To better understand the role of neurotransmitter receptors in neuronal differentiation and maturation a detailed knowledge of their identity, location and function in the plasma membrane of specific neuronal populations during development is required. Combining pre-embedding immunocytochemistry with cell tracking in embryonic brain slice cultures we show that virtually all neurons (~98%) migrating through the lower intermediate zone (LIZ) on their way from the medial ganglionic eminence to the cerebral cortex, express GABABR1. Blockade of GABABRs with a specific antagonist, CGP52432, resulted in a concentration-dependent accumulation of these tangentially migrating neurons in the ventricular/sub-ventricular zones (VZ/SVZ) of the cortex and fewer cells were observed in the cortical plate/marginal zone (CP/MZ) and LIZ. Moreover, they had significantly shorter leading processes compared with similar migrating cells in control slices. Electrophysiological recording in LIZ and CP cells revealed no direct effect of either CGP52432 or the GABABR agonist, baclofen, on resting membrane properties suggesting that the effect of CGP52432 on migration might be mediated through a metabotropic action or the regulation of release of factors controlling migration. These results suggest that GABABRs have an important modulatory role in the migration of cortical interneurons.
The two main neuronal types of the mammalian cerebral cortex, excitatory pyramidal cells and inhibitory non-pyramidal cells, are generated in distinct proliferative zones. Pyramidal neurons arise from the germinal ventricular zone (VZ) and migrate along the processes of radial glia to take up their positions in the developing cortex (Rakic, 1990). Non-pyramidal cells, a diverse group of mostly γ-aminobutyric acid (GABA)-containing interneurons displaying a range of morphologies and molecular identities, originate predominantly in the medial ganglionic eminence (MGE) of the ventral telencephalon (Lavdas et al., 1999; Sussel et al., 1999; Wichterle et al., 1999; Parnavelas, 2000; Marín and Rubenstein, 2001) and follow tangential migratory routes through the intermediate zone (IZ) to reach the cortex (De Carlos et al., 1996; Anderson et al., 1997a,b, 2001; Tamamaki et al., 1997; Letinic et al., 2002). A recent study has demonstrated that the caudal ganglionic eminence is also a source of local-circuit interneurons in addition to inhibitory projection neurons and glia (Nery et al., 2002).
In the mature CNS, physiological actions of GABA are mediated via activation of both ionotropic and metabotropic receptors (Macdonald and Olsen, 1994; Misgeld et al., 1995; Johnston, 1996). Ionotropic GABA receptors (GABAARs and GABACRs) form Cl− channels (Bormann, 1988; Sivilotti and Nistri, 1991; Feigenspan et al., 1993), whereas metabotropic GABA receptors (GABABRs) couple to Ca2+ and K+ channels or adenylate cyclase via Gi/o GTP binding proteins (Kerr and Ong, 1992). Functional GABABRs are heterodimers and so far two receptor subunits, GABABR1 and GABABR2, have been identified (Misgeld et al., 1995; Kaupmann et al., 1997, 1998; Bowery, 1997).
During rat cortical development, GABA is released from cells located close to the target destinations for migrating neurons and it has been shown that embryonic cortical neurons exhibit migratory responses to this neurotransmitter in vitro (Lauder et al., 1986; Behar et al., 1996, 1998, 1999, 2001; Maric et al., 2001). These findings suggest that GABA is a chemoattractant during corticogenesis.
Information concerning the role of GABABRs in the developing cerebral cortex is still sparse; however, several lines of evidence suggest that their activation might play an important role in cortical development. First, functional GABABRs have been identified in the developing cerebral cortex (Janigro and Schwartzkroin, 1988; Cherubini et al., 1991; Gaiarsa et al., 1991; Fukuda et al., 1993). Second, pharmacological studies indicate that GABABR activation stimulates migration of neurons in immature cortical regions (Behar et al., 1998, 2000, 2001). Third, recent experiments using an in vitro chemotaxis assay demonstrated that cortical plate (CP) cells release GABA and taurine, both of which act as a chemoattractant for immature cortical neurons, causing the cells to migrate from the VZ into the CP via activation of GABABRs (Behar et al., 2001). In addition, recent studies have investigated the expression and distribution of GABABRs during early cortical development (Behar et al., 2001; López-Bendito et al., 2002), and demonstrated a particularly high density of the receptor in tangentially orientated neurons in the lower intermediate zone (LIZ) of the cortex (López-Bendito et al., 2002).
