Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The neurotransmitter dopamine (DA) has been implicated in a variety of physiological functions and dysfunctions of dopaminergic systems are involved in several disorders, including Parkinson's disease and schizophrenia (Carlsson, 1987; Zigmond et al., 1990). The effects of DA are mediated by two subfamilies of receptors: the D1-like receptor subtypes (D1, D5) and the D2-like subtypes (D2, D3, D4) (Sibley and Monsma, 1992). Within the D2-like receptor subfamily, the more recently cloned D3 and D4 receptors exhibit restricted patterns of expression relative to the D2 receptor, which appears to be present in most dopaminoceptive areas. The highest levels of D3 receptors are present in the nucleus accumbens (NA), the islands of Calleja, and paleocerebellum with lower levels found in the ventral pallidum (VP) and substantia nigra (SN) (Levesque et al., 1992; Diaz et al., 1995). D3 mRNA is also present in each of these regions (Bouthenet et al., 1991; Diaz et al., 1995).

The physiological function(s) of the D3 receptor remains controversial, but several findings suggest that it may play an important role in neuropsychiatric and neurodegenerative disorders. For example, the density of D3 receptors is significantly elevated in the ventral striatum of cocaine overdose victims, suggesting that it plays a role in cocaine abuse (Staley and Mash, 1996). D3 receptors have also been reported to be increased in schizophrenic subjects (Gurevich et al., 1997), and decreased in Parkinson's disease patients (Ryoo et al., 1998).

"D3 receptor-preferring" compounds produce changes in DA release and turnover, DA cell firing, and modulate locomotor behavior and cocaine self-administration (Levesque, 1996; Levant, 1997). The selectivity of these compounds in vivo has not been established (Burris et al., 1995), making the interpretation of these results difficult. More recently, Pilla et al. (1999) have reported that a D3 partial agonist, which exhibits in vivo selectivity, inhibits cue-controlled cocaine-seeking behavior and has no reinforcing effects. Other studies have allowed inferences to be drawn regarding potential functions of D3 receptors without relying on putative "D3-preferring" drugs. For example, transgenic mice lacking D3 receptors exhibit spontaneous hyperactivity (Accili et al., 1996). Additionally, neonatal lesion of the ventral hippocampus in rats decreases the expression of D3 receptors in the NA and produces increases in locomotor activity (Flores et al., 1996). Prenatal exposure to stress also decreases D3 receptor expression and facilitates the development of locomotor sensitization to amphetamine (Henry et al., 1995). Furthermore, the appearance of the hypolocomotor response to low doses of the D2-like agonist quinpirole (Franz et al., 1996) is correlated with the developmental expression of D3 receptors (Stanwood et al., 1997). Finally, recombinant D3 receptors transfected into MN9D cells are capable of regulating DA synthesis and release (Tang et al., 1994; O'Hara et al., 1996).

The cellular and behavioral functions of D3 receptors will derive, at least in part, from their regional and cellular localizations. The goal of this study was to capitalize on the sensitivity of [¹²⁵I]7-OH-PIPAT to detect the expression of low levels of D3 receptors in additional brain regions and to examine the localization of this receptor using neurochemical lesion techniques. We have demonstrated that the distribution of D3 receptors in rat brain is more widespread than previously appreciated and provided evidence that D3 receptors are located on cell bodies and terminals of dopaminergic and nondopaminergic neurons.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. [¹²⁵I]R-(+)-trans-7-Hydroxy-2-(N-n-propyl-N-3'-iodo-2'-propenyl)aminotetralin ([¹²⁵I]7-OH-PIPAT), [¹²⁵I]NCQ 298, and 4-(2'-methoxyphenyl)-1-[2'-(n-2"-pyridinyl)-p-iodobenzamido]-ethyl-piperazine ([¹²⁵I]p-MPPI) were synthesized as previously described (Burris et al., 1994; Kung et al., 1994, 1995). [³H]SCH 23390, [³H]WIN 35428, and [³H]paroxetine were purchased from DuPont NEN (Boston, MA). [³H]Spiperone was purchased from Amersham (Arlington Heights, IL). (±)-7-OH-DPAT, quinpirole, 1,3-di(2-tolyl)guanidine, quinolinic acid, (±)-8-OH-DPAT, and (+)-butaclamol were obtained from Research Biochemicals International (Natick, MA). 6-Hydroxydopamine (6-OHDA) was obtained from Aldrich (Milwaukee, WI). Fluoxetine was obtained as a gift from Eli Lilly (Indianapolis, IN). All other reagents were purchased from Sigma (St. Louis, MO).

