GABA_A Receptor Subunit Profiles of Tangentially Migrating Neurons Derived From the Medial Ganglionic Eminence

Abstract

During rodent corticogenesis, a sizeable subpopulation of γ-aminobutyric acid (GABA)ergic interneurons arises extracortically from the medial ganglionic eminence (MGE). These neurons progressively acquire responsiveness to GABA in the course of corticopetal tangential migration, a process regulated by ambient GABA and mediated by GABA_A receptors. Here, we combined patch clamp electrophysiology and single-cell reverse transcription–polymerase chain reaction to examine GABA_A receptor expression in green fluorescent MGE-derived (eGFP+) cells in telencephalic slices from gestational day 14.5 BAC-Lhx6 embryos. GABA concentration–response curves revealed lower apparent affinity and efficacy in eGFP+ cells in and around the MGE than their counterparts in the cortex. Pharmacological tests revealed subunit-selective response profiles in the MGE and cortex consistent with differential expression of GABA_A receptor isoforms. Profiling of GABA_A receptor subunit transcripts (α1–5, β1–3, and γ1–3, δ) uncovered increased expression of the α1-, α2-, α5-, γ2-, and γ3-subunit messenger RNAs in the cortex. We propose that the dynamic expression of certain GABA_A receptor subunits contributes to assembling receptor isoforms that confer functional attributes important in regulating the migration and maturation of primordial GABAergic cortical interneurons.

Introduction

Progenitor cells located in the embryonic ganglionic eminences and preoptic area migrate tangentially into the neocortical anlage and give rise to distinct subpopulations of γ-aminobutyric acid (GABA)ergic interneurons in the adult cerebral cortex (Tamamaki et al. 1997; Lavdas et al. 1999; Anderson et al. 2001; Gelman et al. 2009; Miyoshi et al. 2010). Notably, the medial ganglionic eminence (MGE) is the principal source of GABAergic cortical interneurons that contain somatostatin, parvalbumin, calbindin, and, to a lesser extent, neuropeptide-Y (NPY) (Xu et al. 2004; Fogarty et al. 2007). A variety of extrinsic navigational cues orchestrate neuronal migration from the MGE to the dorsal telencephalon (reviewed by Marin and Rubenstein 2003; Wang and Kriegstein 2009). Among them, ambient GABA, present along the tangential migratory path, tonically activates GABA_A receptors in cortex-bound MGE-derived cells, and this has been shown to modulate their migration into the cortical anlage during corticogenesis (Cuzon et al. 2006).

Indeed, GABA has been implicated in cortical development to regulate the proliferation, migration, and differentiation of developing neurons well before morphological signs of synaptogenesis (LoTurco et al. 1995; Behar et al. 1996, 1998, 2000; Ikeda et al. 1997; Manent et al. 2005; Cuzon et al. 2006). MGE-derived cells express functional GABA_A receptors and acquire responsiveness to GABA as they migrate from the MGE into the cortex (Cuzon et al. 2006). During corticogenesis, immature cells express GABA_A receptors that bind GABA with higher affinity than more mature cells, and their activation elicits depolarizing and slowly desensitizing responses (Owens et al. 1999). Numerous studies have demonstrated brain region–specific yet overlapping and developmentally regulated expression of the GABA_A receptor subunits (α1–6, β1–3, γ1–3, δ, ϵ, θ, π, and ρ1–3; e.g., Gambarana et al. 1991; Araki et al. 1992; Laurie et al. 1992; Cheng et al. 2006; Peden et al. 2008; Yu et al. 2009). In addition, the functional and kinetic properties of GABA_A receptor isoforms are subunit dependent (Pritchett et al. 1989; Angelotti and Macdonald 1993; Tia et al. 1996; Hutcheon et al. 2000; Boileau et al. 2003; Lagrange et al. 2007; reviewed by Macdonald and Olsen 1994; Sieghart and Sperk 2002). These considerations, taken together, point to immature neurons expressing functionally different GABA_A receptor isoforms as they mature.

