European Journal of Pharmacology 649 (2010) 161–167

Contents lists available at ScienceDirect

European Journal of Pharmacology

j ourna l homepage: www.e lsev ie r.com/ locate /e jphar

Neuropharmacology and Analgesia

GABAB receptors modulate depolarization-stimulated [3H]glutamate release in slicesof the pars reticulata of the rat substantia nigra

Hernán Cortés c, Francisco Paz a, David Erlij b, Jorge Aceves a, Benjamín Florán a,⁎a Departamento de Fisiología, Biofísica y Neurociencias, Centro de Investigación y de Estudios Avanzados del Instituto Politécnico Nacional, Apartado Postal 14-740,07000 México D.F. Méxicob Department of Physiology, SUNY Downstate Medical Center, Brooklyn, NY 11203, USAc Departamento de Farmacología, CINVESTAV-IPN, Mexico

⁎ Corresponding author. Tel.: +52 55 57473800x513E-mail address: bfloran@fisio.cinvestav.mx (B. Florán

0014-2999/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.ejphar.2010.09.024

a b s t r a c t

a r t i c l e i n f o

Article history:Received 4 June 2010Received in revised form 28 July 2010Accepted 7 September 2010Available online 19 September 2010

Keywords:D1-GABAB interactionGABAB heteroreceptorTurning behaviorSubthalamonigral pathwaySubthalamonigral transmission

GABAB receptors decrease the release of GABA from the striatal terminals within the pars reticulata of thesubstantia nigra by opposing the increase in the release caused by dopamine D1 receptors. The dopamine D1

receptors also increase the release of glutamate from subthalamic terminals in the pars reticulata. BecauseGABAB receptors decrease the glutamate release from these terminals, we have explored if the effect of GABAB

receptors also opposed the effect of the dopamine D1 receptors. The effect of baclofen, a selectiveGABAB-receptor agonist, was tested on the release of [3H]glutamate caused by highly (40 mM) concentratedK+ solutions in slices of the pars reticulata. Baclofen decreased (the concentration causing 50% inhibition, IC50,was 8.15 μM) the increase in the release of the [3H]glutamate caused by the dopamine D1 receptors and it alsodecreased (IC50 was 0.51 μM) this release in the absence of the activation of the dopamine D1 receptors. TheGABAB receptors appear then to inhibit glutamate release in twoways; one dependent on the activation of thedopamine D1 receptors and the other independent of such activation. The protein kinase A-inhibitor H89blocked the increase in the release of the [3H]glutamate caused by the dopamine D1 receptors, though it didnot block the dopamine D1 receptor-independent baclofen inhibition of the release. This finding indicates thatthis inhibition was not via the protein kinase A signal-transduction pathway. N-ethylmaleimide, an alkylatingagent that inactivates pertussis toxin-sensitive Gi proteins, eliminated both the dopamine D1 receptor-dependent and -independent baclofen inhibition, showing that both were mediated by these proteins. Theinjection of baclofen into the pars reticulata of unanesthetized rats caused contralateral rotation, suggesting areduced glutamate release from the subthalamic terminals, thereby stopping the inhibition of the premotorthalamic nuclei, causing locomotion. Our data suggest that GABAB receptors restrain the excitatory input fromthe subthalamic nucleus and stimulate motor behavior.

7; fax: +52 55 7473754.).

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

The pars reticulata of the substantia nigra is rich in dopamine D1

receptors,which are located at the GABAergic striatonigral (Porceddu etal., 1986; Altar and Hauser, 1987; Barone et al., 1987) and theglutamatergic subthalamonigral (Rosales et al., 1997; Ibañez-Sandovalet al., 2006) terminals. Under physiological conditions, these receptorsare activated by dopamine released from dendrites of neurons of thepars compacta of the substantia nigra (Chéramy et al., 1981). Byactivating the dopamine D1 receptors, dopamine stimulates the releaseof GABA from the striatonigral terminals (Floran et al., 1990; Aceves etal., 1995; Rosales et al., 1997) and the release of glutamate fromsubthalamonigral terminals (Rosales et al., 1997), thereby enhancing

the striatonigral (Misgeld, 2004) and subthalamonigral (Ibañez-Sandoval et al., 2006) transmission.

GABAB receptors are located on the striatonigral, pallidonigral, andsubthalamonigral axon terminals (Charara et al., 2000; Boyes andBolam, 2003). They are also present on dopaminergic dendrites withinthe pars reticulata of the substantia nigra (Westernink et al., 1992;Garcia et al., 1997) and, at rather low density, in the soma anddendrites of the neurons of the substantia nigra reticulata (Charara etal., 2000; Boyes and Bolam, 2003). Shen and Johnson (1997) showedthat the selective GABAB-receptor agonist baclofen by activatingpresynaptic GABAB-receptors inhibited the release of glutamate in thepars reticulata. The effect of baclofen may have been mediated by theactivation of GABAB-receptors present on the subthalamic afferents tothe substantia nigra pars reticulata (Boyes and Bolam, 2003).

