D3 dopamine receptors interact with dopamine D1 but not D4 receptors in the GABAergic terminals of the SNr of the rat

Neuropharmacology 67 (2013) 370e378

Contents lists available at SciVerse ScienceDirect

Neuropharmacology journal homepage: www.elsevier.com/locate/neuropharm

D3 dopamine receptors interact with dopamine D1 but not D4 receptors in the GABAergic terminals of the SNr of the rat Refugio Cruz-Trujillo c, Arturo Avalos-Fuentes c, Claudia Rangel-Barajas a, Francisco Paz-Bermúdez a, Arturo Sierra a, Erick Escartín-Perez d, Jorge Aceves a, David Erlij b, Benjamín Florán a, * a

Department of Physiology, Biophysics and Neurosciences, CINVESTAV-IPN, Av. IPN # 2508, San Pedro Zacatenco, México 07360, Mexico Department of Physiology, SUNY Downstate Medical Center, NY, USA Department of Pharmacology, CINVESTAV-IPN, Mexico d Neurobiology of Eating, FES-IZTACALA UNAM, Mexico b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 July 2012 Received in revised form 29 November 2012 Accepted 30 November 2012

The firing rate of substantia nigra reticulata (SNr) neurons is modulated by GABA release from striatonigral and pallidonigral projections. This release is, in turn, modulated by dopamine acting on dopamine D1 receptors at striatonigral terminals and D4 receptors at pallidonigral terminals. In addition, striatal neurons that express D1 receptors also express D3 receptors. In this study we analyzed the possible significance of D3 and D1 receptor colocalization in striatonigral projections. We found that these receptors coprecipitate in SNr synaptosomes suggesting their close association in this structure. D1 agonist SKF 38393 administered alone increased mIPSC frequency in SNr slices and cAMP production in SNr synaptosomes, however, the selective D3 agonist PD 128,907 increased mIPSC frequency and cAMP production only when D1 receptors were concurrently stimulated. The D1 antagonist SCH 23390 blocked completely the effects of the concurrent administration of these agonists while the selective D3 antagonist GR 103691 blocked only the potentiating effects of PD 128,907. These findings further indicate that D1 and D3 receptors are localized in the same structure. The D4 agonist PD 168,077 decreased mIPSCs frequency without changing amplitude, an effect that was blocked by the selective D4 antagonist L 745,870. The effects of D4 receptor stimulation disappeared after lesioning the globus pallidus. D3 agonist PD 128,907 did not reduce mIPSC frequency even in neurons that responded to D4 agonist. In sum, activation of D3 receptors in SNr potentiates the stimulation of transmitter release and cAMP production caused by D1 receptor activation of striatonigral projections while it is without effects in terminals, probably of pallidal origin, that are inhibited by activation of D4 receptors. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: D1eD3 interaction Substantia nigra GABA release Presynaptic D1eD3 receptors Presynaptic D4 receptors

1. Introduction The substantia nigra reticulata (SNr), one of the major output nuclei of the basal ganglia, is composed mainly of GABAergic neurons that project to the relay nuclei of the thalamus, the superior colliculus and the pedunculopontine nucleus (Parent and Hazrati, 1995). Firing of SNr neurons is modulated by GABA release from inhibitory afferents originating in the globus pallidus (GP) and the striatal GABAergic spiny projection neurons (SPNs). Pallidonigral and striatonigral fibers synapse onto separate domains of the somatodendritic tree: striatonigral synapses are found mainly on fine distal dendrites whereas pallidonigral terminals form large

* Corresponding author. Tel.: þ52 55 5747 3800x5137; fax: þ52 55 7473754. E-mail addresses: bfl[email protected], bfloran@fisio.cinvestav.mx (B. Florán). 0028-3908/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neuropharm.2012.11.032

baskets around the soma and proximal dendrites. In addition, intranuclear collaterals provide additional GABAergic innervation (Parent and Hazrati, 1995; Connelly et al., 2010). Activation of dopamine receptors in the SNr has opposing effects on GABA release (Acosta-Garcia et al., 2009; Aceves et al., 2011). These effects arise because two different types of dopamine receptors are segregated between striatonigral and pallidonigral terminals. Of the five G protein-coupled dopamine receptors (GPCR), D1 receptors that increase GABA release are selectively localized in striatonigral projections, while D4 receptors that decrease release are localized in pallidonigral terminals (AcostaGarcia et al., 2009; Aceves et al., 2011). In this study we further examined the distribution and function of dopamine receptors in the GABAergic terminals of the SNr. We were especially interested in determining whether dopamine D3 receptors have a functional role in the modulation of GABA release

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

by striatonigral substance P positive SPNs that coexpress mRNAs for both D3 and D1 receptors (Surmeier et al., 1996), Gerfen and Surmeier (2011) have pointed out that a functional role D3 receptors in striatal neurons has been difficult to establish. We determined whether D1 and D3 receptors are coexpressed in synaptosomes obtained from striatonigral terminals using immunoblot techniques and whether they functionally interact in the modulation of GABAergic transmission. We assayed GABA release by measuring mIPSCs to avoid possible secondary effects produced when action potentials are elicited by electrically stimulating the slices (see Radnikow and Misgeld, 1998; Discussion). Because modulation of GABA release in striatonigral terminals is controlled by the cAMP/PKA cascade (Nava-Asbell et al., 2007), we also determined the effects of D3 receptor activation on cAMP production in synaptosomes of the SNr. In addition we examined whether or not D3 receptors are active in terminals modulated by D4 receptor that probably correspond to pallidonigral projections. A preliminary account of these findings has already been published (Cruz-Trujillo et al., 2008). 2. Methods 2.1. Slice preparation and solutions Experimental procedures were carried out in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals and were approved by the Institutional Animal Care Committees of the CINVESTAV. Brain slices obtained from male Wistar rats (postnatal day 14e21) were used. The rats were anesthetized and decapitated. The brain was quickly removed from the skull and placed in ice-cold artificial cerebrospinal fluid (ACSF) containing (in mM): 124 NaCl, 26 NaHCO3, 2.5 KCl, 1.3 MgCl2, 1.2 NaH2PO4, 2.4 CaCl2, and 10 glucose, pH 7.4 (with 95% O2 and 5% CO2 bubbling through the solution). To prevent swelling of the cell during slicing, NaCl was replaced with choline chloride. Both solutions were continuously oxygenated with the gas mixture. Parasagittal slices (300 mm) containing the SNr were cut with a vibroslicer (Lancer, Technical Products International, St. Louis, MO, USA) and then transferred to normal ACSF at room temperature (ca 25  C) for equilibration. After 1 h a single slice was transferred to a recording chamber continuously superfused with ACSF (1e2 mL/min) at room temperature. To block N-methyl-D-aspartate and non-N-methyl-D-aspartate glutamate receptors, AP-5 (50 mM) and CNQX (10 mM) were added to the superfusion medium. TTX (1 mM) was added to avoid contamination of synaptic responses by action currents.

