GPR37 associates with the dopamine transporter to modulate dopamine uptake and behavioral responses to dopaminergic drugs

Abstract

The orphan G protein-coupled receptor 37 (GPR37) is a substrate of parkin; its insoluble aggregates accumulate in brain samples of Parkinson's disease patients. We report here that GPR37 interacts with the dopamine transporter (DAT) and modulates DAT activity. GPR37 and DAT were found colocalized in mouse striatal presynaptic membranes and in transfected cells and their interaction was confirmed by coimmunoprecipitation assays. Gpr37-null mutant mice showed enhanced DAT-mediated dopamine uptake in striatal membrane samples, with a significant increase in the number of plasma membrane DAT molecules. The null mutant mice also exhibited a decrease in cocaine-induced locomotor activity and in catalepsy induced by dopamine receptor antagonists. These results reveal the specific role of GPR37, a putative peptidergic G protein-coupled receptor, in modulating the functional expression of DAT and the behavioral responses to dopaminergic drugs.

The orphan G protein-coupled receptor 37 (GPR37) is homologous to endothelin (ET_B-R) and bombesin (GRP-R, NMB-R) receptors (1) and it is highly expressed in mammalian brain oligodendrocytes, Purkinje cells, and neurons belonging to the CA3 hippocampal region and to the substantia nigra (SN) pars compacta (2). GPR37 is a substrate of the ubiquitin–protein ligase parkin, and it has been named parkin-associated endothelin-like receptor (PAEL-R) (3). An insoluble form of GPR37 is accumulated in brain samples of Parkinson's disease (PD) patients, and the overexpression of GPR37 in cell cultures, in the absence of parkin, can lead to unfolded protein-induced cell death (3, 4). Little is known about the physiological function of the receptor in the brain and in dopaminergic neurons in particular, although it has been speculated that the aggregation of GPR37 in insoluble complexes is responsible for the preferential loss of SN neurons through the endoplasmic reticulum-specific apoptotic pathway (5, 6). Recent data reported an interaction between GPR37, the head activator neuropeptide and its binding protein (sorting protein-related receptor; SorLA), supporting the hypothesis that GPR37 is involved in neuronal cell survival (7). To investigate the receptor's function, we generated homozygous Gpr37-null mutant mice, which exhibit a reduction in striatal dopamine (DA) content, specific locomotor deficits, and enhanced sensitivity to amphetamine (8).

Several binding partners for the DA transporter (DAT) have been identified, suggesting that a regulated multiprotein complex controls its synthesis, targeting, and expression at specific cellular membrane domains (9). Presynaptic DAT expression is of crucial importance in modulating the synaptic availability of DA at nigrostriatal synapses, and its regulation is dynamically controlled for the maintenance of normal dopaminergic neurotransmission. The increase or decrease of DAT expression in the presynaptic membranes results in decreased or increased synaptic DA concentration, respectively, thus regulating multiple post- and presynaptic signaling pathways mediated by D1-like and D2-like dopamine receptors (10). Although several groups have proposed the physical and functional coupling between DAT and various G protein-coupled receptors, experimental data were reported only for trace amine and DA D2 receptors (11–13). Furthermore, proteins directly involved in PD, such as parkin and α-synuclein, interact with DAT (14–16).

In this study, we show that GPR37 is localized on mouse striatal presynaptic membranes and that it associates with DAT, thus participating in the regulation of DAT expression at the plasma membrane. Moreover, we report specific functional alterations of DAT and striatal DA receptors in Gpr37-null mutant mice, as shown by behavioral analysis after administration of DA receptor antagonists and cocaine. These findings indicate that this putative peptidergic G protein-coupled receptor plays a specific role in the regulation of DAT-mediated nigrostriatal dopaminergic signaling and functional responses to psychostimulant drugs.

Results

GPR37 and DAT Are Associated with the Striatal Presynaptic Plasma Membrane.

Our previous studies (8) suggested a potential role for GPR37 in regulating DA metabolism in mouse SN neurons and in affecting the nigrostriatal dopaminergic signaling pathway. Furthermore previous immunolabeling studies (3) specifically localized the receptor protein in the dopaminergic neurons of the mouse SN pars compacta. To study the membrane localization of GPR37, total membranes, plasma membrane- and synaptic vesicle-enriched fractions were prepared from C57BL/6J adult male mouse striata, as described in Materials and Methods. Fraction samples (Fig. 1A, lanes 1–3) were subjected to immunoblotting using specific antibodies for mouse GPR37, DAT, Na⁺/K⁺ ATPase α1 polypeptide (ATPase α; a plasma membrane marker), and synaptophysin 1 (a synaptic vesicle marker). As shown in Fig. 1A, the GPR37, DAT, and ATPase α1 immunoreactive bands are specifically associated with the striatal plasma membrane fraction, whereas the synaptophysin 1 band is predominant in the synaptic vesicle fraction. In the striatal membranes of wild-type mice, the GPR37-specific monoclonal antibody labels the 50- to 52-kDa band (Fig. 1A), corresponding to the fully denatured, monomeric form of the GPR37 protein (7). No specific GPR37 labeling is found in membrane samples of Gpr37^−/− mice (data not shown).