We have shown that GABAB receptors inhibit the release of GABAfrom the striatonigral axon terminals by preventing the increase inthe release caused by the activation of the dopamine D1 receptors(Nava-Asbell et al., 2007). Here we explored if the GABAB receptors

162 H. Cortés et al. / European Journal of Pharmacology 649 (2010) 161–167

also inhibit the release of glutamate by opposing the facilitation of theglutamate release caused by the activation of the dopamine D1

receptors. In addition, we explored if the activation of the GABAB

receptors in the substantia nigra pars reticulata in the unanesthetizedrat affects the motor activity, as judged by the turning behavior causeby the injection of the GABAB-agonist baclofen into the substantianigra pars reticulata.

2. Materials and methods

2.1. Animals

Brain slices were obtained from male Wistar rats weighing 180 to220 g maintained and handled according to the guidelines of theCINVESTAV-IPN Animal Care Committee, taking all efforts tominimizesuffering and the number of animals used. To avoid the effects ofendogenous dopamine, the rats were treated with reserpine(10 mg/kg, ip) 18 h before the experiments. This reserpine treatmenthas been shown to reduce dopamine levels by 95% in the striatum andby 69% in the substantia nigra pars reticulata (Garcia et al., 1997) andit appears not to affect GABAB receptor pharmacology because it didnot modify the IC50 for baclofen to inhibit GABA release caused by thehigh K+ solution (Floran et al., 1988; Nava-Asbell et al., 2007).

2.2. Preparation of the slices of the pars reticulata of the substantia nigra

After decapitating the rat, the brain was removed and immersed inice-cold artificial cerebrospinal fluid (aCSF) composed of (mM) NaCl,118.25; KCl, 1.75; MgSO4, 1; KH2PO4, 1; NaHCO3, 25; CaCl2, 2; andD-glucose, 10. The brain slices (300-μm thick) containing thesubstantia nigra were obtained with a vibroslicer (Campden Inc.,Cambridge, UK), and were transferred to ice-cold slides. By using astereoscopic microscope, the substantia nigra pars reticulata, identi-fied according to the atlas of Paxinos and Watson (1997), wasmicrodissected.

2.3. [3H]Glutamate release

The experiments of glutamic acid release were done using amethodadapted from Mitchell and Doggett (1980). Once microdissected, theslices were pooled and allowed to equilibrate for 30 min in aCSFmaintained at 37 °C and gassed continuously with O2–CO2 (95:5, v:v).The tissues were incubated for 30 min with 100 nM [3H]glutamate in2-mLaCSF containing200 μMaminooxyacetic acid (to inhibit glutamatedecarboxylase and prevent the conversion of glutamate to GABA)(Kofalvi et al., 2005) and 200 μM dihydrokainic acid (to prevent theuptake of [3H]glutamate by astrocytes) (Kawahara et al., 2002;Bernardinelli and Chatton, 2008). Dihydrokainic acid was present inthemediumonly during the incubation period. At the endof this period,the excess radiolabel was removed by washing twice with aCSF. TheCa2+-free solutions were prepared by substituting all the Ca2+ withMg2+.

The slices were then apportioned randomly between the chambers(80-μL volume) of a superfusion apparatus (20 superfusion chambersin parallel) and superfused with the aCSF at a flow rate of 0.5 mL/min.Each chamber contained three to four slices. The design of thesuperfusion chambers was as described by Aceves and Cuello (1981)except that the electrodes for electrical stimulation were omitted. Amultichannel peristaltic pump (Watson Inc., USA)was used to perfusethe slices. The aCSF for superfusion was contained in reservoirs placedin a constant temperature bath maintained at 38 °C. When the aCSFreached the chambers, its temperature was 36 to 37 °C. To wash outthe [3H]glutamate trapped in the interstitial space, the slices weresuperfused with normal aCSF for 20 min before collecting thefractions for counting radioactivity. Fractions of the superfusatewere collected in a fraction collector every 4 min (each fraction was

2 mL). Four fractions were first collected to determine the releaseunder basal conditions, then [K+] in the aCSF was increased to 40 mM(composition of this solution in mM was NaCl, 81.25; KCl, 38.75;MgSO4, 1; KH2PO4, 1.25; NaHCO3, 25; CaCl2, 2; and D-glucose, 10). Sixmore fractions were collected of the high [K+] medium. All drugswere added to the medium at fraction 2, before changing thesuperfusion to the high [K+] medium, to explore the effects on thebasal release. The radioactivity released into the superfusion mediumin each fraction was measured by liquid scintillation counting. Todetermine the total amount of tritium remaining in the tissue, theslices were collected, treated with 1 mL of 1 M HCl and allowed tostand for 5 h before adding the scintillator and counting theradioactivity in a scintillation counter.

