Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Five dopamine receptor subtypes have been identified so far and may be divided into two subfamilies: the D_1-like receptors (D₁ and D₅) and the D_2-like receptors (D₂, D₃, and D₄) (Missale et al., 1998). The D_2-like receptors also exist as different splice variants, for example, the D_2short and D_2long receptors (Monsma et al., 1989). These variants are generated by alternative splicing of a single gene and differ by the insertion of a 29 amino acid segment in the third intracellular loop in the D_2long receptor. Several responses have been linked to the activation of D_2-like receptors. These include acute responses, such as inhibition of adenylate cyclase, stimulation of K⁺ channels, inhibition of Ca²⁺ channels, and stimulation of sodium/hydrogen ion exchange (Stoof and Kebabian, 1981; Seabrook et al., 1994; Pillai et al., 1998; Coldwell et al., 1999; Ghahremani et al., 1999). Longer term responses, such as those elicited by stimulation of mitogen-activated protein kinase, have also been described (Choi et al., 1999). These responses have been reported for the D_2-like receptors expressed in recombinant systems and probably reflect the activation of different classes of G proteins (Montmayeur et al., 1993; Senogles, 1994; Ghahremani et al., 1999; Obadiah et al., 1999). The relation of the responses observed in recombinant systems to those occurring in native systems is currently unclear, as most native tissues bearing D_2-like receptors contain more than one D_2-like receptor subtype (Strange, 1999).

One of the most important classes of D₂ receptor ligands is the antipsychotic drugs. These comprise a wide range of structural chemical classes with the common ability to act as clinically effective antipsychotic agents and are considered to be antagonists at D_2-like (D₂, D₃, and D₄) receptors. Hall and Strange (1997) suggested that this situation is an oversimplification because they found most of these antagonists act as inverse agonists at D₂ receptors. All the antipsychotic drugs tested, irrespective of their structural class, were inverse agonists and the extent of inverse agonist effect was the same for all the compounds tested (i.e., they all appear as full inverse agonists). There was a good correlation between their potency as inverse agonists and their dissociation constants for binding to the D₂ receptor. Some compounds, which have not so far been shown to possess antipsychotic activity, however, acted as either neutral antagonists (e.g., UH 232) or partial inverse agonists [e.g., AJ 76 (Strange, 1999)].

A limitation of available accounts of the interactions of DA antagonists with D₂ receptors concerns the resolution with which receptor activity has been scored and analyzed. Investigating whether a high-resolution analysis of the different behaviors that central nervous system stimulants induce may offer a more powerful account of the antipsychotic potential of neuroleptic compounds, Koek and Colpaert (1993) demonstrated antipsychotics to differ markedly in terms of the relative doses at which they antagonize the stereotyped behaviors induced by methylphenidate in rats. It was hypothesized that this variation among antipsychotics may be based on differences in the extent to which they exert agonist activity at dopamine receptors. Also, some of the effects of antipsychotic drugs have been shown to occur with only a moderate occupancy (50-75%) of D_2-like receptors and it is difficult to see how these could be achieved simply by antagonism of the effects of endogenous DA (Strange, 1999). Furthermore, it has been proposed that the differences in the affinity of antipsychotic agents for D₂ receptors are entirely determined by how fast they come off the receptor (Seeman and Tallerico, 1998; Kapur and Seeman, 2000). Differences in K_off constants may lead to functionally different kinds of DA blockade. These authors hypothesized that drugs with a high K_off will be faster in blocking D₂ receptors, and once blocked, will provide more access to surges in DA transmission.

