Animal models of sensorimotor gating are used to study the aberrant neurotransmitter systems underlying psychotic disorders. Defined as the neural process responsible for the integration and processing of sensory information, sensorimotor gating is deficient in patients with schizophrenia (Braff et al., 1978). This phenotype can be quantitatively assessed in various species by measuring prepulse inhibition (PPI) of the acoustic startle response (Swerdlow et al., 1991), which is the reduction in response when a startling pulse stimulus is preceded by a weak prepulse stimulus.

The neural substrate for PPI modulation includes the mesocorticolimbic dopamine system (Swerdlow and Geyer, 1998; Swerdlow et al., 2001), especially D₂-like dopamine receptors in the nucleus accumbens (NAc). When administered systemically, both direct and indirect dopamine agonists disrupt PPI in experimental animals (Mansbach et al., 1988), as do selective D₂-like receptor agonists such as quinpirole and 7-hydroxy-dipropylaminotetralin (7-OH-DPAT) (Peng et al., 1990; Varty and Higgins, 1998). The involvement of NAc D₂-like receptors in PPI regulation was confirmed by site-specific infusion of quinpirole into the NAc, which elicited PPI disruption (Wan and Swerdlow, 1993).

Although the effect of acute dopamine agonist treatment on PPI is well defined, there are contradictory reports on the effect of chronic drug treatment. Repeated administration of amphetamine, an indirect-acting dopamine agonist, or apomorphine, a nonselective dopamine agonist, reportedly causes sensitization of PPI disruptive effects (Zhang et al., 1998; Martin-Iverson, 1999). In contrast, repeated administration of the same drugs also has been observed to attenuate or eliminate PPI disruption compared with acute treatment (Mansbach et al., 1988; Druhan et al., 1998; Feifel et al., 2002), as does repeated cocaine treatment (Byrnes and Hammer, 2000). The reason for these behavioral differences is unclear; however, variations in methodology or in animal substrains might contribute to the inconsistent effects. Since the D₂-like receptor is preferentially implicated in PPI regulation, the effect of repeated stimulation might be more clearly defined by the use of a more selective receptor agonist. Therefore, we sought to examine PPI adaptation after repeated treatment with the selective D₂-like receptor agonist, quinpirole.

In addition, we sought to characterize the substrates underlying PPI adaptation following repeated quinpirole treatment by assessing putative changes in D₂-like receptor G protein function. NAc D₂-like receptors are coupled to G_i and G_o proteins (Stoof and Kebabian, 1981), which are critical for PPI modulation (Culm et al., 2003). Chronic cocaine treatment, which attenuates PPI disruption (Byrnes and Hammer, 2000), decreases pertussis toxin-mediated ADP-ribosylation of G_iα and G_oα and reduces G_iα and G_oα immunoreactivity in the NAc (Nestler et al., 1990). In contrast, repeated haloperidol treatment increases D₂-like receptor binding and efficacy of D₂-like receptor G protein coupling as assessed using [³⁵S]GTPγS binding (Geurts et al., 1999), without affecting G_iα or G_oα levels (Meller and Bohmaker, 1996). Therefore, we utilized dopamine-stimulated [³⁵S]GTPγS binding and immunoblots to quantify putative alterations of G proteins coupled to the D₂-like receptor after repeated quinpirole treatment.

Materials and Methods

Experiment I

Animals and Drug Treatment. For all of the following experiments, animals were provided with food and water ad libitum while housed in a climate-controlled facility with 12-h reverse light/dark cycles (lights off at 9:00 AM). Animals were allowed to acclimate to the laboratory for 7 days before handling. All experiments were approved by the Tufts-New England Medical Center Animal Care and Use Committee and conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) weighing 250 to 300 g were habituated to handling and placed into a Startle Monitor behavioral testing chamber (Hamilton-Kinder LLC, Julian, CA) with 70 db ambient noise for 5 min daily on each of 2 days before testing. Animals were treated once daily for 10 consecutive days with the same dose of quinpirole (0.0, 0.05, 0.1, or 0.3 mg/kg, s.c.; Sigma/RBI, Natick, MA) in 0.9% sterile saline.