2.4. Turning behavior

Locally bred male Wistar rats weighing 180 to 200 g wereanesthetized with chloral hydrate (350 mg/kg). A guide cannula (22gauge, 12-mm long) was placed into the substantia nigra parsreticulata of one side for the microinjections. The coordinates usedwere−7.3 mm posterior and 2.2 mm lateral to bregma and−7.5 mmfrom dura, with an angle of 34° with respect to the interaural line. Thecannula was secured in place with dental acrylic glued to the craniumand small stainless steel screws. A wire stylet was inserted into thecannula to prevent clogging.

For rotational behavior rats were tested 5 days after surgery. Asingle dose of baclofen (0.5 μg) or vehicle was injected in a finalvolume of 0.5 μL delivered during 2 min, with the cannula remainingin place three additional minutes. The number of turns per minutewas recorded for 90 min. Only one intracerebral injection (drug orvehicle) were made to any one animal.

At the end of the experiment, the animals were killed by anoverdose of chloral hydrate, and the brains removed and fixed in 10%Formalin for 24 h. Brain sections (150 μm) were obtained with thevibroslicer. Injection sites were assessed by the location of the cannulaon digitized images of the brain slices. These images were super-imposed onto schemes of the brain (Paxinos and Watson, 1997).Experiments with cannula locations not corresponding to thesubstantia nigra pars reticulatra were discarded.

2.5. Drugs

Aminooxyacetic acid (AAOA), R(+)-β-(aminomethyl)-4-cholor-obenzenepropanoic acid hydrochloride (baclofen), N-[2-(p-bromo-cinnamylamino) ethyl]-5-isoquinolinesulfonamide (H-89),2-carboxy-4-isopropyl-3-pyrrolidineacetic acid (dihydrokainicacid), R(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride (SCH 23390), N-ethyl-maleimide (NEM), d-N,α-dimethylphenethylamine hydrochloride(methamphetamine), R(+)-1-phenyl-2,3,4,5-tetrahydro-(1H)-3-benzazepine-7,8-diol HCl (SKF 38393), and reserpine were obtainedfrom Sigma (St. Louis, MO, USA). (RS)-3-amino-2-(4-chlorophenyl)propylsulfonic acid (saclofen) was purchased from Tocris CooksonInc. (Ballwin, MO. USA).

2.6. Data analysis

The radioactivity present in each fraction of perfusatewas calculatedas a fraction of the total radioactivity recovered from that chamber, i.e.the sum of the radioactivity from all fractions collected+the radioac-tivity in the tissue. The effect of drugs on the basal release of the[3H]glutamate was assessed by comparing the fractional release infraction 2 (immediately before exposure of the tissue to the drug) andfraction4 (immediately before exposure to40 mMK+), using thepairedStudent's t-test. Changes in high [K+]-generated [3H]glutamate releasewere assessedby comparing the area under the curvesof the increments

Fig. 1. The release of the [3H]glutamate from the substantia nigra pars reticulata slicescaused by a high concentration (40 mM) of K+ is Ca2+-dependent. (A) Release producedby the high [K+] medium from slices superfused with normal or Ca2+-free solutions. The[3H]glutamate release was expressed as the ratio of the fractional release in fraction four(seeMaterials andmethods). (B) The bar plots are the area under the curve of the increasein the release caused by the high [K+] solution. The area is indicated as AUC. The values arethemean of three independent experiments (5 replicates per experiment) *** Pb0.001 vs.normal Ca2+ (unpaired Student's t-test).

163H. Cortés et al. / European Journal of Pharmacology 649 (2010) 161–167

in the release produced by the exposure to the high [K+] solution, that isthe area between the first and last fractions collected after the change tohigh [K+]. To estimate the increments in the release, we assumed thatthe basal release of [3H]glutamate would remain practically unchangedat the level measured in the fraction immediately preceding thestimulation by the high concentration K+ solution. In experiments inwhich this was tested, the basal release in fraction 10 was 97%±4% ofthat in fraction 4, indicating aminimum change in the basal release. Thebasal fractional release was 0.0044 to 0.0068 times the total radioac-tivity. The release stimulated by the high K+ concentration was 0.0098to0.0115 times the total radioactivity. Thewithin-treatments variabilityin an experiment was greatly reduced by expressing the fractionalrelease as a ratio of the fractional release immediately before changingto the high [K+] medium, i.e. the release in fraction 4 was set to 1). Thesignificance of the effect of drugswas analyzed using a one-way analysisof variance followed by a Tukey–Kramermultiple-comparison post-hoctest using PrismGraphPad4.0 (GraphPad Software, SanDiego, CA, USA).

To determine the dose–response curves, the effects of five doses ofbaclofen were determined in triplicate in each experiment. Theexperiment was repeated three times. These data were fitted with thefour-parameter logistic equation using Prism Graph Pad Software 4.0,which was also used to calculate the IC50 (the concentration causing50% inhibition) and the confidence interval (CI). The significance ofthe effect of drug on the turning behavior was determined using aMann–Whitney U-test comparing the effect of the drug with saline asa control.