371

3-isobutyl-1-metylxantine, 1; all in mM). Aliquots of 250 ml of the synaptosomes were placed in tubes, and the drugs were added in 10 ml volume. Incubation was continued for 15 min and stopped by adding 100 ml of ice-cold trichloroacetic acid (15%) containing unlabeled ATP (2.5 mM) and cAMP (4.5 mM). After a period of 20 min on ice, the tubes were centrifuged (4000 rpm, 5 min, at 4  C) and the supernatants loaded onto Dowex 50W-X4 columns (300 ml per column). A fraction containing [3H]-ATP was eluted with 3 ml of distilled water. A second eluent obtained with 5 ml of distilled water was directly loaded onto neutral alumina columns. Alumina columns were finally eluted with 4 ml of 50 mM TriseHCl buffer pH 7.4 to obtain [3H]-cAMP. The results were expressed as [3H]-cAMP  100]/[3H]cAMPþ[3H]-ATP] and were normalized to basal accumulation. 2.4. Protein preparation, immunoprecipitation, and Western Blot The synaptosomal fraction was incubated with RIPA buffer (sodium orthovanadate 1 mM, sodium pyrophosphate 10 mM, sodium fluoride 100 mM, glycerol 10%, triton X-100, 1%, TriseHCl 50 mM, Na Cl 150 mM, MgCl2 1.5 mM, EGTA 1 mM, SDS 0.1% and sodium deoxycholate 1%) and a complete set of protease inhibitors (Roche Applied Science, México) for 15 min. To obtain the membranes, the synaptosomes were sonicated and protein was quantified by the Bradford method. A sample of 500 mgrs of protein was incubated with D1 (1:250) or D3 (1:250) antibody for 6 h and then added to A-agarose beads (Roche Applied Science, México) during the next 12 h. Then samples were centrifuged 3 times 10,000 RPM/5 min with 500 mL of RIPA buffer. The resultant pellet of the third centrifugation was resuspended in sample buffer (Glycerol 50%, TriseHCl 125 mM, SDS 4% Bromophenol blue 0.08%, b-mercaptoethanol 5%) and heated at 100  C for 10 min. To detect the D1R or D3R the samples were resolved by SDS-PAGE, transferred onto PVDF membranes, and blotted for 1 h at room temperature in Tris buffered saline containing 0.2% Tween 20 and 5% nonfat powdered milk. Membranes were incubated overnight at 4  C with either the anti-D1R antibody (1:2000 dilution) or the anti-D3R antibody (1:500 dilution). D1R and D3R polyclonal antibodies were obtained from Santa Cruz Biotecnology Inc. The specificity of these antibodies has been reported by Fiorentini et al. (2008) and Everett and Senogles (2010). Detections were performed by chemiluminescence (ECL-Plus Amersham) with HRP-conjugated secondary antibodies (1:5000 dilution). 2.5. Lesion of globus pallidus Lesions of the globus pallidus were performed as previously described (AcostaGarcia et al., 2009; Gasca-Martinez et al., 2010). Neonatal rats (postnatal day 7) were anesthetized by hypothermia and placed on a home-made frame under a stereotaxic apparatus (David Kopf), and 50 nL of kainic acid solution (1 mg/mL freshly dissolved in 0.1 M sodium phosphate buffer, pH 7.4) was injected unilaterally into the globus pallidus (coordinates with respect to Bregma: lateral 1.8 mm, anteroposterior 0.7 mm, dorsoventral 4.0 mm). The site of injection was verified by injecting Methylene Blue. In all cases the injection site was located in the central or medial part of the globus pallidus.

2.2. Electrophysiology 2.6. Drugs Neurons were visualized using infrared differential-interference video microscopy with a 40  water-immersion objective (Hamamatsu C2400-50, Hamamatsu Photonics Systems, USA and Axioscop, Carl Zeiss, Oberkochen, Germany). Micropipettes for whole cell recordings were pulled (Sutter Instruments, Novato, CA, USA) from borosilicate glass tubes (1.5 mm outer diameter, WPI, Sarasota, FL, USA) for a final resistance of 2e5 MU, and filled with a solution of the composition (in mM): 130 CsCl, 0.5 CaCl2, 2 MgCl2, 10 HEPES, 5 K2-EGTA, 4 ATP-Mg, and 1 GTP-Na (pH 7.3 adjusted with CsOH; osmolality, 287e290 mOsmL1). This internal solution was used for voltage-clamp recordings. Voltage-clamp recordings were made with an Axopatch 200B amplifier (Axon Instruments, Foster City, CA, USA). Liquid junction potentials (5 mV) were not corrected. Recordings were acquired at 10 kHz using a Digidata 1200 interface (Axon) and pCLAMP software (Axon, v 7.0). The Bessel filter was set at 5 kHz. Access resistance (7e15 MU) was monitored continuously and experiments were abandoned if changes greater than 20% occurred.

40 -Acetyl-N-[4-[4-(2-methoxyphenyl)-1-piperazinyl]butyl]-[1,10 -biphenyl]-4carboxamide (GR 103691), D(-)-2-amino-5-phosphonopentanoic acid (AP-5), 2-Carboxy-3-carboxymethyl-4-isopropenylpyrrolidine (kainic acid), R(-)-7-Chloro-8hydroxy-3-methyl-1-phenyl-2,3,4, 5-tetrahydro-1H-3-benzazepine hydrochloride (SCH 23390), (3-[(4-[4-chlorophenyl]piperazin-1-yl)methyl]-1H pyrrolo[2,3-b] pyridine) hydrochloride (L 745,870), 6-Cyano-7-nitroquinoxaline-2,3-dione (CNQX), bicuculline methiodide, N-[[4-(2-cyanophenyl)-1-piperazinyl]methyl]-3methyl-benzamide maleate (PD 168,077), (RS)-2,3,4,5 tetrahydro-7,8-dihydroxy1-phenyl-1H-3-epine hydrochloride (SKF 38393), (þ)-(4aR,10bR)-3,4,4a,10bTetrahydro-4-propyl-2H,5H-[1]benzopyrano[4,3-b]-1,4-oxazin-9-ol hydrochloride (PD 128,907), tetrodotoxin (TTX), all were obtained from Sigma (St. Louis, MO, USA). The drugs were stored in the freezer as dry aliquots. Stock solutions were prepared just before each experiment and added to the perfusion solution at the final concentration.