Prepulse Inhibition Testing. Startle amplitude was determined using the Startle Monitor behavioral testing system (Hamilton-Kinder, LLC) at baseline and after quinpirole treatment on days 1, 4, 7, and 10. Mean startle amplitude was measured over 100 ms following the presentation of the pulse stimulus in units of newtons. For baseline testing, each animal was exposed to 70 db ambient noise for 5 min followed by a test session consisting of the randomized presentation of 32 trials: 17 pulse trials (40 ms, 120-db pulse) and 15 prepulse trials (five each at 73, 76, and 82 db with a 20-ms prepulse given 100 ms before a 40 ms 120-db pulse). The mean acoustic startle response to pulse alone trials was used to normalize animals into various treatment groups. Starting 2 to 3 days later, the first drug challenge test session began 10 min after quinpirole treatment and consisted of the randomized presentation of 54 trials: 24 pulse trials, 10 73-db prepulse trials, 10 76-db prepulse trials, and 10 82-db prepulse trials. The average intertrial interval was 15 s and percent PPI was calculated using the following equation: 100 – [(mean prepulse response/mean pulse response) × 100]; a higher percent PPI implies greater inhibition of startle response due to presentation of the prepulse. Percent PPI data calculated for each prepulse level and mean pulse response data on the first and last day of testing were analyzed by separate analyses of variance (ANOVAs) with drug treatment as a between-subject factor. Within each treatment group, percent PPI data were analyzed using ANOVA with repeated measures. Post hoc comparisons were conducted using Dunnett's test. Alpha was 0.05.

[³⁵S]GTPγS Binding Analysis. Immediately after PPI testing on day 10, animals were decapitated, and brains were removed and rapidly frozen at –30°C in 2-methylbutane and sectioned at 16 μm using a –20°C cryostat. Sections corresponding to approximately 1.7 mm anterior to bregma were collected, thaw mounted onto gelatin-coated slides, and stored at –80°C for a maximum of 7 days prior to the time of the binding assay. Animals from the high dose group were not assessed because no behavioral adaptations occurred with repeated treatment.

[³⁵S]GTPγS binding was performed as described previously (He et al., 2000). Briefly, sections were preincubated in assay buffer (50 mM Tris-HCl, 3 mM MgCl₂, 0.2 mM EGTA, and 100 mM NaCl, pH 7.4) for 15 min at 25°C followed by a 15-min incubation in the same buffer supplemented with 2 mM GDP (ICN, Costa Mesa, CA). Sections were then incubated in assay buffer containing 2 mM GDP and 50 pM [³⁵S]GTPγS (PerkinElmer Life and Analytical Sciences, Boston, MA) in the absence (basal) or presence of 1, 10, and 100 μM, or 1 mM dopamine (Sigma-Aldrich, St. Louis, MO) for 1 h at 25°C. The sections were washed two times for 3 min at 4°C in 50 mM Tris-HCl (pH 7.4) and briefly rinsed in distilled water. After air drying, slides were coexposed with ¹⁴C radiostandards (ARC-146; American Radiolabeled Chemicals, St. Louis, MO) to X-ray film (Biomax MR; Eastman Kodak Company, Rochester, NY) for 4 days. Autoradiographic images were analyzed using NIH Image (developed by Wayne Rasband, NIMH; available on the Internet at http://rsb.info.nih.gov/nih-image/) to determine the relative amount of ligand binding using a calibration curve constructed in terms of microcuries per gram based on the ¹⁴C radiostandards. Data were expressed as percent binding above basal levels. Nonlinear regression analysis was used to calculate maximal efficacy and log EC₅₀ values from sigmoidal dose-response binding curves using GraphPad Prism (GraphPad Software, San Diego, CA). Maximal efficacy and log EC₅₀ values were compared using a one-way analysis of variance. Log EC₅₀ data are expressed in terms of the EC₅₀ and its 95% confidence interval for each treatment group (the confidence interval is symmetrical on a log scale, but asymmetrical when converted to EC₅₀).

Experiment II

Animals and Drug Treatment. Male Sprague-Dawley rats (Charles River Laboratories) weighing 250 to 300 g were habituated to handling and the behavioral test chambers for 2 days. Animals were treated once daily for 28 consecutive days with the same dose of quinpirole (0.0, 0.05, or 0.1 mg/kg). Treatment groups were normalized according to the mean acoustic startle response observed during baseline testing. The effect of chronic drug exposure on PPI was assessed on days 1, 7, 14, 21, 25, and 28. Percent PPI data calculated for each prepulse level and mean pulse response data on the first and last day of testing were analyzed by separate ANOVAs with drug treatment as a between-subject factor. Within each treatment group, PPI data were analyzed by ANOVA with repeated measures. Post hoc comparisons were conducted using Dunnett's test.