Fig. 1.

GPR37 is enriched in the mouse striatal presynaptic fraction. (A) Western blot analysis of GPR37 and DAT proteins was performed in whole synaptosome (WS, lane 1), plasmalemmal membrane (PM, lane 2), and synaptic vesicle (SV, lane 3) fractions, from the pooled striata of C57BL/6J mice. An aliquot (60 μg) of each fraction was immunoblotted with antibodies specific for GPR37, DAT, and the fraction markers. Both GPR37 and DAT immunoreactive bands are enriched in the plasmalemmal membrane fraction. (B) Western blot analysis of GPR37 and DAT proteins was performed in synaptic membrane fractions, including whole synaptosomes (lane 1), extrasynaptic (lane 2), presynaptic (lane 3), and postsynaptic (lane 4) membrane-enriched fractions, from the pooled striata of C57BL/6J mice. An aliquot (10 μg) of each fraction was immunoblotted by using antibodies specific for the indicated proteins. GPR37 is enriched in the presynaptic fraction. The monoclonal anti-GPR37 antibody labels different N-glycosylation variants of the receptor protein (3).

Pre- or postsynaptic membranes were then fractionated from the synaptic junctional complexes of C57BL/6J adult male mouse striata. Total synaptosomes, the extrasynaptic (pH 6.0-soluble extract) (17) and the pre- and postsynaptic fractions (pH 8.0-soluble and -insoluble extracts) (17) were then subjected to immunoblotting (Fig. 1B, lanes 1–4) using antibodies specific for mouse GPR37, DAT, synaptosomal protein of 25 kDa (SNAP-25, a presynaptic membrane marker), postsynaptic density 95 protein (PSD-95, a postsynaptic membrane marker), and synaptophysin 1 (an extrasynaptic marker). As shown in Fig. 1B, the GPR37 and SNAP-25 immunoreactive bands are enriched in the presynaptic membrane fraction, whereas the DAT band is similarly associated with both the extra- and presynaptic fraction.

Lack of GPR37 Enhances DAT Activity and Increases Plasma Membrane Expression of DAT.

Because GPR37 is enriched in the presynaptic striatal plasma membrane fractions, we tested its potential role in regulating DA uptake at nigrostriatal synapses in vivo. We prepared synaptosomes from the striata of wild-type and Gpr37^−/− adult male littermates and compared the DAT-mediated uptake of radiolabeled [³H]DA (18 nM) (18), whereas nonspecific uptake was measured in the presence of 10 μM nomifensine, a specific inhibitor of monoamine transporter proteins (14). A significant increase of the DAT-mediated DA uptake was detected in the samples obtained from null mutant mice (976.2 ± 56.9 fmol mg⁻¹ min⁻¹), in comparison with wild-type littermates (805.4 ± 1.8 fmol mg⁻¹ min⁻¹; t₍₄₎ = 3.00; P < 0.05; unpaired t test; Fig. 2A).

Fig. 2.

Gpr37^−/− mice exhibit enhanced striatal [³H]DA uptake. (A) [³H]DA (18 nM) uptake by striatal mouse synaptosomes from wild-type (+/+) and Gpr37^−/− (−/−) mice. Data from quadruplicate samples were corrected for unspecific uptake measured in the presence of 10 μM nomifensine and are presented as the mean ± SEM of three separate experiments. (∗, P < 0.05; unpaired t test). (B) Kinetic analysis of [³H]DA uptake in striatal synaptosomes of wild-type (+/+) and Gpr37^−/− male mice. [³H]DA (18 nM) and increasing concentrations of unlabeled DA (from 0.005 to 1 mM) were applied simultaneously. Nonspecific DA uptake was determined in parallel experiments in the presence of 10 μM nomifensine. Data are expressed as [³H]DA uptake in fmol mg⁻¹min⁻¹ and presented as mean ± SEM of three separate experiments, each performed in duplicate. (C) Saturation binding curve of [N-methyl-³H]WIN-35428 to DAT in total striatal extracts from wild-type (+/+) and Gpr37^−/− (−/−) mice. K_d values were 6.55 ± 0.93 × 10⁻⁹ M (+/+) and 5.18 ± 1.10 × 10⁻⁹ M (−/−), whereas B_max values were 2,144 ± 122.3 fmol/mg (+/+) and 1,761 ± 135.3 fmol/mg (−/−, P < 0.05; F test). Data are presented as mean ± SEM of three separate experiments, each performed in duplicate.

