and pharmacological moving target. J. Psychiatry Neurosci. 28, 275e284. Shapiro, D.A., Renock, S., Arrington, E., Chiodo, L.A., Liu, L.X., Sibley, D.R., Roth, B.L., Mailman, R., 2003. Aripiprazole, a novel atypical antipsychotic drug with a unique and robust pharmacology. Neuropsychopharmacology 28, 1400e1411. Tadori, Y., Miwa, T., Tottori, K., Burris, K.D., Stark, A., Mori, T., Kikuchi, T., 2005. Aripiprazole’s low intrinsic activities at human dopamine D2L and D2S receptors render it a unique antipsychotic. Eur. J. Pharmacol. 515, 10e19. Urban, J.D., Vargas, G.A., von Zastrow, M., Mailman, R.B., 2007. Aripiprazole has functionally selective actions at dopamine D2receptor-mediated signaling pathways. Neuropsychopharmacology 32, 67e77. Van Vliet, B.J., Mos, J., Van der Heijden, J.A.M., Feenstra, R., Kruse, C.G., Long, S.K., 2000. DU 12790: a highly potent, atypical dopamine receptor ligand e a putative potent full spectrum antipsychotic with low EPS potential. Eur. Neuropsychopharmacol. 10 (Suppl. 3), S293. Vorobiov, D., Bera, A.K., Keren-Raifman, T., Barzilai, R., Dascal, N., 2000. Coupling of the muscarinic m2 receptor to G protein-activated Kþ channels via Gaz and a receptor-Gaz fusion protein. Fusion between the receptor and Gaz eliminates catalytic (collision) coupling. J. Biol. Chem. 275, 4166e4170. Werner, P., Hussy, N., Buell, G., Jones, K.A., North, R.A., 1996. D2, D3, and D4 dopamine receptors couple to G protein-regulated potassium channels in Xenopus oocytes. Mol. Pharmacol. 49, 656e661. Wiens, B.L., Nelson, C.S., Neve, K.A., 1998. Contribution of serine residues to constitutive and agonist-induced signaling via the D2S dopamine receptor: evidence for multiple, agonist-specific active conformations. Mol. Pharmacol. 54, 435e444. Wilson, J., Lin, H., Fu, D., Javitch, J.A., Strange, P.G., 2001. Mechanisms of inverse agonism of antipsychotic drugs at the D2 dopamine receptor: use of a mutant D2 dopamine receptor that adopts the activated conformation. J. Neurochem. 77, 493e504. Wurch, T., Boutet-Robinet, E.A., Palmier, C., Colpaert, F.C., Pauwels, P.J., 2003. Constitutive coupling of a chimeric dopamine D2/alpha1B receptor to the phospholipase C pathway: inverse agonism to silent antagonism by neuroleptic drugs. J. Pharmacol. Exp. Ther. 304, 380e390.