Brain Research 989 (2003) 135146 www.elsevier.

com / locate / brainres

Research report

Distribution of a GABA B -like receptor protein in the rat central nervous system
K.J. Charles* ,1 , A.R. Calver, S. Jourdain, M.N. Pangalos 2
Neurology Centre of Excellence for Drug Discovery, GlaxoSmithKline, New Frontiers Science Park North, Third Avenue, Harlow, Essex, CM19 5 AW, UK Accepted 5 June 2003

Abstract Using a homology-based bioinformatics approach we have identied the human and rodent orthologues of a novel putative seven transmembrane G protein coupled receptor, termed GABA BL . The amino acid sequence homology of these cDNAs compared to GABA B1 and GABA B2 led us to postulate that GABA BL may be a putative novel GABA B receptor subunit. We have developed a rabbit polyclonal antisera specic to the GABA BL protein and assessed the distribution of GABA BL in the rat CNS by immunohistochemistry. Protein expression was particularly dense in regions previously shown to contain known GABA B receptor subunits. Dense immunoreactivity was observed in the cortex, major subelds of the hippocampus and the dentate gyrus. GABA BL labelling was very conspicuous in the cerebellum, both in the granule cell layer and in Purkinje cells, and was also observed in the substantia gelatinosa and ventral horn motor neurons of the spinal cord. GABA BL immunoreactivity was also noted in a subset of parvalbumin positive hippocampal interneurons. Our data suggest a widespread distribution of GABA BL throughout the rat CNS. 2003 Elsevier B.V. All rights reserved.
Theme: Neurotransmitters, modulators, transporters and receptors Topic: GABA receptors Keywords: GABA B ; Receptor; CNS; Immunohistochemistry

1. Introduction GABA B receptors are members of the group C subfamily of G protein coupled receptors (GPCRs), which also include the metabotropic glutamate receptor, the

calcium-sensing receptor and a group of putative pheromone receptors [3,9]. A distinguishing characteristic of the group C GPCRs is their large N-terminal extracellular domain, which contains the agonist binding site [20]. Functional GABA B receptors are obligate heterodimers

Abbreviations : 7, facial nucleus; Acb, accumbens; AcbC, accumbens core; AcbSh, accumbens shell; AH, anterior hypothalamus; B, nucleus basalis (Meynert); BLA, basolateral amygdala; BNST, bed nucleus of the stria terminalis; Cl, claustrum; CPu, caudate putamen; DMH, dorsal medial hypothalamus; DR, dorsal raphe; gl, granule cell layer of cerebellum; GP, globus pallidus; H, habenula; Hi, hippocampal formation; InC, inferior colliculus; IP, interpeduncular nucleus; LA, lateral amygdala; LAH, lateral anterior hypothalamus; LC, locus coeruleus; LD, laterodorsal thalamic nucleus; LH, lateral hypothalmus; LHb, lateral habenula; LSD, lateral septal nucleus, dorsal; LSI, lateral septal nucleus, intermediate; LSV, lateral septal nucleus, ventral; MG, medial geniculate nucleus; MD, mediodorsal thalamic nucleus; ml, molecular layer of cerebellum; MLF, medial lateral fasciculus; MnR, median raphe; Mo5, motor trigeminal nucleus; MV, medial vestibular nucleus; P, Purkinje cells; Pir, Piriform cortex; Pn, pontine nuclei; Re, reuniens thalamic nucleus; Rh, rhomboid thalamic nucleus; Rt, reticular thalamic nucleus; SG, substantia gelatinosa of spinal cord; SN, substantia nigra; Sp,spinal nucleus; SuG, supercial gray layer of superior colliculus; Th, thalamus; V, ventral thalamic nucleus; VDB, vertical limb of diagonal band; VH, ventral hypothalamus; KLH, keyhole lympet haematoxylin; PAGE, poly acrylamide gel electrophoresis; PBS, phosphate buffered saline * Corresponding author. Tel.: 11-732-274-4620; fax: 11-732-274-4018. E-mail address: charlek1@wyeth.com (K.J. Charles). 1 Present address: Wyeth Research, CN800, Princeton, NJ 08543, USA. 2 Present address: Wyeth Research, CN800, Princeton, NJ 08543, USA. 0006-8993 / 03 / $ see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016 / S0006-8993(03)03163-9