Previous studies have demonstrated that an increase in DAT-mediated DA uptake is caused by an augmentation of the intrinsic affinity of DAT or by an increase in the number of functional transporter molecules expressed at the cell plasma membrane (19, 20). We therefore analyzed the kinetics of DAT-mediated DA uptake, as measured in synaptosomal samples from wild-type and Gpr37^−/− mice at various DA concentrations, according to standard protocols (Fig. 2B and ref. 21). The comparison of the kinetic data showed that the K_m and V_max values for DA uptake in wild-type samples (K_m = 0.93 ± 0.25 × 10⁻⁷ M; V_max = 4,108.0 ± 429.2 fmol mg⁻¹min⁻¹) were in agreement with reported data (22). Instead, the maximum rate of DA uptake was strongly increased in Gpr37^−/− samples (V_max = 6,510.0 ± 241.3 fmol mg⁻¹min⁻¹; F_(1,20) = 22.47, P < 0.001; F test), without a statistically significant variation of the K_m value (K_m = 1.13 ± 0.11 × 10⁻⁷ M). These data show that the lack of GPR37 results in a robust enhancement of DAT-mediated DA uptake, which is not caused by an increase of the transporter affinity for DA. To test whether this effect may be due to a direct increase in the total number of DAT molecules in the striatum, we compared the amount of total DAT-binding sites for [N-methyl-³H]WIN-35428 (a cocaine analog that specifically binds to DAT) (23) in striatal membrane extracts of wild-type and Gpr37^−/− mice. As shown in Fig. 2C, the total number of WIN-35428-binding sites in wild-type samples was similar to the published values (B_max = 2,144.0 ± 122.3 fmol/mg) (24), whereas in the Gpr37^−/− samples, there was a significant decrease (B_max = 1,761.0 ± 135.3 fmol/mg; F_(1,23) = 4.32, P < 0.05; F test) compared with the wild-type samples. Thus, in the absence of the GPR37 protein, the enhancement of DAT-mediated DA uptake at nigrostriatal synapses is not a consequence of an increased total number of DAT molecules, suggesting that GPR37 may inhibit the cell surface expression of active transporter molecules. To examine the hypothesis of a regulation of plasma membrane recruitment of DAT, we performed biotinylation experiments using sulfo-NHS-biotin on striatal synaptosomes of wild-type and Gpr37^−/− mice. Biotinylated proteins were isolated with streptavidin beads and analyzed by Western blot using the DAT antibody. As shown in Fig. 3, the percentage of biotinylated DAT protein at the cell membrane was increased by >30% in Gpr37^−/− samples (wild-type: 39.8 ± 2.9%; Gpr37^−/−: 67.9 ± 4.1%; t₍₆₎ = 5.60; P < 0.01; unpaired t test). These values were calculated upon normalization for the amount of total transporter protein, which was slightly decreased in Gpr37^−/− samples, in agreement with the WIN-35428-binding data (see above). These results indicate an active role of GPR37 in the control of the transporter localization in vivo.

Fig. 3.

Gpr37^−/− mice exhibit increased cell-surface levels of DAT. (A) Cell-surface biotinylation experiments were performed in striatal synaptosomes from wild-type (+/+) and Gpr37^−/− (−/−) mice. Equal aliquots of whole striatal synaptosome samples were incubated with sulfo-NHS-biotin. After sonication, biotinylated proteins were isolated with streptavidin beads. One-tenth of the total lysate fractions was used to detect total immunoreactive DAT, and samples were analyzed by SDS/PAGE, followed by Western blot with the anti-DAT antibody. Results shown are representative of four independent experiments. (B) Quantitation of biotinylated immunoreactive bands expressed as percentage of DAT surface density in wild-type (+/+) and Gpr37^−/− (−/−) striatal synaptosomes. Immunoblots from four separate biotinylation experiments were imaged and quantified, after factoring aliquot volumes and sample handling protocol, mean ± SEM values were plotted (+/+, 39.8 ± 2.9%; −/−, 67.9 ± 4.1%). The Gpr37^−/− samples show a significant increase of biotinylated DAT protein (∗∗, P < 0.01; unpaired t test).

GPR37–DAT Interaction in Transfected HEK-hDAT Cells.