In the present study, kinetic analyses of DA-antagonist interactions were carried out while using a recombinant human dopamine D_2short receptor in a cellular CHO-K1 model system. Receptor activation was monitored by measuring time-dependent Ca²⁺ responses following activation of a chimeric G_q/o protein, because it couples the D_2short receptor efficaciously to the cellular Ca²⁺ signaling pathway. The following questions were addressed: do the putative DA antagonists show intrinsic activity at the presumably unoccupied D_2short receptor (i.e., in the absence of DA), how efficacious does the antagonist-bound D_2short receptor antagonize subsequent activation by DA, and do these antagonists act similarly at the DA-bound D_2short receptor. The Ca²⁺ data indicate that most of the investigated putative DA antagonists act differently, in particular, at the DA-bound D_2short receptor.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Construction of Human Wild-Type and Mutant Thr³⁴³Arg Dopamine D_2short Receptor, Wild-Type, and Chimeric G Protein Genes. The short splice variant of the human dopamine D₂ receptor cDNA (RC: 2.1.DA.02) was cloned by PCR using oligonucleotide primers designed according to the sequence deposited in the GenBank database (accession number S69899). The PCR mixture (50 µl) consisted of 250 ng of reverse-transcribed poly(A⁺) RNA from human whole brain, 350 µM of each dNTP, 400 nM of each primer, and 1 µl of Expand long template DNA polymerase mix in PCR buffer [16 mM (NH₄)₂SO₄, 1.75 mM MgCl₂, 50 mM Tris-HCl (pH 9.2)]. The PCR program consisted of 30 repetitive cycles with a strand separation step at 96°C for 30 s, an annealing step at 60°C for 1 min, and an elongation step at 68°C for 1.5 min. Site-directed mutagenesis of the Thr³⁴³ position (ACT codon) into an Arg residue (AGA codon) was performed using a Quick Change site-directed mutagenesis kit, according to the supplier's instructions. Rat G_o (M17526) and mouse G_q (M55412) protein cDNA were PCR-amplified under similar experimental conditions using gene-specific primers. The chimeric G_q/o and G_q/s proteins were constructed by exchanging the last five amino acids (Glu³⁵⁵-Tyr-Asn-Leu-Val) of a mouse G_q protein by those corresponding to, respectively, a rat G_o (Gly-Cys-Gly-Leu-Tyr) or a rat G_s (Gln-Tyr-Glu-Leu-Leu; M12673) protein. This has been realized by directly inserting the respective nucleotide sequence on the reverse oligonucleotide primer used in a PCR reaction on cloned G_q protein cDNA (Pauwels et al., 2000b). Receptor and chimeric G protein constructions were inserted into a pCR3.1 mammalian expression vector and the nucleotide sequences were fully verified by DNA sequencing and confirmed the respective sequences.

Measurement of Intracellular Ca²⁺ Responses. Subconfluent CHO-K1 cells were transiently transfected with a human D_2short receptor and G_q/o protein plasmid (unless indicated) in an equimolecular amount (10 µg) by electroporation. Cells were assayed between 24 and 48 h upon transfection for intracellular Ca²⁺ responses upon 1-h pulse with 2 µM Fluo-3 fluorescent calcium indicator dye as described (Pauwels et al., 2000a). Either DA or other dopaminergic ligands were assayed for their Ca²⁺ response. Data for Ca²⁺ responses were obtained in arbitrary fluorescence units and were not translated into Ca²⁺ concentrations. A typical DA-mediated Ca²⁺ response displayed two phases: a high-magnitude phase, which was transient, and a low-magnitude phase, which continued throughout the recorded time period (15 min). Fluorescent readings were made every 2 s for the first 3.5 min and subsequently every 5 s for 10 min using a fluorometric imaging plate reader (FLIPR; Molecular Devices, Menlo Park, CA). E_max values were defined as the ligand's maximal high-magnitude response in percentage versus that obtained with 10 µM DA. pEC₅₀ values correspond to a ligand concentration at which 50% of its own maximal high-magnitude Ca²⁺ response was measured. Antagonists were either preincubated for 10 min before DA to prevent the high-magnitude Ca²⁺ phase in the antagonist-bound receptor state, or added 3.5 min upon the stimulation by DA to reverse the low-magnitude Ca²⁺ phase in the DA-bound state. Antagonist capacity (%) of DA-mediated high-magnitude Ca²⁺ response was defined as the property of the ligand (1 µM, added at 10 min before DA) to antagonize the high-magnitude DA response. This was calculated as the surface area between the DA and ligand condition for a period of 4 min upon addition of DA. Reversal capacity (%) of DA-mediated low-magnitude Ca²⁺ response was defined as the property of the ligand (1 µM, added at 3.5 min upon DA addition) to reverse the DA response. This was calculated as the surface area between the DA and ligand condition for a period of 10 min upon addition of the ligand. This latter surface area is expressed for each antagonist in percentage versus the reversal as obtained with 1 µM tropapride. pIC₅₀ values were defined as the ligand concentration to antagonize the high-magnitude Ca²⁺ response or reverse the low-magnitude Ca²⁺ response by 50%. [³H]Sulpiride binding (2.0 nM) and protein levels were determined on intact transfected CHO-K1 cells as described (Pauwels et al., 2000a). Nonspecific [³H]sulpiride binding was determined in the presence of 10 µM (+)-butaclamol.