136

K. J. Charles et al. / Brain Research 989 (2003) 135146

consisting of GABA B1 and GABA B2 subunits [2,10,11,16,17,1921,31]. The GABA B1 receptor subunit is responsible for the binding of agonist (GABA), whilst the GABA B2 subunit is responsible for trafcking of the heterodimer to the cell surface [5,11] and for G protein coupling [10,21,29]. Historically, GABA B receptors are heterogeneous in nature based on their pharmacology [4,18,27]. Thus far no other subunits have been identied and of the reported splice variants for GABA B1 none appear to confer any pharmacological GABA B receptor heterogeneity [6,15,22,30]. GABA B1 has also been shown to be expressed in the absence of GABA B2 in a number of peripheral tissues [6], despite the presence of functional peripheral receptors [27], further strengthening the argument for the existence of additional GABA B receptor subunits. As a result, signicant efforts have been afforded to the discovery of additional mammalian GABA B subunits. A recent report has identied a putative third GABA B receptor subunit in drosophila, termed GABA B3 [23], though as yet no functional activity has been demonstrated. We have recently identied a novel GPCR with homology to both GABA B1 and GABA B2 that we have termed GABA BL , and we have cloned and expressed the human, mouse and rat orthologues of GABA BL (accession numbers: AF488739, AF488740, AF488741). Although this receptor bears signicant sequence homology to GABA B subunits, functional studies have thus far failed to demonstrate any modulation of receptor binding or signalling when GABA BL is expressed transiently in cells with the two known GABA B subunits [7]. We have generated a specic rabbit polyclonal antisera to this novel orphan family C GPCR and report the distribution of GABA BL in the rat CNS.

2.2. Tissue preparation

Male SpragueDawley rats were anaesthetised with sodium pentobarbital prior to being transcardially perfused with 4% paraformaldehyde (PFA) in PBS (pH 7.4). Brains and spinal cords were removed and post-xed in 4% PFA for 24 h before transfer to 30% sucrose for 48 h at 4 8C. Tissues were frozen in isopentane at 240 8C and 35-mm brain and spinal cord sections were cut at approximately 220 8C using a cryostat. Brain and spinal cord sections were stored in cryoprotectant containing 30% glycerol and 30% ethylene glycol in PBS at 220 8C. All procedures involving experimental animals were conducted in accordance with the United Kingdom Animals (Scientic Procedures) Act, 1986 and conformed to GlaxoSmithKline ethical standards.

2.3. Immunohistochemistry
Immunoreactivity was revealed using the ABC detection system as described elsewhere [13]. Briey, all sections were washed thoroughly in PBS, permeabilised in 0.1% Triton X-100 and endogenous peroxidase activity was removed by 1% H 2 O 2 in PBS. Non-specic antibody binding was reduced by pre-incubation in a PBS blocking solution containing 1% bovine serum albumin and 10% normal goat serum. Sections were incubated with primary antisera against GABA BL at 0.86 mg / ml in blocking solution for 48 h at 4 8C. In control experiments, primary antibodies were pre-absorbed with an excess amount of the immunogenic peptide (10 mg / ml) at 4 8C for 72 h prior to tissue incubation. Sections were then incubated with biotinylated goat anti-rabbit antisera (1:200, Vector Labs, Burlingame, CA) followed by a peroxidase-conjugated avidinbiotin complex and 3,39-diaminobenzidine tetrachloride (Vector Labs). Sections were dehydrated in 100% ethanol, immersed in Histolene (CellPath, Leeds, UK) for 10 min and coverslipped with DPX (BDH, Lutterworth, UK). Sections were analysed using a Leica DMR microscope equipped with a Leica DC200 digital camera. Images were prepared using PaintShop Pro Version 7.0 (JASC, Eden Prairie, MN) for grey scale conversion, contrast and brightness adjustments and cropping.

2. Experimental procedures

2.1. Antisera
A rabbit polyclonal antisera was raised against the peptide CREKLQEVLQE (amino acids 527536) [7]. Peptides were synthesised by Research Genetics Inc. (Alabama, USA) and covalently coupled to the carrier keyhole limpet haemocyanin (KLH). Two rabbits were injected with 0.1 mg of the peptideKLH conjugate in Freunds complete adjuvant, and boosted four times with the same amount of antigen in incomplete adjuvant. Terminal bleeds were clotted overnight at 4 8C and the serum separated from blood cells by centrifugation at 80003g. Antigens were immobilised on activated supports to allow for the afnity purication of antisera. Elution from columns was via a pH gradient and fractions were collected and stored in 0.125 M borate buffer. Antisera specicity was conrmed by immunoblotting as described elsewhere [7].

2.4. Immunouorescence
For double labelling studies, tissue sections were initially labelled with anti-GABA BL antisera overnight at room temperature followed by direct detection with Alexa 488 goat anti-rabbit (Molecular Probes, Eugene, OR) at 1:50 in PBS. Sections were then incubated with either mouse anti-GFAP (Sigma) at 1:2000; mouse anti-calretinin; mouse anti-calbindin or mouse anti-parvalbumin (Swant, Bellinzona, CH) all at 1:500 in PBS overnight at room temperature followed by detection with Alexa 633 goat anti-mouse. After washing in PBS sections were mounted

K. J. Charles et al. / Brain Research 989 (2003) 135146

137

onto superfrost plus slides and coverslipped with Citiuor (Citiuor, Canterbury, UK).