Animals and Tissue Preparation. Seventy-four male Sprague-Dawley rats (Charles River, Wilmington, MA) were used for these studies. Tissue for basic pharmacological studies was obtained from rats weighing 250 to 300 g at time of sacrifice. Brains were rapidly removed after decapitation, immediately frozen in 20°C isopentane, and stored at 70°C. Brains were sectioned (20 µm) on a cryostat, thaw-mounted onto gelatin-coated slides, dessicated under vacuum at 4°C for 3 h, and stored at 70°C.

Lesions. Ascending dopaminergic pathways were destroyed by unilateral injection of 6-OHDA into the medial forebrain bundle. Rats (250-300 g) were pretreated with desmethylimipramine (25 mg/kg i.p.) to protect noradrenergic neurons, anesthetized with equithesin (35 mg/kg pentobarbital, 150 mg/kg chloral hydrate), and given unilateral stereotaxic injections of 6-OHDA HBr (8 µg/4 µl 0.9% NaCl, 0.2% ascorbic acid vehicle) into the medial forebrain bundle (AP, 4.7; ML, 1.0; DV, 7.5 from dura). Animals were sacrificed either 7 or 28 days after lesion. Cell bodies of neurons in the NA and their associated projections were destroyed by injection of quinolinate into this nucleus. Animals (275-315 g) were anesthetized and received unilateral injections of quinolinate (150 nmol/0.5 µl, pH 7.5) or vehicle into the NA (AP, 1.5; ML, +1.5; DV, 6.8 from dura) and were sacrificed 8 days later. Serotonergic cells and fibers were destroyed by intraventricular injection of 5,7-dihydroxytryptamine (5,7-DHT). Animals (140-160 g) were pretreated with desmethylimipramine (25 mg/kg i.p) and anesthetized with a ketamine (50 mg/kg)/xylazine (4 mg/kg) cocktail (i.m.). 5,7-DHT (100 µg/10 µl 0.9% NaCl, 0.1% ascorbic acid per side) or vehicle was injected bilaterally into the lateral ventricles and rats were sacrificed 12 days later. All procedures were approved by the University of Pennsylvania Animal Care and Use Committee (Assurance no. 3079-01).

Autoradiographic Procedures. Before incubation with radioligand, tissue sections were thawed and preincubated for 30 min at 30°C in incubation buffer containing 50 mM Tris (pH 7.4), 40 mM NaCl, and 300 µM GTP to promote dissociation of endogenous DA from the receptors. D3 receptors were labeled in adjacent sections with [¹²⁵I]7-OH-PIPAT (0.2 nM) in incubation buffer containing 5 µM 1,3-di(2-tolyl)guanidine to prevent labeling of sigma sites. Incubation time, wash time, and ligand concentration were varied in initial studies to characterize and optimize the binding conditions. In these initial studies as well as in pharmacological displacement experiments, rat forebrain sections were wiped from slides and counted by a gamma counter. All other experiments were analyzed autoradiographically. Incubations with [¹²⁵I]7-OH-PIPAT were conducted at pH 7.0 because this greatly decreased nonspecific binding (NSB) of the ligand. In saturation studies, sections were labeled with 0.04 to 1.2 nM [¹²⁵I]7-OH-PIPAT. NSB was defined with 5 µM (±)-7-OH-DPAT. Sections were incubated for 90 min at room temperature and rinsed at 4°C for 60 to 90 min. For comparison, D3 receptors were also labeled with [³H]7-OH-DPAT (0.25-5.0 nM) in the same buffer and analyzed by the method of Scatchard. NSB was defined with 5 µM 7-OH-PIPAT. All subsequent binding assays were conducted at pH 7.4. D2 receptors were labeled with [¹²⁵I]NCQ 298 (0.05 nM) in incubation buffer containing 30 nM 7-OH-PIPAT to prevent labeling of D3 receptors. Incubations were carried out for 2 h at room temperature and NSB was defined with 2 µM (+)-butaclamol. D1 receptors were labeled with [³H]SCH 23390 (4 nM) in buffer containing 50 mM Tris (pH 7.4), 154 mM NaCl, 2 mM EDTA, 10 mM MgSO₄ and 10 mg/l BSA. NSB was defined with 2 µM (+)-butaclamol. DA uptake sites were visualized using [³H]WIN 35428 (3 nM) in buffer containing 100 mM NaHCO₃ and 30 mM NaH₂PO₄ (pH 8.0). NSB was defined using 10 µM benztropine. 5-HT_1A receptors were labeled with [¹²⁵I]p-MPPI (0.2 nM) in 50 mM Tris (pH 7.4) and NSB was defined with 10 µM 8-OH-DPAT. 5-HT uptake sites were labeled with [³H]paroxetine (0.4 nM) in 50 mM Tris (pH 7.4), 120 mM NaCl, and 5 mM KCl and NSB was defined with 10 µM fluoxetine. After all incubations, slides were rinsed in buffer at 4°C for 1 to 120 min, dipped in ice-cold double distilled H₂O and dried with a stream of warm air.