Statistics. Statistical analysis was performed on antagonist and reversal capacity values using one-way analysis of variance, followed by all pairwise multiple comparison procedures (Tukey's test).

Materials. All molecular biology reagents were purchased from Invitrogen (San Diego, CA), Roche Diagnostics (Indianapolis, IN), Stratagene (La Jolla, CA), and PE Biosystems (Foster City, CA). CHO-K1 cells were obtained from American Type Culture Collection (Rockville, MD). ()-[methoxy-³H]Sulpiride (60-87 Ci/mmol) was obtained from NEN (Les Ulis, France). Fluo-3 was obtained from Molecular Probes (Eugene, OR). DA chlorhydrate, fluphenazine dichlorhydrate, ()-sulpiride, chlorpromazine chlorhydrate, and haloperidol were obtained from Sigma (St. Louis, MO). Bromocripine mesylate, S-(+)-propylnorapomorphine hydrochloride [(+)-NPA], risperidone, clozapine, domperidone, and (+)- and ()-butaclamol chlorhydrate were purchased from Research Biochemicals International (Natick, MA). Lisuride maleate and bromerguride were from Schering (Berlin, Germany). (+)-UH 232 was from Tocris (Baldwin, MO). Nemonapride, tropapride chlorhydrate, S 14066 oxalate, and olanzapine were prepared intra muros.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Ca²⁺ Response as Mediated by Recombinant Human Dopamine D_2short Receptor. In contrast to its lack of effect in nontransfected cells, DA produced a time- and concentration-dependent increase (pEC₅₀ = 8.02 ± 0.09) in the intracellular Ca²⁺ concentration in CHO-K1 cells transiently cotransfected with a human wild-type dopamine D_2short receptor and a chimeric G_q/o protein (Fig. 1A). A high-magnitude Ca²⁺ peak response occurred within 13.2 ± 0.7 s (n = 15) after agonist addition, whereafter the signal decreased at 3 min to 47.9 ± 2.8% of its maximal amplitude. Thereafter, the low-magnitude Ca²⁺ response was maintained for at least the 15-min period during which recordings were made. Assay of the dopamine D_2short receptor alone or by coexpression with either a G_q, G_o, or a chimeric G_q/s protein revealed either no or small DA-mediated Ca²⁺ responses (Fig. 1B). Activation of the dopamine D_2short receptor in the copresence of a G_q/o protein by a series of dopaminergic ligands revealed the following rank order of high-maximal Ca²⁺ responses: DA > lisuride = bromocriptine > (+)-NPA (Fig. 2). This rank order is similar to that obtained by measuring [³⁵S]guanosine-5'-O-(3-thio)triphosphate binding responses in CHO and Ltk fibroblast cells stably transfected with a dopamine D_2short receptor (Gardner et al., 1997; Terasmaa et al., 2000). Each of these ligands also displayed a low-magnitude Ca²⁺ response (Fig. 2D).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Ca2+ responses as obtained with CHO-K1 cells cotransfected with a human D2short receptor and either wild-type or chimeric G proteins. A, cotransfection of human D2short receptor was performed with a chimeric Gq/o protein in CHO-K1 cells as described under Experimental Procedures. In addition to the basal condition, indicated concentrations of DA were applied at minute zero and their effects were monitored every 2 s for 3 min. Curves illustrate a representative experiment performed in quadruplicate. The maximal Ca2+ response as induced by 10 µM DA represents 10,693 ± 407 (n = 89) arbitrary fluorescence units. Transfected cells expressed 1.3 ± 0.2 pmol · mg1 of protein of [3H]sulpiride (2 nM) binding sites. B, transfection of human D2short receptor was performed in the copresence of empty plasmid, or in combination with Gq/o, Go, Gq/s, or Gq protein, and assayed with 10 µM DA. Ca2+ responses were measured as described under Experimental Procedures every 2 s for 3 min. Curves illustrate a representative experiment performed in quadruplicate. AFU, arbitrary fluorescence unit.