2.3. cAMP accumulation assay

2.7. Statistics

cAMP accumulation assays were performed as previously described (Alexander, 1995; Rangel-Barajas et al., 2011). Synaptosomal fractions were isolated from SNr slices. The slices were homogenized in buffer (sucrose, 0.32 M; HEPES, 0.005 M, pH 7.4), and then homogenates were centrifuged at 800  g during 10 min. The resulting supernatant was further centrifuged at 20,000  g during 20 min. From this second centrifugation, the supernatant (S1) was discarded, and the pellet (P1) was resuspended and collocated on the sucrose 0.8 M, HEPES 0.005 M, and buffer (pH 7.4) and was newly centrifuged at 20,000  g during 20 min; finally, the supernatant was discarded, and the new pellet (P2) containing synaptosomes were used. The fraction was incubated with [3H]-adenine (130 nM) during 1 h at 37  C, after this period it was suspended in KrebseHenseleit buffer (composition: NaCl, 127; KCl, 3.73; MgSO4,1.18; KH2PO4, 1.18; CaCl2, 1.8; HEPES, 20; glucose, 11; and

The significance of the effects on mIPSCs in each individual experiment was evaluated applying the KolmogoroveSmirnov two sample test to amplitudes and cumulative inter-event interval distributions (Minianalysis, Synaptosoft). We then verified whether the distribution of values in each experimental series was normal using the Kolmorogrov-Smirnov, D’Agostino & Person omnibus and Shapiro-Wilk normality tests. Because these tests turned out to be positive, we evaluated the statistical significance on frequency and amplitude of mIPSCs using one way ANOVA followed by Tukey’s multiple comparison test. When effects of D1 or D4 antagonist were tested alone student t-test were use to compare with control. The experiments on cAMP production were analyzed using One-Way ANOVA combined with two different post-test analysis; Dunnett’s for comparing with the control data, and Tukey to compare differences between experimental groups.

372

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

3. Results 3.1. D1R and D3R of synaptosomes of SNr coimmunoprecipitate These studies were carried out to determine whether D1R and D3R are coexpressed in the terminals of the SNr. Fig. 1A shows experiments in which synaptosomal proteins were first incubated with the anti-D3R antibody. This antibody immunoprecipitated a z70 kDa species that was recognized by the anti-D1R antibody (lane 2). When the immunoprecipitating antibody was omitted (lane 1) the signal was absent. When the synaptosomal proteins were incubated with anti-D1R antibody it immunoprecipitated protein that was recognized by the anti-D3R antibody. Two major bands, between z60 and z70 kDa, were detected (Fig. 1B, lane 2). This finding is in agreement with the observations of Fiorentini et al. (2008) and of Nimchinsky et al. (1997). These signals were undetectable when the immunoprecipitating antibody was omitted (lane 1). 3.2. Stimulation of dopamine D3 receptors increases the frequency of GABAA-mediated mIPSCs only when coactivated with D1 receptors Whole cell recordings were obtained from 300 mm-thick SNr slices, inwardly going spontaneous synaptic currents were recorded at a holding potential of 80 mV using a pipette filled with a CsClbased internal solution. These events were completely abolished by bath-application of bicuculline (20 mM) and had a reversal potential close to 0 mV in agreement with the equilibrium potential for chloride ions (data not shown). These findings indicate that the currents were mediated by the activation of GABA-A receptors (Curtis et al., 1970). Fig. 2AeC show that bath perfusion with the selective D3 dopamine agonist PD 128,907 (100 nM) alone did not have significant effects on interevent interval (control: 1.62  0.24 Hz; PD 128,077: 1.52  0.18 Hz, n.s., F ¼ 29.68, df ¼ 2, ANOVA followed by Tukey, n ¼ 8) while perfusion with D1 agonist SKF 38393 alone in the same cells (3 mM) significantly increased mIPSC frequency (2.48  0.30 Hz, p < 0.001 when compared to control, F ¼ 29.68, df ¼ 2, ANOVA followed by Tukey, n ¼ 8). The mean mIPSC amplitudes were: 60.72  4.07 pA in control, 59.66  3.52 pA with PD 128,907 and 58.82  3.78 pA in SKF 38393, showing no statistically significant difference (F ¼ 1.32, df ¼ 2, ANOVA followed by Tukey, n ¼ 8). Because the lack of significant effects on mIPSC amplitude were uniform throughout the experiments in this study, description of effects on amplitude are omitted from the rest of the text, however, they are still shown in the figures. The bottom of Fig. 2A shows tracings in control and SKF 38393 treated slices on an expanded time scale, superposition of the tracings shows that the kinetics of the mIPSCs in the presence and absence of the agonist were not altered. Table 1 shows the number and proportion of neurons throughout this study that, on the basis of the SmirnoveKolmogorov test,

exhibited a significant change in mIPSC interevent interval during different ligand applications. Administering SKF 38393 alone was effective in 83% of the cells (82/98), while PD 128,907 administered alone was always without effects (0/21). Because previous evidence (Schwartz et al., 1998; Fiorentini et al., 2008; Marcellino et al., 2008) suggests that stimulation of D3 receptors is effective when they are coactivated concomitantly with D1 receptors, we tested the effects of adding the selective D3 agonist PD 128,907 (100 nM) to SKF 38393 containing perfusion fluid (Figs. 2DeF and 3AeC). In the experiments of Fig. 2DeF addition of SKF 38303 increased frequency from 1.64  0.15 Hz to 2.35  0.20 Hz (p < 0.001, F ¼ 71.94, df ¼ 3, ANOVA followed by Tukey, n ¼ 11). When next PD 128,907 was administered together with SKF 38393, frequency increased to 3.14  0.24 Hz (p < 0.001, F ¼ 71.94, df ¼ 3, ANOVA followed by Tukey, n ¼ 11). The table shows that 58% of the cells were further stimulated (33/57) when PD 128,907 was added to the SKF 38393 containing fluid. Finally, when the selective D1 antagonist SCH 23390 (100 nM) was also included in the perfusion fluid the increase in frequency was reversed to a value that was not significantly different to the control (1.57  0.14 Hz 1.64  0.15 Hz, F ¼ 71.94, df ¼ 3, ANOVA following Tukey, n ¼ 11). In other experiments SCH 23390 administered alone did not affect significantly amplitude or interevent interval (mean amplitude control 66.93  5.63, SCH 23390 65.29  6.28, not significant difference t ¼ 1.40, df ¼ 4 student t test; mean frequency control 1.85  0.12 Hz, SCH 23390 1.86  0.18 Hz, not significant difference, t ¼ 0.046, df ¼ 4, student t test, n ¼ 5 data not shown in graphs). Fig. 3AeC shows the effects of the selective D3 receptor antagonist GR 103691 on the combined effects of SKF 38393 and PD 128,907. Again SKF 38393 shifted cumulative interevent interval distributions to the left (control: 1.53  0.10 Hz; SKF 38393: 2.29  0.14 Hz, p < 0.001, F ¼ 63.01, df ¼ 3, ANOVA followed by Tukey, n ¼ 11). Next, PD 128,907 further shifted to the left the cumulative interevent interval distribution control (SKF 38393: 2.29  0.14 Hz; SKF 38393 þ PD 128077: 2.93  0.21 Hz, p < 0.001, F ¼ 63.01, df ¼ 3, ANOVA followed by Tukey, n ¼ 11). When the selective D3 blocker GR 103691 (50 nM) was also included in the perfusion solution the combined effects of PD 128,907 and SKF 38393 were reduced to a level (2.27  0.14 Hz) that was not significantly different from that observed when SKF 38393 was administered alone (F ¼ 63.01, df ¼ 3, ANOVA followed by Tukey, n ¼ 11). Fig. 3DeF shows that the potentiating effects of PD 128,907 were reversed on perfussion with aCSF that contained only SKF 38393 for approximately 15 min (control 1.63  0.15 Hz; SKF 39393 þ PD 128,907 3.25  0.27; SKF 39393 2.47  0.22; n ¼ 11). 3.3. Stimulation of dopamine D3 receptors increase [3H]-cAMP production only when coactivated with D1 receptors D3 dopamine receptors are a subtype of the D2 family that interacts with Gi/o proteins inhibiting cAMP formation (Missale