Densitometry. Labeled sections were apposed to LKB Ultrofilm or Amersham Hyperfilm-³H in X-ray cassettes and exposed at room temperature for 4 h to 8 weeks, depending on the radioligand and the receptor density. A plastic tritium standard calibrated with tissue sections containing either ³H or ¹²⁵I was included in each cassette as previously described (Artymyshyn et al., 1990). Films were developed in Kodak GBX developer (3 min), rinsed in water (20 s), fixed in Kodak GBX fixer (6 min), and rinsed in cool water for 15 to 20 min. The autoradiograms were analyzed using a Macintosh-based image processing system using NIH Image 1.47 software. Optical density was converted to femtomoles per milligram of protein based on calibration curves generated from the tritium-containing standards. Adjacent sections stained with cresyl violet were used to identify anatomical structures.

Data Analysis. Kinetic rate constants of association and dissociation were determined by pseudo first-order transformation of association and dissociation binding data using unweighted linear regression analysis. Competition curves were analyzed by nonlinear regression for a one-site fit using an iterative curve-fitting program. IC₅₀ values were transformed to K_i values using the method of Cheng and Prusoff. Maximum receptor density and K_d values were determined by Scatchard transformation of saturation-specific binding data using unweighted linear regression analysis. B_max values based on additional single-point binding assays were estimated using the equation B_max = B(L + K_d)/L.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Characterization of [¹²⁵I]7-OH-PIPAT Binding to D3 Receptors in Brain Sections. In general, [¹²⁵I]7-OH-PIPAT bound to rat brain sections with high affinity and displayed a low level of nonspecific binding. The time course of association was relatively rapid and monophasic (Fig. 1A) and yielded a linear pseudo first-order rate plot (Fig. 1A, inset). Specific binding reached equilibrium within 90 min. The rate constant of association (k₊₁) determined from the pseudo first-order rate plot was 3.6 × 10⁷ M¹ min¹. The kinetics of dissociation was evaluated by allowing [¹²⁵I]7-OH-PIPAT to equilibrate with sections taken from the NA for 90 min and then transferring sections to containers containing buffer without [¹²⁵I]7-OH-PIPAT (Fig. 1B). A first-order plot of the data was linear, indicating that dissociation was monophasic (Fig. 1B, inset). The rate constant of dissociation (k₁) calculated from the first-order plot of these data was 0.0071 min¹. The affinity, or K_d, for [¹²⁵I]7-OH-PIPAT binding determined from the ratio of the kinetic rate constants (k₁/k₊₁) was 0.20 nM. These results are consistent with a simple, reversible bimolecular interaction between [¹²⁵I]7-OH-PIPAT and D3 receptors in rat brain sections. The pharmacological identity of the sites labeled with [¹²⁵I]7-OH-PIPAT was confirmed by measuring the inhibition of the binding of [¹²⁵I]7-OH-PIPAT by several competing ligands (Fig. 1C). Competition assays demonstrated a rank order of potency [7-OH-DPAT (K_i = 1.1 nM) > quinpirole (K_i = 5.2 nM) > domperidone (K_i = 13.1 nM) > dopamine (K_i = 25.3 nM), > clozapine (K_i = 191 nM)] consistent with labeling of the D3 receptor (Fig. 1C; Table 1).