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Comparison between dose-dependent dopaminergic ligand-induced Ca2+ responses at D2short receptor in the copresence of a Gq/o protein in CHO-K1 cells. High- and low-magnitude Ca2+ responses were measured as described under Experimental Procedures. A-C, high-magnitude Ca2+ response: Emax values are expressed as a percentage of the respective high-magnitude Ca2+ response induced by 10 µM DA. Curves were constructed using mean values ± S.E.M. obtained in 5 to 12 independent transfection experiments. D, time-dependent Ca2+ responses mediated by the indicated ligands at maximally effective concentrations. Curves illustrate a representative experiment performed in quadruplicate.

Action of Putative Antagonists at the Unoccupied (Dopamine-Free) Dopamine D_2short Receptor. Within a series of investigated putative DA antagonists (Table 1), (+)-UH 232 and bromerguride demonstrated positive intrinsic activity at the D_2short receptor (Fig. 3). A high-, but not a low-magnitude, Ca²⁺ response (Fig. 3C) was observed upon activation of the D_2short receptor by (+)-UH 232 and bromerguride in contrast to DA, lisuride, bromocriptine, and (+)-NPA (Fig. 2D). The agonist effect as mediated by (+)-UH 232 and bromerguride seems specific to the D_2short receptor because it was not observed in nontransfected cells. The other ligands being investigated did not show intrinsic activity at 1 µM or lower concentrations; although small (<15%) effects on the basal Ca²⁺ response were observed at 10 µM. These effects do not correlate with their affinity for the D_2short receptor (Seeman and Tallerico, 1998). In further experiments, antagonists were analyzed at maximally 1 µM so as to avoid nonspecific effects on Ca²⁺ signaling. Assay of these ligands (data not shown) at a facilitating mutant D_2short Thr³⁴³Arg receptor (Wilson et al., 1999) displayed an enhanced maximal response for partial agonists (i.e., bromerguride by 53% and (+)-UH 232 by 59%), however, none of the other putative antagonists enhanced or attenuated the basal Ca²⁺ signal at this mutant D_2short receptor.

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Ca2+ responses by putative dopamine antagonists at D2short receptor in the copresence of a Gq/o protein in CHO-K1 cells. High- and low-magnitude Ca2+ responses were measured as described under Experimental Procedures. A and B, high-magnitude Ca2+ response: Emax values are expressed as a percentage of the respective high-magnitude Ca2+ response induced by 10 µM DA. Curves were constructed using mean values ± S.E.M. obtained in five independent transfection experiments. C, time-dependent Ca2+ responses mediated by the indicated ligands at maximally effective concentrations. Curves illustrate a representative experiment performed in quadruplicate. The DA curve was taken from Fig. 2D.