3. Results

3.1. Effect of Ca2+-free solutions on [3H]glutamate release caused byhigh K+ concentration solutions

As illustrated in Fig. 1, omitting Ca2+ from the medium decreasedthe release produced by the high [K+] solution by 84%±5% (Fig. 1B),but the basal release was not affected (compare the release infractions 1 and 2 with those in fractions 3 and 4 in Fig. 1A).

3.2. Effect of baclofen on the dopamine D1 receptor-facilitated [3H]glutamate release

As Fig. 2A shows, activation of the dopamine D1 receptors with SKF38393 increased the high [K+]-stimulated [3H]glutamate release by70%±18%. The GABAB-agonist baclofen not only blocked the increasein the release produced by the dopamine D1 agonist but in addition itinhibited the release by 44%±6% when measured under controlconditions. The effect of baclofen was specific, because it was blockedby the GABAB-antagonist saclofen, which by itself did not modify therelease.

To verify whether the GABAB receptors had an inhibitory effectthemselves, the effect of baclofen was studied in the absence of thestimulation of the release by dopamine D1 receptors. In this conditionbaclofen inhibited the high [K+]-stimulated [3H]glutamate release by45%±5% (Fig. 2B). The inhibition was specific because it was blockedby saclofen, which again itself had no effect. From the dose–responsecurves (Fig. 3), the estimated IC50 for the baclofen inhibition of theincrease in release produced by the dopamine D1 receptors was8.15 μM (CI 2.14 to 29.16 μM), whereas it was 0.51 μM (CI 0.31 to0.82 μM) for inhibiting the release in the absence of the activation ofthe dopamine D1 receptors.

3.3. Effect of baclofen in the presence of inhibition of protein kinase A

The increased depolarization-stimulated release of neurotransmit-ters causedby thedopamineD1 receptors is via the cAMP-protein kinaseA (PKA) signal-transduction pathway (Arias-Montano et al., 2007). Toexplore whether the dopamine D1-receptor-independent baclofen

inhibition of the release of glutamate was via the PKA, the effect ofbaclofen was studied in the presence of H89, a PKA inhibitor. As seen inFig. 4, the H89 blocked the increase in the release caused by theactivation of the dopamine D1 receptors with SKF 38393 but it did notblock the baclofen inhibition of the K+-stimulated [3H]glutamaterelease, indicating that the inhibition was not via the cAMP-PKA-signaltransduction. The addition of H89 alone did not modify the release.

3.4. Effect of baclofen in the presence of the inhibition of pertussis toxin-sensitive guanine nucleotide binding proteins

The GABAB receptors suppress the transmitter release throughthe guanine nucleotide binding proteins (G-proteins) inhibition ofthe Ca2+ currents (Dolphin and Scott, 1987). To explore theparticipation of the G-proteins in the baclofen inhibition of therelease of the glutamate, we tested if N-ethylmaleimide (NEM), asulfhydryl-alkylating agent that affects SH-containing proteins andinactivates pertussis toxin-sensitive G-proteins (Shapiro et al.,1994), prevented the inhibition of the glutamate release caused bythe GABAB receptors. As shown in Fig. 5, NEM eliminated thedopamine D1 receptor-dependent and the dopamine D1 receptor-independent baclofen inhibition of the high [K+]-stimulated [3H]glutamate release, indicating that two types of inhibition weremediated by the G-proteins.

Fig. 2. Effect of baclofen (100 μM) on the generated [3H]glutamate release from thesubstantia nigra pars reticulata slices in the presence (A) or absence (B) of the activation ofthe D1 dopamine receptors by SKF 38393 (1 μM). Baclofen inhibits the dopamineD1-stimulated and the non- dopamine D1-dependent [3H]glutamate release. In both cases,saclofen (1 μM) blocked the baclofen inhibition. Saclofen, by itself, had no effect. In eachcase, values are the mean±SE of three independent experiments (5 replicates perexperiment). In A, *** Pb0.001 vs. control, ## Pb0.01 vs. SKF 38393 and vs. control. In B, **Pb0.01 vs. control.

Fig. 3. The baclofen inhibition of the [3H]glutamate release from the substantia nigrapars reticulata slices is dose-dependent. The dose–response curves obtained in thepresence of SKF 38393 (1 μM) was obtained subtracting the values obtained in theabsence of the agonist from those obtained in its presence. In this manner the two typesof inhibition were isolated. The dose–response curves in the presence of SKF areindicated by the open circles; the curve in its absence, by closed circles. In the presenceof the agonist, the calculated IC50 was 8.15 μM (CI 2.14 to 29.16 μM); in the absence, itwas 0.51 μM (CI 0.31 to 0.82 μM).

Fig. 4. Effect of the PKA inhibitor H89 (10 μM) on the effect of SKF 38393 (1 μM) andbaclofen (100 μM) on the produced [3H]glutamate release from the substantia nigrapars reticulata slices. The H89 had no effect by itself. Bars are means±SE of threeindependent experiments (four replicates per experiment). ** Pb0.01 vs control.