Western Blot Analysis of G_iαand G_sαProtein. Following PPI testing on day 28, animals were decapitated, and their brains were removed and frozen at –30°C in 2-methylbutane. A subset of brains from each treatment group were sectioned using a –20°C cryostat to a level corresponding to 1.7 mm anterior to bregma; the remaining brains were utilized for [³⁵S]GTPγS binding analysis (described below). A unilateral 2-mm wide micropunch (1-mm deep) of the NAc was obtained and homogenized in ice-cold homogenization buffer (50 mM Tris, pH 7.4, 1 mM dithiothreitol, 1 mM EGTA, 10 μg/ml leupeptin, and 20 μg/ml aprotinin). Homogenized tissue was centrifuged at 10,000g, and the resulting pellet was resuspended in 100 μl of homogenization buffer. Protein content in tissue homogenates was calculated using a protein assay kit (Bio-Rad, Hercules, CA), and the samples were stored at –80°C before analysis.

Aliquots of tissue homogenates were separated by SDS-polyacrylamide gel electrophoresis using 12% polyacrylamide gels. Seven and 14 μg of total protein were loaded per lane for G_iα and G_sα, respectively. All experimental samples were analyzed in duplicate. Recombinant G_iα or G_sα protein (Santa Cruz Biotechnology, Santa Cruz, CA) was used as calibration standards on each gel at concentrations of 50, 100, and 200 ng/well for G_iα and 40, 80, and 160 ng/well for G_sα. The recombinant G_iα protein utilized migrates with a molecular weight of 42 kDa, approximately 1 kDa larger than endogenous G_iα due to the addition of an epitope tag.

After gel electrophoresis, proteins were transferred to membranes (Immobilon-P; Millipore, Bedford, MA) then incubated overnight at 4°C in 5% blocking solution (5% dry milk, 1× TBS/0.05% Tween). Membranes were incubated for 1 h at room temperature in antisera specific for either the carboxy terminus of G_iα1–3 of rat origin (1:500 dilution, C-10; Santa Cruz Biotechnology, Inc.) or the carboxy terminus of G_sα of rat origin (1:400 dilution, C-18; Santa Cruz Biotechnology, Inc.). Blots were incubated for 1 h at room temperature in a 1:10,000 dilution of horseradish peroxidase-conjugated donkey anti-rabbit IgG (Pierce Biotechnology, Rockford, IL) in 1% blocking solution and washed twice in 1× TBS/0.05% Tween and once in 1× TBS. Immunolabeled bands were detected using chemiluminescence (ECL Plus; Amersham Biosciences, Piscataway, NJ). Quantification software (Quantity One; Bio-Rad Laboratories) was utilized to generate a standard curve of density × pixel area versus nanograms of recombinant G_α protein. These standard curves were used to calculate the relative immunoreactivity of bands representing G_iα with a molecular weight of 40 to 41 kDa or G_sα with a molecular weight of 42 and 47 kDa (Self et al., 1994). Data were compared using a one-way ANOVA.

[³⁵S]GTPγS Binding Analysis. Immediately after PPI testing on day 28, animals were decapitated, and brain tissue was processed as described in Experiment I. Following preincubation, sections were incubated in assay buffer containing 2 mM GDP (ICN) and 50 pM [³⁵S]GTPγS (PerkinElmer Life and Analytical Sciences) in the absence (basal) or presence of 100 μM dopamine (Sigma-Aldrich) for 1 h at 25°C. Only one concentration of dopamine (100 μM), which elicits maximal [³⁵S]GTPγS binding (He et al., 2000), was utilized in this assay because the results of Experiment I revealed that repeated quinpirole treatment did not affect maximal efficacy or EC₅₀ values. After sections were washed, dried, and exposed to X-ray film, the relative amount of ligand binding was determined using a calibration curve based on ¹⁴C radiostandards coexposed to film. Data were expressed as percent binding above basal. The effect of chronic quinpirole administration on basal and dopamine-stimulated [³⁵S]GTPγS binding was assessed using one-way ANOVA.