In this study, we tested the hypothesis that primordial cortical interneurons express multiple GABA_A receptor subunits that can form functionally distinct receptor isoforms in the course of tangential migration. To this end, we combined single-cell reverse transcription–polymerase chain reaction (RT-PCR) with whole-cell patch clamp recording to compare the expression profile of GABA_A receptor subunit transcripts and the pharmacological properties of GABA-activated responses between cortex-bound MGE-derived cells migrating in the subpallium and their cohorts that have arrived in the cortical anlage. Our results revealed dynamic changes consistent with MGE-derived cells favoring the expression of more mature patterns of GABA_A receptor subunits and functional receptor isoforms as they migrate from their site of origin to their destinations in the cortex.

Materials and Methods

All procedures involving animals were in full compliance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and approved by the Dartmouth Medical School Institutional Animal Care and Use Committee. Transgenic mice were generated by National Institute of Neurological Disorders and Stroke Gene Expression Nervous System Atlas Bacterial Artificial Chromosome (BAC) Transgenics Project using a BAC vector in which the entire transcription unit and associated regulatory domain for the Lhx6 gene controlled the expression of eGFP (Gong et al. 2003). Since the transcription factor Lhx6 identifies postmitotic neurons generated from the MGE and is required for cortical migration and specification of these cells (Lavdas et al. 1999; Alifragis et al. 2004; Liodis et al. 2007), they fluoresce green, facilitating their identification in acute slices for electrophysiological analysis and harvesting for single-cell expression profiling. The eGFP+-expressing MGE-derived cells are heretofore referred to as eGFP+ cells.

Acute Embryonic Slice Preparation

On embryonic day 14.5 (E14.5), dams were asphyxiated with CO₂, and fetuses were removed by caesarian section. BAC-Lhx6 embryos were genotyped by the presence of eGFP fluorescence in the mouth region, visualized using ultraviolet (UV) goggles, since Lhx6 has been implicated in tooth development and palate formation (Grigoriou et al. 1998; Zhang et al. 2002; Denaxa et al. 2009). The brains of BAC-Lhx6 GFP embryos were isolated and immersed in ice-cold oxygenated artificial cerebral spinal fluid (aCSF) containing (in millimolars) NaCl 124, KCl 5.0, MgCl₂ 2.0, CaCl₂ 2.0, glycine 0.01, NaH₂PO₄ 1.25, NaHCO₃ 26, and glucose 10. The brains were then embedded in 3.5% low-melting point agarose (Invitrogen), and coronal slices (250 μm) from the anterior half of the cerebral hemisphere were obtained using a vibroslicer (WPI). For consistency, only slices in which the MGE and lateral ganglionic eminence are demarcated by the ganglionic sulcus and clearly distinguishable were used.

Electrophysiology

The slices were stored at room temperature in a reservoir of oxygenated aCSF prior to electrophysiological recording. Embryonic slices were transferred to a recording chamber, stabilized by an overlaying platinum ring strung with plastic string mesh, and maintained at 32–34 °C on a heated stage fit onto an upright microscope (BX51WI; Olympus). Slices were perfused at a rate of 0.5 mL/min with oxygenated aCSF. eGFP+ cells were identified under fluorescence illumination and Hoffman Modulation Optics (Modulation Optics) using a 40× water immersion objective (3-mm working distance; Olympus). Real-time images were captured using an analog video camera attached to a video frame grabber (Integral Technologies) and displayed on a computer monitor, which also aided the navigation and placement of the drug and recording pipettes. Patch clamp recording pipettes were pulled from borosilicate glass capillaries (1.5-mm outer diameter and 0.86-mm inner diameter; Sutter Instrument Co.) and filled with recording solution composed of (in millimolars) KCl 140, CaCl₂ 1.8, MgCl₂ 1.0, N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid 5.0, and Mg²⁺ ATP 3.0. When filled with recording solution, the patch pipettes had resistances of 8–10 MΩ. Recordings were made using an AxoPatch 200A amplifier (Molecular Devices). Whole-cell membrane currents were filtered at 5 kHz, digitized using Clampex v9.0 and analyzed with Clampfit v9.0 (Molecular Devices). Statistical analysis was performed using Sigma Stat 3.0 (SPSS Inc.). Mean peak current amplitude of drug-evoked currents was analyzed using Student's t-Test. Data were reported as mean ± standard error of the mean (mean ± SEM).

