Pre- and postsynaptic localization of GABAB receptors in the basal ganglia in monkeys

GABAB receptors in the monkey basal ganglia

Pergamon PII: S0306-4522(99)00409-1

Neuroscience Vol. 95, No. 1, pp. 127–140, 2000 127 Copyright q 1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/00 $20.00+0.00

www.elsevier.com/locate/neuroscience

PRE- AND POSTSYNAPTIC LOCALIZATION OF GABAB RECEPTORS IN THE BASAL GANGLIA IN MONKEYS A. CHARARA,*† C. HEILMAN,† A .I. LEVEY† and Y. SMITH*†‡ *Division of Neuroscience, Yerkes Regional Primate Research Center, 954 Gatewood Road NE, Atlanta, GA 30329, U.S.A. †Department of Neurology, Emory University, Atlanta, GA 30329, U.S.A.

Abstract—GABAergic neurotransmission involves ionotropic GABAA and metabotropic GABAB receptor subtypes. Although fast inhibitory transmission through GABAA receptors activation is commonly found in the basal ganglia, the functions as well as the cellular and subcellular localization of GABAB receptors are still poorly known. Polyclonal antibodies that specifically recognize the GABABR1 receptor subunit were produced and used for immunocytochemical localization of these receptors at the light and electron microscope levels in the monkey basal ganglia. Western blot analysis of monkey brain homogenates revealed that these antibodies reacted specifically with two native proteins corresponding to the size of the two splice variants GABABR1a and GABABR1b. Preadsorption of the purified antiserum with synthetic peptides demonstrated that these antibodies recognize specifically GABABR1 receptors with no cross-reactivity with GABABR2 receptors. Overall, the distribution of GABABR1 immunoreactivity throughout the monkey brain correlates with previous GABAB ligand binding studies and in situ hybridization data as well as with recent immunocytochemical studies in rodents. GABABR1-immunoreactive cell bodies were found in all basal ganglia nuclei but the intensity of immunostaining varied among neuronal populations in each nucleus. In the striatum, interneurons were more strongly stained than medium-sized projection neurons while in the substantia nigra, dopaminergic neurons of the pars compacta were much more intensely labeled than GABAergic neurons of the pars reticulata. In the subthalamic nucleus, clear immunonegative neuronal perikarya were intermingled with numerous GABABR1-immunoreactive cells. Moderate GABABR1 immunoreactivity was observed in neuronal perikarya and dendritic processes throughout the external and internal pallidal segments. At the electron microscope level, GABABR1 immunoreactivity was commonly found in neuronal cell bodies and dendrites in every basal ganglia nuclei. Many dendritic spines also displayed GABABR1 immunoreactivity in the striatum. In addition to strong postsynaptic labeling, GABABR1-immunoreactive preterminal axonal segments and axon terminals were frequently encountered throughout the basal ganglia components. The majority of labeled terminals displayed the ultrastructural features of glutamatergic boutons and formed asymmetric synapses. In the striatum, GABABR1-containing boutons resembled terminals of cortical origin, while in the globus pallidus and substantia nigra, subthalamic-like terminals were labeled. Overall, these findings demonstrate that GABAB receptors are widely distributed and located to subserve both pre- and postsynaptic roles in controlling synaptic transmission in the primate basal ganglia. q 1999 IBRO. Published by Elsevier Science Ltd. Key words: GABAB receptor, striatum, globus pallidus, subthalamic nucleus, substantia nigra, immunohistochemistry.

restricted to the subthalamic nucleus (STN). 91,93 GABA is, therefore, the major inhibitory neurotransmitter that regulates neuronal activity within the basal ganglia. The effects of GABA in the CNS are mediated by two distinct receptor subtypes, GABAA and GABAB, characterized by their pharmacological 45 and electrophysiological 14 properties. The GABAA receptor is a heterooligomeric protein which gates a chloride channel and possesses modulatory sites for benzodiazepines and a variety of other substances. 61,70 The GABAB receptors belong to the family of seven transmembrane domain metabotropic receptors that use G proteins to regulate ion channels and intracellular second messengers. 15,16 Whereas stimulation of GABAA receptors produces fast postsynaptic increases in Cl 2 conductance which can be blocked by bicuculline, 20 activation of GABAB receptors 45 induces either slow postsynaptic increase in K 1 conductance, 37,67 or presynaptic decrease in Ca 21 conductance. 31 Thus, the synaptic release of GABA in a particular brain region can mediate distinct and complex electrophysiological effects depending on the relative abundance and distribution of the two major subtypes of GABA receptors in relation to the release sites of transmitter. Until recently, knowledge of the distribution of GABAB receptors in the CNS has been limited to autoradiographic ligand binding studies in rats 16,17,26,39,106 and pigeons. 102 Although some of these studies showed that GABAB receptor

The basal ganglia are composed of several synaptically interconnected subcortical nuclei which play important roles in regulating various aspects of psychomotor behaviors, 3–5,63 and are central to the pathophysiology of common human movement disorders that range from hypokinetic (e.g., Parkinson’s disease) to hyperkinetic (Huntington’s chorea) dysfunction. 1,2,4,27,105 The extrinsic innervation of the basal ganglia is primarily glutamatergic originating in the neocortex and the thalamus. By contrast, the output pathways as well as the intrinsic synaptic circuits of the basal ganglia use mainly GABA as transmitter. 73,74,97 GABAergic neurons predominate in the striatum, globus pallidus and substantia nigra pars reticulata (SNr) whereas glutamatergic neurons are ‡To whom correspondence should be addressed. Tel.: 1 1-404-727-7519; fax: 1 1-404-727-3278. E-mail address: [email protected] (Y. Smith) Abbreviations: ABC, avidin–biotin complex; BSA, bovine serum albumin; EDTA, ethylenediaminetetra-acetate; GABAA, GABA receptor subtype A; GABAB, GABA receptor subtype B; GABABR1, GABAB receptor subunit R1; GABABR2, GABAB receptor subunit R2; GABABR1a, GABAB receptor R1 isoform a; GABABR1b, GABAB receptor R1 isoform b; GPe, globus pallidus, external segment; GPi, globus pallidus, internal segment; IR, immunoreactivity; NGS, normal goat serum; PB, phosphate buffer; PBS, phosphate-buffered saline; SDS, sodium dodecyl sulfate; SNc, substantia nigra pars compacta; SNr, substantia nigra pars reticulata; STN, subthalamic nucleus; TBS, Tris-buffered saline; VTA, ventral tegmental area. 127

128

A. Charara et al.

binding sites are present in basal ganglia nuclei, little is known about the cellular and precise synaptic localization of this receptor subtype. However, there is abundant evidence from pharmacological and electrophysiological studies for presynaptic effects of GABAB receptors on the release of GABA and other neurotransmitters in the striatum, the globus pallidus and the substantia nigra in rats. 6,21,68,69,71,76,83,86,90,101 The recent cloning of GABABR1 receptors 49 has enabled, by in situ hybridization and light microscope immunocytochemistry, the cellular localization of GABAB receptor through the rat CNS 36,49,62,87 and the monkey thalamus. 66 However, these studies did not provide information regarding the subcellular localization of receptors. Therefore, we have produced and characterized affinity-purified rabbit antibodies interacting specifically with GABABR1 receptors to elucidate the cellular and subcellular localization of GABABR1 receptors in the monkey basal ganglia. Our findings demonstrate that GABAB receptors are widely distributed and located to subserve pre- and postsynaptic functions in the primate basal ganglia. EXPERIMENTAL PROCEDURES

Generation and characterization of polyclonal GABABR1 receptor antiserum GABABR1 receptor antibodies were generated in rabbits against a synthetic peptide corresponding to the predicted intracellular C-terminal domain of the GABABR1 receptor common to the two splice variants GABABR1a and GABABR1b. 49 The antibodies were then affinity-purified. The peptide was synthesized on an applied Biosystem Peptide Synthesizer. An N-terminal lysine was added to the peptide to facilitate coupling to the carrier protein. After synthesis, the peptide was purified using reverse-phase high-performance liquid chromatography. The sequence of the peptide was NH2RGPSEPPDRLSCDGSRVHLLYK-COOH. The purified peptide was coupled to thyroglobulin using glutaraldehyde prior to immunization. The peptide conjugate was mixed with Freund’s adjuvant and rabbit polyclonal antisera were generated against this conjugate (Covance Res. Prod., Denver, PA). Ten to 24 days after each booster immunization, blood was obtained and clotted. The serum was then collected by centrifugation and stored at 2208C. Antibodies were affinity-purified on a column prepared by coupling bovine serum albumin (BSA)-conjugated peptide to Affi-Gel 10 (BioRad, Richmond, CA) prior to their use in immunoblotting and immunocytochemistry. Tissue. Tissue was collected from freshly obtained monkey liver and brain dissected on ice. The experiments were performed according to the National Institute of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize animal suffering and to reduce the number of animals used. The brains were hemisected, and the frontal cortex, the thalamus and the cerebellum were microdissected. The tissues were immediately homogenized with a Brinkmann Polytron (Brinkmann Instruments, Westbury, NY) in ice-cold 10 mM Tris–HCl, 1 mM EDTA (TE, pH 7.4). The homogenates were collected and stored frozen at 2808C until use. Immunoblot analysis. The specificity of the antibodies was evaluated by immunoblotting of monkey brain and liver membranes (100 mg). Aliquots of proteins were subjected to sodium dodecyl sulfate (SDS)–polyacrylamide gel electrophoresis (7.5% acrylamide) and transferred by electroblotting to polyvinylidene fluoride membranes (Immobilon P, Millipore, Bedford, MA). The blots were blocked with 5% non-fat dry milk, 0.1% Tween 20 in Tris-buffered saline (TBS) (20 mM Tris–HCl 1 137 mM NaCl, pH 7.4) at room temperature for 1 h, and then incubated overnight at 48C with affinity-purified GABABR1 antibodies (1 mg/ml) in blocking buffer. The blots were then rinsed and incubated, for 1 h at room temperature, with horseradish peroxidase-conjugated goat anti-rabbit IgG (Bio-Rad laboratories, Hercules, CA), diluted 1:10,000 in blocking buffer. After several washes in TBS, the immunoreactive proteins were