Results

PPI Adaptation after Repeated Quinpirole Administration. In Experiment I, acute quinpirole treatment significantly reduced PPI at all doses examined. Increasing quinpirole doses reduced PPI by 30, 31, and 79% compared with vehicle treatment following a 76-db prepulse [F(3,93) = 16.7, p ≤ 0.005] (Fig. 1). Repeated administration of the lowest dose (0.05 mg/kg) attenuated the ability of quinpirole to reduce PPI. By day 10, PPI had increased significantly by 34 and 38% compared with day 1 following 76- and 82-db prepulses, respectively (Fig. 1 and Table 1). PPI increased upon repeated treatment with the intermediate dose (0.1 mg/kg) following 76- and 82-db prepulses, but remained significantly lower than vehicle after 73- and 76-db prepulses (Table 1). Repeated treatment with the highest dose (0.3 mg/kg) did not affect PPI disruption following any prepulse level. Neither acute nor repeated quinpirole treatment altered mean startle responses to pulse trials in the low or intermediate dose groups; however, mean startle response was significantly reduced on day 10 in the high dose group (Table 2).

The effect of longer drug exposure on PPI was investigated in the second experiment. Again, PPI was disrupted by acute quinpirole treatment; increasing doses of quinpirole reduced PPI on day 1 of treatment by 34 and 46% compared with vehicle treatment following 76-db prepulses [F(2,65) = 4.2, p ≤ 0.05] (Fig. 2). In fact, significant PPI disruption occurred on day 1 at all prepulse levels except after 73-db prepulses in the low dose treatment group (Table 3). PPI increased gradually with repeated quinpirole treatment (Fig. 2). Both quinpirole doses produced full recovery of normal PPI levels by day 28, when PPI did not differ among treatment groups at any prepulse level [73-db prepulse: F(2,64) = 1.1, p = 0.3; 76-db prepulse: F(2,64) = 0.5, p = 0.6; 82-db prepulse: F(2,64) = 2.4, p = 0.1] and was significantly higher than day 1 after both quinpirole doses following 76- and 82-db prepulses (Table 3). Neither acute nor 28 days of repeated quinpirole treatment affected mean startle responses to pulse trials (Table 4).

Effect of Repeated Quinpirole Administration on G Protein Level and Function in the NAc. There was no significant effect of 10-day quinpirole treatment on the maximum efficacy of dopamine-mediated [³⁵S]GTPγS binding nor on EC₅₀ values in either the NAc core or shell (Fig. 3 and Table 5). For example, maximal dopamine efficacy in the NAc core after 0.05 and 0.1 mg/kg quinpirole treatment was 93 and 81% of vehicle group values, while EC₅₀ values decreased by 12% after the low dose and increased by 13% after the intermediate quinpirole dose.

Similarly, 28-day quinpirole exposure had no effect on dopamine-mediated [³⁵S]GTPγS binding in the NAc core or shell (Fig. 4). Binding increased slightly in both regions after quinpirole treatment compared with vehicle treatment, but there was no significant effect of treatment on either basal (Fig. 4, inset) or dopamine-stimulated [³⁵S]GTPγS binding.

The effect of 28-day quinpirole treatment on NAc G proteins was assessed further by measuring levels of G protein immunoreactivity (Fig. 5). There were no significant differences between treatment groups in the amount of G_iα protein, nor were treatment effects detected in the levels of G_sα protein (Table 6).

Discussion

Attenuation of D₂-Like Agonist-Induced PPI Disruption. Acute quinpirole treatment disrupted PPI in agreement with prior studies (Peng et al., 1990; Caine et al., 1995; Wan et al., 1996), although the doses effective herein were lower than in most previous reports. This could be due to methodological differences, such as the use of slightly different prepulse levels or different drug sensitivity across rat strains and suppliers. For example, similar doses of quinpirole (0.03 and 0.1 mg/kg) fail to disrupt PPI in male Sprague-Dawley rats purchased from Harlan (Indianapolis, IN) (Peng et al., 1990; Caine et al., 1995; Wan et al., 1996), but quinpirole significantly disrupts PPI in male Wistar rats at a dose of 0.01 mg/kg (Varty and Higgins, 1998). Analogous strain and supplier-based differences in sensitivity to PPI-disruptive effects of other dopamine agonists have been documented previously (Swerdlow et al., 2000).