Dopamine Action at Antagonist-Bound D_2short Receptor. Preincubation of transfected CHO-K1 cells for 10 min to increasing concentrations of the putative antagonist tropapride before 10 µM DA exposure indicated full antagonism of the high-magnitude Ca²⁺ response at 0.1 µM and higher concentrations (Fig. 4A). A comparison between the maximal magnitude of antagonism by a series of putative DA antagonists at 1 µM for the DA-mediated high-magnitude Ca²⁺ response is provided in Fig. 4B. The quantification of the magnitude of the ligands' antagonism versus 10 µM DA and their potencies is summarized in Table 1. (+)-Butaclamol, bromerguride, and nemonapride did antagonize the DA response to a same extent (p = N.S.) as tropapride. The other ligands being investigated were less effective as antagonist of the high-magnitude Ca²⁺ response. These ligands can be respectively classified in three different classes on basis of their degree of antagonist capacity: fluphenazine, S 14066, haloperidol, risperidone, and domperidone because their maximal antagonist capacity was between 65 and 85% compared with that of tropapride; ()-sulpiride, which acted as a weak partial antagonist; and ()-butaclamol, chlorpromazine, (+)-UH 232, and olanzapine, which were virtually inactive as clozapine at a DA concentration of 10 µM. Otherwise, these latter compounds, with the exception of ()-butaclamol, antagonized partially the high-magnitude Ca²⁺ response mediated by submicromolar concentrations of DA (Table 1). Figure 5 illustrates the concentration-response curves for the dopaminergic ligands displaying more than 60% of antagonist capacity for the high-magnitude Ca²⁺ response mediated by 10 µM DA. Besides the observed different magnitudes of antagonist capacity, these ligands demonstrated 89-fold differences in antagonist potency (Table 1). In addition, peak Ca²⁺ values as observed in the DA-mediated high-magnitude response appeared slower in the presence of increasing concentrations of either haloperidol or risperidone compared with, for instance, those mediated by S 14066 (Fig. 5).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Antagonism of dopamine-mediated high-magnitude Ca2+ response by putative dopamine antagonists. A, antagonism of DA (10 µM)-mediated high-magnitude Ca2+ response by the indicated concentrations of tropapride in CHO-K1 cells cotransfected with D2short receptor and Gq/o protein. Tropapride was added 10 min before 10 µM DA. High-magnitude Ca2+ responses were measured as described under Experimental Procedures, and curves illustrate a representative experiment performed in quadruplicate, of nine independent transfection experiments. B, antagonism of DA (10 µM)-mediated high-magnitude Ca2+ response by 1 µM of the indicated ligands in CHO-K1 cells cotransfected with D2short receptor and Gq/o protein. Ca2+ responses were measured as in A, and quantified as percentage remaining of the surface area of the Ca2+ response obtained with 10 µM DA alone. Surface areas are expressed in percentage as mean values ± S.E.M. of 5 to 14 independent transfection experiments. Antagonist capacity values are not statistically different for aversus clozapine and for bversus nemonapride. AFU, arbitrary fluorescence unit.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Kinetics of antagonism of dopamine-mediated high-magnitude Ca2+ response by putative dopamine antagonists. Antagonism of DA (10 µM)-mediated high-magnitude Ca2+ responses by the indicated concentrations of ligands was performed as described in the legend to Fig. 4A. Ligands were added 10 min before 10 µM DA. Ca2+ responses were measured as described under Experimental Procedures. Curves illustrate a representative experiment performed in quadruplicate. Mean pIC50 and antagonist capacity values ± S.E.M. are summarized in Table 1. AFU, arbitrary fluorescence unit.