Fig. 1. Coimmunoprecipitation of D1R and D3R in SNr synaptosomes. A. Representative coIP of D1R by the anti-D3R antibody (lane 2). Lane 1 was run in the absence of the precipitating antibody. B, CoIP of D3R by the anti-D1R antibody. Lane 1 was run in the absence of the precipitating antibody. 500 mgrs of proteins were used in each IP that was repeated 4 times.

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

373

Fig. 2. A, B and C. The D3 selective agonist PD 128,907 administered alone does not modify spontaneous GABA release by presynaptic terminals. A. Representative traces of mIPSCs. The top tracing was recorded during the control period. The next tracing was recorded during the application of 100 nM of PD 128,907. The third tracing was recorded during the separate application of the selective D1 agonist SKF 38393 (3 mM). The bottoms are expanded tracings comparing the time courses of mIPSC treated with SKF 38393. (a, was recorded during control conditions; b during the action of the agonist and c is a superimposition of a and b). B. Cumulative probability distributions for inter-event interval (top left) and current amplitude (bottom left) of GABAergic mIPSCs shown in (A). C. Dot plots (n ¼ 8) show results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each condition in a single experiment are connected by lines. D, E and F. The D3 selective agonist PD 128,907 increased spontaneous GABA release when D1 receptors were simultaneously coactivated with SKF 38393. D. Shows representative traces recorded successively during control conditions and then the application of SKF 38393 alone, SKF 38393 and PD 128,907 together and finally when the selective D1 antagonist SCH 23390 was given together with the two agonists. E. Cumulative probability distributions for inter-event interval (top left) and current amplitude (bottom left) of GABAergic mIPSCs shown in (E). F. Dot plots summarize (n ¼ 11) results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each individual condition in a single experiment are connected by lines. ***p < 0.001, with respect to control; ###p < 0.001, with respect SKF 38393 alone; n.s, not significant difference.

et al., 1998). Since the activity of the cAMP/PKA cascade is a major determinant of GABA release in nigrostriatal terminals (RangelBarajas et al., 2008), we examined the effect of D3 receptor stimulation on cAMP production in synaptosomes prepared from slices of the SNr to find whether changes in this signaling cascade parallel the effects on mIPSCs. Control cAMP production was always taken as 100%. As shown in Fig. 4A, activation of D1 receptors with SKF 38393 (3 mM) Table 1 Number of neurons that exhibited a change in mIPSC frequency during agonist treatment.

1. 2. 3. 4. 5.

Treatment

Effect

No effect

Total

%

D1 D3 D1 D4 D3

82 0 33 27 0

16 21 24 9 8

98 21 57 36 8

83 0 58 75 0

(SKF 38393) (PD 128,907) þ D3 (PD 168,077) þ D4

The numbers for neurons used to construct this table were collected throughout the study and may be different from the n values shown in individual figures because the latter are only from data using a single protocol.

significantly increased cAMP production to 125  3.7%, p < 0.01 with respect to control, F ¼ 45.22, df ¼ 3 ANOVA followed by Dunnett’s (n ¼ 4 experiments, 4 replicates per experiment). PD 128,907 (100 nM) added alone didn’t significantly modify cAMP production (101  3%, ns. not statistically different from control, F ¼ 45.22, df ¼ 3 ANOVA followed by Dunnett’s n ¼ 4 experiments, 4 replicates per experiment). However, when PD 128,907 was added concurrently with SKF 38393, it increased cAMP production to a value significantly higher (152  8.1%, F ¼ 45.22, p < 0.001 with respect the control and p < 0.01 with respect SKF 38393) than the level reached when SKF 38393 was administered alone. Antagonism of D1 receptors with SCH 23390 (100 nM) reduced the effects of administering SKF 38393 alone or given together with PD 128,907 to a level that was not significantly different from the control (Fig. 4B, 98  5%, no statistically significant with respect the control, F ¼ 22.30, df ¼ 4 ANOVA followed by Dunnett’s n ¼ 4 experiments, 4 replicates per experiment). In contrast, the selective D3 antagonist GR 106391 (50 nM) reduced the response produced by administering both agonists together to a level that was not significantly different from that found when SKF 38393 was given

Fig. 3. The effects of the D3 agonist PD 128,907 on spontaneous GABA release are inhibited by the selective D3 antagonist GR 103691 and are reversible. A, B and C. The D3 selective agonist PD 128,907 increased spontaneous GABA release when D1 receptors were simultaneously coactivated with SKF 38393. A. Shows representative tracings recorded first during control conditions and then the application of SKF 38393 alone, followed by SKF 38393 and PD 128,907 together and finally the two agonists plus the selective D3 antagonist GR 103691. B. Cumulative probability distributions for inter-event interval (top left) and current amplitude (bottom left) of GABAergic mIPSCs shown in A. C. Dot plots summarize (n ¼ 8) results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each experimental condition in a single experiment are connected by lines. D, E and F. Reversibility of PD 128,907 effects. D. The top tracing was recorded during the control period. The next tracing was recorded during the application of 100 nM PD 128,907 together with SKF 38393 (3 mM). During the bottom tracing the slice was perfused with a solution containing SKF 38393 only. E. Cumulative probability distributions for inter-event interval (top left) and current amplitude (bottom left) of GABAergic mIPSCs shown in D. F. Dot plots summarize (n ¼ 8) results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each condition in a single experiment are connected by lines. ***p < 0.001, with respect the control; ###p < 0.001 with respect SKF 39393; n.s. not significant difference.