To investigate whether the activation or blockade of GABABRs can modulate the tangential migration of cortical interneurons we have used in vitro embryonic organotypic cultures, in combination with immunohistochemical techniques. We have also used whole-cell and perforated patch-clamp recordings to analyze the electrophysiological properties of the migratory cells as well as the functionality of their GABABRs. Our findings indicate that blockade of GABABRs modifies the distribution of tangential migratory cells as well the length of their leading process. The lack of a consistent direct effect of baclofen or CGP52432 on the resting membrane potential and input resistance in LIZ and CP cells raises the possibility that the effect on migration by CGP52432 seen in tangentially migratory interneurons could be indirect via an effect on neighboring cells or mediated through intracellular messenger systems and/or the regulation of release of factors controlling migration. These findings provide strong support that signaling through GABABRs modulates neuronal migration in the developing cortex.
Materials and Methods
Thirty-three pregnant Wistar rats, obtained from the Animal House Facilities of the Department of Human Anatomy and Genetics in Oxford, were used in the present study. The care and handling of the animals prior to and during the experimental procedures followed European Union and UK Home Office regulations, and were approved and supervised by the Animal Care and Use Committee of the institution. We have endeavored to minimize both the suffering and the number of animals used in this study.
Pregnant rats at embryonic day (E) 15 stage of gestation (n = 33) were killed by cervical dislocation. Fetuses were collected by cesarean section of the dam. The fetuses (n = 91) were rapidly removed and placed in artificial cerebrospinal fluid (ACSF), in which NaCl was substituted with equiosmolar sucrose, at 4°C. The composition of the ACSF was (in mM): KCl, 3; NaHCO3, 24; NaH2PO4, 1.25; MgSO4, 2; CaCl2, 2; d-glucose 10; and sucrose, 252. The brains were dissected from the skull and embedded in a 3% solution of type VII low-gelling agarose (Sigma) in Gey’s balanced salt solution (GBSS, Gibco), pH 7.4, at 40°C. Brains were then cut coronally at 300 μm with a vibrating blade microtome (Leica VT1000S, Nussloch, Germany). Slices containing the MGE and primordial cortex were placed onto Millicell 0.45 mm, black gridded, 13 mm membranes (Millipore) in 30 mm Petri dishes containing 250 ml of DMEM/F12 (Life Technologies) with 2.4 mg/ml d-glucose, 5 μl/ml N2 supplement (Gibco), 0.1 mM glutamine, 5 units/ml penicillin, 5 μg/ml streptomycin, and 10% fetal calf serum (FCS, Gibco) for 1 h at 37°C with 5% CO2. The procedures described above were performed under sterile conditions.
Injection of Fluorescent Tracer
To examine whether the tangentially migratory neurons express GABABRs during their migration, the MGE (n = 220) of the two hemispheres derived from E15 brains was injected with 0.7 μm tungsten-M10 particles (Bio-Rad) impregnated with 1 mM 5-(and-6)-(((4-chloromethyl)benzoyl)amino)tetramethylrhodamine (Cell-Tracker Orange CMTMR; Molecular Probes) using a glass micropipette. These labeled cultures provided a reliable in vitro assay to test the effect of agonist or antagonist at GABABRs.
Drugs and Control
To determine the role of GABABRs in tangential migration, we used a specific agonist (RS)-Baclofen ((RS)-4-amino-3-(4-chlorophenyl)butanoic acid; Tocris) or antagonist CGP52432 (3-[[[(3,4-dichlorophenyl)methyl]-amino]propyl] diethoxymethyl)phosphinic acid; Tocris) at GABABRs that was added to the culture media 1 h after placement of the CMTMR. Three experimental groups were used: (i) control group grown in culture medium alone; (ii) agonist group, 100 μM of the specific GABABR agonist, baclofen was added to the culture medium; and (iii) antagonist group, different concentrations (10 nM, 50 nM, 200 nM, 1 μM, 5 mM or 60 mM) of the selective GABABR antagonist CGP52432 were added to the culture medium. Slices from the three experimental groups were cultured at 37°C in a mixture of 5% CO2 and 95% air for 24 h. Slices were then fixed in 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) for 2 h at 4°C, and subsequently rinsed in PBS, coverslipped in PBS/glycerol solution (1:1), and observed with a fluorescent microscope (Leica DMR) or a laser-scanning confocal microscope (Leica TCS SP1).
To establish the relative frequency of CMTMR-containing cells in control conditions and after GABABR agonist or antagonist treatment, we carried out a quantitative analysis in different embryonic cortical compartments. First, in order to anatomically define those compartments, we counterstained freshly fixed slices with the chromatin stain bis-benzimide (10 min in 2.5 μg/ml solution in 0.1 M phosphate buffer; Sigma). The use of the bis-benzimide allowed us to distinguish embryonic cortical compartments that we subsequently analyzed in three groups: MZ/CP, LIZ and SVZ/VZ. Neuronal counts and immunohistochemical experiments were carried out on those three defined cortical compartments (see Fig. 3G–I).