visualized with enhanced chemoluminescence (Amersham, Buckinghamshire, U.K.). For preadsorption experiments, antibodies were preadsorbed with 10 mg/ml homologous peptide for 1 h at room temperature. Immunocytochemistry Animals and preparation of tissue. Three adult male rhesus monkeys (3–5 kg; Yerkes Regional Primate Research colony) and one male (850 g) squirrel monkey (Saimiri sciureus; California Regional Primate Research Center) were deeply anesthetized with an overdose of pentobarbital and perfused transcardially with cold oxygenated Ringer solution, followed by fixative containing 4% paraformaldehyde and 0.1% glutaraldehyde in phosphate buffer (PB; 0.1 M, pH 7.4). After fixative perfusion, the brains were washed with PB, taken out from the skull and cut into 10 mm-thick blocks in the frontal plane. Tissue sections through the rostrocaudal extent of the basal ganglia were then obtained with a Vibratome (60 mm-thick), collected in cold phosphate-buffered saline (PBS; 0.01 M, pH 7.4) and treated with sodium borohydride (1% in PBS) for 20 min. The sections were placed in a cryoprotectant solution (PB, 0.05 M, pH 7.4, containing 25% sucrose and 10% glycerol) for 20 min, frozen at 2808C for 20 min, thawed and washed in PBS before being processed for immunocytochemistry Localization of GABAB receptor immunoreactivity at the light microscope level. The immunocytochemical localization of GABABR1 receptor subtype was performed using the avidin–biotin complex (ABC) method. 46 After blocking non-specific sites with 10% normal goat serum (NGS) and 1% BSA in PBS for 1 h at room temperature, the sections were incubated for two days at 48C in the primary antibody (1–3 mg/ml) solution. The sections were then washed in PBS, incubated for 90 min in biotinylated goat anti-rabbit IgG (1:200; Vector Labs, Burlingame, CA), rinsed again in PBS, and finally incubated for an additional 90 min in the ABC solution (1:100, Vectastain Standard kit, Vector Labs). All immunoreagents were diluted in PBS containing 1% NGS, 1% BSA and 0.1% Triton X-100. Sections were then rinsed in PBS and Tris buffer (0.05 M, pH 7.6) before being placed in a solution containing 0.025% 3,3 0 -diaminobenzidine tetrahydrochloride (Sigma Chemicals, St Louis, MO), 0.01 M imidazole (Fisher Scientific, Norcross, GA) and 0.006% H2O2 for 10 min. The reaction was stopped by repeated washes in PBS. Finally, the sections were mounted on gelatin-coated slides, dehydrated in alcohol, immersed in toluene and a coverslip was applied with Permount. Localization of GABAB receptor immunoreactivity at the electron microscope level. The sections were processed for the visualization of GABABR1 receptors according to the protocol described above, except that Triton X-100 was not included in the solutions. They were washed in PB (0.1 M, pH 7.4) and postfixed in osmium tetroxide (1% in PB) for 20 min. This was followed by washing in PB and dehydration in a graded series of ethanol and propylene oxide. Uranyl acetate (1%) was added to the 70% ethanol for 35 min to improve the contrast in the electron microscope. The sections were then embedded in resin (Durcupan ACM, Fluka, Ft Washington, PA) on microscope slides and placed in the oven for 48 h at 608C. Areas of interest were selected, cut out from the slides and glued on the top of resin blocks. Serial ultrathin sections were then cut on an ultramicrotome, collected on to Pioloform-coated single copper grids, stained with lead citrate 77 and analysed with the electron microscope. Control experiments The specificity of labeling was tested by incubation with solutions in which the affinity-purified antibodies were replaced by pre-immune rabbit serum. Other control experiments were performed in which the primary antibody was preadsorbed with the homologous GABABR1 peptide or the GABABR2 peptide (NSPEHIQRLSLQLPILHHAYL; the amino acids in italics are similar in R1). 58 Analysis of data Transverse sections containing basal ganglia structures were taken from each of the four monkeys for both light and electron microscope analyses. To determine the size of cell bodies, only neuronal perikarya in which nucleus and proximal dendrites could be clearly visualized

GABAB receptors in the monkey basal ganglia

129

< 130,000 mol. wt protein and a < 100,000 mol. wt protein, in homogenates from monkey brains including the frontal cortex, thalamus, and cerebellum (Fig. 1A). These results were consistent with the predicted molecular weight of proteins corresponding to GABABR1a and GABABR1b receptor subunits on the basis of their amino acid sequences. 49 Lightly-labeled bands with molecular weights .140,000 were seen in the cerebellum and the frontal cortex, most likely representing receptor aggregates. The two proteins were differentially expressed in the brain structures examined. No proteins were labeled in tissue from liver which has been shown to lack GABAB receptors (Fig. 1A). Preadsorption of purified antibodies with homologous peptide totally abolished all immunoreactive bands (Fig. 1B) except for the light band visible at 49,000 mol. wt in the thalamus which is most probably due to the presence of proteins specifically recognized by the secondary antibodies (Fig. 1A, B). Immunostaining in the basal ganglia

Fig. 1. Antiserum against GABABR1 receptors. (A) Western blot analysis demonstrating the specificity of antibodies against GABABR1 receptors. Adult monkey brain tissue homogenates (100 mg membrane proteins) were subjected to SDS–polyacrylamide gel electrophoresis and immunoblotted with affinity-purified antibodies (1 mg/ml) raised against a synthetic peptide corresponding to the C-terminal region of GABABR1 receptors. GABABR1 antibodies detected proteins of < 130,000 and < 100,000 mol. wts corresponding, respectively, to the molecular weight predicted for GABABR1a and GABABR1b receptor subunits. Both proteins are found in the cerebellum, the frontal cortex and the thalamus. The remaining bands seen in some regions are likely to represent a receptor aggregate except for the band visible at 49,000 mol. wt in the thalamus which is most probably due to proteins specifically detected by the secondary antibodies (see panel B). No band was seen for liver membrane proteins. (B) Immunoreactivity is completely abolished when antibodies are preadsorbed with the synthetic GABABR1 peptide (10 mg/ml) prior to immunoblotting. Note that the 49,000 mol. wt band in the thalamus remains labeled indicating that it is due to the secondary antibodies. Molecular weight standards are indicated on the left (in × 10 3 mol. wt).

were selected. The measurements were performed at different rostrocaudal levels of each structure with a Leitz ocular micrometer using a × 40 lens. A total of 22 blocks of tissue were selected for the electron microscope analysis (four blocks from the putamen, two from the caudate nucleus, three from the external pallidal segment (GPe), three from the internal pallidal segment (GPi), four from the STN, three from the substantia nigra pars compacta (SNc), and three from the SNr). One block was taken from each structure in the three rhesus monkeys. In addition, one block from the putamen and one block from the STN were taken from the squirrel monkey. The blocks from the striatum were all cut from levels rostral to the anterior commissure, while blocks from the SNc and SNr were selected from sections corresponding to the middle third of the substantia nigra, where SNc and SNr are most easily distinguishable from each other. The ultrastructural analysis was carried out on ultrathin sections collected from the surface of each block where the staining was optimal. The number of immunoreactive terminals and their postsynaptic targets were recorded. The size of these terminals was measured from electron micrographs. RESULTS

Immunoblotting Polyclonal antibodies directed against GABABR1 receptors were characterized by the western blot technique with membranes from monkey brain and liver. The purified antibodies reacted specifically with two distinct native proteins, a