No GPR37-specific antibodies are currently available for immunofluorescence or immunoprecipitation protocols with mouse tissue samples ex vivo (3, 4). We therefore studied the possible interactions between GPR37 and DAT in a HEK293 cell line, which stably expresses the human DA transporter (HEK-hDAT) (25).

We transiently transfected HEK293 or HEK-hDAT cells with a full-length human GPR37 cDNA encoding a fused carboxy-terminal influenza virus hemagglutinin (HA) peptide sequence (GPR37-HA), which does not affect the normal synthesis and intracellular localization of GPR37 (3). Confocal immunofluorescence experiments showed that, when expressed alone, GPR37 and DAT localized quite diffusely throughout the cells, with distributed intracellular and plasma membrane fluorescent immunolabeling [supporting information (SI) Fig. 7], as already reported in studies with HEK293 and other transfected cell lines (3, 7, 14, 15). When GPR37 was expressed in HEK-hDAT cells, both proteins colocalized extensively in discrete intracellular and plasma membrane areas, suggesting the formation of GPR37–DAT complexes (SI Fig. 7).

We assayed the interaction between GPR37 and DAT by reciprocal coimmunoprecipitation in HEK-hDAT cell lysates, after transfection with or without the full-length GPR37-HA cDNA. The immunoprecipitation with a HA-specific antibody, followed by immunoblotting with a DAT antibody, showed that the DAT protein was quantitatively coprecipitated only when the GPR37-HA vector was transfected (Fig. 4). In addition, the DAT antibody coprecipitated the GPR37-HA protein only in lysates obtained from transfected cells (Fig. 4), thus demonstrating that the specific coprecipitation was independent of the order in which the antibodies were used. Control experiments demonstrated that the DAT protein was equally immunoprecipitated by the DAT-specific monoclonal antibody in both transfected and mock-transfected cell samples, whereas the HA-specific antibody precipitated the GPR37-HA protein only in transfected cell lysates (Fig. 4).

Fig. 4.

GPR37 coprecipitates with DAT in transfected HEK293-hDAT cells. HEK293-hDAT cells were transfected with linearized plasmid vector or the GPR37-HA construct. Immunoprecipitations (IP) were performed with anti-HA antibody (Upper) or the anti-DAT monoclonal antibody (Lower), followed by Western blot detection (IB) with anti-DAT or anti-HA, respectively. Total cell lysates were analyzed by Western blotting using specific antibodies to show the expression levels of DAT and GPR37-HA. A representative result from three experiments is shown.

Limited Variation of Striatal DA Receptor Levels in Gpr37^−/− Mice.

We performed saturation radioligand-binding experiments with striatal membranes from wild-type and Gpr37^−/− animals to assess genotype-dependent variations in the levels of other crucial components of the nigrostriatal dopaminergic signaling pathway. When assayed with specific radiolabeled antagonists, total binding levels (B_max values; Table 1) of both D1- and D2-like dopaminergic receptors were slightly increased in the null mutant samples, although these variations were not statistically significant, whereas the D2-like receptor affinity was significantly decreased in the Gpr37^−/− samples (Table 1). Ligand-binding experiments with a specific antagonist of the brain vesicular monoamine transporter (VMAT2) showed no significant, genotype-dependent difference in the value of both the affinity constant and total binding sites (Table 1). These data suggest that, in Gpr37^−/− mice, the enhanced DAT-mediated DA uptake, with consequent reduced levels of striatal DA, may induce a limited, compensatory increase in the number of D1 and D2 receptors and a decreased affinity of D2 receptor for DA antagonists.

Table 1.

Parameters of ligand binding in striatal membrane extracts of wild-type and Gpr37^−/− mice

Protein	Genotype	Bmax, fmol/mg	Kd, M
D1R	Wild type	3,706.0 ± 300.1	7.22 ± 1.34 × 10−10
	Gpr37−/−	4,087.0 ± 427.8	6.81 ± 1.44 × 10−10
D2R	Wild type	424.9 ± 34.0	2.98 ± 0.63 × 10−10
	Gpr37−/−	500.9 ± 86.9	6.94 ± 2.30 × 10−10*
VMAT2	Wild type	1,774.0 ± 149.8	6.72 ± 1.19 × 10−9
	Gpr37−/−	1,500.0 ± 135.4	5.49 ± 1.10 × 10−9

Saturation assays of radioligand binding to DA receptors and VMAT2 vesicular transporter in striatal membrane extracts were performed as described in Materials and Methods. The values are means ± SEM calculated by nonlinear curve fitting of saturation binding data from three to five independent experiments, each performed in duplicate (∗, F_(1,56) = 5.03, P < 0.05, compared with wild type; F test). D1R, dopamine D1 receptor; D2R, dopamine D2 receptor; VMAT2, vesicular monoamine transporter 2.