Antagonist Action at Dopamine-Bound D_2short Receptor. Figure 6A illustrates the effect of increasing concentrations of tropapride on the reversal of the low-magnitude Ca²⁺ response by preexposure of transfected CHO-K1 cells for 3.5 min to 10 µM DA. Tropapride (1 µM) reversed within 142 ± 23 s (n = 11) the low-magnitude Ca²⁺ signal to the basal level that had also been observed before DA stimulation. This reversal effect was fully maintained for the recorded time period of 10 min. Nemonapride (1 µM) approached most closely (p = N.S.) the maximal reversal obtained with tropapride (Fig. 6B). S 14066, risperidone, and haloperidol, at the concentration of 1 µM, reversed the low-magnitude Ca²⁺ response by 64, 68, and 79%, respectively. The other ligands were either less or not efficacious in reversing the low-magnitude Ca²⁺ response induced by 10 µM DA; 1 µM bromerguride, domperidone, and ()-sulpiride reversed the Ca²⁺ response by only 27 to 41%. Larger reversal effects (82-87%) were obtained with these compounds in experiments using a DA concentration of only 1 µM (Fig. 6B). A similar enhanced reversal effect on the low-magnitude Ca²⁺ response mediated by either 1 or 0.1 µM DA was observed with 1 µM clozapine, (+)-UH 232, fluphenazine, (+)-butaclamol, olanzapine, and chlorpromazine in contrast to 1 µM the inactive ()-enantiomer of butaclamol (Fig. 6B). Figure 6C shows for the herein investigated ligands the weak relationship between their magnitude of preventing effect of the high-magnitude Ca²⁺ response and their magnitude of reversing effect of the low-magnitude Ca²⁺ response. Figure 7 illustrates the kinetics for the reversal of the DA responses by these compounds. S 14066 clearly showed besides a smaller, also a slower onset of reversal of the DA response at 10 µM compared with the response of tropapride (Fig. 7A). (+)-Butaclamol, bromerguride, and fluphenazine displayed at submicromolar concentrations of DA a slower onset of reversal compared with tropapride; however, they could under these conditions attain almost maximal reversal at 10 min of incubation (Fig. 7, B and C).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Reversal of dopamine-mediated low-magnitude Ca2+ response by putative dopamine antagonists. CHO-K1 cells were transfected with D2short receptor and Gq/o protein. DA was applied at 1 min and 3.5 min later the putative antagonist was added and Ca2+ response was followed every 5 s for 10 min. A, reversal of low-magnitude Ca2+ response induced by 10 µM DA by increasing concentrations of tropapride. Curves illustrate a representative experiment, performed in quadruplicate, of four independent transfection experiments. The Ca2+ fluorescence value at the beginning of the reversal stage was set equal to zero, and this artifactually makes the original baseline (before compound addition) negative. B, reversal of low-magnitude Ca2+ response induced by various DA concentrations (, 0.1 µM; , 1 µM; and , 10 µM) by 1 µM of indicated dopaminergic ligands. Reversal capacity of low-magnitude Ca2+ response was defined as the property of the ligand to reverse the respective DA response as performed in A. This was calculated as the surface area between the respective DA and ligand condition for a period of 10 min upon addition of the ligand. Surface areas for each of the DA concentrations are expressed in percentage versus the respective tropapride (1 µM) condition. Values represent mean ± S.E.M. of 5 to 19 independent transfection experiments. Reversal capacity values at 10 µM DA are not statistically different for aversus clozapine and bversus nemonapride. C, relationship between the ligands' magnitude of preventing effect of the high-magnitude Ca2+ response and their magnitude of reversing effect of the low-magnitude Ca2+ response. Data as obtained with 10 µM DA were taken from Figs. 4B and 6B. A weak correlation (r2 = 0.43, p < 0.01) was obtained. AFU, arbitrary fluorescence unit.