View larger version (15K): [in this window] [in a new window]   Fig. 1.   A and B, association (A) and dissociation (B) plots of [125I]7-OH-PIPAT (0.20 nM) binding to D3 receptors in the NA. Dissociation was initiated after a 90-min incubation at room temperature. The linear first-order rate plots are shown as insets for each graph. C, inhibition of the binding of [125I]7-OH-PIPAT to D3 receptors by prototypical agonists in the NA. Each point represents the mean ± S.E.M. of five to six determinations.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Saturation isotherms of the binding of [125I]7-OH-PIPAT to D3 receptors in the NA (A), SN (B), and median raphe nucleus (C). The linear Scatchard transformations are shown as insets for each graph. Despite markedly different levels of expression, the Kd for [125I]7-OH-PIPAT is consistent across regions. Each point represents the mean and S.E.M. of four to six determinations.

Distribution of D3 Receptors. Autoradiograms of coronal rat brain sections at various levels labeled with [¹²⁵I]7-OH-PIPAT show the characteristic distribution of D3 receptors (Fig. 3). High levels of binding are generally restricted to the shell of the nucleus accumbens (Fig. 3B) and lobules 9 and 10 of the cerebellum (Fig. 3F). Labeling is also seen in the islands of Calleja (Fig. 3, A-C), ventral caudate-putamen (Cpu) (Fig. 3B), ventral pallidum (Fig. 3C), and substantia nigra (Fig. 3D). Interestingly, a low level of D3 receptor binding is present in the median and dorsal raphe nuclei of the midbrain (Fig. 3E). D3 receptor binding could not be detected in cortical regions except for a very low density of sites within the parietal region (Fig. 3C). Only the very lateral portion of the ventral tegmental area (bordering the SN) showed D3 receptor labeling (Fig. 3D). D2 receptors were labeled in adjacent sections with [¹²⁵I]NCQ 298 and their distribution and densities were consistent with previous reports (data not shown; Bouthenet et al., 1991; Boyson et al., 1986). In general, D3 receptors are expressed at lower densities and exhibit a more restricted pattern of expression than D2 receptors.

View larger version (161K): [in this window] [in a new window]   Fig. 3.   Autoradiograms of rat brain sections labeled with [125I]7-OH-PIPAT (0.20 nM) at the level of the NA (A and B), VP (C), SN (D), raphe nuclei (E), and cerebellum (F). Nap, rostral pole of nucleus accumbens; Nas, shell of nucleus accumbens; Nac, core of nucleus accumbens; ICj, islands of Calleja; Atg, anterior tegmental nucleus, DR, dorsal raphe nucleus; MnR, median raphe nucleus; 9, lobule 9 of cerebellum; 10, lobule 10 of cerebellum. The regions of the nucleus accumbens, substantia nigra, and median raphe that were measured for the Scatchard analysis in Fig. 2 are encircled with a solid line.

Effect of 6-OHDA Lesion. Ascending dopaminergic pathways were selectively destroyed by administration of the neurotoxin 6-OHDA into the medial forebrain bundle. The extent of 6-OHDA lesion was evaluated by measuring the loss of DA uptake sites with [³H]WIN 35428 (Fig. 4). [³H]WIN 35428 binding was decreased by greater than 90% in both the NA and CPu by 28 days after 6-OHDA lesion. In contrast, a significant increase (+26%) in [¹²⁵I]NCQ 298 binding to D2 receptors in the NA was detected 28 days after 6-OHDA lesion (Fig. 5A). A similar increase in [¹²⁵I]NCQ 298 binding to D2 receptors was observed in the CPu and was also observed with the nonselective D2-like ligand [³H]spiperone (data not shown). D2 receptor binding was greatly reduced in the SN at 7 (51%) and 28 (69%) days postlesion, which is consistent with the loss of dopaminergic cells in these regions (Fig. 5A). [¹²⁵I]7-OH-PIPAT binding to D3 receptors was significantly decreased in both the NA and SN after 6-OHDA lesion (Fig. 5B). The magnitude of the decrease in D3 expression was similar at both time points in the SN (45% and 51%, at 7 and 28 days postlesion, respectively), but was somewhat less pronounced at the earlier time point in the NA (32 and 51%, respectively). D3 expression was unchanged in the VP after 6-OHDA lesion (Fig. 5B).