Gpr37^−/− Mice Show Lower Cocaine-Induced Locomotor Activity.

Given the experimental evidence of the GPR37–DAT interaction, we assessed the possible alterations of a specific DAT-mediated behavioral paradigm after the genetic ablation of GPR37. Therefore, Gpr37^−/− male mice and their wild-type littermates were tested in an open field arena after the acute systemic administration of saline, 10, or 20 mg/kg of cocaine, whose stimulatory effects result from its direct inhibition of DAT-mediated DA reuptake and consequent increase in locomotor activity (26). As shown in Fig. 5, Gpr37^−/− mice displayed a reduced locomotor response to the highest dose of cocaine, compared with their wild-type littermates, in particular during the initial time intervals after the injection (Fig. 5; Tukey's HSD for unequal n after repeated-measures ANOVA; genotype × treatment × time factors; F_(8,16) = 2.01, P < 0.05). A further post hoc analysis of the interaction between the treatment × genotype factors also showed that the increase of locomotor activity was significantly dose-dependent in the wild-type animals but not in the Gpr37^−/− mice (Fig. 5 Inset, Fisher's least-squares difference test). Moreover, null mutant mice treated with the highest dose of cocaine showed (SI Fig. 8) a higher percentage (33%, 6 of 18) of low responders (below the 25th percentile) compared with the wild-type littermates (14%, 2 of 14), and a lower percentage (11%, 2 of 18) of high responders (above the 75th percentile) compared with wild-type (36%, 5 of 14).

Fig. 5.

Cocaine-induced locomotor activity. Gpr37^−/− mice (−/−) showed a lower response to the administration of 20 mg/kg cocaine compared with their wild-type (+/+) littermates (∗∗, P < 0.01; ∗∗∗, P < 0.001; post hoc Tukey's HSD for unequal n). Symbols represent the mean (±SEM) distance traveled every 10 min after the i.p. injection of either saline (+/+, n = 10; −/−, n = 10), 10 mg/kg cocaine (+/+, n = 10; −/−, n = 10), or 20 mg/kg cocaine (+/+, n = 14; −/−, n = 18). Data are expressed as percentage of baseline locomotor activity measured before the injection (last 10-min interval of habituation phase, data not shown). (Inset) Effect of cocaine treatment on Gpr37^−/− mice locomotor activity. Bars represent the mean (±SEM) distance traveled in 90 min, expressed as percentage of baseline locomotor activity measured before the injection for each genotype and compared across treatment groups. (##, P < 0.01, Gpr37^−/− vs. wild type, 20 mg/kg cocaine; ∗∗, P < 0.01; ∗∗∗, P < 0.001; post hoc Fisher's least-squares difference test).

Reduced Catalepsy by D1 and D2 Receptor Antagonists in Gpr37^−/− Mice.

To study the functional significance of the GPR37 ablation and the consequent increase in DAT-dependent DA uptake, we also compared in vivo the cataleptic effects induced by selective antagonists of D1- and D2-like DA receptors, as a functional paradigm of the nigrostriatal signaling pathway. Fig. 6 shows the cataleptic effects induced in Gpr37^−/− mice and their wild-type littermates by the systemic administration of the D1 receptor antagonist SCH23390 (Fig. 6A) or the D2 receptor antagonist haloperidol (Fig. 6B). Gpr37^−/− mice showed a reduced cataleptic response to the administration of each antagonist (SCH23390: F_(1,15) = 11.43, P < 0.005; haloperidol: F_(1,18) = 3.88, P = 0.06). The difference between genotypes in the time spent in a cataleptic posture after the injection of both drugs was dose-dependent (SCH23390: F_(3,45) = 306.88, P < 0.001; haloperidol: F_(3,54) = 27.76, P < 0.001). Additionally, the repeated-measures ANOVA revealed a significant effect of the interaction genotype × dose of both drugs (SCH23390: F_(3,45) = 11.49, P < 0.001; haloperidol: F_(3,54) = 3.66, P < 0.05). The Gpr37^−/− mice showed a significantly reduced cataleptic response at the dose of 0.1 mg/kg SCH23390 and 0.5 mg/kg haloperidol (Fig. 6).

Fig. 6.