3. Results

3.1. Immunohistochemistry
The specicity of anti-GABA BL antisera has been assessed by immunoblotting and pre-absorption immunohistochemistry, as described elsewhere [7]. GABA BL immunoreactivity (IR) was widely distributed throughout the rat brain and spinal cord (Fig. 1). GABA BL

protein expression was seen throughout the neocortex, layers IIVI (Table 1), the caudate putamen, thalamus and hypothalamus. GABA BL IR was particularly intense in the pyramidal cell layers of the hippocampus (CA13) and also in the granule cells of the dentate gyrus. The supercial gray region of the superior colliculus also demonstrated GABA BL IR, as did the reticular and pontine nuclei, the locus ceruleus and the facial nucleus. Immunoreactivity was conspicuous in the granule cell layer of the cerebellum, and in the deep cerebellar nuclei. More detailed results are given below. Results of GABA BL IR are summarised in Tables 13 and are based upon subjective measures of the apparent IR density. Immunoreactivity

Fig. 1. Photomicrographs showing GABA BL protein expression in coronal sections throughout the rat brain. GABA BL was highly expressed in the hippocampus (C,D), piriform cortex (AC), hypothalamus (B,C), interpeduncular nucleus (E), supercial grey of the superior colliculus (E), raphe nuclei (E) and was very conspicuous in the cerebellum (F). For abbreviations see list. Scale bar55 mm.

138

K. J. Charles et al. / Brain Research 989 (2003) 135146

Table 1 Semi-quantitative analysis of GABA BL protein expression in the rat cortex Cortical area Cortical layer I Orbital Motor Parietal Visual Piriform Insular Somatosensory Auditory 1 1 1 1 1 1 1 1 II 1111 111 1111 11111 11111 111 1111 111 III 111 111 111 111 111 111 111 111 IV 1111 1111 1111 1111 11 1111 11111 1111 V 1111 1111 111 111 111 1111 111 111 VI 1111 111 11 11 111 111 11 111

Many, but not all neurons were positive for GABA BL receptor protein in each structure; none (2), very low ( 1 ), low ( 11 ), moderate ( 111 ), dense ( 1111 ) or very dense ( 11111 ).

has been scored as none (2), very low ( 1 ), low ( 11 ), moderate ( 111 ), dense ( 1111 ) and very dense ( 111 11 ).

3.2. Cerebral cortex

GABA BL IR was detected throughout the neocortex in neuronal somata, though the intensity of IR varied according to the cortical layer and region studied (Table 1). Throughout all cortical areas studied layer I showed only minimal IR for GABA BL protein in small scattered cell body proles consistent with the sparse occurance of neurons in this layer. Layer II of all cortical areas, in contrast, showed moderate to dense IR for GABA BL receptor protein, with the visual and piriform corticies demonstrating very dense GABA BL IR. The large pyramidal neurons of layer IV were also dense in their GABA BL protein expression, except in the somatosensory cortex where IR was very dense (Fig. 2A) and the piriform cortex where GABA BL IR was low. Notably there was a total absence of dendritic labelling in these large pyramidal neurons, instead IR was located in the membrane and cytoplasm of the soma, a trend that was observed throughout all cortical layers and regions. Protein expression in layers V and VI was variable according to the region studied. Parietal, visual, piriform and somatosensory cortices demonstrated low GABA BL protein expression, in contrast to dense IR noted in orbital cortex.

accumbens, GABA BL IR was restricted to the membrane of cell bodies. Fewer somata were labelled in the globus pallidus than in other regions of the basal ganglia. GABA BL IR in the globus pallidus was once again restricted to cell membranes, though the actual intensity of this staining was greater than seen in the caudate putamen. Septal GABA BL IR was dense in the lateral septal regions (Fig. 1A) being expressed mainly in cell bodies but also to a lesser degree in the neuropil. GABA BL IR in the bed nucleus of the stria terminalis (BNST) was moderate.

3.4. Amygdala
GABA BL immunoreactivity in the amygdala varied according to subnuclei (see Table 2 and Fig. 1C). In the basolateral amygdala GABA BL IR was dense, and was located primarily to the cell membrane of cell bodies. The intercalated masses also showed dense GABA BL IR, however, the lateral and central nuclei demonstrated only moderate GABA BL IR, again in a cell body restricted manner.