For quantitative assays, samples were taken from cultures prepared from 58 embryos using between 6 and 20 different cultures for each condition. Quantification was carried out under fluorescence microscope from slices chosen in a systematic manner from each animal and experiment. Randomly selected fields, using a 250 mm2 reticule under a 40× objective lens, were then photographed from the three cortical compartments with a Leica DC500 digital camera. A minimum of 10 fields were chosen in each of 6–10 separate experiments. The total number of CMTMR-containing cells identified in each of the three selected cortical compartment under the different treatments was counted. Since the distributions of the different samples for each treatment did not differ from each other (Kolmogorov–Smirnov test, P > 0.63), the data were pooled. These data were converted to percentage of CMTMR-containing cells in each embryonic cortical compartment.
For the estimation of possible morphological changes in the CMTMR positive cells after application of GABABR agonist or antagonist, the length of the leading processes of 90 CMTMR-containing cells identified in each embryonic cortical compartment was measured. The leading processes were measured in control conditions and in slices subjected to GABABR agonist and antagonist treatment using a digitizing table and appropriate software (SigmaScan Pro, SPSS Inc., Chicago, IL). Measurements were carried out under fluorescence microscope from slices chosen in a systematic manner from three animals. Fifteen randomly selected fields from the three layers were then photographed with a Leica DC500 digital camera. The data were expressed as the mean length ± standard deviation (SD) of the leading processes in each embryonic cortical layer.
The Kolmogorov–Smirnov non-parametric test was used to examine whether samples taken for each animal and experiment were from a homogeneous population. All remaining statistical comparisons were performed using Student’s t-test. The differences were considered significant at the level of P < 0.05. Data are presented as mean ± SD, unless otherwise indicated. The dose–response data were analyzed by using a Hill equation to fit a sigmoidal function through the data points, using the built-in iterative Levenberg-Marquardt algorithm in Igor Pro (Wavemetrics, Lake Oswego, OR).
To further characterize the expression of GABABRs in CMTMR-labeled cells, following organotypic culturing and injection of CMTMR in the MGE, 20 slices originating from 10 fetal brains were used for immuno-histochemical analysis.
Antibodies and Controls
An affinity-purified polyclonal antibody against GABABR1a/b was raised in rabbits. The characteristics and specificity of this antibody have been described elsewhere (Kulik et al., 2002).
To test method specificity in the procedures for light microscopy, the primary antibody was omitted or replaced by 5% (v/v) normal serum of the species of the primary antibody. Under these conditions no selective labeling was observed. Some sections were incubated with the primary antibody and without secondary antibodies to test for cross-reactivity of the secondary antibody and the CMTMR. No cross-labeling was detected.
Immunohistochemistry for CMTMR Cells Expressing GABABR1
Selected slice cultures that were previously analyzed and photographed with a fluorescence microscope were removed from the slides, embedded in 4% agarose and subsequently re-sectioned at 60 mm with a microtome (Leica VT1000S). Sections were incubated in 50 mM tris-buffered saline (TBS) containing 0.1% Triton X-100 and 10% Normal Goat Serum (NGS) at room temperature for 1 h, and they were then incubated overnight at 4°C with an affinity-purified polyclonal antibody against GABABR1a/b, at a final protein concentration of 1–2 μg/ml, in TBS containing 1% NGS. After washing in TBS, sections were incubated with Cy2-conjugated goat anti-rabbit antibody diluted 1:200 in TBS for 2 h. Immunolabeled sections were mounted in PBS/glycerol and observed using a laser-scanning confocal microscope (Leica TCS SP1). Images were stored and analyzed using appropriate software supplied by the microscope manufacturer (Leica). Brightness and contrast were adjusted for the whole frame using appropriate software.