Drug Application

GABA (0.1–500 μM; Sigma), diazepam (3 μM; Sigma), L655,708 (10 μM; Tocris), loreclezole (10 μM; Sigma), methyl-6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate (DMCM, 100 nM; Tocris), Ro15-4513 (1 μM; Tocris), 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP, 10 μM; Tocris), zinc chloride (100 μM; Sigma), and zolpidem (100 nM; Tocris) were dissolved in aCSF, stored as frozen stock, and diluted to working concentrations with aCSF immediately prior to each recording session. The drug solutions were loaded into separate barrels of an 8-barrel drug pipette assembly and applied using regulated pulses of pressure (≤3 psi, Picospritzer; General Valve Corporation) within 10 μm of the soma of the cell under study. The timing and the duration of the pressure pulses were controlled by a digital timing unit (Pulse train 1831; WPI). One of the barrels of the multibarrel assembly was routinely filled with aCSF, which was applied between drug applications to clear drugs from the vicinity of the cell. This also served to control for any mechanical artifacts that occur occasionally due to bulk flow.

Profiling Expression of GABA_A Receptor Subunit Messenger RNAs

Expression profiles of candidate GABA_A receptor subunit messenger RNAs were obtained from embryonic MGE and neocortical tissue as well as individual eGFP+ cells employing a RT-PCR–based protocol modified from that published previously (Yeh et al. 2002; Cuzon et al. 2006). For profiling embryonic tissue, E13.5–E14.5 embryonic brains were isolated, and a parasagittal incision was made bilaterally along the dorsal rostral-to-caudal extent of the cortex, thereby splaying the cortical mantle and revealing the V-shaped ridges comprising the lateral, medial, and caudal ganglionic eminences on the floor of the lateral ventricles. The MGE and cortical tissue were microdissected and processed separately. Tissue was homogenized in TRI reagent (Molecular Research Center Inc.); RNA was extracted using bromochloropropane; precipitated by adding isopropanol, ammonium acetate (3 N), and glycogen (5 mg/mL); washed in 75% ethanol; and then solubilized in ribonuclease-free water. First-strand cDNA was synthesized by the addition of RT reaction mix consisting of reverse transcriptase (RT-SSIII; Invitrogen), 5× first-strand buffer, 2.5 mM dNTPs, 300 ng Oligo dT, RNasin inhibitor, and 100 mM dithiothreitol in a final volume of 30 μL and incubated for 1 h at 42 °C.

For profiling single cells, individual cells were aspirated into the glass-recording pipette. The tip of the pipette was then broken inside an RT reaction tube, and the contents were expelled by positive pressure into a PCR reaction tube. First-strand cDNA was synthesized by the addition of RT reaction mix in a final volume of 20 μL and incubated overnight at 37 °C. The RT-SSIII was inactivated at 90 °C for 15 min. The first-round multiplex PCR reaction was performed for 20 cycles (94 °C for 20 s, 60 °C for 60 s, and 72 °C for 60 s) in a cocktail of primer sets for GFP, β-actin, and GABA_A receptor α1–5, β1–3, γ1–3, and δ-subunits (10 pM each). Three microliters of the first-round PCR product was then used as template for the second-round of PCR amplification in the presence of individual primer sets.

PCR amplification of reverse transcribed cDNA template obtained from tissue samples or first-round PCR product from single cells was performed using a programmable mastercycler (Mastercycler ES realplex; Eppendorf) in a solution containing 2× Sybr Greener mix (Eppendorf), 50 pmol of either GABA_A receptor subunit–specific primers (α1–5, β1–3, γ1–3, and δ) or β-actin, and 1 μL of cDNA. A standard curve was generated for each PCR run using adult mouse whole brain cDNA at concentrations ranging between 100 μg/mL and 1 pg/mL and probed for β-actin. Samples were run in duplicates. A set of “no-RT” controls with water added in lieu of the sample were routinely included and run in parallel. A 15-μL aliquot of the reaction product was then electrophoresed in 1% agarose parallel to a molecular weight ladder and visualized under UV illumination after staining with ethidium bromide. The GABA_A receptor subunit–specific primer sequences and expected molecular weight of the PCR amplicons were the same as those previously reported (Liu and Burt 1998; Cuzon et al. 2006). Semiquantitative expression profiling data were expressed as mean ± SEM, and significant differences were determined by Student's t-test.