164 H. Cortés et al. / European Journal of Pharmacology 649 (2010) 161–167

3.5. Effect of baclofen on motor behavior

The injection of baclofen into the substantia nigra pars reticulatareverses akinesia in reserpine-treated rats (Johnston and Duty, 2003).To explore whether the baclofen-generated locomotion was becausethe rats were treated with reserpine, we explored whether the sameeffect was produced by the injection of baclofen into the substantianigra pars reticulata of normal rats, where endogenous dopamine ispresent and being able to exert a tonic activation of dopamine D1

receptors. As shown in Fig. 6, baclofen unilaterally injected into thesubstantia nigra pars reticulta produced contralateral turning (88±13turns in 80 min), indicating that the GABAB receptors within thesubstantia nigra pars reticulata stimulate locomotion even in normalrats.

4. Discussion

GABAB receptors on the glutamatergic terminals in the substantianigra pars reticulata are thought to function as heteroreceptors,

regulating glutamate release, and thus regulating the strength of theexcitatory transmission to the neurons of the substantia nigrareticulata. Consistent with this hypothesis, our results have shownthat GABAB receptors regulate, via the G-proteins, the depolarization-stimulated [3H]glutamate release, first by inhibiting the facilitation ofglutamate release caused by the dopamine D1 receptors and second byinhibiting the release of the amino acid via a dopamine D1 receptor-independent mechanism. The inhibitory effect of the GABAB receptorson glutamate release appears to promote locomotion as judged by thecontralateral turning caused by the injection of the GABAB receptoragonist baclofen in the substantia nigra pars reticulata.

4.1. Source of the [3Hglutamate release

The depolarization caused by the high K+ concentration releasesglutamate from the nerve terminals and glial cells (Bernath, 1992).Because the release of the glutamate from the glial cells is notCa2+-dependent (Bernath, 1992), the strong Ca2+ dependency

Fig. 5. The NEM eliminates the baclofen inhibition of the high [K+]-generated[3H]glutamate release from the substantia nigra pars reticulata slices. Theconcentration of the drugs is NEM, 10 μM; baclofen, 100 μM; SKF 38393, 1 μM.Values are the mean±SE of three independent experiments (3 replicates perexperiment). ** Pb0.01 vs. control.

165H. Cortés et al. / European Journal of Pharmacology 649 (2010) 161–167

(Fig. 1) suggests that, under the present conditions, the [3H]glutamatewas released mainly from the nerve terminals. The presence ofdihydrokainic acid during incubation of the slices would have reducedthe uptake of the [3H]glutamate into the glial cells (Kawahara et al.,2002; Bernardinelli and Chatton, 2008), thereby reducing the releasefrom glial cells caused by the high K+ solution. The reversed activity ofthe glutamate transporter is not likely to have caused the release of the[3H]glutamate because the glutamate transport is not Ca2+ dependent(Szatkowski et al., 1990), further supporting the idea that the releasewas via the Ca2+-dependent exocytosis.

Although the substantia nigra pars reticulata receives glutamateprojections from the cortex (Naito and Kita, 1994), the pedunculo-pontine nucleus (Takakusaki et al., 1996), and the zona incerta (Heiseand Mitrofanis, 2004), the drastic fall (about 80%) in glutamate levelsin the substantia nigra pars reticulata after forming the lesion of thesubthalamic neurons with kainic acid (Rosales et al., 1997) indicates amuch higher density of subthalamic terminals, suggesting that theseterminals were the main source of the released [3H]glutamate.

Fig. 6. Baclofen injected into the substantia nigra pars reticulata promotes locomotion.Baclofen (0.5 μg/0.5 μL) was injected into the substantia nigra pars reticulata of theright side of normal rats. The animals received either baclofen or saline only once. Thevalues are the mean±SE of six rats in each case. Significant difference (Pb0.05) fromcontrol values; Mann–Whitney U-test.

4.2. Inhibition of the dopamine D1 receptor-mediated facilitation of the[3H]glutamate release by the GABAB receptors

Although it was known that the dopamine D1 receptors on thesubthalamic afferents in the substantia nigra pars reticulata facilitatethe glutamate release (Rosales et al., 1997) and the subthalamonigraltransmission (Ibañez-Sandoval et al., 2006), and that the GABAB

receptors inhibit glutamatergic transmission (Shen and Johnson,1997), it was unknown if the two types of receptors interact. Ourresults show that the GABAB receptors interact with the dopamine D1

receptors preventing the increase in glutamate release caused by thedopamine D1 receptors (Fig. 2). The two receptors also interact inmodulating the GABA release from the striatonigral terminals (Nava-Asbell et al., 2007). In both cases the effect of the activation of theGABAB receptors is via the G-proteins, most probably through the αi