Immunohistochemistry for GABABRs Using the Immunoperoxidase Method
Fetuses (n = 10) at E16 were collected by cesarean section after anesthesia of the dam with an intraperitoneal injection of a Rompun/Imalgene mixture (0.1 ml/kg BW). Animals were deeply anesthetized by hypothermia and then perfused with 4% PFA in PBS thought the heart (López-Bendito et al., 2001, 2002). After perfusion, tissue blocks containing the neocortex were dissected and washed thoroughly in 0.1 M phosphate buffer (PB, pH 7.4). Coronal 60 μm sections were cut with a microtome (Leica VT1000S) and collected in 0.1 M PB. For the immuno-peroxidase method, sections were incubated in 10% normal goat serum (NGS) in 50 mM Tris buffer (pH 7.4) containing 0.9% NaCl (TBS), with 0.2% Triton X-100, for 1 h. Sections were then incubated at 4°C for 48 h with affinity-purified polyclonal antibodies against GABABR1a/b at a final protein concentration of 1–2 μg/ml in TBS containing 1% NGS. After several washes in TBS, the sections were incubated at room temperature for 2 h in biotinylated goat anti-rabbit IgG (Vector Laboratories, Burlingame, CA) diluted 1:50 in TBS containing 1% NGS. They were then transferred into avidin–biotin–peroxidase complex (ABC kit, Vector Laboratories) diluted 1:100 and left for 2 h at room temperature. Peroxidase enzyme activity was revealed using 3,3¢-diaminobenzidine tetrahydrochloride (DAB; 0.05% in TB, pH 7.4) as the chromogen and 0.01% H2O2 as the substrate. Finally, the sections were air-dried and coverslipped prior to observation in a Leica DMRS photomicroscope equipped with differential interference contrast optics.
Electrophysiology for GABABRs in Embryonic Organotypic Culture Slices
Intracellular Recordings and Data Analysis
Patch-clamp recordings were obtained using conventional procedures (Sakmann and Stuart, 1995). Briefly, brain slices (300 mm) were prepared from E16 embryos (n = 72) as described above for organotypic cultures, and were maintained for at least 1 h in culture medium (equilibrated with 5% CO2 and 95% O2) at 37°C before recording commenced. They were then transferred one by one to a standard submerged-style recording chamber and superfused at a rate of 2–3 ml/min with ACSF of the following composition (in mM): NaCl, 126; KCl, 3; NaH2PO4, 1.25; MgSO4, 2, CaCl2, 2; NaHCO3, 26; and d-glucose, 10. All recordings were made at room temperature (22–25°C). Cells were identified using an upright fixed-stage microscope (Zeiss Axioskop, Jena, Germany) fitted with differential interference contrast optics and an infrared-sensitive video camera (Newvicon; Hamamatsu, Tokyo, Japan).
Current-clamp recordings using whole-cell and perforated patch configurations were obtained with an Axoclamp-2B amplifier (Axon Instruments, Union City, CA) in bridge mode. Recording pipettes with fine tips (pipette resistance, 8–15 MΩ for whole-cell recording; 6–7 MΩ for perforated patch-clamp recording) were pulled from borosilicate glass tubing. The pipette solution (pH, 7.35; 285–295 mOsm) contained (in mM): K gluconate, 130; HEPES, 40; NaCl, 4; MgATP, 4; and GTP, 0.3; with addition of biocytin, 5 mg/ml. pH was adjusted with KOH. For perforated patch recording, the tip was filled with pipette solution as above, omitting biocytin, and backfilled with solution of the same composition supplemented with 25 mg/ml gramicidin (Sigma), in 0.1% DMSO. Typical final access resistances were 30–70 MΩ (whole-cell recording) and 50–100 MΩ (perforated patch recording). Junction potentials were not corrected. Capacitance compensation was maximal and bridge balance was monitored and adjusted as required. The voltage responses to negative and positive current steps were recorded, filtered at 1 kHz, digitized at 2 kHz, stored on computer on-line (ITC-16, Instrutech, Long Island, NY), and subsequently analyzed off-line using custom-made procedures in Igor Pro (WaveMetrics, Lake Oswego, OR). To test the electrophysiological properties of GABABRs in tangentially migratory and CP cells, the following compounds were added to ACSF during the recording: 300 μM (RS)-Baclofen (Tocris) and 5 mM CGP 52 432 (Tocris). All results are given as mean ± SD. Statistical significance was assessed using a two-tailed unpaired Student’s t-test (P < 0.05). After recording, the slices were fixed in 4% PFA in PBS for 4 h. Subsequently, the slices were incubated in Cy2-conjugated streptavidin diluted 1:500 in 0.3% Triton X-100 in TBS for 3 h. Finally, the sections were mounted in PBS/glycerol and observed using a laser-scanning confocal microscope (Leica TCS SP1).
GABABRs Are Expressed in the Tangentially Migratory Neurons
We have previously demonstrated with immunohistochemical methods that GABABR1 is present in cells oriented tangentially along the LIZ of the neocortex, suggesting that this receptor is expressed in the migrating neurons presumably originating from the MGE (López-Bendito et al., 2002). To test this hypothesis we conducted experiments in organotypic cultures prepared from E15 rat brain. We labeled the tangentially migratory cells by placing a small number of tungsten particles impregnated with the cell tracker CMTMR in the MGE (Alifragis et al., 2002), and then performed immunohistochemical experiments for GABABRs on the same cultured slices.