Results

Apparent Potency and Affinity of GABA_A Receptors in eGFP+ Cells Change With Advancing Migration

In acute E14.5 BAC-Lhx6 slices, an incremental series of GABA concentrations (0.1–500 μM) were focally applied to eGFP+ cells in the region of the MGE or the intermediate zone of the cortex (Fig. 1A). The current traces in the inset of Figure 1B illustrate representative whole-cell current responses to 100 μM GABA obtained from a cell in the region of the MGE (upper trace) and from one recorded in the cortex (lower trace). The mean peak amplitudes of GABA-activated whole-cell current responses were normalized to those monitored in eGFP+ cells in the MGE region and plotted semilogarithmically as a function of the GABA concentrations tested (0.1–500 μM; Fig. 1B). In the same slices, GABA at concentrations ≥10 μM consistently evoked responses of larger amplitude in eGFP+ cells recorded in the cortex compared with those elicited in their counterparts in the MGE region. Consequently, the concentration–response curve constructed from eGFP+ cells in the cortex shifted leftward (EC₅₀ of 188.0 μM in and around the MGE vs. 30.7 μM in cortex), reflecting an increase in both apparent potency and affinity to GABA. This implies increased expression of GABA_A receptors, either of the same or of the different isoforms, in the course of tangential migration.

Figure 1.

MGE-derived primordial GABAergic cortical interneurons acquire sensitivity to GABA as they migrate into the cortex. (A) An eGFP+ cell in the intermediate zone of an E14.5 BAC-Lhx6 slice visualized under epifluoresence and Hoffman Modulation. The recording pipette located on the right is used to monitor whole-cell current responses to drugs applied by an 8-barrel drug pipette assembly located on the left. Scale bar = 10 μm. (B) GABA concentration–response curves for eGFP+ cells recorded in the MGE area and intermediate zone of the cortex of acute slices obtained from E14.5 BAC-Lhx6 embryos. The amplitude of responses to each concentration of GABA was normalized to the maximal response amplitude recorded in the MGE. Inset: current responses to 100 μM GABA applied to eGFP+ cells located in the MGE (top trace) and the cortex (bottom trace).

GABA_A Receptor Subunit Transcripts in the Developing MGE and Cortex

Since subunit composition can account for functional and kinetic differences in GABA_A receptor properties (Verdoorn et al. 1990; Hutcheon et al. 2000; Devor et al. 2001), we asked whether the difference in concentration–response profiles to GABA reflected a regionally dependent variation in the expression of GABA_A receptor subunits. We first analyzed the expression pattern of 12 GABA_A receptor subunit transcripts (α1–5, β1–3, γ1–3, and δ) in the MGE and neocortical tissue microdissected from E14.5 brains. The α6- and ρ(1–3) subunits were not profiled since their expression is largely limited to the cerebellum and visual system, respectively (Varecka et al. 1994; Yeh et al. 1996; Albrecht et al. 1997; Alakuijala et al. 2005). Semiquantitative comparison, with the abundance of each GABA_A receptor subunit transcript normalized to that of β-actin in the same sample, revealed a conspicuous increase in the expression of α1-, α2-, α5-, γ2-, and γ3-subunit transcripts in tissue derived from the cortex compared with the MGE tissue (Fig. 2).

Figure 2.

Upregulated expression of GABA_A receptor subunit transcripts in the developing cortex. Semiquantitative RT-PCR expression of GABA_A receptor subunit transcripts (α1–5, β1–3, γ1–3, and δ) of tissue microdissected from E14.5 BAC-Lhx6 MGE (n = 14) and cortex (n = 14). Data are normalized to β-actin and expressed as mean ± SEM. Asterisks denote a significant difference in relative abundance between the MGE and cortex (Student's t-test; p < 0.05).