The general features of GABABR1 immunolabeling in the basal ganglia were similar in the four animals used in this study. GABABR1 immunoreactivity (IR) was virtually absent when the affinity-purified antibody was omitted from the incubation solution or preadsorbed with the homologous peptide (Fig. 2B). Moreover, preadsorbtion of the affinitypurified antibodies with the synthetic GABABR2 receptor peptide had no effect on GABABR1-IR indicating that GABABR1 antibodies do not cross-react with GABABR2 receptors (Fig. 2C). Altogether, with the high degree of specificity demonstrated in the western blot analysis, these data indicate that the antiserum used in this study reacts selectively with GABABR1 receptors. Overall, the pattern of GABABR1 immunostaining was relatively similar among basal ganglia nuclei. Moderate to strong IR was found in subsets of neuronal elements including perikarya, dendritic processes, dendritic spines and axon terminals. Occasionally, GABABR1-IR was also detected in cell bodies and thin processes of astrocytes (Fig. 5B, C). Striatum. At the light microscope level, the striatum displayed moderate IR for GABABR1 receptors (Fig. 3A). The staining was homogeneous and relatively similar throughout the caudate nucleus, putamen and nucleus accumbens. There was no patchy distribution of immunostaining, which suggests that GABABR1-IR is not differentially expressed in the patch-matrix compartments (Fig. 3A). At high magnification, virtually all neuronal perikarya appeared to be immunostained but to varying degree of intensity; a subpopulation of darkly-labeled large neuronal perikarya (maximum diameter ˆ 24 ^ 3 mm; n ˆ 120) clearly stood out among a large number of lightly-labeled medium-sized cell bodies (maximum diameter ˆ 14 ^ 2 mm; n ˆ 120) (Fig. 3B). Most immunoreactive perikarya gave rise to two to four dendritic processes that were weakly labeled and, thus, cannot be followed for long distance in the striatal tissue. In addition to neuronal perikarya and dendrites, strong GABABR1 immunostaining was associated with punctate elements and thin axon-like processes (Fig. 3B). The ultrastructural analysis confirmed the light microscope observations that GABABR1-IR was expressed in two types of neuronal perikarya. The majority displayed the typical ultrastructural features of medium-sized projection

130

A. Charara et al.

endoplasmic reticulum (Fig. 3C). This was also the case in the remaining basal ganglia components. Large and small dendrites that received symmetric and asymmetric synaptic inputs from labeled and unlabeled boutons as well as many dendritic spines also displayed strong immunoreactivity (Fig. 4A–C). In dendrites, the reaction product was strongly associated with microtubules, whereas in spines it usually filled the whole element (Figs 3C, 4A–C). However, nonimmunoreactive dendritic spines in the vicinity of stronglylabeled elements were frequently encountered (Fig. 4A). In addition to postsynaptic elements, GABABR1-IR was found in many myelinated and unmyelinated axonal segments as well as in terminal boutons (Fig. 4A–E). The examination of 70 labeled terminals revealed that they formed asymmetric synapses primarily with the head of dendritic spines (76%) (Fig. 4A, B) but also with dendritic processes (24%) (Fig. 4C), and displayed the ultrastructural features of corticostriatal and/or thalamostriatal boutons. 51,80,81,95 Their maximum diameter ranged from 0.6 to 1.7 mm, and they contained a large number of small round synaptic vesicles and mitochondria (Fig. 4A–C). All dendritic processes and 55% of dendritic spines contacted by GABABR1-containing terminals were also immunoreactive for GABABR1 receptor. Another much rarer type of presynaptic terminal (n ˆ 6) formed “en passant” type axonal varicosities which were filled with large pleomorphic vesicles and formed symmetric synapses with immunolabeled dendrites (Fig. 4D).

Fig. 2. Specificity of GABABR1-IR in the monkey basal ganglia. (A) Lowpower micrograph showing GABABR1-IR in the putamen. (B) Section of the putamen incubated with the GABABR1 antiserum preadsorbed with the homologous peptide (10 mg/ml). (C) Section of the putamen incubated with the GABABR1 antiserum preadsorbed with the synthetic GABABR2 peptide (10 mg/ml). Note that preadsorbtion with the homologous peptide abolished specific immunostaining whereas preadsorption with GABABR2 peptide had no effect on the distribution and intensity of GABABR1-IR in neurons. IC, internal capsule. Scale bars ˆ 1 mm.

neurons, 29,92 i.e. they contained a large nucleus with a smooth membrane surrounded by a rim of cytoplasm (Fig. 3C). A second population of larger immunoreactive cell bodies were typical of striatal interneurons, 12,30 with abundant cytoplasm surrounding a deeply-invaginated nucleus. The immunoperoxidase reaction product was associated preferentially with organelles involved in proteins synthesis, maturation and transport such as the Golgi apparatus and

Globus pallidus. In both the internal (GPi) and external (GPe) segments of the globus pallidus, virtually all cell bodies exhibited a moderate GABABR1-IR (Fig. 3A). The majority of immunoreactive cells had a medium- to large-sized multipolar perikaryon (maximum diameter ˆ 25 ^ 5 mm; n ˆ 60) and were uniformly distributed throughout the rostrocaudal extent of the pallidal complex (Fig. 5A). Other larger immunoreactive neurons (maximum diameter ˆ 34 ^ 7 mm; n ˆ 30), which probably belong to the basal forebrain cholinergic cell group, were scattered along the lateral borders of GPe and GPi but were especially abundant in the internal and external medullary laminae. At the electron microscope level, GABABR1-IR was enriched in postsynaptic neuronal elements including perikarya and dendritic shafts of various sizes. The most striking presynaptic labeling in both pallidal segments was found in numerous unmyelinated preterminal axonal segments (Fig. 5B, D, E). Moreover, a population of immunoreactive terminals (n ˆ 33) that had a maximum diameter ranging from 0.9 to 2 mm and contained numerous small round synaptic vesicles as well as two or three mitochondria were also frequently encountered in GPe and GPi (Fig. 5B, C). In the cases where the synaptic specialization could be seen (n ˆ 18), it was of the asymmetric type. The main postsynaptic targets of these boutons were GABABR1-immunoreactive dendritic shafts (Fig. 5B–D). The ultrastructural features and pattern of synaptic organization of these terminals strikingly resembled those of boutons arising from the subthalamic nucleus (Fig. 5B, C). 88,96,97 Unlabeled boutons, morphologically similar and close to the GABABR1-immunoreactive terminals, formed asymmetric synapses with strongly-labeled dendrites. In a few cases (n ˆ 12), lightly-labeled terminals that displayed the ultrastructural features of striatal boutons 88 were also seen in both pallidal segments and formed symmetric axodendritic synapses (Fig. 5D).

GABAB receptors in the monkey basal ganglia

131

Fig. 3. Immunohistochemical detection of GABABR1 in the monkey striatum. (A) The putamen (PUT) and the external segment of the globus pallidus (GPe) display moderate to intense GABABR1 immunoreactivity. (B) GABABR1-IR is present in numerous medium-sized neuronal perikarya (arrowheads) and some large-sized neuronal cell bodies (arrow) and dendrites in the caudate nucleus. Note the numerous punctate elements in the neuropil (open arrowheads). (C) A low-power electron micrograph of a GABABR1-immunoreactive perikaryon (PER) that displays the ultrastructural features of a medium-sized projection neuron, i.e. large unindented nucleus surrounded by a rim of cytoplasm, in the putamen. Note the proximal dendrites (DEN) emerging from the soma. The peroxidase deposit displays a patchy distribution (large arrowheads) being frequently associated with vesicular organelles attached to saccules of the Golgi apparatus (in cell body) and microtubules (in dendrites). No particular aggregation of peroxidase reaction product was found along the plasma membrane. Scale bars ˆ 250 mm (A); 50 mm (B); 2 mm (C).

Subthalamic nucleus. The STN harbored a multitude of GABABR1-immunoreactive neuronal perikarya embedded within a lightly-stained neuropil composed of thin punctate structures and varicose processes (Fig. 6A, B). The labeled cell bodies (maximum diameter ˆ 21 ^ 3 mm; n ˆ 30), intermingled with immunonegative perikarya, were evenly distributed throughout the rostrocaudal extent of the STN (Fig. 6B). In the electron microscope, prominent immunoperoxidase labeling for GABABR1 receptors was found in neuronal perikarya and dendrites (Fig. 6C–E). In addition, scattered myelinated pre-terminal axonal segments and a subpopulation of axon terminals (n ˆ 38) displayed strong GABABR1-IR. The labeled boutons had a maximum diameter ranging from 0.8–2 mm, and were packed with small round electron-lucent synaptic vesicles (Fig. 6D, E). In the 22 cases where the synaptic specialization could be seen, it was of the asymmetric type and the postsynaptic targets were GABABR1-immunoreactive dendrites (Fig. 6D–E). Immunonegative boutons forming asymmetric synapses and large

GPe-like terminals forming symmetric synapses were frequently seen in the vicinity of the GABABR1-containing terminals (Fig. 6C, D). Substantia nigra. In the substantia nigra, there was a sharp contrast between the intensity of neuronal immunostaining in the pars compacta and pars reticulata (Fig. 7A). Moderate to intense labeling was found throughout the SNc (Fig. 7B), mainly in large multipolar neurons (maximum diameter ˆ 26 ^ 5 mm; n ˆ 30) whereas neuronal perikarya in the SNr were very lightly stained (Fig. 7C). In fact, the SNc stood out as the most intensely GABABR1immunoreactive component of the basal ganglia. Strong IR for GABABR1 receptor was also found in neuronal perikarya and dendrites in the ventral tegmental area (VTA) (Fig. 7A). The ultrastructural analysis was consistent with observations in the light microscope. In fact, the labeling in the SNr was very similar to that found in GPe and GPi, i.e. dendrites and many preterminal unmyelinated axons as well