Most of the CMTMR-labeled cells in the LIZ were bipolar with tangentially oriented processes (Figs 1 and 2). Their morphology was very similar to that found in previous tracing and immunocytochemical studies on tangentially migrating cells (O’Rourke et al., 1995; Tamamaki et al., 1997; Lavdas et al., 1999). Most of these labeled neurons possessed a long leading process, which in some cases was branched.
Immunoreactivity for GABABR1 observed after 24 h in cultured re-sectioned material (Figs 1C,D and 2) was indistinguishable from that observed in freshly fixed E16 slices labeled using a pre-embedding immunoperoxidase method (Fig. 1A,B). At E16, we found strong immunoreactivity for GABABR1 in the neocortical primordium. The cortical plate (CP), marginal zone (MZ) and subplate (SP) were immunoreactive for GABABR1 (Fig. 1A,B). In addition, GABABR1 immunoreactivity was also present in the tangentially migratory cells in the LIZ (Fig. 1A,B). In the re-sectioned material, the receptor was similarly localized at the plasma membrane and in the cytoplasm of cells located in the LIZ, MZ, CP and SP (Figs 1C,D and 2I,K). The CMTMR labeling was observed mainly in the nucleus and cytoplasm (Fig. 2J,K) of cells located in the LIZ with tangential orientation. Virtually all CMTMR-labeled cells, ~98% (n = 260 CMTMR-labeled cells examined in n = 4 cultures of E15 rats), expressed the GABABR1 subunit (Figs 1D and 2F,G,H,K). This result indicates that GABABRs are expressed in migrating cells originating in subpallial areas.
Blockade of GABAB Receptors Changes the Distribution of Migrating Neurons in Organotypic Cultures
Recent pharmacological studies have provided evidence that activation of GABA receptors by endogenous GABA can modify cortical neuronal migration (Behar et al., 2000, 2001). To test whether GABABRs play a role in the migration of the tangentially migratory neurons, we investigated the effect of the application of a specific antagonist at GABABRs, CGP52432, in organotypic cortical slice cultures. We placed tungsten particles impregnated with 1 mM CMTMR in the MGE (Fig. 3A–C), the main source of cortical tangentially migratory GABAergic neurons (Anderson et al., 1997a; Lavdas et al., 1999), in E15 rat brain slices to label that specific neuronal population. We incubated the slices for 24 h under three conditions: (i) standard culture medium, (ii) medium with the addition of 100 μM of the specific GABABR agonist, baclofen, and (iii) medium with the addition of different concentrations of CGP52432 (10 nM, 50 nM, 200 nM, 1 μM, 5 mM or 60 mM). After incubation, the slices were fixed and counterstained with bis-benzimide to enable identification of cortical layers based on their cell density (Fig. 3G–I). In the presence of 200 nM, 1 μM, 5 mM or 60 mM CGP52432 we observed significantly more CMTMR-labeled cells located in the VZ/SVZ relative to baclofen and control conditions (Fig. 3E; Fig. 4). Moreover, after CGP52432 treatment, less CMTMR positive cells were found in the MZ/CP or LIZ in comparison with control and baclofen treatment (Fig. 4). No differences in the percentage of CMTMR cells per cortical compartment were observed between control and baclofen treatment (Fig. 4).
To investigate whether, in the presence of CGP52432, the expression of the GABABR1 subunit in the CMTMR positive cells changed, we performed immunofluorescence experiments for GABABR1 in culture sections that were incubated for 24 h with CGP52432 (Fig. 2G,H). The percentage of GABABR1 positive CMTMR-labeled cells after CGP52432 treatment did not differ from control (control, 98%; n = 257 CMTMR cells examined; CGP52432 60 μM, 98%; n = 155 CMTMR cells examined). GABABR1 expression was mainly localized in the cytoplasm and plasma membrane (Fig. 2I) of LIZ, MZ, CP and SP cells and no differences in immunoreactivity or localization were found under this treatment condition in comparison with control slices.