3.5. Hippocampus and habenula

In the hippocampus GABA BL IR was striking in its dense / very dense labelling of the pyramidal and granule cells, respectively (Fig. 1C,D and Fig. 3). The pyramidal cells of the CA1 sub region were densely IR for GABA BL with labelling restricted to the cell body layer. In the stratum oriens and radiatum of the CA1 region GABA BL IR was restricted to only very few interneurons, with no dendritic or neuropilic labelling observed (Fig. 3A). A similar pattern of IR was also noted in the CA3 pyramidal cells with dense GABA BL IR restricted to the cell body layer (Fig. 3B). GABA BL IR was very dense in the denate gyrus with IR in the cell bodies of granule cells, and also in occasional interneurons in the hilus (Fig. 3C). The habenula was very densely immunoreactive for GABA BL in the medial nucleus (Fig. 1C), with only moderate IR in the lateral region. The interpeduncular nucleus of the mid

3.3. Basal ganglia, nucleus accumbens and septal nuclei

GABA BL IR was noted throughout the regions of the basal ganglia and nucleus accumbens of the rat (Fig. 1A). In the caudate putamen IR was restricted to the cell somata of neurons with a cell diameter consistent with that of medium spiny interneurons (20 mm) (Fig. 2B). Labelling in these cells was mainly restricted to the cell membrane, with a weak cytoplasmic expression also noted. GABA BL IR was also observed in the nucleus accumbens, though the shell demonstrated dense IR in contrast to the core where protein expression was moderate. In both regions of the

K. J. Charles et al. / Brain Research 989 (2003) 135146 Table 2 Semi-quantitative analysis of GABA BL protein expression in the rat brain Brain regions Basal ganglia Structure Caudate-putamen Accumbens shell Accumbens core Globus pallidus GABA BL 111 1111 111 11 1111 11 1 1 1111 111 111 1111 1111 1111 11111 11 11111 111 11111 1111 1111 1111 111 1111 1111 111 1111 111 1111 1111 1111 111 111 1111 11111 1111 111 1111 11111 111 11111 11 1111 1111 1111

139

Table 3 Semi-quantitative distribution of GABA BL protein expression in control rat lumbar spinal cord Lamina I II III IV V VI VII VIII IX X GABA B1a 1111 1111 111 111 11 11 111 111 1111 1111

Septum BNST Diagonal band Amygdala Bas. Meynert n. Basolateral Lateral Central Intercalated mass CA1 CA3 Dentate gyrus interneurons Medial Lateral Medial Lateral Ventral Reticular thalamus Dorsal Ventromedial Lateral Arcuate Tuber cinereum Superior colliculus Inferior colliculus Interpeduncular n. Ventral tegmental a Substantia nigra pc Substantia nigra pr Pontine nuclei Spinal trigeminal n. Locus coeruleus Vestibular nuclei Dorsal raphe nuclei Median raphe nuclei Olivary nuclei Molecular layer Purkinje cells Granule layer Deep nuclei

Hippocampus

Many, but not all neurons were positive for GABA BL receptor protein in each structure; none (2), very low ( 1 ), low ( 11 ), moderate ( 111 ), dense ( 1111 ) or very dense ( 11111 ).

Habenula Interpeduncular Thalamus

positive astrocytes in the hippocampus did not express GABA BL protein (Fig. 5DF).

3.6. Thalamus
Most regions of the thalamus demonstrated moderate to dense GABA BL IR (Fig. 1BD). Medial thalamic nuclei were densely immunoreactive for GABA BL as were the lateral and ventral nuclei, with labelling mainly observed in the cell body membrane. The reticular thalamus demonstrated moderate GABA BL IR (Fig. 2D), here GABA BL protein was expressed in cell bodies and little or no dendritic or neuropilic IR was observed.

Hypothalamus

Mid brain

Pons Rhombenceph.

3.7. Hypothalmus
Expression levels of GABA BL in the rat hypothalamus were moderate to dense according to the nucleus (see Table 2 and Fig. 1B,C). The dorsal and ventromedial hypothalamic nuclei both demonstrated dense GABA BL IR as did the median eminence. Immunoreactivity in the arcuate nucleus was also dense (Fig. 2C) with GABA BL protein observed in the cell membrane and cytoplasm. Moderate GABA BL expression was noted in the lateral nucleus and in the tuber cinereum. In general throughout the hypothalamus GABA BL IR was present in cell bodies but not in dendrites or neuropil.