Catalepsy bar test. Response of Gpr37^−/− (−/−) and their wild-type (+/+) littermates to catalepsy induced by the systemic administration of the D1 receptor antagonist SCH23390 (+/+, n = 9; −/−, n = 8) (A) and the D2 receptor antagonist haloperidol (+/+, n = 11; −/−, n = 9) (B). Symbols represent the mean (±SEM) time spent with both front paws resting on the bar; a cut off time was set at 120 s (∗, P < 0.05; ∗∗, P < 0.01; Gpr37^−/− vs. wild type; post hoc Fisher's least-squares difference). Differences between doses are not portrayed. BL, baseline latency to move from the bar before the first injection.

Discussion

We report here that the orphan G protein-coupled receptor GPR37 is found colocalized with DAT in mouse striatal presynaptic membranes and in transfected cell membranes and the two proteins coimmunoprecipitate from transfected cell extracts (Figs. 1 and 4 and SI Fig. 7). Furthermore, the genetic ablation of GPR37 in homozygous null mutant mice provokes the marked augmentation of DAT-mediated DA uptake activity in the striatum (Fig. 2), with an increased expression of DAT at the synaptic plasma membrane (Fig. 3) and a significant reduction of cocaine-induced locomotor activity (Fig. 5) and of catalepsy induced by DA receptor antagonists (Fig. 6). These data indicate that the functional interaction between DAT and GPR37 can modulate the plasma membrane expression of the transporter protein and, consequently, DAT-mediated dopaminergic signaling and functional responses to dopaminergic drugs.

Experimental evidences suggest that DAT undergoes rapid internalization from specific plasma membrane domains to intraneuronal vesicular stores in response to protein kinase C activation by treatment with phorbol esters or neurokinin peptides (27, 28). This process does not apparently require direct phosphorylation of DAT, as reported for the down-regulation of ligand-activated G protein-coupled receptors (29). Therefore, the GPR37 activation by cognate neuropeptide ligands might be critical for modulating DAT expression at the plasma membrane. GPR37 is considered to be involved in the neuropeptide regulation of cell survival during brain development and in the differentiation of glial cells and neurons (3–7), and it interacts with the morphogenetic head activator neuropeptide and its binding protein SorLA upon expression in transfected culture cells (7). Because the peptide activation of GPR37 would cause the receptor down-regulation by phosphorylation and internalization (28), this could, in turn, result in the increased internalization of DAT and other proteins interacting with GPR37 (8). We are setting up specific experimental assays, aiming at defining the possible correlation between levels of GPR37 phosphorylation and plasma membrane expression of DAT.

The GPR37 gene is ubiquitously expressed in human and rodent brain and its transcripts are particularly abundant in corpus callosum and SN (1). Immunohistochemical labeling has localized GPR37 to oligodendrocytes associated with fiber tracts as well as SN dopaminergic neurons, hippocampal neurons in the CA3 region, and cerebellar Purkinje cells (3). In this study, using fractionated synaptic junctional complexes, we show that GPR37 is enriched in the striatal presynaptic membrane fraction (Fig. 1). GPR37 is a substrate of the parkin ubiquitin–protein ligase, and its insoluble aggregates are found accumulated in brain samples of PD patients. The experimental overexpression of the receptor protein in neuronal cell cultures and in Drosophila and mouse dopaminergic neurons in vivo leads, in absence of parkin, to selective cell death (3–6). In a previous paper we reported that Gpr37-null mutant mice exhibit a reduction in striatal DA content, specific locomotor deficits, and enhanced sensitivity to amphetamine (8), indicating the functional alteration of the nigrostriatal dopaminergic signaling pathway. In this study, we detected a strong enhancement of the DAT-mediated DA uptake in Gpr37-null mutant mice, associated with a significant increase in the percentage of transporter molecules translocated to the plasma membrane (Figs. 2 and 3). Therefore, we propose that GPR37 and DAT interact in a functional complex to regulate the membrane sorting and trafficking of the transporter. Our hypothesis is supported by the colocalization of GPR37 with DAT in striatal presynaptic membranes and in transfected cell membranes (Fig. 1 and SI Fig. 7). We also show that GPR37 and DAT coimmunoprecipitate from transfected cell extracts (Fig. 4), although caution is necessary in the interpretation of these experiments (30).