The Effect of CGP52432 Does Not Depend on Rostro-caudal Telencephalic Level
To test whether the increase of CMTMR-positive cells in the VZ/SVZ caused by CGP52432 is dependent upon any particular rostro-caudal telencephalic level we prepared serial brain slices in which pharmacological studies using the same drug conditions as described above were carried out. CMTMR-labeled cells were found accumulated in the VZ/SVZ after the CGP52432 treatment at caudal as well as at rostral brain levels (Fig. 4A,B). Fewer CMTMR-positive cells were found in the MZ/CP and LIZ in slices treated with the GABABR antagonist at both rostro-caudal levels (Fig. 4). No differences in the percentage of CMTMR-labeled cells located in different cortical compartments were observed between control and baclofen situations at rostral and caudal levels (Fig. 4).
The CGP52432 Concentration–Effect Relation Is Consistent with an Action Via GABABRs
To investigate whether the effect of CGP52432 on tangential migration is mediated via specific antagonism at GABABR we quantified the number of CMTMR positive cells per cortical compartment using different concentrations of CGP52432. A Hill equation was used to fit a sigmoidal function through the data points. The estimated concentration of CGP52432 that gave half of the maximal effect was 31 nM for LIZ and 125 nM for SVZ/VZ (Fig. 5). This compares to the reported IC50 of 85 nM at rat cortical GABAB autoreceptors (Lanza et al., 1993). Thus, the result is consistent with an effect via blockade of GABABR.
We tested whether the block by CGP52432 could be overcome by an agonist by adding baclofen (300 mM) in combination with CGP52432 (at its IC50 concentration, 85 nM; n = 51). CMTMR injections were performed at the E15 MGE as was described before. After 24 h in vitro, no significant differences were observed in the percentage of CMTMR-labeled cells located in different cortical compartments between cultures treated with baclofen plus CGP52432 and control cultures (control: MZ, 20.4 ± 1.3%; LIZ, 56.2 ± 1.8%; SVZ/VZ, 23.3 ± 1.5%; CGP52432 85 nM + baclofen: MZ, 19.5 ± 1.8%; LIZ, 51.3 ± 1.6%; SVZ/VZ, 29.2 ± 1.3%; P = 0.1). Compared to the concentration–effect curve for CGP52432, these data appear to suggest a partial reversal of the CGP52432 effect; however, this result did not reach significance and was not investigated further.
GABABR Blockade Alters Neurite Morphology in Tangential Migratory Neurons
Recent studies have demonstrated that activation of AMPA-type glutamate receptors causes retraction of neurites in migrating neurons in the LIZ (Poluch et al., 2001). Since the same neuronal populations express GABABRs, we were interested in determining whether the effect of GABABR antagonists on migration might be mediated via similar morphogenetic processes. Thus, we measured the length of the leading process of CMTMR positive cells in the three experimental conditions. After treatment with 5 μM CGP52432 of E15 rat brain slices for 1 day in vitro (DIV), we found a significant decrease in the length of the leading process of CMTMR positive cells (Fig. 6) in the LIZ (control, 78.3 ± 13.7 μm; CGP52432, 61.9 ± 11.8 mm; P < 0.001) and in the VZ (control, 78.5 ± 13.3 μm; CGP52432, 61.7 ± 10.9 mm; P < 0.001), but not in the MZ (control, 77.9 ± 15.3 μm; CGP52432, 75.6 ± 15.8 mm; P = 0.65).
Lack of Direct Effect on Passive Membrane Properties by GABABR Agonists and Antagonists
In an attempt to investigate whether the effect of the GABABR antagonist on migration in tangential neurons could be mediated via a direct effect on electrical membrane properties, we monitored the membrane potential as well as voltage response to negative and positive current steps during wash-in of either baclofen (300 μM) or CGP52432 (5 mM) in E16 rat cortical slices. No consistent effects could be seen in LIZ cells (n = 7 and 12; Fig. 7Ai, Bi, Ci, Di, E). Similarly, we detected no consistent effect in CP cells at this developmental stage (n = 2 and 12; Fig. 7Aii, Bii, Cii, Dii, E). As a positive control, we investigated the effect of baclofen on layer V pyramidal neurons in somatosensory cortex in slices from P14 rats. In contrast to the embryonic slices, there was a robust hyperpolarization associated with a decrease in input resistance in these neurons in agreement with previous reports from neocortical pyramidal cells [n = 4; Fig. 7Aiii, Biii, Ciii (Connors et al., 1988)].
One possible explanation for the lack of response to baclofen in tangential migratory cells could be rapid ‘wash-out’ of cytoplasmic components into the recording pipette during whole-cell recording because of the small volume of these embryonic cells. To exclude this possibility we performed perforated-patch recordings (Marty and Neher, 1995) using the antibiotic gramicidin. After wash-in of 300 mM baclofen we did not observe any significant change of the membrane potential or input resistance in LIZ cells (–49.6 ± 8.4 mV and 3.26 ± 1.14 GΩ, respectively, 10 min after wash-in of baclofen compared with −54.7 ± 8.4 mV and 3.48 ± 0.92 GΩ in control; n = 4). These results strengthen the conclusion using whole-cell patch recordings that baclofen has no direct effect on resting membrane properties in tangential migratory cells at E16.