Fig. 4. A. The D3 selective agonist PD 128,907 does not modify cAMP accumulation in slices of the SNr when applied alone. However, the stimulation of cAMP produced by stimulation of dopamine D1 receptors with SKF 38393 is further enhanced when D3 receptors are simultaneously stimulated. A. Effects of SKF 38393 administered alone (**p < 0.01 with respect to control), PD 128,907 alone (n.s., non significant with respect to control) and both agonists together, (***p < 0.001, with respect the control; ##p < 0.01, with respect to SKF 38393 alone). B. Effect of SKF 38393 alone (*p < 0.05 with respect to control). Effects of SKF 38393 administered together with PD 128,907 (***p < 0.001 with respect to control), when the two agonists were administered in the presence of the D1 selective blocker SCH 23390 the accumulation of cAMP was not significantly different from control. When the two agonists were administered together with the D3 selective blocker GR 103691, the accumulation of cAMP was not significantly different than that measured in SKF 38393 alone but it was higher than that seen in control (**p < 0.01) and significantly lower when compared to the level seen with the two agonists together (#p < 0.05). C. Forskolin-stimulated cAMP accumulation (***p < 0.001) was not modified by D3 agonist PD 128,907 n ¼ 4 experiments, 4 replicates per experiment.

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

375

alone (126  4%, p < 0.01 with respect the control, p < 0.05 with respect to SKF 38393 þ PD 128,907, F ¼ 22.30, df ¼ 4 ANOVA followed by Tukey n ¼ 4 experiments, 4 replicates per experiment). Activation of Gi/o proteins decreases forskolin stimulated cAMP production (Insel and Ostrom, 2003). To examine whether stimulation of D3 receptors in terminals of the SNr activate Gi/o proteins we determined the effects of PD 128,907 on forskolin-stimulated cAMP production (Liu et al., 2009). PD 128,907 didn’t significantly modify forskolin stimulated cAMP production (forskolin 220  14% vs. forskolin þ PD 128,907, 228  16%, no significant difference between them, F ¼ 43.42, df ¼ 3 ANOVA followed by Tukey n ¼ 4 experiments, 4 replicates per experiment). 3.4. Stimulation of D4 receptors reduces transmitter release from pallidonigral terminals The selective D4 antagonist L 745,870 (100 nM) administered alone did not modify either mIPSCs interevent interval or amplitude (mean amplitude control: 51.31  7.48: þL 745,870: 50.69  7.46, n.s., t ¼ 1.01, df ¼ 5 student t test; mean frequency control: 1.43  0.27 Hz; þL 745,870: 1.51  0.37 Hz, n.s., t ¼ 0.46, df ¼ 5 student t test, n ¼ 6 cells data). In other experiments the selective D4 agonist PD 168,077 (100 nM) (Fig. 5) was added together with the D4 antagonist and no effects were observed (2.00  0.31e1.92  0.30 Hz, n.s., F ¼ 15.56, df ¼ 3 ANOVA followed by Tukey, n ¼ 8). However, when L 745,780 was washed leaving PD 168,077 alone in the solution, mIPSCs interevent interval was reduced from 1.92  0.31 to 1.38  0.22 Hz (p < 0.001, F ¼ 15.56, df ¼ 3 ANOVA following Tukey, n ¼ 8). When the slice was subsequently washed with a solution that did not contain the D4 ligands the frequency increased to 1.90  0.31 Hz (not significantly different from control: 2.00  0.31, F ¼ 15.56, df ¼ 3 ANOVA following Tukey, n ¼ 8). To determine the site of the D4 receptors, the neurons of the globus pallidus were lesioned by injecting kainic acid into the nucleus. We have shown previously that the injection of kainic acid reduces the neuronal population of the globus pallidus by about 80% (see Acosta-Garcia et al., 2009; Gasca-Martinez et al., 2010). When the effects of activating D4 receptors with the selective agonist PD 168,077 on mIPSC frequency were compared in the SNr of intact and lesioned sides of the same brain the lesion prevented the D4 agonist effect. In the nonlesioned side: control, 2.42  0.36; PD 168,077, 1.79  0.30 (p < 0.001, F ¼ 72.82, df ¼ 2 ANOVA followed by Tukey, n ¼ 10; recovery, 2.35  0.33 ns with respect control). On the lesioned side, control, 2.17  0.33; PD 168,077, 2.13  0.36 (ns, not significant, F ¼ 0.43, df ¼ 2 ANOVA followed by Tukey, n ¼ 9) These results confirm that D4 receptors responsible for the inhibition of GABAergic transmission in the SNr are primarily located in pallidal terminals. 3.5. Stimulation of D3 receptors doesn’t modify mISPCs in synapses that are inhibited by D4 receptor stimulation Because we didn’t find any inhibitory effects of D3 agonist PD 128,907 given alone, we also examined its effects in cells that showed an inhibitory response to the D4 agonist PD 168,077. We tested the effects when the agonists were administered both separately (Fig. 6AeC) or together (Fig. 6DeF). As previously describe PD 128,907 alone did not significantly inhibit mISPCs frequency (Control: 2.06  0.16 Hz; PD 128,907: 1.96  0.16 Hz); however PD 168,077 alone had a significant inhibitory effect (to 1.42  0.13 Hz, p < 0.001, F ¼ 63.82, df ¼ 2 followed by Tukey, n ¼ 8). When PD 168,077 and PD 128,907 were given concurrently (Fig. 6DeF), the frequency was reduced significantly from 1.82  0.30 Hz to 1.27  0.22 Hz (, p < 0.001, F ¼ 37.47, df ¼ 2 ANOVA

Fig. 5. The selective D4 antagonist L 745,870 blocks the inhibition of spontaneous GABA release produced by the selective D4 agonist PD 168,077. A. Representative traces of mIPSCs. The top tracing was recorded before the application of ligands. Then, successively from top to bottom, the perfusion fluid contained L 745,870 þ PD 168,077, then PD 168,077 alone and finally the control solution. B. Cumulative probability distributions for inter-event interval (top) and current amplitude (bottom) of GABAergic mIPSCs shown in (A). Dot plots summarize (n ¼ 8) results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each individual condition in a single experiment are connected by lines. ***p < 0.001 with respect control.

followed by Tukey). When subsequently PD 168,077 was present alone in the perfusion fluid the frequency stayed at nearly the same depressed value 1.24  0.23 Hz. This value was significantly lower than the control (p < 0.001, F ¼ 37.47, df ¼ 2 ANOVA followed by Tukey, n ¼ 8). PD 168,077 had inhibitory effects in 75% of the cells tested (Table 1) compared with the lack of effects of PD 128,907 administered alone (0/21, 0%). 4. Discussion The most interesting finding of this study is that stimulation of D3 receptors in SPNs that project to the SNr augment the increase of GABA release produced by D1 receptor stimulation. D3 receptor activation also enhanced the stimulation of cAMP produced by D1 receptor activation in the SNr. In addition, we didn’t find an inhibitory action of D3 receptor stimulation on either GABA release or cAMP production even in those terminals where D4 receptor stimulation inhibits GABA release.