132

A. Charara et al.

Fig. 4. Electron micrographs showing pre- and postsynaptic immunoreactivity for GABABR1 in the striatum. (A) Some dendritic spines (SP) and dendrites (DEN) display strong GABABR1-IR in the putamen, whereas other dendritic spines in their vicinity are devoid of immunostaining (USP). Note the presence of a GABABR1-immunoreactive bouton (b1) forming an asymmetric synapse (arrowheads) with an unlabeled spine. Asterisks (*) indicate four terminals devoid of GABABR1-IR. (B) A GABABR1-immunoreactive bouton (b1) forming an asymmetric synapse (arrowhead) with a labeled dendritic spine (SP) surrounded by two immunolabeled dendrites (DEN) in the putamen. (C) shows a GABABR1-immunoreactive terminal (b1) and a non-immunoreactive bouton (*) that form asymmetric synapses (arrowheads) with a labeled dendrite (DEN) in the putamen. Another labeled terminal of which the synaptic specialization could not be established is indicated (b2). (D) GABABR1-immunoreactive vesicle-filled axon (ax1) that forms an “en passant” type symmetric synapse (open arrow) with a labeled dendrite in the monkey putamen. Another labeled axon cut in the transverse plane is indicated (ax2). (E) A GABABR1-immunoreactive myelinated axon (ax) in the vicinity of an unlabeled myelinated axon (*) in the putamen. Scale bars ˆ 0.5 mm.

as a population of subthalamic-like axon terminals (n ˆ 41) forming asymmetric synapses with GABABR1-immunoreactive dendrites (Fig. 7E, F) were immunostained. One of these boutons also established an axo-axonic contact with an unlabeled terminal (Fig. 7E). In the SNc, the labeling was found predominantly in neuronal perikarya and dendritic shafts. Rare preterminal axons, and axon terminals (n ˆ 13) which, in some cases (n ˆ 5), formed asymmetric synapses with GABABR1-immunoreactive dendrites were also seen (Fig. 7D). In addition, a few lightly-labeled terminals (n ˆ 15) packed with large pleomorphic vesicles were seen

in the SNc and SNr, but none of these boutons formed clear synaptic contact. DISCUSSION

The results of this study provide the first description of the distribution of GABABR1-IR in the basal ganglia of primates at the light and electron microscope level. Our findings show that GABABR1-IR is widely expressed in perikarya, dendrites and spines of basal ganglia neurons, but the degree of immunostaining varied among neuronal populations in each

GABAB receptors in the monkey basal ganglia

133

Fig. 5. GABABR1 immunostaining the monkey globus pallidus. (A) A high-power micrograph showing a GABABR1-immunoreactive neuron that lays in a punctate neuropil (small arrows) in the external segment of the globus pallidus (GPe). (B) GABABR1-IR is present in numerous preterminal axonal segments (ax) and a subthalamic-like bouton (b1) which forms an asymmetric synapse (arrowhead) with an immunoreactive dendrite (DEN) in GPe. Note the presence of a non-immunoreactive bouton (*), and a labeled glial cell process (GL) in the same field. (C) A GABABR1-immunolabeled subthalamic-like bouton (b1) forming an asymmetric synapse with a dendrite (DEN) in the internal segment of the globus pallidus (GPi). A non-immunoreactive bouton (*) and a labeled glial cell process (GL) are also indicated. (D) A GABABR1-containing striatal-like terminal (b1) which forms a symmetric axodendritic synapse (open arrow) in GPe. Note the presence of labeled preterminal unmyelinated axon (ax) in the same field. (E) shows numerous vesicle-filled preterminal axonal segments (ax) immunoreactive for GABABR1 in GPe. Scale bars ˆ 50 mm (A); 0.25 mm (B, D); 0.5 mm (C, E).

basal ganglia structure. In the striatum, interneurons were more strongly labeled than medium-sized projection neurons while in the substantia nigra, dopaminergic neurons in the SNc and VTA were more darkly stained than GABAergic neurons in the SNr. Similarly, clear immunonegative neuronal perikarya among numerous GABABR1-positive cells were seen in the STN. In addition to being expressed postsynaptically, GABABR1IR was frequently found in preterminal axonal segments and axon terminals in basal ganglia nuclei. A common finding to all basal ganglia structures was that most of the immunoreactive terminals displayed the ultrastructural features of glutamatergic boutons and formed asymmetric synapses, which strongly suggests that GABAB receptors play an important role as heteroreceptors modulating glutamatergic neurotransmission in the primate basal ganglia. Moreover, a subpopulation of immunoreactive striatal-like boutons were also found in the globus pallidus and SNr, suggesting a potential role as autoreceptor that regulates GABAergic striatofugal transmission in basal ganglia output structures. These findings are discussed in relation to previous anatomical, neurochemical, and electrophysiological studies of the localization and functions of GABAB receptors in the basal ganglia.

Characterization and specificity of GABAB receptor antibodies We have developed antibodies to localize immunocytochemically GABABR1 receptors at light and electron microscope level. Western blot analysis revealed that these antibodies selectively recognized two proteins with molecular weights consistent with those of the two splice variants GABABR1a and GABABR1b 49 in monkey brain. Preadsorption of the antibodies with the homologous peptide abolished immunoreactive bands, and sections incubated with preadsorbed antibodies were devoid of immunostaining. Together, these observations demonstrate that the affinitypurified antibodies used in this study react in a highly specific manner with the targeted epitope on the GABABR1 protein. Recent cloning studies demonstrated the existence of another GABAB receptor named GABABR2. 48,50,58,104 The sequence of this receptor subtype has 30–40% homology with the GABABR1 receptor, 48,50,58 which opens up the possibility that antibodies used in the present study also recognized the GABABR2 receptors. However this is unlikely to be the case for the following reasons: (i) the amino acid sequence of GABABR1 used to produce our antibodies differs from the

134

A. Charara et al.

Fig. 6. GABABR1 immunostaining the monkey subthalamic nucleus (STN). (A) Low-power micrograph showing moderate GABABR1-IR in the STN. (B) An immunonegative neuronal perikaryon (arrowheads) intermingled with numerous GABABR1-immunoreactive cell bodies. Note the numerous punctate elements in the STN neuropil (arrows). (C) shows a non-immunoreactive bouton (*) apposed to a large GABABR1-immunoreactive dendrite (DEN). (D) A GABABR1-labeled bouton (b1) forms an asymmetric synapse (arrowhead) with an immunoreactive dendrite (DEN1). Note the presence of a non-immunoreactive bouton (*) forming an asymmetric synapse (arrowhead) with another immunolabeled dendrite (DEN2) in the same field. (E) shows a GABABR1positive bouton (b1) that forms two asymmetric axodendritic synapses (arrowheads). Scale bars ˆ 250 mm (A); 50 mm (B); 1 mm (C); 0.5mm (D); 0.25 mm (E).

corresponding sequence of GABABR2 by 16 amino acids (R2: NSPEHIQRLSLQLPILHHAYL; the amino acids in italics are similar in R1); 58 (ii) the spatial and temporal expression of GABABR1 and GABABR2 differs in many rat brain structures. 58 For instance, whereas most striatal neurons display moderate to high GABABR1-IR (present study; Refs 36, 62 and 87) and express high level of GABABR1 mRNA, the striatum is almost devoid of GABABR2 mRNA in adult rats; 48,50,58 (iii) preadsorption of the GABABR1 antiserum with the GABABR2 peptide had no effect on the distribution and intensity of the GABABR1 immunostaining throughout the monkey basal ganglia. Together, these data demonstrate that the GABABR1 antiserum used in the present study is highly specific and does not cross-react with the GABABR2 subunit. Using western blot and immunocytochemistry, we found that, overall, the pattern of GABABR1-IR in various regions of the monkey brain is largely consistent with the regional expression of the GABABR1 mRNAs 48–50,58 and the distribution of GABAB receptor binding sites previously described in

rats. 17,25,26 Moreover, the pattern of distribution of GABABR1-immunoreactive neurons described in the present study is in keeping with that recently shown in rats using other GABABR1 antisera, 36,62,87 which further supports the specificity of our GABABR1 antiserum. Pre- and postsynaptic GABAB receptors in the striatum Our results demonstrate that GABABR1-IR is expressed in putative glutamatergic terminal boutons forming asymmetric axospinous (76%) and axodendritic (24%) synapses in the monkey striatum. The two major sources of terminals forming asymmetric synaptic junctions in the striatum are the cerebral cortex and the caudal intralaminar thalamic nuclei 32,51,80,92,95 which both contain GABABR1-IR neurons (present study; Refs 36, 62 and 87). However, that the thalamostriatal projection terminates predominantly on dendrites whereas corticostriatal afferents innervate mostly dendritic spines 32,95 strongly suggest that the majority of GABABR1-immunoreactive terminals seen in the striatum arise from the cerebral cortex.