The lack of effect of ligands at GABABRs on membrane properties of tangentially migratory neurons is consistent with previous studies, reporting no effect of baclofen on [Ca2+]i in IZ migratory interneurons (Behar et al., 1996; Soria and Valdeolmillos, 2002). We conclude that the effect of CGP52432 on neuronal migration is mediated via mechanisms distinct from direct changes in membrane properties of tangentially migratory neurons.
We have reported here that exposure to CGP52432, a specific GABABR antagonist, induced a change in the distribution of tangentially migratory neurons originating in the MGE. We also showed that this drug induced a reduction of the length of their leading process. This leads us to propose that early in development, before the differentiation of synapses, GABABRs may have a role in modulating the migration of tangentially migratory interneurons.
Tangential Interneurons Express GABABRs during Their Migration
The expression of GABABRs in different populations of cortical cells at very early stages of embryonic development has been described earlier (Behar et al., 2001; López-Bendito et al., 2002). We reported the expression of the GABABR subunit, GABABR1, in cells located at the LIZ (López-Bendito et al., 2002). In the present study, we have shown that, in fact, GABABR1 is expressed in the plasma membrane of cells migrating from the MGE towards the cortex. This was demonstrated by labeling the cells originating from the MGE with a cell-tracker, CMTMR, followed by immunostaining for the GABABR1 subunit.
Several studies have shown that migrating interneurons in the LIZ express a variety of ionotropic receptors, including AMPA-type glutamate receptors with high calcium permeability (Métin et al., 2000; Poluch et al., 2001) (López-Bendito, Luján and Fairén, unpublished observations), NMDA-type glutamate receptors and GABAA receptors (Soria and Valdeolmillos, 2002), the activation of which lead to changes in [Ca2+]i. In the present study, we demonstrate that the same cell population also expresses the metabotropic receptor GABABR1. It has been shown that GABA is capable of both stimulating and arresting the migration of embryonic cortical neurons via different signaling mechanisms and GABA-Rs (Behar et al., 1998). Therefore, it might be possible that activation of GABABRs is necessary for a proper navigation through different cortical compartments and for the acquisition of the final position in the cortical circuitry.
Blockade of GABABRs Changes the Distribution of Tangential Migratory Cortical Interneurons
We have shown in the present study that in vitro application of CGP52432 in cortical embryonic slices results in a concentration dependent change in the cortical distribution of tangentially migratory neurons originating in the MGE. More CMTMR-positive cells were found in the VZ/SVZ areas of the cortex whereas fewer were found at the MZ, CP and LIZ following treatment with CGP52432. The effect was similar at rostral and caudal telencephalic levels. The concentration–effect curve was consistent with an effect via GABABR. The estimated concentration giving half of the maximal effect was close to the IC50 value of the drug at GABA autoreceptors in postnatal rat cerebral cortex (Lanza et al., 1993), suggesting that the effect of CGP52432 might be mediated through the regulation of release of GABA or other factors controlling migration. Moreover, the GABABR agonist baclofen reduced the estimated inhibitory effect of CGP52432 at IC50, although this reduction did not reach significance. Altogether, these results strongly suggest that the change in distribution of tangential migrating neurons in the cortex after CGP52432 treatment is indeed due to block of GABABRs.
Previous studies using BrdU pulses in saclofen-treated culture slices, another GABABR antagonist, have shown an accumulation of BrdU positive cells at the level of VZ/LIZ and fewer cells were found in the CP (Behar et al., 2001). The conclusion of these authors was that saclofen is able to inhibit the radial migration of postmitotic VZ cells into the CP via GABABRs but the authors did not investigate whether tangentially migratory cells could be also affected. A ventricle-directed migration of LIZ cells has also been demonstrated (Nadarajah et al., 2002). It is possible that certain migratory LIZ cells need the activation of GABABRs to be able to ‘escape’ from the SVZ/VZ to enter into the CP. Therefore, the blockade of GABABRs could produce an accumulation of those neurons at the lower cortical compartments. Because we only tested the effect of CGP52432 after 1 DIV it might be possible that the accumulation of CMTMR-labeled cells at the VZ is not a permanent effect and the cells might be only delayed from the normal timing of their entry into the CP. Long time course cultures and time-lapse studies may help resolve this question.