GABA_A Subunit Transcripts in Individual Migrating MGE-Derived Cells Before and After Entry Into the Cortex

The above results demonstrating differential expression of certain GABA_A receptor subunit transcripts were obtained from microdissected MGE and cortical tissue, both of which contained heterogeneous populations of neuronal and nonneuronal cell types. To optimize selection of migrating MGE-derived cells in the acute telencephalic slices that are destined to become GABAergic cortical interneurons, we targeted eGFP+ cells situated in the area proximal to the corticostriate juncture (CSJ) with leading edge directed dorsally toward the cortex and excluded those located away from the CSJ and appearing to migrate ventrally in the direction of the striatum. We characterized the pharmacological properties of GABA-mediated responses using agents that have been reported to modulate GABA_A receptor function in a subunit-dependent manner (Table 1) and profiled the expression of GABA_A receptor subunits in these cells. The results of these pharmacological and expression profiling parameters were compared between individual eGFP+ corticopetal GABAergic interneurons and those that have already migrated into the developing cortex.

Table 1.

Pharmacological agents employed to assess expression of GABA_A receptor isoforms

Compound	Action	Reference
Zolpidem	Positive allosteric modulator α1- and γ-subunit–containing GABAA receptors.	Sieghart (1995); Mohler et al. (1996)
DMCM	Inverse agonist at α1-subunit–containing GABAA receptors.	Sieghart (1995); Mohler et al. (1996)
Ro15-4513	Weak partial inverse agonist at α4-subunit–containing GABAA receptors.	Wisden et al. (1991); Wafford et al. (1996); Knoflach et al. (1996)
L655-708	Inverse agonist at α5-subunit–containing GABAA receptors.	Pritchett and Seeburg (1990)
Loreclezole	Positive allosteric modulator of β2- or β3-subunit–containing GABAA receptors.	Stevenson et al. (1995)
Diazepam	Positive allosteric modulator of γ-subunit–containing GABAA receptors.	Pritchett et al. (1989); Yang et al. (1995); Saxena and Macdonald (1996)
Zinc	Inhibits GABAA receptor function at receptors without the γ-subunit.	Draguhn et al. (1990)
THIP	GABAA receptor agonist selective for δ-subunit–containing receptor isoforms.	Adkins et al. (2001); Brown et al. (2002)

α-Subunits

Zolpidem potentiates GABA-activated current responses in GABA_A receptor isoforms containing the α1- and γx-subunits (Sieghart 1995; Mohler et al. 1996). DMCM, on the other hand, is an inverse agonist with greatest affinity for α1-subunit–containing GABA_A receptor isoforms (Sieghart 1995; Mohler et al. 1996). The current traces in Figure 3A illustrate that focal application of GABA (100 μM; 5 s) elicited modest slowly desensitizing responses in eGFP+ cells proximal to the CSJ as reported previously (Cuzon et al. 2006). The majority of these eGFP+ cells displayed GABA responses that were insensitive to potentiation by zolpidem (100 nM; 16 of 18 cases; Fig. 3A1) or attenuation by DMCM (100 nM; 16 of 18 cases; Fig. 3A2). The mean peak amplitude of the GABA response during exposure to zolpidem was statistically similar to control (Fig. 3C; 106 ± 6.2%; P = 0.26) as was the DMCM-induced attenuation (Fig. 3C; 93.8 ± 7.4%; P = 0.34). By contrast, zolpidem potentiated (18 of 20 cases; Fig. 3B1) and DMCM attenuated (16 of 18 cases; Fig. 3B2) GABA responses monitored in the majority of eGFP+ cells recorded in the cortex. Zolpidem potentiated the mean amplitude of the GABA-activated current responses by 143.4 ± 1.9% (Fig. 3C; P = 0.03), while DMCM suppressed the same GABA responses by 68.2 ± 8.6% (Fig. 3C; P < 0.001).

Figure 3.