Cerebellum

Many, but not all neurons were positive for GABA BL receptor protein in each structure; none (2), very low ( 1 ), low ( 11 ), moderate ( 111 ), dense ( 1111 ) or very dense ( 11111 ).

brain, to which the habenula projects, was also dense for GABA BL protein expression (Fig. 1D). GABA BL protein was expressed on parvalbumin positive interneurons in the CA1 stratum pyramidale of the hippocampus, predominantly in cell bodies (Fig. 4AC). Calbindin positive interneurons in the CA3 regions were also positive for GABA BL , however, the calbindin positive mossy bres did not show co-expression with GABA BL (Fig. 4DF). GABA BL protein was also expressed calretinin positive interneurons in the CA1 and other regions of the rat hippocampus (Fig. 5AC). However, GFAP

3.8. Midbrain structures

GABA BL IR varied in the superior colliculus in a layer specic manner. In the supercial grey layer GABA BL IR was very dense, whereas IR was moderate in all other areas, though all exhibited the same pattern of expression in that only cell body labelling was observed (Fig. 1D,E). In the inferior colliculus GABA BL IR was homogenous throughout all regions in the neuronal somata. GABA BL IR in the substantia nigra was dense in the pars reticulata, whereas it was only moderate in the pars compacta. In both

140

K. J. Charles et al. / Brain Research 989 (2003) 135146

Fig. 2. Photomicrographs showing GABA BL immunoreactivity in layer V of the somatosensory cortex (A), medium spiny neurons of the caudate putamen (B), arcuate nucleus of the hypothalamus (C), and the reticular thalamic nucleus (D). Scale bars550 mm.

nuclei of the substantia nigra GABA BL labelling was restricted to cell bodies with no neuropilic IR observed. GABA BL labelling of the ventral tegmental area was also noted, in this region of the midbrain IR was moderate in cell body proles only.

3.9. Pons, rhombencephalon and cerebellum

In the monoaminergic neurons of the pons GABA BL IR was very dense (Fig. 1E) being restricted once again to cell bodies. In the spinal trigeminal and vestibular nuclei dense GABA BL IR was noted, though the locus coeruleus exhibited only moderate IR. Raphe nuclei exhibited varying intensities of GABA BL IR, with the dorsal raphe demonstrating very dense protein expression (Figs. 1E and 6A), moderate GABA BL IR was also present in the median raphe, and was dense in the nucleus of the solitary tract (Fig. 6B). The cerebellum demonstrated low to dense GABA BL IR depending on the layer. The molecular layer demonstrated only low GABA BL IR, which was restricted to small cell bodies, presumably those of stellate cells (Fig. 6C). The

Purkinje cells revealed dense GABA BL protein expression with IR restricted to the cell body and none noted in the dendrites projecting into the molecular cell layer. Dense GABA BL IR was observed in the granule cell layer of the cerebellum (Fig. 6C). The deep cerebellar nuclei also exhibited dense GABA BL IR again in cell bodies but not dendrites or neuropil.

3.10. Spinal cord

GABA BL protein expression in the spinal cord was widespread (Table 3 and Fig. 7A). Supercial laminae of the dorsal horn exhibited dense GABA BL IR in small cell bodies with no neuropilic expression noted (Fig. 7B). In the deeper laminae GABA BL IR was less intense and in laminae V and VI was only low. The deep laminae of the ventral horn expressed moderate to dense GABA BL IR, with the large motor neurons of lamina IX demonstrating dense membrane bound IR (Fig. 7C). Lamina X around the central canal also exhibited dense protein expression. GABA BL IR was also observed in the white matter of the spinal cord.

K. J. Charles et al. / Brain Research 989 (2003) 135146

141

Fig. 3. Photomicrographs showing GABA BL protein expression in the rat hippocampus. GABA BL IR was dense in the CA1 (A), CA3 (B) and in the dentate granule layer and hilus (h) (C). Scale bars550 mm.

4. Discussion The required heterodimerisation of two distinct GABA B receptor subunits for the formation of functional receptors coupled with the observed discrepancies between tissue expression in the two known subunits has led to an intense search for other novel subunits [16,17,19,31]. We have identied through database mining a novel orphan seven transmembrane protein with homology to and structural characteristics of the GABA B receptor subunits and the recently reported Drosophila GABA B3 subunit [23]. The protein expression of GABA BL IR in the rat brain and spinal cord was reminiscent of that previously observed

with antisera specic to GABA B1a , GABA B1b and GABA B2 receptor subunits [8] and was in agreement with GABA BL mRNA analysis [7]. Highest levels of GABA BL immunoreactivity were seen in areas that also demonstrated high levels of GABA B receptor subunits, namely the cortex, thalamus, hypothalamus, hippocampus, the granule cell layer of the cerebellum and throughout the spinal cord, which correlates well with previous Taqman data for the GABA BL protein [7]. Areas known to be modulated by GABA also demonstrated GABA BL protein expression. For example, the inhibitory interneurons of the reticular thalamus which are known to be GABAergic and involved in the thalamo-cortical loop [1], expressed mod-