subunit, as shown by the elimination of the baclofen inhibition by theNEM (Shapiro et al., 1994). The dopamine D1 receptors stimulateadenylyl cyclase through the Gαs proteins thereby, via the PKA,opening the P–Q type calcium channels, thus increasing theneurotransmitter release (Arias-Montano et al., 2007). By inhibitingadenylyl cyclase activity, the GABAB receptors would prevent thestimulation of the enzyme by the dopamine D1 receptors. Blockingactivation of the PKA with H89 eliminates the dopamine D1 receptor-mediated facilitation of the release of both glutamate and GABA,indicating a similar signalization pathway for both. In contrast, thedopamine D1 receptor-independent inhibition produced by theGABAB receptors was not affected by H89 (Fig. 4), showing that itwas not mediated via a PKA signaling.

The IC50 for the baclofen inhibition of glutamate release (8.15 μM)is slightly greater than that for inhibiting GABA release (1 μM; Nava-Asbell et al., 2007). Shen and Johnson (1997) found a similar IC50 forinhibiting both the GABA and glutamate transmission (0.6 and0.78 μM). Because the consequence of the inhibition of the GABArelease on the activity of the pars reticulata neurons would beopposed to that of the inhibition of the glutamate release, which effectwould predominate? It is possible that the activation of the GABAB

receptors in the GABA and glutamate terminals does not occursimultaneously. In addition, because the effect of the GABAB receptorsrequires depolarization of the axon terminals (Nava-Asbell et al.,2007; our results), we would expect that the inhibition is exerted onthe pathway being active at a given moment. In the absence ofdopamine, as in Parkinson's Disease, only the inhibitory effect of theGABAB receptors on glutamate release would operate, because theactivity of the striatonigral projection is reduced whereas the activityof the subthalamic one is enhanced (DeLong, 1990).

4.3. Dopamine D1 receptor-independent inhibition of glutamate releaseby GABAB receptors

Baclofen inhibited the release of the [3H]glutamate in the absence ofthe activation of the dopamine D1 receptors, suggesting that the GABAB

receptors per se inhibit the release of glutamate. As in the inhibition ofthe release of GABA, the inhibition was eliminated by the NEM (Fig. 5),indicating that it was also mediated by the G-proteins. The presynapticGABAB receptors appear to suppress the transmitter release through theG-protein-coupled inhibition of calcium currents (Takahashi et al.,1998). Apparently the βγ complex of the heterotrimeric G-proteinsinteract with the calcium channel α1 subunit thereby suppressing theactivity of the channels (De Waard et al., 1997). Accordingly, thedopamine D1 receptor-independent inhibition of glutamate release bythe GABAB receptorswould bemediated by the βγ complexwhereas, asdescribed above, the dopamine D1 receptor-dependent inhibition ofrelease would be via the αi subunit of the G-proteins.

An important issue that remains is the mechanism by which thesepresynaptic heteroreceptors are activated. The fact that the axo-axonic synapses are rare in the basal ganglia rule out the possibility of

166 H. Cortés et al. / European Journal of Pharmacology 649 (2010) 161–167

a direct synaptic release of the transmitter. It is more likely that theyare activated via a volume transmission. Once released, GABA woulddiffuse out of the synaptic cleft and activate the extra- and presynapticGABAB heteroreceptors. The released GABA can originate from anumber of sources; striatonigral and striatopallidal terminals, recur-rent collaterals from the nigrothalamic neurons, and from the GABAinterneurons.

Synaptic inputs in vivo may be able to control the release of GABAand glutamate such that the presynaptic inhibition does not occursimultaneously at both the glutamate and GABA nerve terminals. Thatbaclofen generates contralateral turning would indicate that theactivation of the heteroreceptors would predominate over that of theautoreceptors on striatonigral terminals. Two conditions would favorthe activation of the heteroreceptors; the higher affinity of theheteroreceptor, specially those receptors inhibiting the release ofglutamate independently of the activation of dopamine D1 receptors(IC50=0.5 μM) over that of the autoreceptor on striatonigralterminals (IC50=3.6 μM; Nava-Asbell et al., 2007) and the activityof the subthalamonigral pathway derived from the spontaneous firingof the subthalamic neurons. The GABAB receptors mediating the directeffect would be also activated first that those mediating the inhibitionof the dopamine D1-receptor-mediated increase in glutamate release(IC=8.15 μM). Under conditions in vitro, it seems that GABA is notbeing tonically released from the GABA terminals because theaddition of saclofen did not affect the release of glutamate.