GABAB receptors in the monkey basal ganglia

135

Fig. 7. GABABR1 immunostaining in the substantia nigra. (A) A low-power micrograph showing strong labeling in the substantia nigra pars compacta (SNc) and the ventral tegmental area (VTA) in comparison to a low to moderate immunoreactivity in the substantia nigra pars reticulata (SNr). (B and C) compare the intensity of GABABR1 immunostaining associated with neuronal perikarya in SNc (B) and SNr (C). (D) A GABABR1-labeled bouton (b1) that forms an asymmetric synapse (arrowhead) with a labeled dendrite (DEN) in the SNc. Note the presence of an immunonegative (*) terminal that makes a synaptic contact (arrowhead) with the same dendrite. (E) Many labeled preterminal axonal segments (ax) and an immunoreactive bouton (b1) that forms an asymmetric synapse (arrowhead) with a GABABR1-positive dendrite (DEN) in the SNr. Note the presence of axon terminals devoid of GABABR1-IR in the neuropil (asterisks). Small arrows indicate an axo-axonic contact between the GABABR1-immunoreactive bouton (b1) and a non-immunoreactive terminal (b2). (F) A GABABR1containing bouton (b1) that forms an asymmetric synapse (arrowheads) with an immunoreactive dendrite (DEN) in the SNr. Note the presence of two axon terminals devoid of GABABR1 immunoreactivity (asterisks). Scale bars ˆ 1 mm (A); 25 mm (B, C); 0.5 mm (D); 1 mm (E, F).

On the other hand, since thalamic afferents from rostral intralaminar nuclei provide a substantial input to dendritic spines in rats, 108 tract-tracing studies combined with GABABR1 immunocytochemistry should be carried out before ruling out that some of these boutons arise from the thalamus. The existence of presynaptic GABAB receptors on glutamatergic terminals in the striatum is controversial. For instance, some investigators reported significant losses of striatal GABAB receptor binding sites following lesion of cortical afferents, 52,64 whereas others have shown that GABAB receptor level remained unchanged after decortication, but was markedly reduced after striatal ibotenate lesions in rats. 25 Similarly, application of GABAB receptor

agonists has been shown to depress glutamatergic excitation in spiny striatal neurons in vitro 68,69,84 and in vivo. 18,19 However, intracellular recordings from rat caudate neurons showed that systemic or intracaudate administration of baclofen, a GABAB agonist, did not reduce excitatory postsynaptic potentials in response to cortical or thalamic stimulation, but rather blocked evoked hyperpolarizations, suggesting that baclofen exerts a direct effect on intracaudate circuitry rather than on glutamatergic afferents. 107 Although these discrepancies may result from technical differences in the extent of cortical lesions or stimulations, properties of radioligands, location and/or conditions of cell recording, they indicate that a better knowledge of the exact localization of GABAB

136

A. Charara et al.

receptors is a prerequesite for the interpretation of further functional studies. Another much less frequent type of GABABR1-containing terminals encountered in the monkey striatum formed “en passant” type symmetric synapses and displayed the ultrastructural features of either dopaminergic terminals from the substantia nigra, 7,35,95 or GABA-containing intrinsic striatal boutons. 13,78 In line with these observations, pharmacological evidence indicated that baclofen reduces the evoked release of dopamine 6,76 and potentiates the release of metenkephalin 83 in striatal slices. Baclofen has also been shown to decrease the level of extracellular dopamine in the striatum in vivo. 98 Future double immunocytochemical studies are essential to characterize the exact neurotransmitter content of this subset of GABABR1-containing terminals. At present, very little is known about potential functions of striatal postsynaptic GABAB receptors. In vitro studies showing that application of GABAB agonists in striatal slices depressed the synaptic potentials of striatal neurons indicate that at least one population of GABA-releasing neurons within the striatum express postsynaptic GABAB receptors. 85 In keeping with these functional studies, GABABR1-IR was found to be expressed postsynaptically in numerous projection neurons and interneurons in the monkey striatum. There was no patchy distribution of immunostaining. These results are different from those of a recent study showing that the distribution of some GABAA receptor subunits varied between the striosome and the matrix compartments of the monkey striatum; whereas b2/3 subunits were homogeneously distributed and confined to large-sized immunoreactive neurons, the a1 subunit was found in high density in the matrix compartment where it was mostly expressed by medium-sized aspiny neurons. 103 Another finding of the present study is that the cell bodies of interneurons were more strongly labeled than those of output cells which is in keeping with recent findings in rats showing that the GABABR1-IR was particularly abundant in a small population of neurons scattered throughout the striatum. 62 Whether the differential intensity of perikaryal labeling is reflected by a larger density of subsynaptic GABAB receptors in interneurons than projection neurons remains to be established. Although subsets of dendrites and spines displayed strong immunoreactivity, we could not provide any information on the synaptic localization of the receptors in the present study because of the diffuse nature of the peroxidase reaction product. Immunogold studies are in progress to characterize the subsynaptic localization of GABAB receptors at individual synapses on striatal neurons. Presynaptic GABAB heteroreceptors on putative glutamatergic subthalamopallidal and subthalamonigral terminals An interesting finding of the present study common to both pallidal segments and SNr was the localization of GABABR1 receptors in terminal boutons forming asymmetric axodendritic synapses. Based on their morphological features and pattern of synaptic organization, most of these boutons very probably arose from the STN, 56,89,96,97 though minor sources of asymmetric synaptic inputs such as the pedunculopontine nucleus and the thalamus 27,81 cannot be excluded. If subthalamopallidal and subthalamonigral terminals, do indeed express GABAB receptors, they could potentially be used as targets to reduce the overflow of glutamate released by STN

neurons in basal ganglia output structures in Parkinson’s disease. 1,9,28 In support of this hypothesis, application of baclofen was found to decrease the efflux of glutamate in the rat globus pallidus in vivo 90 and reduce the evoked synaptic currents mediated by glutamate in the rat SNr in vitro. 86 Another common GABABR1-immunoreactive presynaptic element found in the globus pallidus and the SNr were small unmyelinated axons which, in some cases, contained numerous electron-lucent vesicles. Although the source and neurotransmitter content of these axons remain to be established, these observations suggest that GABABR1 receptors might be expressed in the preterminal portion of some axons flowing through the globus pallidus and SNr (see below). The STN also contained a substantial number of putative glutamatergic terminals that displayed GABABR1-IR in monkeys. The exact origin of these boutons is not clear. Terminals with a similar ultrastructure were labeled after anterograde tracer injections in the pedunculopontine nucleus, cerebral cortex, and intralaminar thalamic nuclei. 10,11,41,65 Further studies combining tract-tracing methods with immunocytochemistry are essential to clarify this issue. Solving this problem and elucidating the functions of these GABAB heteroreceptors is particularly important to understand the pathophysiology of Parkinson’s disease, since abnormal regulation of glutamatergic afferents has been proposed to underlie the increased firing rate of STN neurons in Parkinson’s disease. 24,60 Together with our observations in the globus pallidus, these findings suggest that GABAB agonists might have beneficial therapeutic effects in hypokinetic disorders by both decreasing the release of glutamate from extrinsic afferents to the STN and reducing the glutamate overflow from STN terminals in GPi and SNr. Moderate to strong GABABR1-IR was also associated with neuronal perikarya and dendrites in the two pallidal segments and the STN. Although the role of postsynaptic GABAB receptors in these brain regions is not yet known, neurons in those structures receive massive GABAergic inputs from the striatum (to GPe and GPi) and GPe (to GPi and STN) 43,44,53,55,88,89,94,96,97 and express a1, b2/3, and g2 GABAA receptor subunit immunoreactivity. 23,103 These observations strongly suggest that, in addition to the well-known GABAA-mediated inhibitory effects, 54,75 GABAB receptors might be involved in modulating transmission at these synapses. High resolution immunogold studies are in progress to elucidate the subsynaptic localization of GABABR1 receptors associated with striatal and GPe inputs to GPi and STN neurons. GABAB receptors on striatopallidal and striatonigral afferents Although most GABABR1-containing terminals encountered in GPe, GPi and SNr displayed the ultrastructural features of glutamatergic STN boutons, a small number of lightly-labeled terminals formed symmetric synapses and resembled striatal terminals, which is consistent with the low level of GABABR1 immunoreactivity associated with striatal output neurons. 22,97 However, because they were so rare and faintly labeled, it was difficult to provide much information on these terminals in the present study. On the other hand, recent in vivo 40,71,101 and in vitro 21,34,42,86 electrophysiological data support the expression of presynaptic GABAB autoreceptors on striatal, pallidal and intranigral GABAergic