After 10 days of repeated treatment, quinpirole-mediated PPI disruption was significantly attenuated by 0.05 mg/kg treatment, but not at higher doses. Higher doses of quinpirole have been reported to significantly disrupt PPI in addition to reducing mean startle responses to pulse trials (Peng et al., 1990), as observed herein following repeated treatment with 0.3 mg/kg quinpirole. Extending the treatment period to 28 days, however, significantly increased PPI following both 0.05 and 0.1 mg/kg quinpirole doses. The time course of behavioral attenuation following repeated quinpirole treatment differed slightly between experiments. This might be the result of substrain differences, as these animals were obtained from different locations operated by the same supplier.

This reversal of PPI disruption in response to repeated exposure to a selective D₂-like receptor agonist suggests that tolerance occurred to the effect of quinpirole. These findings are consistent with earlier studies demonstrating the development of tolerance to the disruptive effects of cocaine, amphetamine, or apomorphine upon repeated treatment (Druhan et al., 1998; Byrnes and Hammer, 2000; Feifel et al., 2002). In fact, such PPI tolerance coexists with cocaine-induced motor sensitization (Byrnes and Hammer, 2000), which is thought to involve D₁-like receptor supersensitivity (Henry et al., 1998). However, PPI tolerance was blocked by concurrent haloperidol administration (Feifel et al., 2002), suggesting that D₂-like receptor stimulation is critical. Furthermore, the effect of repeated quinpirole treatment has similar effects on locomotor activity; acute quinpirole dose dependently reduces locomotor activity, but such inhibition is attenuated upon repeated administration (Rowlett et al., 1995). Thus, behavioral tolerance upon repeated treatment with dopamine agonists might be due to a selective action at D₂-like receptors. It is possible that hepatic metabolism of quinpirole is induced by repeated treatment, perhaps leading to reduced brain exposure to the drug. No current data exist to address this possibility, which is unlikely because tolerance to PPI disruption also occurs upon repeated treatment with dopamine agonists from other chemical classes (e.g., cocaine, amphetamine, and apomorphine; Druhan et al., 1998; Byrnes and Hammer, 2000; Feifel et al., 2002).

Receptor G Protein Function Associated with PPI Tolerance. Repeated treatment with a higher quinpirole dose (i.e., 1 mg/kg) is known to down-regulate D₂-like receptor density in the NAc (Subramaniam et al., 1992) and increase striatal dopamine synthesis and turnover (Rowlett et al., 1995; Sullivan et al., 1998; Koeltzow et al., 2003). Furthermore, repeated cocaine treatment reduces levels of G_iα and G_oα protein in the NAc (Nestler et al., 1990). Such alteration of receptors or G proteins mediating the actions of quinpirole could produce PPI tolerance. In fact, quinpirole-mediated PPI disruption was reduced when NAc G_i/G_o proteins were functionally inactivated by pertussis toxin (Culm et al., 2003). Therefore, we evaluated dopamine-stimulated receptor G protein function as well as G protein levels following repeated quinpirole treatment.

The functional coupling between receptors and G proteins in the NAc was investigated following repeated quinpirole treatment using [³⁵S]GTPγS binding analysis, which has been used to detect changes in receptor-effector coupling after chronic drug exposure. For example, chronic haloperidol administration increases D₂ agonist-mediated [³⁵S]GTPγS binding to striatal membranes (Geurts et al., 1999). This assay can be used in rat brain sections to examine regional specificity of drug effects (He et al., 2000) and a low concentration of Mg²⁺ in the assay buffer, which preferentially excludes G_s proteins (Waeber and Moskowitz, 1997), permits selective visualization of D₂-like receptor activity by stimulation with dopamine, as confirmed previously (Culm et al., 2003).