4.4. Pathophysiological implications

An increased output from the substantia nigra pars reticulataneurons appears to determine rigidity (Double and Crocker, 1995)and bradykinesia of Parkinson's disease (DeLong, 1990; Robertson,1992), whereas a decreased output from the substantia nigra parsreticulata is linked to hyperkinetic disorders (DeLong, 1990).Modulation of the glutamate release in the substantia nigra parsreticulata could be useful in the treatment of these disorders. Thepresynaptic inhibition of glutamate release from the subthalamoni-gral pathway would be expected to reduce the output from thesubstantia nigra pars reticulata neurons, thus improving Parkinson-ism (Chesselet, 1996). Here we have found that the intranigraladministration of baclofen promotes locomotion, which is consistentwith a reduced glutamate release from the subthalamonigralterminals. By blocking the dopamine D1 receptor-mediated stimula-tion of the release of GABA from the striatonigral terminals (Nava-Asbell et al., 2007), the GABAB receptors may be able to controlhyperkinetic disorders.

In conclusion, our work demonstrates that the GABAB receptors,most likely on the subthalamic terminals within the substantia nigrapars reticulata, modulate glutamate release by opposing the increasein release produced by the activation of the dopamine D1 receptorsand by directly inhibiting the release of the excitatory amino acid. Theinhibition of the release of the glutamate would promote locomotion,reverse akinesia (Johnston and Duty, 2003), and regulate muscle tone(Hemsley et al., 2002).

Acknowledgements

This work was supported by a grant (50427Q) from Conacyt(Mexico) to J. A. Thanks to Dr. Ellis Glazier for editing this English-language text.

References

Aceves, J., Cuello, A.C., 1981. Dopamine release induced by electrical stimulation ofmicrodissected caudate-putamen and substantia nigra of the rat brain. Neurosci-ence 10, 2069–2075.

Aceves, J., Floran, B., Sierra, A., Mariscal, S., 1995. D-1 receptor mediated modulation ofthe release of γ-aminobutyric acid by endogenous dopamine in the basal ganglia ofthe rat. Prog. Neuropsychopharmacol. Biol. Psychiatry 19, 727–739.

Altar, C.A., Hauser, K., 1987. Topography of substantia nigra innervation by D1 receptor-containing striatal neurons. Brain Res. 410, 1–11.

Arias-Montano, J.A., Floran, B., Floran, L., Aceves, J., Young, J.M., 2007. Dopamine D1

receptor facilitation of depolarization-induced release of gamma-amino-butyricacid in rat striatum is mediated by the cAMP/PKA pathway and involves P/Q-typecalcium channels. Synapse 61, 310–319.

Barone, P., Tucci, I., Parashos, S.A., Chase, T.N., 1987. D-1 dopamine receptor changesafter striatal quinolinic acid lesion. Eur. J. Pharmacol. 138, 141–145.

Bernardinelli, Y., Chatton, J.Y., 2008. Differential effects of glutamate transporterinhibitors on the global electrophysiological response of astrocytes to neuronalstimulation. Brain Res. 1240, 47–53.

Bernath, S., 1992. Calcium-independent release of amino acid neurotransmitters: factor artifact? Prog. Neurobiol. 38, 57–91.

Boyes, J., Bolam, J.P., 2003. The subcellular localization of GABAB receptor subunits inthe rat substantia nigra. Eur. J. Neurosci. 18, 3279–3293.

Charara, A., Heilman, T.C., Levey, A.I., Smith, Y., 2000. Pre- and postsynaptic localizationof GABAB receptors in the basal ganglia in monkeys. Neuroscience 95, 127–140.

Chéramy, A., Leviel, V., Glowinski, J., 1981. Dendritic release of dopamine in thesubstantia nigra. Nature 289, 537–542.

Chesselet, M.F., 1996. Basal gaglia and movement disorders: an update. TrendsNeurosci. 19, 417–422.

De Waard, M., Liu, H., Walker, D., Scott, V.E.S., Gurnett, C.A., Campbell, K.P., 1997. Directbinding of G-protein beta gamma complex to voltage-dependent calcium channels.Nature 385, 446–450.

DeLong, M.R., 1990. Primate models of movement disorders of basal ganglia origin.Trends Neurosci. 13, 281–285.

Dolphin, A.C., Scott, R.H., 1987. Calcium channel currents and their inhibition by (−)−baclofen in rat sensory neurones: modulation by guanine nucleotides. J. Physiol.386, 1–17.

Double, K.L., Crocker, A.D., 1995. Dopamine receptotors in the subtantia nigra areenvolved in the regulation of muscle tone. Proc. Natl. Acad. Sci. USA 5, 1669–1673.

Floran, B., Silva, I., Nava, C., Aceves, J., 1988. Presynaptic modulation of the release ofGABA by GABAA receptors in pars compacta and by GABAB receptors in parsreticulata of the rat substantia nigra. Eur. J. Pharmacol. 150, 277–286.

Floran, B., Aceves, J., Sierra, A., Martinez-Fong, D., 1990. Activation of D1 dopaminereceptors stimulates the release of GABA in the basal ganglia of the rat. Neurosci.Lett. 116, 136–140.

Garcia, M., Floran, B., Arias-Montaño, J.A., Young, J.M., Aceves, J., 1997. Histamine H3receptor activation selectively inhibits dopamine D1 receptor-dependent [3H]GABA release from depolarization-stimulated slices of rat substantia nigra parsreticulata. Neuroscience 80, 241–249.