137

GABAB receptors in the monkey basal ganglia

afferents to SNc and/or SNr neurons in rats. A possible explanation for this discrepancy between anatomical and electrophysiological features is that presynaptic GABAB autoreceptors are expressed in the preterminal segment of axonal boutons rather than in the terminals themselves. Numerous GABABR1-immunoreactive small unmyelinated axons were, indeed, encountered in GPe, GPi and SNr. If those axons belong to striatal neurons they provide multiple sites where GABA might activate presynaptic GABAB autoreceptors. It is worth noting that such a preterminal localization was recently described for the metabotropic glutamate receptor mGluR2 in the monkey striatum. 72 GABAB receptors in the substantia nigra pars compacta Electrophysiological studies indicate that GABAB receptors play an important role in regulating the activity of midbrain dopaminergic neurons in rats, 33,59,71,79,82,84,99–101 but the exact mechanisms and localization of GABAB receptors that mediate those effects are controversial. For instance, systemic administration of baclofen shifted the firing pattern of midbrain dopaminergic neurons from a bursting mode to a more regular single spike activity, 32 which is consistent with the inhibitory effects of local intranigral injections of baclofen on dopamine release in the rat striatum. 82 In contrast, Paladini et al. 71 recently showed that local pressure application of GABAB antagonists in the rat SNc regulates the firing rate of dopaminergic neurons by blocking presynaptic GABAB autoreceptors on striatal, pallidal and intranigral GABAergic afferents. Our findings showed that the VTA and SNc were the most intensely labeled basal ganglia structures for GABABR1 receptor in monkeys, which support a strong expression of postsynaptic GABAB receptors in those brain regions, though post-embedding immunogold must be carried out to verify this possibility. These observations are consistent with recent in situ hybridization 49 and immunocytochemical 36,62,87 studies showing that GABABR1 receptors are highly expressed in the SNc and VTA compared to the SNr. However, previous autoradiographic studies have shown that binding sites for GABAB receptors were always higher in the SNr than the SNc. 17,26,38 These discrepancies could be explained by the fact that GABAB binding sites in the SNr may originate from postsynaptic receptors located on dendrites of SNc neurons, or from presynaptic GABAB receptors on axons and terminals arising from extrinsic sources. Together, these physiological and anatomical data strongly support the hypothesis that GABAB receptors activation is important in modulating the activity of midbrain dopaminergic neurons. However, the exact mechanism of action and relative importance of pre- versus postsynaptic

GABAB receptors in mediating these effects remains to be established. CONCLUSIONS

One of the most important findings of this study is that GABABR1-IR is expressed in putative glutamatergic afferents throughout the monkey basal ganglia. This strongly supports a major role for GABAB heteroreceptors in modulating glutamatergic transmission in the basal ganglia circuitry. Furthermore, the fact that GABAB receptors are expressed by putative subthalamic terminals and glutamatergic afferents to STN neurons open up the possibility of using GABAB agonists as therapeutic agents to silence overactive STN neurons in Parkinson’s disease. An important issue that remains to be established is the mechanism by which these presynaptic heteroreceptors are activated. That axo-axonic synapses are very rare in the basal ganglia rule out the possibility of direct synaptic release of transmitter. Another possibility would be that, once released, GABA diffuses out of the synaptic cleft and activates extra- and presynaptic GABAB heteroreceptors. 8 Evidence for such a paracrine mode of GABAB receptor activation was, indeed, demonstrated in the rat hippocampus. 47 The efficacy of such a non-specific mode of transmission largely depends on the extent to which GABA can diffuse and the affinity of GABAB receptors for its transmitter. Although such information is still lacking for basal ganglia structures, it is worth noting that presynaptic GABAB receptors were found to have a much higher affinity for GABA than GABAA receptors in the rat hippocampus. 109 Another interesting issue raised by these studies is the relationship between GABAergic synaptic inputs and the localization of GABAB receptor subtypes. Our findings clearly indicate that GABAB receptors are expressed postsynaptically by most neurons in the basal ganglia. Whether these postsynaptic receptors are located in the main body or perisynaptic to GABAergic synapses is critical to understand the mechanism of activation of these receptors. Recent immunocytochemical data revealed that GABABR1 receptors are located extrasynaptically, with no labeling at the active zones, to GABAergic synapses in the rat cerebellum. 36,57 Preand post-embedding immunogold studies are currently in progress in our laboratory to verify whether this extrasynaptic localization of GABAB receptors also applies to basal ganglia neurons in primates. Acknowledgements—The authors thank Jean-Franc¸ois Pare´ for skilful technical assistance, and Frank Kiernan for photography. This research was supported by NIH Grants RR 00165 and R01 NS37323-01. Ali Charara holds a Fellowship from the FCAR.

REFERENCES

1. Albin R. L., Young A. B. and Penney J. B. Jr (1989) The functional anatomy of basal ganglia disorders. Trends Neurosci. 12, 366–375. 2. Albin R. L., Young A. B. and Penney J. B. Jr (1995) The functional anatomy of disorders of the basal ganglia. Trends Neurosci. 18, 63–64. 3. Alexander G. E., Delong M. R. and Strick P. L. (1986) Parallel organization of functionally segregated circuits linking basal ganglia and cortex. A. Rev. Neurosci. 9, 357–381. 4. Alexander G. E., Delong M. R. and Crutcher M. D. (1992) Do cortical and basal ganglionic motor areas use motor programs to control movement? Behav. Brain Res. 15, 656–665. 5. Alheid G. F. and Heimer L. (1988) New perspectives in basal forebrain organization of special relevance for neuropsychiatric disorders: the striatopallidal, amygdaloid and corticopetal components of the substantia innominata. Neuroscience 27, 1–39. 6. Arias-Montano J. A., Martinez-Fong D. and Aceves J. (1992) GABAB receptor activation partially inhibits N-methyl-d-aspartate-mediated tyrosine hydroxylase stimulation in rat striatal slices. Eur. J. Pharmac. 218, 335–338. 7. Arluison M., Dietl M. and Thibault J. (1984) Ultrastructural morphology of dopaminergic nerve terminals and synapses in the striatum of the rat using tyrosine hydroxylase immunoreactivity: topographical study. Brain Res. Bull. 13, 269–285.