View larger version (43K): [in this window] [in a new window]   Fig. 7.   Kinetics of reversal of dopamine-mediated low-magnitude Ca2+ response by putative dopamine antagonists. Reversal of DA-mediated low-magnitude Ca2+ response by the indicated concentrations of ligands was performed as described in the legend to Fig. 6. DA (10, 1, or 0.1 µM) was applied at 1 min and 3.5 min later increasing concentrations of ligand were added. Ca2+ response was followed every 5 s for 10 min. The Ca2+ fluorescence value at the beginning of the reversal stage was set equal to zero, and this artifactually makes the original baseline (before compound addition) negative. Curves illustrate a representative experiment performed in quadruplicate. Mean pIC50 and reversal capacity values ± S.E.M. are summarized in Table 1. A, reversal of 10 µM DA. B, reversal of 1 µM DA. C, reversal of 0.1 µM DA. T, reversal of the corresponding low-magnitude Ca2+ response by 1 µM tropapride; AFU, arbitrary fluorescence unit.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The present study characterizes the antagonist properties of a large series of dopaminergic ligands that have previously been defined as putative antagonists at dopamine D₂ receptors. These ligands were investigated under three different activation states of the recombinant human D_2short receptor, i.e., the unoccupied DA-free receptor, the DA-bound receptor, and the antagonist-bound receptor before activation by DA. The reversal of the low-magnitude Ca²⁺ phase in the DA-bound receptor state is of particular interest to investigate the properties of dopaminergic antagonists because this experimental condition approaches the best a hyperdopaminergic receptor state. Certain DA pathways have been postulated to be overactive in schizophrenia (Niedermier and Nasrallah, 1997), although no one has been able to prove that a true hyperdopaminergic state exists in schizophrenia (Goldstein, 1999). Although the evidence of such DA hyperactivity is not completely persuasive, all currently available antipsychotic drugs effective in schizophrenia block dopamine receptors with varying degrees of selectivity (Goldstein, 1999). We suggest that besides receptor selectivity, the degree of antagonism at dopamine receptors may be another factor to consider in the antipsychotic medication. Our Ca²⁺ data suggest a wide spectrum in the magnitude of antagonist capacity as observed by analyses of DA-antagonist interactions at the D_2short receptor while measuring time-dependent Ca²⁺ responses. The relevance of this in understanding the antipsychotic activity of these drugs is difficult to evaluate at the current stage, nonetheless it appears clear that this series of compounds interacts in different and multiple ways with the dopamine D₂ receptor. Two dopaminergic ligands, tropapride and nemonapride, were capable to fully antagonize and reverse, respectively, the high- and low-magnitude phases of the 10 µM DA-mediated Ca²⁺ response. Haloperidol, risperidone, and S 14066 also shared both antagonist and reversal properties but with a significantly lower magnitude. The other dopaminergic ligands being investigated displayed a weaker capacity to reverse the low-magnitude Ca²⁺ response, although some of them were efficacious to antagonize the DA-mediated high-magnitude Ca²⁺ response when bound to the receptor before DA. To the best of our knowledge, this is the first demonstration that these dopaminergic ligands can be differentiated at the recombinant human dopamine D_2short receptor by taking into account their antagonist properties.

It has previously been reported that the chimeric G_q/i/o proteins can convert the coupling of G_i/o protein-coupled receptors to the phospholipase C pathway (Liu et al., 1995; Conklin et al., 1996); they therefore are suitable to monitor receptor-mediated Ca²⁺ responses. Ca²⁺ mobilization by dopamine D₂ receptors has been demonstrated in CHO-K1 and Ltk fibroblast cells stably expressing the receptor; this response was sensitive to the G_i/o protein-inactivating agent pertussis toxin (Hayes et al., 1992; Liu et al., 1992). Hence, this Ca²⁺ response is likely to be mediated by -subunits of G_i/o proteins. In the present study, a robust DA-mediated Ca²⁺ response was exclusively observed in the copresence of a G_q/o protein; it consisted of a rapid, transient response with a high-magnitude phase followed by a low-magnitude phase, which continued for the recorded time period (15 min). Both phases could be fully antagonized by tropapride, indicating both phases of this Ca²⁺ process need to be mediated by the dopamine D_2short receptor. The lack of Ca²⁺ response in nontransfected cells together with an activation profile as observed with several dopaminergic agonists similar to that reported (Gardner et al., 1997; Terasmaa et al., 2000) further indicate that the herein described Ca²⁺ responses are mediated by the recombinant dopamine D_2short receptor. Besides the investigated dopaminergic agonists, bromerguride and (+)-UH 232 displayed weak positive intrinsic activity at the dopamine D_2short receptor, which could be enhanced at the facilitating mutant Thr³⁴³Arg D_2short receptor. Both enantiomers of UH 232 have also been reported as partial agonists by measuring the extracellular acidification rate at the D_2long receptor stably transfected in CHO cells (Coldwell et al., 1999). Monitoring forskolin-stimulated cAMP accumulation, Hall and Strange (1997) suggested (+)-UH 232 to be a weak partial inverse agonist at the stably transfected D_2short receptor rather than a truly neutral antagonist. It cannot be excluded that these minor differences in intrinsic activities for UH 232 reflect effector-dependent features. The other putative DA antagonists being investigated did not display any measurable form of intrinsic activity as assayed in the absence of agonist at the unoccupied D_2short receptor. Remarkably, bromerguride, in contrast to (+)-UH 232, when applied before DA was almost fully capable (95%) to antagonize the DA-mediated high-magnitude Ca²⁺ response despite this compound acting as a partial agonist at the unoccupied D_2short receptor. Otherwise, haloperidol and risperidone, which are presumably free of intrinsic activity at the unoccupied D_2short receptor, were not fully effective as antagonists on both the low- and high-magnitude Ca²⁺ phase. Therefore, these Ca²⁺ data suggest that dopaminergic compounds act differently at the unoccupied dopamine D_2short receptor and dopamine-bound receptor.