Heise, C.E., Mitrofanis, J., 2004. Evidence for a glutamatergic projection from the zonaincerta to the basal ganglia of rats. J. Comp. Neurol. 468, 482–495.

Hemsley, K.M., Farrall, E.J., Crocker, A.D., 2002. Dopamine receptors in the subthalamicnucleus are involved in the regulation of muscle tone in the rat. Neurosci. Lett. 317,123–126.

Ibañez-Sandoval, O., Hernandez, A., Floran, B., Galarraga, E., Tapia, D., Valdiosera, R.,Erlij, D., Aceves, J., Bargas, J., 2006. Control of the subthalamic innervation ofsubstantia nigra pars reticulata by D1 and D2 dopamine receptors. J. Neurophysiol.95, 1800–1811.

Johnston, T., Duty, S., 2003. GABAB receptor agonists reverse akinesia followingintranigral or intracerebroventricular injection in the reserpine-treated rat. Br. J.Pharmacol. 139, 1480–1486.

Kawahara, K., Hosoya, R., Sato, H., Tanaka, M., Nakajima, T., Iwabuchi, S., 2002. Selectiveblockade of astrocytic glutamate transporter GLT-1 with dihydrokainate preventsneuronal death during ouabain treatment of astrocyte/neuron cocultures. Glia 40,337–349.

Kofalvi, A., Rodrigues, R.J., Ledent, C., Mackie, K., Vizi, E.S., Cunha, R.A., Sperlagh, B., 2005.Involvement of cannabinoid receptors in the regulation of neurotransmitter releasein the rodent striatum: a combined immunochemical and pharmacologicalanalysis. J. Neurosci. 25, 2874–2884.

Misgeld, U., 2004. Innervation of the substantia nigra. Cell Tissue Res. 318, 107–114.Mitchell, P.R., Doggett, N.S., 1980. Modulation of striatal glutamic acid release by

dopaminergic drugs. Life Sci. 26, 2073–2081.Naito, A., Kita, H., 1994. The cortico-nigral projection in the rat: an anterograde tracing

study with biotinylated dextran amine. Brain Res. 637, 317–322.Nava-Asbell, C., Paz-Bermudez, F., Erlij, D., Aceves, J., Florán, B., 2007. GABAB receptor

activation inhibits dopamine D1 receptor-mediated facilitation of GABA release insubstantia nigra pars reticulata. Neuropharmacology 53, 631–637.

Paxinos, G., Watson, C., 1997. The Rat Brain in Stereotaxic Coordinates. Academic Press,San Diego, CA.

Porceddu, M.L., Giorgi, O., Ongini, E., Mele, S., Biggio, G., 1986. 3H-SCH 23390 bindingsites in the rat substantia nigra: evidence for a presynaptic localization andinnervation by dopamine. Life Sci. 39, 321–328.

Robertson, H.A., 1992. Dopamine receptor interactions: some implications for thetreatment of Parkinson's disease. Trends Neurosci. 15, 201–206.

Rosales, M.G., Martinez-Fong, D., Morales, R., Nunez, A., Flores, G., Gongora-Alfaro, J.L.,Floran, B., Aceves, J., 1997. Reciprocal interaction between glutamate and dopaminein the pars reticulata of the rat substantia nigra: a microdialysis study.Neuroscience 80, 803–810.

Shapiro, M.S., Wollmuth, L.P., Hille, B., 1994. Modulation of Ca2+ channels by PTX-sensitive G-proteins is blocked by N-ethylmaleimide in rat sympathetic neurons. J.Neurosci. 14, 7109–7116.

167H. Cortés et al. / European Journal of Pharmacology 649 (2010) 161–167

Shen, K.Z., Johnson, S.W., 1997. Presynaptic GABAB and adenosine A1 receptors regulatesynaptic transmission to rat substantia nigra reticulata neurones. J. Physiol. 505,153–163.

Szatkowski, M., Barbour, B., Attwell, D., 1990. Non-vesicular release of glutamate fromglial cells by reversed electrogenic glutamate uptake. Nature 348, 443–446.

Takahashi, T., Kajikawa, Y., Tsujimoto, T., 1998. G-protein-coupled modulation ofpresynaptic calcium currents and transmitter release by a GABAB receptor. J.Neurosci. 18, 3138–3146.

Takakusaki, K., Shiroyama, T., Yamamoto, T., Kitai, T., 1996. Cholinergic andnoncholinergic tegmental pedunculopontine projection neurons in rats relealedby intracellular labeling. J. Comp. Neurol. 371, 345–361.

Westernink, B.H.C., Santiago, M., De Vries, J.B., 1992. In vivo evidence for a concordantresponse on terminal and dendritic dopamine release during intranigral infusion ofdrugs. Naunyn-Schmiedeberg´s Arch. Pharmacol. 346, 637–643.