It has become evident over the recent years that a complex protein network is involved in regulating the functional expression of DAT at the perisynaptic membrane. Several DAT-interacting proteins have been identified (for a recent review, see ref. 9), and, whereas the physical and functional coupling between DAT and various G protein-coupled receptors have been hypothesized, experimental data were reported only for the trace amine receptors and the dopamine D2 receptor (11–13). We show here that a putative peptidergic G protein-coupled receptor, GPR37, plays an important role in the regulation of DAT activity. Other proteins involved in PD pathogenesis interact with DAT, including DJ-1, which enhances DAT activity (31), and α-synuclein and parkin, which exert an inhibitory effect (14–16). So GPR37 and parkin functionally interact between themselves (3) and with DAT, suggesting that they may be required to correctly modulate DAT membrane targeting and DAT-mediated DA uptake, to control critical alterations of DAT metabolism that may lead to selective degeneration of SN dopaminergic neurons. Levels of DAT, as measured by brain imaging techniques in vivo, are indeed altered in PD (32), and the increased DA uptake may play a crucial role in determining susceptibility to PD. In a previous study (8), however, we showed that the lack of GPR37 resulted in the resistance to the Parkinsonian neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Although this result cannot be directly linked to the increased plasma membrane expression of DAT, it is possible to postulate the existence of other mechanisms of intracellular protection that could enhance the resistance to toxic agents and oxidative stress (33).

The concentration of DA at nigrostriatal synapses is the most important factor in controlling DA-induced locomotor responses. Cocaine acts as a stimulant by directly blocking DAT-mediated DA uptake, thereby elevating extracellular synaptic DA levels and inducing an increase in locomotor activity (26). Although control mice showed a significant dose-dependent increase of locomotor activity after an acute injection of cocaine, Gpr37^−/− mice showed a very low response even to the highest dose of cocaine (Fig. 5). The lower locomotor response to the same single dose of 20 mg/kg cocaine exhibited by null mutant mice compared with their wild-type littermates is likely to result from the increased number of active DAT molecules available on the presynaptic membrane of the mutant mice, thus emphasizing the role of the GPR37 receptor in regulating the plasma membrane expression of DAT and in modulating its function in vivo.

The null mutant mice and their wild-type littermates exhibited different individual responses to cocaine administration (SI Fig. 8). Other studies hypothesized that individual differences in drug-induced locomotor response could involve differential membrane expression of DAT (34, 35). Interestingly, the Gpr37^−/− mice showed a higher locomotor activity induced by the administration of a high dose of amphetamine, another DAT-specific stimulant drug (8). Given that amphetamine and cocaine are thought to produce an elevation of extracellular DA by different modes of action on DAT (36, 37), the differential sensitivity of Gpr37^−/− mice to these drugs indicates that this mutant strain can be instrumental to study in vivo the dissociation of the two mechanisms of action upon use of specific behavioral tests.

Recent studies (6) reported that the overexpression of GPR37 in transgenic mice is associated with increased DA content in the striatum. Conversely, we showed (8) that Gpr37^−/− mice have reduced levels of striatal DA, which could result from the observed increase (Fig. 3) of DAT-mediated DA uptake. This might induce a compensatory increase of the number and sensitivity of nigrostriatal DA receptors. Indeed, we report a trend for higher B_max values of both subtypes of receptors, as seen in DA-deficient mice (38), and a significantly lower affinity of D2 receptors for an antagonist of DA binding (Table 1). The up-regulation of D2 receptor-binding sites and D1 receptor supersensitivity was also observed after striatal denervation (39–41). These effects might explain the reduced behavioral response of Gpr37^−/− mice, compared with wild-type, to the administration of D1 and D2 receptor antagonists in the catalepsy bar test (Fig. 6).

In summary, this study provides insight into the modulation of DAT-mediated DA uptake by a putative peptidergic G protein-coupled receptor. Our findings suggest that GPR37, although interacting with parkin and other proteins involved in PD, is also a crucial component of the multiprotein complex required for the control of DAT-mediated nigrostriatal signaling and response to psychostimulants. Gpr37^−/− mice will be instrumental in carrying out detailed in vitro and in vivo analysis, which will model critical modifications of DAT metabolism that may lead to selective degeneration of SN dopaminergic neurons and physiopathological alterations induced by drugs of abuse.

Materials and Methods

Information about materials and additional description of methods is provided in SI Materials and Methods.

Animals.

Male and female homozygous Gpr37^−/− mice and their wild-type littermates (10–16 weeks old) were used (8).

Subcellular Fractionations and Western Blot Analysis.

Synaptic vesicles were isolated from the striatum of C57BL/6J adult males (n = 8) essentially as described (42). Presynaptic and postsynaptic protein samples were prepared from the striatum of adult male C57BL/6J mice (n = 12); synaptosomes were prepared by Percoll gradient purification (43) and were solubilized by sequential extraction as reported (17). Protein samples were separated by SDS/PAGE and analyzed by Western blot.

Synaptosomal [³H]DA-Uptake Experiments.