View larger version (29K): [in this window] [in a new window]   Fig. 4.   Effect of 6-OHDA lesion on binding of [3H]WIN 35428 to DA uptake sites. Administration of 6-OHDA into the medial forebrain bundle produced a significant decrease in the number of uptake sites in the CPu and NA indicative of a large loss of DA terminals. Data expressed as percentage of control ± S.E.M. Control data were grouped across the 7- and 28-day time points for presentation because there was no difference between them. For the purposes of statistical analysis these values were kept separate. Each bar represents the mean and S.E.M. from seven animals. *P < .05, **P < .01 by paired t test.

View larger version (54K): [in this window] [in a new window]   Fig. 5.   Effect of 6-OHDA lesion on binding of [125I]NCQ 298 to D2 receptors (A) and [125I]7-OH-PIPAT to D3 receptors (B). Control values for [125I]NCQ 298 binding are 45.7 ± 2.57, 12.1 ± 0.76, and 30.7 ± 2.14 fmol/mg in the NA, VP, and SN, respectively. Control values for [125I]7-OH-PIPAT binding are 27.1 ± 2.26, 21.4 ± 1.54, and 8.17 ± 0.53 fmol/mg in the NA, VP, and SN, respectively. Each bar represents the mean and S.E.M. from seven animals. *P < .05, **P < .01 by paired t test.

Effect of Quinolinic Acid Lesion. Intrinsic neurons of the NA and their associated projections were destroyed by administration of quinolinic acid. The success of this lesion was validated by measuring D1 receptor expression with [³H]SCH 23390 (Fig. 6). [³H]SCH 23390 binding was greatly decreased in the NA and SN compared with control. The lesion also spread into the CPu. This pattern of loss has been attributed to a loss of projection neurons within the NA and Cpu, which express D1 receptors on their cell bodies and terminals (Altar and Hauser, 1987; Filloux et al., 1991). In contrast, no change was observed for [¹²⁵I]NCQ 298 (Fig. 7A) or [³H]spiperone (data not shown) binding to D2-like receptor subtypes. Specific D3 receptor expression measured with [¹²⁵I]7-OH-PIPAT was extensively decreased in both the NA (50%) and SN (55%) after this lesion (Fig. 7B). [¹²⁵I]7-OH-PIPAT binding was also significantly decreased (51%) in the VP after quinolinate lesion (Fig. 7B).

View larger version (40K): [in this window] [in a new window]   Fig. 6.   Effect of quinolinate lesion of the NA on binding of [3H]SCH 23390 to D1 receptors. Data expressed as percentage of control ± S.E.M. Each bar represents the mean and S.E.M. from six animals. **P < .01 by unpaired t test.

View larger version (68K): [in this window] [in a new window]   Fig. 7.   Effect of quinolinate lesion on binding of [125I]NCQ 298 to D2 receptors (A) and [125I]7-OH-PIPAT to D3 receptors (B). Control values for [125I]NCQ 298 binding are 50.2 ± 3.78, 13.1 ± 0.75, and 32.0 ± 1.57 fmol/mg in the NA, VP, and SN, respectively. Control values for [125I]7-OH-PIPAT binding are 35.6 ± 0.93, 22.13 ± 0.82, and 7.10 ± 0.54 fmol/mg in the NA, VP, and SN, respectively. Each bar represents the mean and S.E.M. from six animals. **P < .01 by unpaired t test.