Expression of GABA_A receptor α-subunits in individual BAC-Lhx6 eGFP+ cells in the area before the CSJ and cortex. (A and B) Whole-cell responses of eGFP+ cells in the area proximal to the CSJ (A) or the cortex (B) to focal coapplication of GABA either alone (100 μM; black traces) or in conjunction with zolpidem (100 nM; A1 and B1; red traces), DMCM (100 nM; A2 and B2; red traces), Ro15-4513 (1 μM; A3 and B3; red traces), and L655,708 (10 μM; A4 and B4; red traces). (C) Percentage change in the amplitude of GABA-induced current responses in the presence of GABA_A receptor modulators in eGFP+ cells examined in the area proximal to the CSJ (black bars) and cortex (gray bars). Data are expressed as mean ± SEM. The dashed line denotes no change from the current elicited by the application of GABA alone. (D and E) Example of individual eGFP+ cells harvested from the area proximal to the CSJ (D) and the cortex (E) and profiled for their expression of GABA_A receptor α1–5 subunits. The darker bands in the molecular weight (MW) ladder lanes denote the 600-bp position. Each arrowhead points to the predicted size of the amplicon detected for the corresponding α-subunit transcript, which are also given at the top of each lane. (F) Semiquantitative determination of the abundance of GABA_A receptor α-subunit transcripts (α1–5) relative to β-actin in the same eGFP+ cells harvested from the area proximal to the CSJ and the cortex. Data are expressed as mean ± SEM. Asterisk denotes a significant difference in the normalized GABA + modulator response amplitude between eGFP+ cells from the area proximal to the CSJ and the cortex (C) or transcript abundance (F) relative to that found in the CSJ (p < 0.05, Student's t-test).

Ro15-4513 (1 μM), the partial inverse agonist for α4-subunit–containing recombinant GABA_A receptors (Wisden et al. 1991; Knoflach et al. 1996; Wafford et al. 1996) attenuated GABA-activated current responses in the majority of eGFP+ cells all along the tangential migratory route (CSJ: Fig. 3A4; 17 of 17 cells and cortex: Fig. 3B4; 17 of 20). The mean magnitude of the Ro15-4513-induced attenuation in cells recorded in the area proximal to the CSJ was not significantly different from that recorded in cells located in the cortex (Fig. 3C; proximal to CSJ: 81.6 ± 5.2% and cortex: 78.0 ± 11.0%; P = 0.27). The α5-subunit–selective inverse agonist L655-708 (10 μM; Pritchett and Seeburg 1990) also attenuated GABA responses in eGFP+ cells regardless of location (proximal to CSJ: 69.5 ± 4.5%; 14 of 14 cells and cortex 41.5 ± 9.6%; 15 of 15 cells). However, eGFP+ cells in the cortex were attenuated to a significantly greater extent than their counterparts proximal to the CSJ (Student's t-test; P = 0.01). These results, taken together, suggest that migrating eGFP+ cells already express functional α4- and α5-subunit–containing GABA_A receptor isoforms prior to entry into the cortex and begin to incorporate the α1-subunit in assembling functional GABA_A receptors once they have reached the cortical anlage.

Expression profiling of individual eGFP+ cells proximal to the CSJ revealed the presence of the α2- (31 of 36 cells), α4- (30 of 36 cells), and α5- (30 of 36 cells) subunits in a majority of the eGFP+ cells (Fig. 3D). The α1-subunit transcript was notably absent in 27 of 36 cells (Fig. 3D). However, the α1-subunit transcript was consistently present in eGFP+ cells located in the cortex of the same acute slice preparation (Fig. 3E; 44 of 44 cells). Overall, although the α1-subunit was detectable in a few cells harvested proximal to the CSJ, its expression was dramatically upregulated in individual eGFP+ cells situated in the cortex as were the α2-, α3-, and α5-subunits (Fig. 3F). It should be noted that the increase in the α3 transcript was not evident in MGE- and cortex-derived tissues (Fig. 2). It may be that α3 transcript expression is limited to the MGE-derived eGFP+ cells (proximal to CSJ: 31 of 36 cells and cortex: 43 of 44 cells), masking a relatively modest (albeit significant) increase in relative abundance when assayed at the tissue level.