The failure to detect consistent changes in resting membrane properties in LIZ cells following either baclofen or CGP52432 treatment combined with the lack of detectable GABABR2 subunit of the GABABRs (López-Bendito et al., 2002), considered necessary for fully functional GABABRs (Jones et al., 1998; Kaupmann et al., 1998; White et al., 1998; Kuner et al., 1999), raises the question of whether the effect seen might be indirect, e.g. through blockade of GABABRs regulating release of factors controlling migration. We have previously shown that both subunits of GABABRs are expressed in CP cells at early embryonic stages (López-Bendito et al., 2002). Moreover, the presence of close appositions between elongating cortical axons and GABAergic IZ cells has been reported (Métin et al., 2000). At their presynaptic location in mature neurons, GABABRs have been shown to decrease neurotransmitter release (Bowery, 1993; Misgeld et al., 1995), and thus it is conceivable that blockade of GABABRs could enhance glutamate release, which via activation of AMPA receptors on tangential migratory neurons could arrest migration (Métin et al., 2000; Poluch et al., 2001).
Another possibility, however, is that even without changes in the membrane potential, GABABRs could be mediating an effect through changes in intracellular messengers not readily detectable with electrophysiological methods, maybe involving a novel and not yet identified subunit.
Involvement of GABABRs in Neurite Outgrowth
We demonstrated here that pharmacological GABABR manipulation affects tangentially migratory neurons, not only at the level of the route of migration but also the length of their leading process. After CGP52432 treatment, CMTMR cells located in the LIZ and VZ had a shorter leading process compared with controls. Similar changes in the neurite length of those cells were observed after activation of AMPA receptors (Poluch et al., 2001). These IZ cells express GluR1, but few if any GluR2 subunits of AMPA receptors (König et al., 1998; Métin et al., 2000). We do not yet know whether the reduction in the length of the leading process after GABABR blockade is causally linked to the effect on the migration. Thus, the accumulation of migrating cells in the VZ after CGP52432 treatment might be due to the shortening of the leading process, or, to the contrary, the effect of the GABABR antagonist on the leading process might be a consequence of the accumulation of cells at the VZ. However, the possibility that these are two independent effects cannot be ruled out at present. Further experiments are needed to solve this problem. Whatever the significance of the reduction of the length of the leading process may be, it has been proposed that if an endogenous agonist is present in vivo at an appropriate concentration, it is likely to increase cytosolic calcium concentration ([Ca2+]i) that might lead, directly or indirectly, to neurite retraction. Therefore, the possibility exists that the changes observed in neurite length after blockade of GABABRs might be mediated indirectly through enhancement of glutamate release and subsequent activation of AMPA receptors present in the migrating neurons. Alternatively, however, a direct effect of GABABRs through changes in cytosolic Ca2+ can not be ruled out, although the failure of baclofen to modulate [Ca2+]i in LIZ cells makes this possibility appear less likely (Soria and Valdeolmillos, 2002).
Possible Role of GABABRs In Vivo
Our results revealed that CGP52432 in rat embryonic cortex, most likely acting via GABABRs, alters the migration of tangential interneurons. This finding suggests that cortical cells in situ and maybe even the tangential migratory neurons themselves release factors that stimulate neurons to migrate in a particular manner. Our results support the idea that neurotransmitter receptors play a role in regulating migration early in development before synapses are established. A range of different molecules have been described recently to be involved in this early cortical process, such as Slit in repelling lateral ganglionic eminence neurons (Zhu et al., 1999) and semaphorins 3A-3F and neurotrophins in the guidance of LIZ tangential cells from the MGE towards the cortex (Polleux et al., 2002; Tamamaki et al., 2003). Neurotransmitter receptors may act in concert with these guidance mechanisms.
Recently, mice lacking the GABABR1 subunit of the GABABRs have been produced, showing behavioral, biochemical and functional alterations in the brain (Prosser et al., 2001; Schuler et al., 2001). Future in vivo and in vitro studies in these mice might help understand the role of GABABRs in early stages of cortical development.
We are grateful to Prof. John Parnavelas, Ms Rosalind Carney and Mr Daniel Blakey for their thoughtful comments on a previous version of this manuscript. This work was supported by Grants from The European Community (QLRT-1999-30158), The Wellcome Trust (063 974/B/01/Z) and the Human Frontier Science Program (RGP0107/2001).
Address correspondence to Dr Guillermina López-Bendito, Department of Human Anatomy & Genetics, University of Oxford, South Park Road, Oxford OX1 3QX, UK. Email: firstname.lastname@example.org.