Utilizing a range of dopamine concentrations, we demonstrated that 10-day quinpirole exposure did not alter receptor G protein coupling, having no significant effect on the maximum efficacy of dopamine nor on EC₅₀ values (Table 5). Increasing dopamine concentrations produced a gradual increase in [³⁵S]GTPγS binding in all treatment groups, and dopamine-stimulated [³⁵S]GTPγS binding levels approached a plateau at 100 μM, as observed previously by He et al. (2000). However, dopamine in solution can undergo autooxidation resulting in the production of reactive quinone derivatives (Slivka and Cohen, 1985), which could interfere with [³⁵S]GTPγS binding. If so, then the absolute values obtained for maximum efficacy and EC₅₀ in Table 5 might be artifactually reduced in all treatment groups. Nevertheless, the resulting sigmoidal curves were indistinguishable, indicating that receptor G protein coupling was not significantly altered after repeated quinpirole administration.

In an additional experiment, 100 μM dopamine-stimulated [³⁵S]GTPγS binding was assessed after a 28-day quinpirole treatment period. Basal levels of [³⁵S]GTPγS binding were slightly higher than in the assay of tissue from animals treated with quinpirole for only 10 days, which might have contributed to the relatively lower dopamine-stimulated [³⁵S]GTPγS percent binding above basal values in all groups of this experiment. Nevertheless, no alteration of [³⁵S]GTPγS binding was observed after 28 days of quinpirole treatment, as neither basal nor dopamine-stimulated [³⁵S]GTPγS binding differed from the control group (Fig. 4).

NAc G proteins were further investigated after a 28-day quinpirole treatment using quantitative Western blot analysis. Although repeated treatment with an indirect dopamine agonist reduced NAc G_iα and G_oα levels (Nestler et al., 1990), the relatively low dose and selectivity of the D₂-like dopamine agonist utilized herein had no effect on NAc G_iα or G_sα protein levels (Table 6). Thus, repeated treatment had no effect on either G protein level or D₂-like receptor G protein efficacy, so it is unlikely that repeated quinpirole administration altered NAc D₂ receptor level.

Alterations in Dopaminergic Signaling. These data indicate that repeated stimulation of D₂-like receptors using quinpirole produces behavioral tolerance to sensorimotor-gating deficits in the absence of receptor or G protein changes, suggesting that the locus of dopaminergic adaptation might be at the intracellular level. In support of this hypothesis, it was recently reported that extracellular dopamine release in the NAc is similarly reduced by both acute and repeated quinpirole treatment, suggesting that repeated quinpirole does not alter presynaptic D₂ autoreceptor sensitivity and implicating postsynaptic neuroadaptations in the resulting behavioral changes (Koeltzow et al., 2003).

Continuous exposure to G_i/G_o protein-coupled receptor agonists leads to supersensitivity of the cAMP signaling cascade (Thomas and Hoffman, 1987), also known as heterologous sensitization of adenylate cyclase. For example, upregulation of the cAMP signaling pathway develops after chronic administration of opiates and cocaine (Nestler and Aghajanian, 1997; Self et al., 1998), and chronic morphine treatment increases the expression and activity of cAMP-dependent protein kinase A (PKA) in the locus coeruleus (Lane-Ladd et al., 1997). Alterations in the level of G_iα or G_sα are not required for heterologous sensitization (Palmer et al., 1997; Watts et al., 1999), so the investigation of downstream targets in the cAMP pathway might provide additional insight into the mechanisms underlying the development of behavioral tolerance following repeated quinpirole exposure. Acute quinpirole treatment reduces cAMP accumulation (Johnston et al., 2002), as well as forskolin-stimulated dopamine and cyclic adenosine 3′,5′-monophosphate-regulated phosphoprotein phosphorylation (Nishi et al., 1997). Therefore, PPI disruption induced by acute quinpirole treatment could be related to changes in adenylate cyclase or PKA function, and the reversal of PPI disruption upon repeated treatment might be due to compensatory changes in the same enzymes. In fact, antipsychotic treatments that reduce PPI disruption (Swerdlow and Geyer, 1998) also alter PKA expression and function (Dwivedi et al., 2002).

In the present study, long term D₂-like agonist treatment was associated with the reversal of sensorimotor gating deficits, an adaptation that would be advantageous in reducing symptoms of schizophrenia. The resulting functional change in the endogenous dopamine system might have therapeutic potential for the treatment of schizophrenia. For example, the dopamine partial agonist, aripiprazole, appears to be an effective treatment that reduces both positive and negative symptoms in schizophrenia (Kane et al., 2002). Thus, full or partial dopaminergic agonists might produce benefits by attenuating D₂-like receptor signaling without altering receptor G protein function.