138

A. Charara et al.

8. Attwell D., Barbour B. and Szatkowski M. (1993) Nonvesicular release of neurotransmitter. Neuron 11, 401–407. 9. Bergman H., Wichmann T. and Delong M. R. (1990) Reversal of experimental Parkinsonism by lesions of the subthalamic nucleus. Science 249, 1436–1438. 10. Bevan M. D. and Bolam J. P. (1995) Cholinergic, GABAergic, and glutamate-enriched inputs from the mesopontine tegmentum to the subthalamic nucleus in the rat. J. Neurosci. 15, 7105–7120. 11. Bevan M. D., Francis C. M. and Bolam J. P. (1995) The glutamate enriched cortical and thalamic input to neurons in the subthalamic nucleus of the rat: convergence with GABA-positive terminals. J. comp. Neurol. 361, 491–511. 12. Bolam J. P., Wainer B. H. and Smith A. D. (1984) Characterization of cholinergic neurons in the rat neostriatum. A combination of choline acetyltransferase immunocytochemistry, Golgi-impregnation and electron microscopy. Neuroscience 12, 711–718. 13. Bolam J. P., Powell J. F., Wu J. Y. and Smith A. D. (1985) Glutamate decarboxylase-immunoreactive structures in the rat neostriatum: a correlated light and electron microscopic study including a combination of Golgi impregnation with immunocytochemistry. J. comp. Neurol. 237, 1–20. 14. Bormann J. (1988) Electrophysiology of GABAA and GABAB receptor subtypes. Trends Neurosci. 11, 112–116. 15. Bowery N. G. (1989) GABAB receptors and their significance in mammalian pharmacology. Trends pharmac. Sci. 10, 401–407. 16. Bowery N. G., Price G. W., Hudson A. L., Hill D. R., Wilkin G. P. and Thurnbull M. L. (1984) GABA receptor multiplicity. Visualization of different receptor subtypes in the mammalian CNS. Neuropharmacology 23, 219–231. 17. Bowery N. G., Hudson A. L. and Price G. W. (1987) GABAA and GABAB receptor site distribution in the rat nervous system. Neuroscience 20, 365–383. 18. Calabresi P., Mercuri N. B., Demurtas M. and Bernardi G. (1990) Endogeneous GABA mediates presynaptic inhibition of spontaneous and evoked excitatory synaptic potentials in the rat neostriatum. Neurosci. Lett. 118, 99–102. 19. Calabresi P., Mercuri N. B., Demurtas M. and Bernardi G. (1991) Involvement of GABA systems in feedback regulation of glutamate- and GABAmediated synaptic potentials in rat neostriatum. J. Neurophysiol. 440, 581–599. 20. Callachan H., Cottrell G. A., Hather N. Y., Lambert J. J., Nooney J. M. and Peters J. A. (1987) Modulation of the GABAA receptor by progesterone and metabolites. Proc. R. Soc. Lond. 231, 359–369. 21. Chan P. K. Y., Leung C. K. S. and Yung W. H. (1998) Differential expression of pre- and postsynaptic GABAB receptors in rat substantia nigra pars reticulata neurones. Eur. J. Pharmac. 349, 187–197. 22. Chang H. T., Wilson C. J. and Kitai S. T. (1981) Single neostriatal efferent axons in the globus pallidus: a light and electron microscopy study. Science 213, 915–918. 23. Charara A. and Smith Y. (1998) Subsynaptic distribution of GABA-A receptor subunits in the globus pallidus and subthalamic nucleus in monkeys. Soc. Neurosci. Abstr. 24, 1650. 24. Chesselet M. F. and Delfs J. M. (1996) Basal ganglia and movement disorders: an update. Trends Neurosci. 19, 417–422. 25. Chu D. C. M., Penney J. B. and Young A. B. (1987) Autoradiographic evidence for post-synaptic GABAB receptors in rat striatum and hippocampus. Soc. Neurosci. Abstr. 13, 952. 26. Chu D. C. M., Albin R. L., Young A. B. and Penney J. B. (1990) Distribution and kinetics of GABAB binding sites in rat central nervous system: a quantitative autoradiographic study. Neuroscience 34, 341–357. 27. Clarke N. P., Bolam J. P. and Bevan M. D. (1996) Glutamate enriched inputs from the mesopontine tegmentum to the entopeduncular nucleus in the rat. Eur. J. Neurosci. 8, 1363–1371. 28. Delong M. R. (1990) Primate models of movement disorders of basal ganglia origin. Trends Neurosci. 13, 281–285. 29. DiFiglia M., Pasik P. and Pasik T. (1976) A Golgi study of neuronal types in the neostriatum of monkeys. Brain Res. 114, 245–256. 30. DiFiglia M. and Aronin N. (1982) Ultrastructural features of immunoreactive somatostatin neurons in the rat caudate nucleus. J. Neurosci. 2, 1267–1274. 31. Dolphin A. C. and Scott R. H. (1986) Inhibition of calcium currents in cultured rat dorsal root ganglion neurones by (2)-baclofen. Br. J. Pharmac. 88, 213–220. 32. Dube´ L., Smith A. D. and Bolam J. P. (1988) Identification of synaptic terminals of thalamic or cortical origin in contact with distinct medium-size spiny neurons in the rat neostriatum. J. comp. Neurol. 267, 455–471. 33. Engberg G., Kling-Petersen T. and Nissbrandt H. (1993) GABAB-receptor activation alters the firing pattern of dopamine neurons in the rat substantia nigra. Synapse 15, 229–238. 34. Floran B., Silva I., Nava C. and Aceves J. (1988) Presynaptic modulation of the release of GABA by GABAA receptors in pars compacta and by GABAB receptors in pars reticulata of the rat substantia nigra. Eur. J. Pharmac. 150, 277–286. 35. Freund T. F., Powell J. F. and Smith A. D. (1984) Tyrosine hydroxylase-immunoreactive boutons in synaptic contact with identified striatonigral neurons, with particular reference to dendritic spines. Neuroscience 13, 1189–1215. 36. Fritschy J.-M., Meskenaite V., Weinmann O., Honer M., Benke D. and Mohler H. (1999) GABAB-receptor splice variants GB1a and GB1b in rat brain: developmental regulation, cellular distribution and extrasynaptic localization. Eur. J. Neurosci. 11, 761–768. 37. Ga¨hwiler B. H. and Brown D. A. (1985) GABAB-receptor-activated K 1 current in voltage-clamped CA3 pyramidal cells in hippocampal cultures. Proc. natn. Acad. Sci. U.S.A. 82, 1558–1562. 38. Garant D., Sperber E. and Moshe S. (1992) The density of GABAB binding sites in the substantia nigra is greater in pups than in adults. Eur. J. Pharmac. 214, 75–78. 39. Gehlert D. R., Yamamura H. I. and Wamsley J. K. (1985) Gamma-aminobutyric acid B receptors in the rat brain: quantitative autoradiographic localization using [ 3H](2)-baclofen. Neurosci. Lett. 56, 183–188. 40. Giralt M. T., Bonanno G. and Raiteri M. (1990) GABA terminal autoreceptors in the pars compacta and in the pars reticulata of the rat substantia nigra are GABAB. Eur. J. Pharmac. 175, 137–144. 41. Hartman-Von Monakow K., Akert K. and Ku¨nzle H. (1978) Projections of the precentral motor cortex and other cortical areas of the frontal lobe to the subthalamic nucleus in the monkey. Expl Brain Res. 33, 395–403. 42. Ha¨usser M. A. and Yung W. H. (1994) Inhibitory synaptic potentials in guinea-pig substantia nigra dopamine neurones in vitro. J. Physiol. 479, 401–422. 43. Hazrati L.-N. and Parent A. (1992) The striatopallidal projection displays a high degree of anatomical specificity in the primate. Brain Res. 592, 213–227. 44. Hazrati L.-N., Parent A., Mitchell S. and Haber S. (1990) Evidence for interconnections between the two segments of the globus pallidus in primates: a PHA-L anterograde tracing study. Brain Res. 533, 171–175. 45. Hill D. R. and Bowery N. G. (1981) 3H-Baclofen and 3H-GABA binding to bicuculline-insensitive GABAB site in rat brain. Nature 290, 149–152. 46. Hsu S. M., Raine L. and Fanger H. (1981) Use of avidin–biotin–peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J. Histochem. Cytochem. 21, 557–580. 47. Isaacson J. S., Solı´s J. M. and Nicoll R. A. (1993) Local and diffusion synaptic actions of GABA in the hippocampus. Neuron 10, 165–175. 48. Jones K. A., Borowsky B., Tamm J. A., Craig D. A., Durkin M. M., Dai M., Yao W. J., Johnson M., Gunwaldsen C., Huang L. Y., Tang C., Shen Q., Salon J. A., Morse K., Laz T., Smith K. E., Nagarathnam D., Noble S. A., Branchek T. A. and Gerald C. (1998) GABAB receptors function as heteromeric assembly of the subunits GABABR1 and GABABR2. Nature 396, 674–679. 49. Kaupmann K., Huggel K., Heid J., Flor P. J., Bischoff S., Mickel S. J., McMaster G., Angst C., Bittiger H., Froestl W. and Bettler B. (1997) Expression cloning of GABAB receptors uncovers similarity to metabotropic glutamate receptors. Nature 396, 239–246.