376

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

Fig. 6. The D3 selective agonist PD 128,907 does not modify spontaneous GABA release in neurons that respond to the selective D4 agonist 168,077. A, B and C. A. Representative traces during the control period (top), and the separate applications of PD 128,907 (middle) and PD 168,077 (bottom). B. Cumulative probability distributions for inter-event interval (top) and current amplitude (bottom) of GABAergic mIPSCs shown in (A). C. Dot plots summarize (n ¼ 8) results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each individual condition in a single experiment are connected by lines ***p < 0.001 with respect to control, ###p < 0.001 with respect to PD 128,907. D. Representative tracings during the control period (top), and the applications of PD 128,907 together with PD 168,077 (middle) and PD 168,077 alone (bottom). E. Cumulative probability distributions for inter-event interval (top) and current amplitude (bottom) of GABAergic mIPSCs shown in (D). F. Dot plots summarize (n ¼ 8) results of mIPSC frequency (top) and amplitude (bottom) determinations. Values for each individual condition in a single experiment are connected by lines. ***p < 0.001, with respect to control, n.s., not significant.

4.1. D3 and D1 receptors are colocalized in striatonigral terminals The observation that activation of dopamine D3 receptors increased frequency of GABAergic mIPSCs without changing their amplitude implies that the effect is produced by enhanced transmitter release and not by an effect on the postsynaptic membrane. The finding that D1 and D3 receptors are colocalized in synaptosomes of the SNr (Fig. 1) taken together with the finding that D3 receptor activation stimulate GABA release and cAMP production only when coactivated with D1 receptors, strongly implies that both receptors are located in the same population of terminals. These terminals are, almost certainly, the projections of SPNs because dopamine receptor types in the GABAergic terminals of the SNr are anatomically segregated: dopamine D1 receptors are found in striatonigral projections, while dopamine D4 receptors are confined to pallidonigral terminals (Acosta-Garcia et al., 2009). Electrophysiological evidence is in agreement with this conclusion (Aceves et al., 2011). Other evidence is also in agreement with this proposal, mRNAs determinations show that the messenger for D1 and D3 receptors

are selectively coexpressed in substance P positive SPNs (Surmeier et al., 1996). Also, D3 protein is found in SNr, but not its mRNA, further suggesting an external origin of nigral D3 receptors (Diaz et al., 1995). Thus our observations taken together with other findings in the literature show that the colocalization of D1 and D3 receptors in striatonigral neurons has a well-defined role that hitherto hadn’t been identified (Gerfen and Surmeier, 2011). 4.2. Atypical signaling by D3 receptors In our experiments the effects of D3 dopamine receptor stimulation are atypical for a member of the D2 family of dopamine receptors. In general, it is agreed that D2 receptors preferentially interact with Gi/o proteins inhibiting cAMP formation (Missale et al., 1998). Studies, mainly in the limbic system, show that activation of D3 receptors act as typical members of the D2 group of receptors reducing signaling through inhibition of the cAMP/PKA cascade (Schwartz et al., 1998). Schwartz et al. (1998), however, indicated that signaling by D3 receptors can occur both through the typical coupling with Gi/o proteins and through synergism with D1 receptors. The present results and our parallel determinations of

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

[3H] GABA release (Avalos et al., 2010), are the first to show that coactivation of D1 and D3 receptors stimulate both the production of cAMP (Fig. 4) and transmitter release (Figs. 2 and 3) in presynaptic nerve endings. Our immunoprecipitation determinations in the SNr (Fig. 1) indicate that D1 and D3 receptors are expressed together. Prior studies using striatal proteins showed a similar pattern of coimmunoprecipitation (Fiorentini et al., 2008). Other studies showed that D3 receptor stimulation increased the affinity of D1 receptors in membranes extracted from the striatum (Marcellino et al., 2008). Based on these findings and observations in transfected cells Marcellino et al. (2008) and Fiorentini et al. (2008) have proposed that the formation of D1eD3 receptor heteromers may be the mechanism of their functional interactions. However, there are investigators that consider that the conclusion that dimerization is the mechanism of pharmacological interactions between GPCRs has to be taken with caution. Their concern arises because changes in downstream signaling pathways may also explain the findings (Chabre et al., 2009). 4.2.1. Role of endogenous dopamine on the effects of receptor antagonists on miniature IPSCS and evoked IPSCs In our experiments D1 antagonist SCH 23390 or D4 antagonist L 745,870 didn’t have effects on mIPSCs unless given concurrently with the corresponding agonist. In contrast, Aceves et al. (2011) showed that antagonists of D1 and D4 receptors modify eIPSCs in the absence of added agonists. Differences in the action of dopamine antagonists on mIPSCs and eIPSCs are not without precedent (Radnikow and Misgeld, 1998). GABA release in the SNr is modulated by dopamine released by dendrites of the SNc (Aceves et al., 1995). Under certain experimental conditions, release of endogenous dopamine may be sufficiently enhanced to modulate synaptic transmission. Working in slices of the SNr, Radnikow and Misgeld (1998) showed that D1 antagonists given alone depress eIPSCs but do not affect mIPSCs. They proposed that when eIPSCs are elicited, dopamine in the synaptic cleft reaches sufficient levels to modulate GABA release from striatonigral terminals and while when action potential firing is blocked, dopamine levels are so low that they don’t affect mIPSCs. A similar role for endogenous dopamine has been suggested in the modulation of glutamate release in the SNr. Ibanez-Sandoval et al. (2006) found that eEPSCs in the SNr are modified by D1 and D2 antagonists in the absence of added agonists and concluded that under their experimental conditions endogenous dopamine modulates glutamate release. In agreement with the conclusion of Radnikow and Misgeld (1998), we suggest that the different effects on mIPSCs (present results), and eIPSCs (Aceves et al., 2011) of D1 and D4 antagonists are due to different levels of dopamine in the synaptic cleft. 4.3. D4 receptors in pallidonigral projections. Absence of D3 receptor inhibitory effects We found that the selective D4 agonist PD consistently reduced mIPSCs frequency without affecting their amplitude; this effect was blocked by the selective D4 antagonist L 745,870 (Fig. 5). As in previous experiments (Acosta-Garcia et al., 2009), lesions of the GP eliminated the responses produced by D4 receptor activation. These results are in agreement with the conclusion of AcostaGarcia et al. (2009) who identified D4 responding terminals as pallidonigral projections. Similar conclusions were also reached by Aceves et al., (2011). In contrast with the consistent inhibitory effects of the D4 agonists (Figs. 5 and 6) we didn’t find an inhibitory effect of the D3 agonist PD 128,907 on mIPSCs, even in cells that responded to