142

K. J. Charles et al. / Brain Research 989 (2003) 135146

Fig. 4. Confocal images showing GABA BL protein co-localisation in the rat hippocampus. GABA BL was expressed in the CA1 stratum pyramidale (A) as were occasional parvalbumin positive interneurons (B), the merged image (C) shows GABA BL protein expressed on parvalbumin positive interneurons (arrows). CA3 stratum pyramidale cells expressed GABA BL protein (D), the stratum radiatum of the CA3 expressed calbindin positive projections and scattered calbindin-positive interneurons (E), the merged image shows co-localisation (arrows) in the interneurons but not in the mossy bres (F).

erate levels of GABA BL in cell soma, although no IR was observed in the neuropil. Transient calcium current activation has been shown to be regulated by pre- and post-

synaptic GABA B receptors in the hypothalamus [26], and dense GABA BL IR was observed in such regions. In addition, an area of intense GABA BL immunoreactivity in

K. J. Charles et al. / Brain Research 989 (2003) 135146

143

Fig. 5. Confocal images showing GABA BL protein expression in the hippocampus. GABA BL was expressed on the pyramidal cells of the CA1 region (A), calretinin positive interneurons were also scattered throughout the pyramidal cell layer (B), the merged image shows GABA BL protein expressed on the calretinin positive interneurons (arrows) (C). GABA BL protein expression in the CA1 region (D) was not present on GFAP positive astrocytes (E) in that or any other region of the rat hippocampus as shown by a separation of the two uorescent channels.

the current study was the cerebellum, the Purkinje cells of which show inhibited voltage-gated calcium currents upon application of the GABA B receptor agonist baclofen [24]. Throughout the rat brain and spinal cord GABA BL

protein expression was observed on neuronal cell body proles and not in the surrounding neuropil, a pattern of localisation indicative of post-synaptic receptor expression. However, GABA BL was expressed in the medial habenula

144

K. J. Charles et al. / Brain Research 989 (2003) 135146

Fig. 6. Photomicrographs showing GABA BL IR in the hind brain. GABA BL IR was dense in the dorsal raphe (A), the nucleus of the solitary tract (B). In the cerebellum the granule cell layer (gl) and Purkinje cells were densely immunoreactive for GABA BL (C), a few scattered cells, presumably stellate cell bodies, in the molecular cell layer (ml) also showed GABA BL IR. Scale bars550 mm.

which projects via the fasiculus retroexus to the interpeduncular nucleus, which also demonstrated dense GABA BL IR. Expression in these structures may indicate both pre- and post-synaptic expression. Clues as to the potential inhibitory / excitatory nature of this orphan receptor can be inferred from its cerebellar distribution. The Purkinje cells, which receive excitatory input from ascending mossy bres, as well as inhibitory

input from surrounding inhibitory interneurons, demonstrate strong GABA BL protein expression, as do the granule cells, which act as excitatory interneurons. Golgi cells situated close to the Purkinje cells in the granule cell layer act as inhibitory interneurons. GABA BL IR was observed on these interneurons suggesting a potential inhibitory role of GABA BL in this system. Further evidence pointing towards an inhibitory role of this orphan

K. J. Charles et al. / Brain Research 989 (2003) 135146

145

receptor is its expression on inhibitory interneurons of the hippocampus. GABA BL protein was expressed on parvalbumin-positive inhibitory interneurons in all regions of the hippocampus, which represent approximately half of the GABAergic neurons in the principle cell layers of the hippocampus [12]. Parvalbumin-positive interneurons represent basket or chandelier cells. Chandelier and basket cells are known to innervate the somata and initial axon segment of the principle cells of the hippocampus, the organisation of their dendritic tree suggests that they are in a position to receive excitatory inputs from all major afferent sources, and the axon collaterals of these cells in the CA1 project horizontally to pyramidal cells in the stratum pyramidale [12]. These cells are characteristically GABAergic in nature, and as such play a large role in maintaining hippocampal tone by inhibition. Calbindin, an intracellular Ca 2 1 -binding protein which represents a marker of a different subset of neurons in the hippocampus, is found throughout the dentate gyrus, including the mossy bres of these dentate granule cells, and does not represent basket or chandelier interneurons. Calbindin positive neurons are believed to innervate pryamidal cell dendrites, consistent with the mossy bre labelling we have observed in this study. Calbindin positive interneurons are also found throughout the CA13 subregions and represent GABAergic interneurons with distant projections [12]. In the CA3 calbindin positive interneurons were observed in the stratum oriens, pyramidale and radiatum. These interneurons were found to contain GABA BL protein as seen by immunouorescence. The calretinin positive interneurons visualised in the CA1 region of the rat hippocampus by immunouorescence probably represent the spine-free sub-population, which are described as interneurons specialised to innervate other interneurons [12]. GABA BL protein expression was observed on these interneurons in the CA1 and also with calretinin positive interneurons in the CA3 (not shown) which probably represent the spiny sub-population of calretinin positive interneurons thought to innervate principal cell dendrites but with mossy bre innervation. Thus with regard to the calretinin-positive sub-population GABA BL is expressed on both spiny and aspiny hippocampal interneurons. GABA B receptor subunits are expressed on GFAP positive astrocytes in the hippocampus [14,25,28]. In contrast, GABA BL protein was not observed on such cells and therefore we assume GABA BL has no inuence on normal astrocytic function. In conclusion, we have demonstrated for the rst time