GABAB receptors in the monkey basal ganglia

139

50. Kaupmann K., Malitschek B., Schuler V., Heid J., Froestl W., Beck P., Mosbacher J., Bischoff S., Kulik A., Shigemoto R., Karschin A. and Bettler B. (1998) GABAB-receptor subtypes assemble into functional heteromeric complexes. Nature 396, 683–687. 51. Kemp J. M. and Powell T. P. (1970) The cortico-striate projection in the monkey. Brain 93, 525–546. 52. Kilpatrick G. J., Muhyaddin M. S., Roberts P. J. and Woodruff G. N. (1983) GABAB binding sites on rat striatal synaptic membranes. Br. J. Pharmac. 78, (Suppl.) 6P. 53. Kincaid A. E., Penney J. B., Young A. B. and Newman S. W. (1991) Evidence for a projection from the globus pallidus to the entopeduncular nucleus in the rat. Neurosci. Lett. 128, 121–125. 54. Kita H. (1992) Responses of globus pallidus neurons to cortical stimulation: intracellular study in the rat. Brain Res. 589, 84–90. 55. Kita H. (1993) GABAergic circuits of the striatum. Prog. Brain Res. 99, 51–70. 56. Kita H. and Kitai S. T. (1987) Efferent projections of the subthalamic nucleus in the rat: light and electron microscopic analysis with the PHA-L method. J. comp. Neurol. 260, 435–452. 57. Kulik A., Nomura S., Yamaru Y., Malitschek B., Kuhn R., Bettler B. and Shigemoto R. (1998) Subcellular localization of metabotropic gammaaminobutyric acid receptor GABABR1 immunoreactivity in the rat brain. Soc. Neurosci. Abstr. 24, 1588. 58. Kuner R., Ko¨hr G., Gru¨newald S., Eisenhardt G., Bach A. and Kornau H. C. (1999) Role of heteromer formation in GABAB receptor function. Science 283, 74–77. 59. Lacey M. G., Mercuri N. B. and North R. A. (1988) On the potassium conductance increase activated by GABAB and dopamine D2 receptors in rat substantia nigra neurones. J. Physiol., Lond. 401, 437–453. 60. Levy R., Hazrati L.-N., Herrero M. T., Vila M., Hassani O. K., Mouroux M., Ruberg M., Aseni H., Agid Y., Feger J., Obeso J. A., Parent A. and Hirsch E. C. (1997) Re-evaluation of the functional anatomy of the basal ganglia in normal and parkinsonian states. Neuroscience 76, 335–343. 61. Macdonald R. L. and Olsen R. W. (1994) GABAA receptor channels. A. Rev. Neurosci. 17, 569–602. 62. Margeta-Mitrovic M., Mitrovic I., Riley R. C., Jan L. Y. and Basbaum A. I. (1999) Immunocytochemical localization of GABAB receptors in the rat central nervous system. J. comp. Neurol. 405, 299–321. 63. Marsden C. D. (1984) Function of the basal ganglia as revealed by cognitive and motor disorders in Parkinson’s disease. Can. J. neurol. Sci. 11, 129–135. 64. Moratalla R. and Bowery N. G. (1991) Chronic lesion of corticostriatal fibers reduces GABAB but not GABAA binding in rat caudate putamen—an autoradiographic study. Neurochem. Res. 16, 309–315. 65. Moriizumi T., Nakamura Y., Kitao Y. and Kudo M. (1987) Ultrastructural analyses of afferent terminals in the subthalamic nucleus of the cat with a combined degeneration and horseradish peroxidase tracing method. J. comp. Neurol. 265, 159–174. 66. Mun˜oz A., Huntsman M. M. and Jones E. G. (1998) GABAB receptor gene expression in monkey thalamus. J. comp. Neurol. 394, 118–126. 67. Newberry N. R. and Nicoll R. A. (1988) A bicuculline-resistant inhibitory post-synaptic potential in rat hippocampal pyramidal cells in vitro. Nature 332, 156–158. 68. Nisenbaum E. S., Berger T. W. and Grace A. A. (1992) Presynaptic modulation by GABAB receptors of glutamatergic excitation and GABAergic inhibition of neostriatal neurons. J. Neurophysiol. 67, 477–481. 69. Nisenbaum E. S., Berger T. W. and Grace A. A. (1993) Depression of glutamatergic and GABAergic synaptic responses in striatal spiny neurons by stimulation of presynaptic GABAB receptors. Synapse 14, 221–242. 70. Olsen R. W. and Tobin A. J. (1990) Molecular biology of GABAA receptors. Fedn Am. Socs exp. Biol. J. 4, 1469–1480. 71. Paladani C. A., Celada P. and Tepper J. M. (1999) Striatal, pallidal, and pars reticulata evoked inhibition of nigrostriatal dopaminergic neurons is mediated by GABAA receptors in vivo. Neuroscience 89, 799–812. 72. Paquet M. and Smith Y. (1997) Differential distribution of metabotropic glutamate receptors (mGluRs) in the monkey striatum. Soc. Neurosci. Abstr. 23, 188. 73. Parent A. and Hazrati L.-N. (1995) Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamo-cortical loop. Brain Res. Rev. 20, 91–127. 74. Parent A. and Hazrati L.-N. (1995) Functional anatomy of the basal ganglia. II. The place of the subthalamic nucleus and external pallidum in basal ganglia circuitry. Brain Res. Rev. 20, 128–154. 75. Precht W. and Yoshida M. (1971) Blockage of caudate-evoked inhibition of neurons in the substantia nigra by picrotoxin. Brain Res. 32, 229–233. 76. Reimann W. (1983) Inhibition by GABA, baclofen and gabapentin of dopamine release from rabbit caudate nucleus are there common or different sites of action?. Eur. J. Pharmac. 94, 341–344. 77. Reynolds E. S. (1963) The use of lead citrate at high pH as an electron opaque stain in electron microscopy. J. Cell Biol. 17, 208–212. 78. Ribak C. E., Vaughn J. E. and Roberts E. (1979) The GABA neurons and their axon terminals in rat corpus striatum as demonstrated by GAD immunocytochemistry. J. comp. Neurol. 187, 261–284. 79. Romero J., De Miguel R., Ramos J. A. and Fernandez-Ruiz J. J. (1998) The activation of cannabinoid receptors in striatonigral GABAergic neurons inhibited GABA uptake. Life Sci. 62, 351–363. 80. Sadikot A. F., Smith Y., Bolam J. P. and Parent A. (1992) Efferent connections of the centromedian and parafascicular thalamic nuclei in the squirrel monkey: a light and electron microscopic study of the thalamostriatal projection in relation to striatal heterogeneity. J. comp. Neurol. 320, 228–242. 81. Sadikot A., Parent A. and Franc¸ois C. (1992) Efferent connections of the centromedian and parafascicular thalamic nuclei in the squirrel monkey: a PHA-L study of subcortical projections. J. comp. Neurol. 315, 137–159. 82. Santiago M. and Westerink B. H. C. (1993) The role of GABA receptors in the control of nigrostriatal dopaminergic neurons: dual-probe microdialysis study in awake rats. Eur. J. Pharmac. 219, 175–181. 83. Sawynok J. and LaBella F. S. (1981) GABA and Baclofen potentiate the K 1-evoked release of methionine-enkephalin from rat striatal slices. Eur. J. Pharmac. 70, 103–110. 84. Seabrook G. R., Howson W. and Lacey M. G. (1990) Electrophysiological characterization of potent agonists and antagonists at pre- and postsynaptic GABAB receptors in rat brain slices. Br. J. Pharmac. 101, 949–957. 85. Seabrook G. R., Howson W. and Lacey M. G. (1991) Subpopulations of GABA-mediated synaptic potentials in slices of rat dorsal striatum are differentially modulated by presynaptic GABAB receptors. Brain Res. 562, 332–334. 86. Shen K. Z. and Johnson S. W. (1997) Presynaptic GABAB and adenosine A1 receptors regulate synaptic transmission to rat substantia nigra reticulata neurones. J. Physiol. 505, 153–163. 87. Shigemoto R., Kulik A., Tamaru Y., Malitschek B., Kuhn R. and Bettler B. (1998) Immunocytochemical distribution of GABABR1 in the rat CNS. Soc. Neurosci. Abstr. 24, 1587. 88. Shink E. and Smith Y. (1995) Differential synaptic innervation of neurons in the internal and external segments of the globus pallidus by the GABA- and glutamate-containing terminals in the squirrel monkey. J. comp. Neurol. 358, 119–141. 89. Shink E., Bevan M. D., Bolam J. P. and Smith Y. (1996) The subthalamic nucleus and the external pallidum—two tightly interconnected structures that control the output of the basal ganglia in the monkey. Neuroscience 73, 335–357. 90. Singh R. (1990) GABA-B receptors modulate glutamate release in the rat caudate and globus pallidus. Soc. Neurosci. Abstr. 16, 1041. 91. Smith Y. and Parent A. (1988) Neurons of the subthalamic nucleus in primates display glutamate but not GABA immunoreactivity. Brain Res. 453, 353–356. 92. Smith A. D. and Bolam J. P. (1990) The neural network of the basal ganglia as revealed by the study of synaptic connections of identified neurones. Trends Neurosci. 13, 259–265.

140

A. Charara et al.

93. Smith Y., Parent A., Seguela P. and Descarries L. (1987) Distribution of GABA-immunoreactive neurons in the basal ganglia of the squirrel monkey (Saimiri sciureus). J. comp. Neurol. 259, 50–61. 94. Smith Y., Bolam J. P. and Von Krosigk M. (1990) Topographical and synaptic organization of the GABA-containing pallidosubthalamic projection in the rat. Eur. J. Neurosci. 2, 500–511. 95. Smith Y., Bennett B. D., Bolam J. P., Parent A. and Sadikot A. F. (1994) Synaptic relationships between dopaminergic afferents and cortical or thalamic input in the sensorimotor territory of the striatum in monkey. J. comp. Neurol. 344, 1–19. 96. Smith Y., Whichmann T. and Delong M. R. (1994) Synaptic innervation of neurones in the internal pallidal segment by the subthalamic nucleus and the external pallidum in monkeys. J. comp. Neurol. 343, 297–318. 97. Smith Y., Bevan M. D., Shink E. and Bolam J. P. (1998) Microcircuitry of the direct and indirect pathways of the basal ganglia. Neuroscience 86, 353–387. 98. Smolders I., De Klippel N., Sarre S., Ebinger G. and Michotte Y. (1995) Tonic GABAergic modulation of striatal dopamine release studied by in vivo microdialysis in the freely moving rat. Eur. J. Pharmac. 284, 83–91. 99. Sugita S., Johnson S. W. and North R. A. (1992) Synaptic inputs to GABAA and GABAB receptors originate from discrete afferent neurons. Neurosci. Lett. 134, 207–211. 100. Tepper J. M., Martin L. P. and Anderson D. R. (1995) GABAA receptor-mediated inhibition of rat substantia nigra dopaminergic neurons by pars reticulata projection neurons. J. Neurosci. 15, 3092–3103. 101. Tepper J. M., Paladini C. A. and Celada P. (1998) GABAergic control of the firing pattern of substantia nigra dopaminergic neurons. Adv. Pharmac. 42, 694–699. 102. Veenman C. L., Albin R. L., Richfield E. K. and Reiner A. (1994) Distribution of GABAA, GABAB, and benzodiazepine receptors in the forebrain and midbrain of pigeons. J. comp. Neurol. 344, 161–189. 103. Waldvogel H. J., Fritschy J.-M., Mohler H. and Faull R. L. M. (1998) GABAA receptors in the primate basal ganglia: an autoradiographic and a light and electron microscopic immunohistochemical study of the a1 and b2,3 subunits in the baboon brain. J. comp. Neurol. 397, 297–325. 104. White J. H., Wise A., Main M. J., Green A., Fraser N. J., Disney G. H., Barnes A. A., Emson P., Foord S. M. and Marshall F. H. (1998) Heterodimerization is required for the formation of a functional GABAB receptor. Nature 396, 679–682. 105. Wichmann T. and Delong M. R. (1993) Pathophysiology of parkinsonian motor abnormalities. Adv. Neurol. 60, 53–61. 106. Wilkin G. P., Hudson A. L., Hill D. R. and Bowery N. G. (1981) Autoradiographic localization of GABAB receptors in rat cerebellum. Nature 294, 584–587. 107. Wilson J. S. and Wilson J. A. (1985) Baclofen attenuates hyperpolarizing not depolarizing responses of caudate neurons in cat. Brain Res. 342, 396–400. 108. Xu Z. C., Wilson C. J. and Emson P. C. (1991) Restoration of thalamostriatal projections in rat neostriatal grafts: an electron microscopic analysis. J. comp. Neurol. 303, 22–34. 109. Yoon K. W. and Rothman S. M. (1991) The modulation of rat hippocampal synaptic conductances by baclofen and gamma-aminobutyric acid. J. Physiol. 442, 377–390. (Accepted 2 August 1999)