Synaptosomes were prepared from the striatum of Gpr37^−/− and wild-type littermate adult male mice (n = 4 for each genotype/experiment), as described (44). For each experimental determination, 30 μg of pooled synaptosomes were prewarmed at 37°C for 5 min, before addition of 18 nM [³H]DA for 5 min (18). Nonspecific [³H]DA uptake was determined in the presence of 10 μM nomifensine, as described (18). Kinetic assays were performed by incubating synaptosomes with increasing concentrations of unlabeled DA (0.005–1 μM) and a constant concentration of [³H]DA (18 nM).

Biotinylation and Quantification of DAT at the Plasma Membrane.

Total protein aliquots (250 μg) from crude striatal synaptosomes were incubated for 1 h at 4°C in 500 μl of 1.5 mg/ml sulfo-NHS-biotin in PBS/Ca/Mg buffer [138 mM NaCl/2.7 mM KCl/1.5 mM KH₂PO₄/9.6 mM Na₂HPO₄/1 mM MgCl₂/0.1 mM CaCl₂ (pH 7.3)]. Samples were then lysed by sonication in Triton X-100 buffer [10 mM Tris (pH 7.4)/150 mM NaCl/1 mM EDTA/1.0% Triton X-100/complete protease inhibitors (Roche, Indianapolis, IN)], and cellular debris was removed by centrifugation. Biotinylated and nonbiotinylated proteins were separated in the presence of agarose beads conjugated to streptavidin (Pierce, Rockford, IL) in Triton X-100 buffer for 1 h at room temperature.

Cell Culture and Transient Transfection with GPR37 Expression Construct.

HEK293 cells and HEK-hDAT cells were maintained in Dulbecco's modified Eagle's medium with 10% FBS, 2 mM l-glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin; 400 μg/ml geneticin (G418) was added only to HEK-hDAT cells. Full-length human GPR37 cDNA containing an HA peptide tag at its carboxyl terminus (GPR37-HA) was obtained from total human brain RNA (45) and cloned in the pcDNA3 expression vector (Invitrogen, Carlsbad, CA). The linearized GPR37-HA construct or vector plasmid were transfected with LipofectAMINE 2000 Reagent (Invitrogen) into cells grown to 70–80% confluency. Subsequent experiments were performed 48 h after transfection.

Immunopurification.

Tranfected cells were solubilized as described (3). The soluble extracts were incubated with the anti-DAT monoclonal antibody or anti-rat IgG (Pierce) in the presence of G-Sepharose beads (Sigma, St. Louis, MO) or with anti-HA affinity matrix (Roche) following the manufacturer's protocol. Immune complexes were then washed five times and subjected to SDS/PAGE and Western blotting. Blots were probed with the anti-DAT monoclonal antibody or rat anti-HA peroxidase-conjugated antibody (clone 3F10; Roche) according to the producer's instructions.

Radioligand-Binding Experiments.

Total membrane homogenates were prepared from striatum of Gpr37^−/− and wild-type littermate male mice (n = 4 for each genotype/experiment), and aliquots of each group were incubated with tritiated ligands, as described (46).

Behavioral Tests.

Cocaine-induced locomotor activity: male mice were tested in an open field arena (1,848 cm²) by using a video tracking system (View Point, Champagne-au-Mont-d'Or, France). After 30 min of habituation, mice were injected i.p. with saline, 10, or 20 mg/kg cocaine hydrochloride. Locomotor activity was measured for 90 min after the injection, and it was expressed as the percentage of the baseline distance traveled in the last 10-min interval before the injection. Catalepsy bar test: female mice were observed for cataleptic behavior induced by the systemic administration of either D1- or D2-like receptor antagonist in a transparent plastic box (43 × 26 × 18 cm). The cataleptic response was measured as the duration of an abnormal upright posture in which the forepaws of the mouse were placed on a horizontal bar (0.4-cm-diameter steel rod, covered with nonslippery tape) 3.5 cm above the box floor. In two separate experiments, after assessment of the baseline (BL) latency to move from the bar, female mice were injected with either the D1 receptor antagonist SCH-23390 (0.03, 0.1, and 0.3 mg/kg s.c.) or the D2 receptor antagonist haloperidol (0.125, 0.25, and 0.50 mg/kg i.p.), by using a cumulative dose regimen in which each animal was given increasing doses of antagonists and tested after each injection. The latency to move at least one of the two forepaws was recorded 15 min after injection in the SCH-23390 experiment and 30 min after injection in the haloperidol experiment. A cutoff time was set at 120 s (47).

Abbreviations

DA

dopamine

DAT

dopamine transporter

GPR37

G protein-coupled receptor 37

HA

influenza virus hemagglutinin

PAEL-R

parkin-associated endothelin-like receptor

PD

Parkinson's disease

SN

substantia nigra.