Fig. 7. Photomicrographs showing GABA BL protein expression in the rat spinal cord (A). GABA BL IR was dense in the dorsal horn (B) and also in the ventral horn motor neurons (C). Some white matter IR was also evident (A). Scale bars A55 mm; B,C550 mm.

146

K. J. Charles et al. / Brain Research 989 (2003) 135146 [10] B. Duthey, S. Caudron, J. Perroy, B. Bettler, L. Fagni, J.P. Pin et al., A single subunit (GB2) is required for G protein activation by the heterodimeric GABAB receptor, J. Biol. Chem. 277 (2002) 3236 3241. [11] A.K. Filippov, A. Couve, M.N. Pangalos, F.S. Walsh, D.A. Brown, S.J. Moss, Heteromeric assembly of GABA(B)R1 and GABA(B)R2 receptor subunits inhibits Ca(21 ) current in sympathetic neurons, J. Neurosci. 20 (2000) 28672874. [12] T.F. Freund, G. Buzsaki, Interneurons in the hippocampus, Hippocampus 6 (1996) 347470. [13] G. Hervieu, P.C. Emson, The localisation of somatostatin receptor 1 sst1 immunoreactivity in the rat brain using an N-terminal specic antibody, Neuroscience 85 (1998) 12631284. [14] E. Hosli, L. Hosli, Evidence for GABAB-receptors on cultured astrocytes of rat CNS: autoradiographic binding studies, Exp. Brain. Res. 80 (1990) 621625. [15] S. Isomoto, M. Kaibra, Y. Sakurai-Yamashita, Y. Nagayama, Y. Uezono, K. Yano et al., Cloning and tissue distribution of novel splice variants of the rat GABAB receptor, Biochem. Biophys. Res. Commun. 253 (1998) 1015. [16] K.A. Jones, B. Borowsky, J.A. Tamm, D.A. Craig, M.M. Durkin, M. Dai et al., GABA(B) receptors function as a heteromeric assembly of the subunits GABA(B)R1 and GABA(B)R2, Nature 396 (1998) 674679. [17] K. Kaupmann, K. Huggel, J. Heid, P.J. Flor, S. Bischoff, S.J. Mickel et al., GABA(B)-receptor subtypes assemble into functional heteromeric complexes, Nature 396 (1998) 683687. [18] D.I. Kerr, J. Ong, GABAB receptors, Pharmacol. Ther. 67 (1995) 187246. [19] R. Kuner, G. Kohr, S. Grunewald, G. Eisenhardt, A. Bach, H.C. Kornau, Role of heteromer formation in GABAB receptor function, Science 283 (1999) 7477. [20] B. Malitschek, C. Schweizer, M. Keir, J. Heid, W. Froestl, J. Mosbacher et al., The N-terminal domain of gamma-aminobutyric Acid(B) receptors is sufcient to specify agonist and antagonist binding, Mol. Pharmacol. 56 (1999) 448454. [21] M. Margeta-Mitrovic, Y.N. Jan, L.Y. Jan, Function of GB1 and GB2 subunits in G protein coupling of GABAB receptors, Proc. Natl. Acad. Sci. USA 98 (2001) 1464914654. [22] S.C. Martin, S.J. Russek, D.H. Farb, Human GABA(B)R genomic structure: evidence for splice variants in GABA(B)R1 but not GABA(B)R2, Gene 278 (2001) 6379. [23] M. Mezler, T. Muller, K. Raming, Cloning and functional expression of GABA(B) receptors from Drosophila, Eur. J. Neurosci. 13 (2001) 477486. [24] I.M. Mintz, B.P. Bean, GABA B receptor inhibition of P-type Ca 2 1 channels in the central neurons, Neuron 10 (1993) 889898. [25] M. Nilsson, P.S. Eriksson, L. Ronnback, E. Hansson, GABA induces Ca 2 1 transients in astrocytes, Neuroscience 54 (3) (1993) 605614. [26] K. Obrietan, A. Van den Pol, GABA B receptor mediated regulation of glutamate-activated calcium transients in hypothalamic and cortical neuron development, J. Neurophysiol. 81 (1999) 94102. [27] J. Ong, D.I. Kerr, GABA receptors in peripheral tissues, Life Sci. 46 (1990) 14891501. [28] C. Patte, P. Gandolfo, J. Leprince, J-L. Thoumas, M. Fontaine, H. Vaudry et al., GABA inhibits endozapine release from cultured rat astrocytes, Glia 25 (1999) 404411. [29] M.J. Robbins, A.R. Calver, A.K. Fillipov, W.D. Hirst, R.B. Russell, M.D. Wood et al., GABA(B2) is essential for G-protein coupling of the GABA(B) receptor heterodimer, J. Neurosci. 21 (2001) 8043 8050. [30] D.A. Schwarz, G. Barry, S.D. Eliasof, R.E. Petroski, P.J. Conlon, R.A. Maki, Characterization of gamma-aminobutyric acid receptor GABAB(1e), a GABAB(1) splice variant encoding a truncated receptor, J. Biol. Chem. 275 (2000) 3217432181. [31] J.H. White, A. Wise, M.J. Main, A. Green, N.J. Fraser, G.H. Disney et al., Heterodimerization is required for the formation of a functional GABA(B) receptor, Nature 396 (1998) 679682.