Effect of 5,7-DHT Lesion. Serotonergic neurons were selectively destroyed by intraventricular administration of the neurotoxin 5,7-DHT. The success of this lesion was confirmed by 79 and 68% decreases in [¹²⁵I]p-MPPI binding to 5-HT_1A receptors in the dorsal and median raphe nuclei, respectively (Fig. 8A). [³H]Paroxetine binding to 5-HT uptake sites was also decreased by approximately 90% in the hippocampus and amygdala after 5,7-DHT, suggesting a large loss of serotonergic projections (Fig. 8B). 5,7-DHT lesion produced no changes in D2 or D3 expression in sections taken from the level of the NA, VP, and SN (Table 3). [¹²⁵I]NCQ 298 and [¹²⁵I]7-OH-PIPAT binding was also examined in the dorsal and median raphe nuclei after 5,7-DHT to test whether D2 and/or D3 receptors are expressed on serotonergic cells and no changes in the binding of either radioligand were observed (Table 3).

View larger version (45K): [in this window] [in a new window]   Fig. 8.   Effect of intraventricular 5,7-DHT lesion on binding of [125I]p-MPPI to 5-HT1A receptors in the raphe nuclei (A) and [3H]paroxetine binding to 5-HT uptake sites in terminal regions (B). Each bar represents the mean and S.E.M. from seven animals. **P < .01 by unpaired t test.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Methodological Considerations. We directly compared the distributions of D2 and D3 receptor subtypes using [¹²⁵I]NCQ 298 and [¹²⁵I]7-OH-PIPAT autoradiography. Selectivity of the ligands used is always an important concern in the interpretation of pharmacological studies. In fact, [¹²⁵I]7-OH-PIPAT is capable of labeling both D2 and D3 subtypes (Kung et al., 1994; Burris et al., 1995). This is also true for all other D3 receptor ligands currently available. In the presence of guanine nucleotides, however, a selective labeling of GTP-insensitive D3 receptors is observed. For example, in the absence of guanine nucleotides the affinities of [¹²⁵I]7-OH-PIPAT for D2 and D3 expressing HEK-293 cell membranes are 0.7 and 0.1 nM, respectively, but in the presence of 5'-guanylylimidodiphosphate the D2 receptor K_d is at most 2.3 nM (Burris et al., 1994). The lack of magnesium and inclusion of sodium ions in our labeling buffer further minimizes high-affinity binding to D2 receptors in our studies. [¹²⁵I]NCQ 298 is also capable of labeling both D2 and D3 receptors (Filtz et al., 1993; Burris et al., 1995). In our assays, we have prevented [¹²⁵I]NCQ 298 binding to D3 receptors by adding guanine nucleotides and a low concentration of 7-OH-PIPAT (30 nM) to selectively block GTP-insensitive D3 receptors. We cannot, however, rule out a small contribution of D3 receptor binding to the [¹²⁵I]NCQ 298 signal.

Pharmacology and Anatomical Distribution of [¹²⁵I]7-OH-PIPAT Binding to D3 Receptors. Our experiments demonstrate the utility of [¹²⁵I]7-OH-PIPAT as a radioligand for the study of the dopamine D3 receptor. The specificity of [¹²⁵I]7-OH-PIPAT labeling was confirmed by kinetic, saturation, and competition analyses. Scatchard analyses revealed a single high-affinity binding site. K_d values were similar across regions and comparable to those observed previously in rat brain homogenates and transfected cells (Burris et al., 1994; Kung et al., 1994). Association and dissociation pseudo first-order rate plots were linear, consistent with a simple bimolecular interaction with a single receptor protein, and resulted in similar estimates of the dissociation constant. Displacement curves provided additional evidence of the selective binding of [¹²⁵I]7-OH- PIPAT to D3 receptors in these in vitro assays.

Effect of Lesions on D2 and D3 Receptor Expression. 6-OHDA lesion of dopaminergic cells and fibers produced changes in the expression of DA uptake sites and D2-like receptors that are consistent with previous reports (Creese and Snyder, 1979; Joyce, 1991)_. The extensive loss of DA uptake sites in terminal regions is consistent with a highly successful lesion. The loss of D2 receptors in the SN suggests that D2 receptors are expressed on dopaminergic neurons in the SN. The increases in D2 receptors in the NA and CPu after 6-OHDA lesion were also expected based on previous work and suggest that intrinsic neurons of NA and/or nondopaminergic afferents to the NA express D2 receptors, which up-regulate in response to loss of DA innervation. These conclusions are supported by immunohistochemical studies (Levey et al., 1993; Sesack et al., 1994).