377

D4 agonist PD 168,077 (Fig. 6) or on the forskolin stimulated cAMP production. In contrast, Aceves et al. (2011) proposed that activation of D3 receptors inhibits GABA release from pallidonigral terminals. They base their conclusion on two of their findings: the agonist quinelarone decreased while the antagonist U-9914A increased the amplitude of the eIPSC produced by pallidonigral projection stimulation. The literature for quinelarone (Newman-Tancredi et al., 1997; Coldwell et al., 1999; Newman-Tancredi et al., 2002,) indicates that the concentration (500 nM) used by Aceves et al. (2011) activates not only D3 receptors but also D2 and D4 receptors. The antagonist U 99194A is more selective (Waters et al., 1993). In addition, other studies have failed to find control of palidal neurons by D3 receptors. Shin et al. (2003) and Hernandez et al. (2006) found that electrophysiological properties of these cells are modulated by D4 receptors but not by D2/D3 receptors. Determinations of [3H] GABA release in slices of the SNr also failed to detect an inhibitory effect of D3 receptor stimulation on GABA release in the SNr (Avalos et al., 2010). The evidence of molecular biology studies points out in the same direction since expression of D3 receptors in pallidal structures is either weak or nonexistent. (Sokoloff et al., 1990; Bouthenet et al., 1991; Levesque et al., 1992; Levant, 1998; Ridray et al., 1998). As discussed above, effects of dopamine antagonists administered in the absence of added agonists imply significant release of endogenous dopamine. One possibility is that U 99194A may antagonize the effects of endogenous dopamine on D3 receptors located outside pallidonigral terminals themselves. A site for such receptors could be the GABAergic projections that the limbic system sends to the SNr (Dray and Oakley, 1977; Walass and Fonnum, 1980; Strahlendorf and Barnes, 1983). D3 receptors are expressed at high density in the limbic system where they inhibit the PKA cascade (Schwartz et al., 1998). Thus, it is possible that the electrical stimuli in the very large slices used by Aceves et al. (2011) might have been effective in activating such fiber population. Evidently, the elucidation of the mechanism of action of U991494A on IPSCs evoked by electrical stimulation has still to be resolved by additional experiments. 5. Conclusions The present results, taken together with findings on the segregation of D1 and D4 receptors in the GABAergic terminals of the SNr (Acosta-Garcia et al., 2009; Aceves et al., 2011) and coexpression of D1 and D3 receptors in substance P positive SPNs (Surmeier et al., 1996), show that D3 receptors are functionally active in striatonigral terminals potentiating the stimulatory effects of D1 receptors. In contrast mIPSCs from D4 expressing terminals, presumably of pallidonigral origin, are not inhibited by D3 receptor activation. Acknowledgments The work was supported by a grant (152326) from CONACyT (México) to BF. References Aceves, J., Floran, B., Sierra, A., Mariscal, S., 1995. D-1 receptor mediated modulation of the release of gamma-aminobutyric acid by endogenous dopamine in the basal ganglia of the rat. Prog. Neuropsychopharmacol. Biol. Psychiatry 19, 727e 739. Aceves, J.J., Rueda-Orozco, P.E., Hernández, R., Plata, V., Ibañez-Sandoval, E., Galarraga, E., Bargas, J., 2011. Dopaminergic presynaptic modulation of nigral afferents: its role in the generation of recurrent bursting in substantia nigra pars reticulate neurons. Front. Syst. Neursosci. 5, 1e10. Acosta-Garcia, J., Hernandez-Chan, N., Paz-Bermudez, F., Sierra, A., Erlij, D., Aceves, J., Floran, B., 2009. D4 and D1 dopamine receptors modulate GABA

378

R. Cruz-Trujillo et al. / Neuropharmacology 67 (2013) 370e378

release in the substantia nigra pars reticulata of the rat. Neuropharmacology 57, 725e730. Alexander, S.P., 1995. The measurement of cyclic AMP levels in biological preparations. Methods Mol. Biol. 41, 79e89. Avalos, A., Rangel, C., Silva, I., Cruz, R., Aceves, J., Erlij, D., Floran, B.G., 2010. Intracelular Ca2þ regulates dopamine D3 receptor effects on GABA reléase by striatonigral terminals of the rat via activation of CAMKII. Neuroscience. Abstr. 741.12. Bouthenet, M.L., Souil, E., Martres, M.P., Sokoloff, P., Giros, B., Schwartz, J.C., 1991. Localization of dopamine D3 receptor mRNA in the rat brain using in situ hybridization histochemistry: comparison with dopamine D2 receptor mRNA. Brain Res. 564, 203e219. Chabre, M., Deterre, P., Antonny, B., 2009. The apparent cooperativity of some GPCRs does not necessarily imply dimerization. Trends Pharmacol. Sci. 30, 182e 187. Coldwell, M.C., Boyfield, I., Brown, A.M., Stemp, G., Middlemiss, D.N., 1999. Pharmacological characterization of extracellular acidification rate responses in human D2 (long), D3 and D4.4 receptors expressed in Chinese hamster ovary cells. Br. J. Pharmacol. 127, 1135e1144. Connelly, W.M., Schulz, J.M., Lees, G., Reynolds, J.N., 2010. Differential short-term plasticity at convergent inhibitory synapses to the substantia nigra pars reticulata. J. Neurosci. 30, 14854e14861. Cruz-Trujillo, R., Sierra, A., Valdiosera, R., Erlij, D., Floran, B., Aceves, J., 2008. Dopamine modulates via D4 receptors pallido-nigral transmisión to substantia nigra. Neuroscience. Abstr. 274.15. Curtis, D.R., Duggan, A.W., Felix, D., Johnston, G.A., 1970. GABA, bicuculline and central inhibition. Nature 226, 1222e1224. Diaz, J., Levesque, D., Lammers, C.H., Griffon, N., Martres, M.P., Schwartz, J.C., Sokoloff, P., 1995. Phenotypical characterization of neurons expressing the dopamine D3 receptor in the rat brain. Neuroscience 65, 731e745. Dray, A., Oakley, N.R., 1977. Projections from nucleus accumbens to globus pallidus and substantia nigra in the rat. Experientia 34, 68e70. Everett, B., Senogles, S., 2010. D3 dopamine receptor signals to activation of phospholipase D through a complex with Rho. J. Neurochem. 112, 963e971. Fiorentini, C., Busi, C., Gorruso, E., Gotti, C., Spano, P., Missale, C., 2008. Reciprocal regulation of dopamine D1 and D3 receptor function and trafficking by heterodimerization. Mol. Pharmacol. 74, 59e69. Gasca-Martinez, D., Hernandez, A., Sierra, A., Valdiosera, R., Anaya-Martinez, V., Floran, B., Erlij, D., Aceves, J., 2010. Dopamine inhibits GABA transmission from the globus pallidus to the thalamic reticular nucleus via presynaptic D4 receptors. Neuroscience 169, 1672e1681. Gerfen, C.R., Surmeier, D.J., 2011. Modulation of striatal projection systems by dopamine. Annu. Rev. Neurosci. 34, 441e466. Hernandez, A., Ibanez-Sandoval, O., Sierra, A., Valdiosera, R., Tapia, D., Anaya, V., Galarraga, E., Bargas, J., Aceves, J., 2006. Control of the subthalamic innervation of the rat globus pallidus by D2/3 and D4 dopamine receptors. J. Neurophysiol. 96, 2877e2888. Ibanez-Sandoval, O., Hernandez, A., Floran, B., Galarraga, E., Tapia, D., Valdiosera, R., Erlij, D., Aceves, J., Bargas, J., 2006. Control of the subthalamic innervation of substantia nigra pars reticulata by D1 and D2 dopamine receptors. J. Neurophysiol. 95, 1800e1811. Insel, P.A., Ostrom, R.S., 2003. Forskolin as a tool for examining adenylyl cyclase expression, regulation, and G protein signaling. Cell. Mol. Neurobiol. 23, 305e314. Levant, B., 1998. Differential distribution of D3 dopamine receptors in the brains of several mammalian species. Brain Res. 800, 269e274. Levesque, D., Diaz, J., Pilon, C., Martres, M.P., Giros, B., Souil, E., Schott, D., Morgat, J.L., Schwartz, J.C., Sokoloff, P., 1992. Identification, characterization, and localization of the dopamine D3 receptor in rat brain using 7-[3H]hydroxy-N, N-di-n-propyl-2-aminotetralin. Proc. Natl. Acad. Sci. U. S. A. 89, 8155e8159. Liu, X.Y., Mao, L.M., Zhang, G.C., Papasian, C.J., Fibuch, E.E., Lan, H.X., Zhou, H.F., Xu, M., Wang, J.Q., 2009. Activity-dependent modulation of limbic dopamine D3 receptors by CaMKII. Neuron 61, 425e438.