the distribution of a novel GABA B -like protein by immunohistochemistry in the rat CNS. GABA BL is widely distributed with highest intensity of immunoreactivity seen in the hippocampus, amygdala, habenula, hypothalamus, raphe, cerebellum and dorsal horn of the spinal cord. GABA BL is expressed on certain interneurons of the hippocampus, though this study has by no means exhausted the potential subpopulations of interneurons within the CNS. GABA BL appears to be predominantly expressed in neuronal somata, though dendritic expression cannot be ruled out due to the detection methods utilised within this manuscript. Functional assessment of this receptor has, however, failed to produce a pharmacological response to GABA B receptor ligands even when co-transfected with GABA B1 or GABA B2 [7]. The possibility remains that this orphan receptor heterodimerises with an as yet undiscovered protein to form a functional receptor, alternatively the ligand for this novel GABA B -like receptor is indeed not GABA but one yet to be identied. The distribution of GABA BL in the rodent may provide valuable clues as to its potential role in physiology and may direct studies for the identication of potential protein partners, however, until the endogenous ligand for this GABA B -like protein is identied it is impossible to assign a role to its function in the CNS.

References
[1] P. Arcelli, C. Frassoni, M.C. Regondi, S. DeDias, R. Spreaco, GABAergic neurons in mammalian thalamus: A marker of thalamic complexity? Brain Res Bull 42 (1996) 2737. [2] D. Benke, M. Honer, C. Michel, B. Bettler, H. Mohler, Gammaaminobutyric acid type B receptor splice variant proteins GBR1a and GBR1b are both associated with GBR2 in situ and display differential regional and subcellular distribution, J. Biol. Chem. 274 (1999) 2732327330. [3] B. Bettler, K. Kaupmann, N. Bowery, GABAB receptors: drugs meet clones, Curr. Opin. Neurobiol. 8 (1998) 345350. [4] G. Bonanno, M. Raiteri, Multiple GABAB receptors, Trends Pharmacol. Sci. 14 (1993) 259261. [5] A.R. Calver, M.J. Robbins, C. Cosio, S.Q.J. Rice, A.J. Babbs, W.D. Hirst et al., The C-terminal domains of the GABA(b) receptor subunits mediate intracellular trafcking but are not required for receptor signalling, J. Neurosci. 21 (2001) 12031210. [6] A.R. Calver, A.D. Medhurst, M.J. Robbins, K.J. Charles, M.L. Evans, D.C. Harrison et al., The expression of GABA(B1) and GABA(B2) receptor subunits in the CNS differs from that in peripheral tissues, Neuroscience 100 (2000) 155170. [7] A.R. Calver, D. Michalovic, T.T. Testa, M.J. Robbins, C. Jaillardd, J. Hill et al., Molecular cloning and characterisation of a novel GABAB-related G-protein coupled receptor, Mol. Brain. Res 110 (2003) 305317. [8] K.J. Charles, M.L. Evans, M.J. Robbins, A.R. Calver, R.A. Leslie, M.N. Pangalos, Comparative immunohistochemical localisation of GABA(B1a), GABA(B1b) and GABA(B2) subunits in rat brain, spinal cord and dorsal root ganglion, Neuroscience 106 (2001) 447467. [9] Y. Cheng, R. Lotan, Molecular cloning and characterization of a novel retinoic acid-inducible gene that encodes a putative G proteincoupled receptor, J. Biol. Chem. 273 (1998) 3500835015.