Fluphenazine, (+)-butaclamol and bromerguride could antagonize the high-magnitude Ca²⁺ response when incubating the cells with antagonist before DA. However, the capacity of these antagonists to reverse, at the same concentration of DA (10 µM), the low-magnitude Ca²⁺ phase was weak or absent for the recorded time period of 10 min. In addition, the time course of their reversal effect at lower DA concentrations was slower compared with that of tropapride. Therefore, they appear to bind slowly to the D_2short receptor. In contrast, haloperidol, risperidone, and nemonapride are likely to bind rapidly to the D_2short receptor as observed for tropapride. Indeed, no major differences were observed in their capacity to either antagonize or reverse the high- and low-magnitude Ca²⁺ responses. Moreover, the corresponding onset time for reversal of the DA response was, in contrast to S 14066, very similar to that of tropapride. Hence, the investigated DA antagonists not only act differently in their capacity to antagonize but also in their onset time of action, when interacting with the DA-bound D_2short receptor. Kapur and Seeman (2000) recently argued that the differences in the affinity of antipsychotic agents are entirely determined by how fast they come off the D₂ receptor. Accordingly, differences in K_off constants may lead to functionally different ways of DA blockade. This working hypothesis based on ligand-D₂ receptor interactions was approached by measuring ligand binding affinities and the amount of receptor occupation versus time for the rat striatal D₂ receptor. A major difference with our study is the absence of comparison of antagonists at the DA-bound D₂ receptor when analyzing antagonist-D₂ receptor interactions. In addition, our study was performed with a recombinant human D_2short receptor on intact cells and analyzed with a functional approach by measuring a Ca²⁺ response instead of determining ligand binding affinities on a membrane preparation. Also, the accuracy of measuring occupancy at the D₂ receptor by a ligand highly depends on the choice of the radioligand used to label the receptor (Seeman and Tallerico, 1998). The observed differences in reversal capacity between the herein investigated DA antagonists may reflect distinct binding pockets at the D₂ receptor for these antagonists. In case the antagonist binding site overlaps with the site of DA, reversal of the DA-mediated low-magnitude Ca²⁺ phase will depend considerably on the competition between the antagonist and DA. Otherwise, a more distinct antagonist binding site will likely be less dependent on the presence of DA. Receptor mutagenesis studies may further examine this hypothesis. A subgroup of substituted benzamide drugs [i.e., ()-sulpiride versus nemonapride] has been shown to be specifically affected by mutation of a histidine³⁹³ in the sixth transmembrane domain of the rat D_2long receptor (Woodward et al., 1994), suggesting a different binding interface.

In conclusion, the results of this study show that different DA antagonists display distinct capacities for inhibition of DA-mediated Ca²⁺ responses via a G_q/o protein. It appears crucial to compare these ligands at different activation states of the D_2short receptor. In particular, the comparison between DA-bound receptor state versus antagonist-bound receptor state may reveal distinct antagonist properties, which appear otherwise undetectable. Further efforts should be concentrated on activation of the D₂ receptor subtypes by natural G_i/o proteins to confirm the herein observed differences in antagonist properties of antipsychotic drugs.