Marcellino, D., Ferre, S., Casado, V., Cortes, A., Le Foll, B., Mazzola, C., Drago, F., Saur, O., Stark, H., Soriano, A., Barnes, C., Goldberg, S.R., Lluis, C., Fuxe, K., Franco, R., 2008. Identification of dopamine D1eD3 receptor heteromers. Indications for a role of synergistic D1eD3 receptor interactions in the striatum. J. Biol. Chem. 283, 26016e26025. Missale, C., Nash, S.R., Robinson, S.W., Jaber, M., Caron, M.G., 1998. Dopamine receptors: from structure to function. Physiol. Rev. 78, 189e225. Nava-Asbell, C., Paz-Bermudez, F., Erlij, D., Aceves, J., Floran, B., 2007. GABA(B) receptor activation inhibits dopamine D1 receptor-mediated facilitation of GABA release in substantia nigra pars reticulata. Neuropharmacology 53, 631e637. Newman-Tancredi, A., Audinot, V., Chaput, C., Verriele, L., Millan, M.J., 1997. [35S] Guanosine-5’-O-(3-thio)triphosphate binding as a measure of efficacy at human recombinant dopamine D4.4 receptors: actions of antiparkinsonian and antipsychotic agents. J. Pharmacol. Exp. Ther. 282, 181e191. Newman-Tancredi, A., Cussac, D., Audinot, V., Nicolas, J.P., De Ceuninck, F., Boutin, J.A., Millan, M.J., 2002. Differential actions of antiparkinson agents at multiple classes of monoaminergic receptor. II. Agonist and antagonist properties at subtypes of dopamine D(2)-like receptor and alpha(1)/alpha(2)adrenoceptor. J. Pharmacol. Exp. Ther. 303, 805e814. Nimchinsky, E.A., Hof, P.R., Janssen, W.G., Morrison, J.H., Schmauss, C., 1997. Expression of dopamine D3 receptor dimers and tetramers in brain and in transfected cells. J. Biol. Chem. 272, 29229e29237. Parent, A., Hazrati, L.N., 1995. Functional anatomy of the basal ganglia. I. The corticobasal ganglia-thalamo-cortical loop. Brain Res. Brain Res. Rev. 20, 91e127. Radnikow, G., Misgeld, U., 1998. Dopamine D1 receptors facilitate GABAA synaptic currents in the rat substantia nigra pars reticulata. J. Neurosci. 18, 2009e2016. Rangel-Barajas, C., Silva, I., Garcia-Ramirez, M., Sanchez-Lemus, E., Floran, L., Aceves, J., Erlij, D., Floran, B., 2008. 6-OHDA-induced hemiparkinsonism and chronic L-DOPA treatment increase dopamine D1-stimulated-GABA release and [(3)H]-cAMP production in substantia nigra pars reticulata of the rat. Neuropharmacology 55, 704e711. Rangel-Barajas, C., Silva, I., Lopez-Santiago, L.M., Aceves, J., Erlij, D., Floran, B., 2011. L-DOPA-induced dyskinesia in hemiparkinsonian rats is associated with up-regulation of adenylyl cyclase type V/VI and increased GABA release in the substantia nigra reticulata. Neurobiol. Dis. 41, 51e61. Ridray, S., Griffon, N., Mignon, V., Souil, E., Carboni, S., Diaz, J., Schwartz, J.C., Sokoloff, P., 1998. Coexpression of dopamine D1 and D3 receptors in islands of Calleja and shell of nucleus accumbens of the rat: opposite and synergistic functional interactions. Eur. J. Neurosci. 10, 1676e1686. Schwartz, J.C., Diaz, J., Bordet, R., Griffon, N., Perachon, S., Pilon, C., Ridray, S., Sokoloff, P., 1998. Functional implications of multiple dopamine receptor subtypes: the D1/D3 receptor coexistence. Brain Res. Brain Res. Rev. 26, 236e 242. Shin, R.M., Masuda, M., Miura, M., Sano, H., Shirasawa, T., Song, W.J., Kobayashi, K., Aosaki, T., 2003. Dopamine D4 receptor-induced postsynaptic inhibition of GABAergic currents in mouse globus pallidus neurons. J. Neurosci. 23, 11662e 11672. Sokoloff, P., Giros, B., Martres, M.P., Bouthenet, M.L., Schwartz, J.C., 1990. Molecular cloning and characterization of a novel dopamine receptor (D3) as a target for neuroleptics. Nature 347, 146e151. Strahlendorf, H.K., Barnes, C.D., 1983. Control of substantia nigra pars reticulate neurons by nucleus accumbens. Brain Res. Bull. 11, 259e263. Surmeier, D.J., Song, W.J., Yan, Z., 1996. Coordinated expression of dopamine receptors in neostriatal medium spiny neurons. J. Neurosci. 16, 6579e6591. Walass, I., Fonnum, F., 1980. Biochemical evidence for g-aminobutyrate containing fibres from the nucleus accumbens to substantia nigra and ventral tegmental area in the rat. Neuroscience 5, 63e72. Waters, N., Svensson, K., Haadsma-Svensson, S.R., Smith, M.W., Carlsson, A., 1993. The dopamine D3-receptor: a postsynaptic receptor inhibitory on rat locomotor activity. J. Neural Transm. Gen. Sect. 94, 11e19.