β-Subunits

Loreclezole is a broad-spectrum anticonvulsant with positive allosteric action on β2- or β3-subunit–containing GABA_A receptor isoforms (Wafford et al. 1994; Wingrove et al. 1994). Loreclezole (10 μM) potentiated GABA (100 μM) responses of eGFP+ cells in the area proximal to the CSJ (Fig. 4A; 126.0 ± 5.6%; P = 0.04, n = 18) and cortex (Fig. 4B; 130.2 ± 8.3%; P = 0.03; n = 17). Expression profiling in individual eGFP+ cells indicated varied patterns of β1-, β2-, and β3-subunit expression. Figure 4D illustrates an eGFP+ cell harvested from the area proximal to the CSJ that expressed the β2- and β3-subunit transcripts (β1-subunit in 21 of 26 cells, β2-subunit in 32 of 36 cells, and β3-subunit in 28 of 36 cells; Fig. 4D). The example of another cell harvested from the cortex displayed transcripts for all three β-subunits (β1: 23 of 26 cells, β2: 40 of 44 cells, and β3: 42 of 44 cells; Fig. 4E). Of the β-subunit transcripts, the expression of the β3 transcript increased significantly in cortical eGFP+ cells compared with those from the area proximal to the CSJ (Fig. 4F; area proximal to the CSJ: n = 36 and cortex: n = 44).

Figure 4.

Expression of GABA_A receptor β-subunits in individual BAC-Lhx6 eGFP+ cells in the area proximal to the CSJ and cortex. (A and B) Whole-cell current responses of eGFP+ cells in the area proximal to the CSJ (A) or the cortex (B) to focal application of GABA either alone (100 μM; black traces) or in conjunction with loreclezole (10 μM; red traces). (C) Percentage change in the amplitude of GABA-induced current responses in the presence of loreclezole in eGFP+ cells examined in the area proximal to the CSJ (black bars) and cortex (gray bars). Data are expressed as mean ± SEM. The dashed line denotes no change from the current response amplitude elicited by application of GABA alone. (D and E) Expression profiling of the GABA_A receptor β1–3 subunits in individual eGFP+ cells located in the area proximal to the CSJ (D) and the cortex (E). The darker bands in the molecular weight (MW) ladder lanes correspond to the 600-bp position. The arrowheads point to PCR-amplified products of the corresponding β-subunit transcripts, which are also given at the top of each lane. (F) Semiquantitative determination of the abundance of GABA_A receptor β-subunit transcripts (β1–3) relative to β-actin in the same eGFP+ cells harvested from the area proximal to the CSJ (black bars) and the cortex (gray bars). Data are expressed as mean ± SEM. Asterisk in (F) denotes a significant increase in cortical cells of transcript abundance relative to that detected in the CSJ (P < 0.05, Student's t-test).

γ- and δ-Subunits

Since eGFP+ cells proximal to the CSJ expressed GABA_A receptors with lower affinity to GABA (Fig. 1B) and since the γ2- and γ3-subunit transcripts are upregulated in the cortex (Fig. 2), we hypothesized that MGE-derived cells differentially incorporated γ-subunits with advancing migration and maturation. Zinc inhibits GABA_A receptor function in recombinant receptor isoforms lacking γ-subunits, while diazepam potentiates GABA action at receptors containing γ-subunits (Pritchett et al. 1989; Kirshek et al. 1998; Hosie et al. 2003). In the present study, zinc chloride (100 μM) attenuated GABA responses monitored in 12 of 14 eGFP+ cells recorded in the area proximal to the CSJ (Fig. 5A1 and C; 73.1 ± 5.5%; P < 0.01). The same 12 cells were insensitive to potentiation by diazepam (3 μM; Fig. 5A2 and C; 109.8 ± 9.3%; P = 0.39; 12 of 14). The remaining 2 cells tested in the area proximal to the CSJ exhibited zinc and diazepam response profiles that resembled those of cortical eGFP+ cells. Specifically, in the cortex, the great majority of eGFP+ cells (14 of 15 cells) displayed GABA-induced currents that were insensitive to modulation by zinc (Fig. 5B1; 93.3 ± 5.7%; P = 0.30) but were potentiated by diazepam (Fig. 5B2; 150.0 ± 17.0%; P = 0.004). These results indicated that MGE-derived cells switch from expressing GABA_A receptor isoforms with γ-less to γ-containing pharmacology in the course of corticopetal migration. However, expression profiling of the family of γ-subunits (γ1–γ3) yielded a different outcome. Specifically, all 36 eGFP+ cells harvested and profiled in the area proximal to the CSJ and all 44 of their counterparts in the cortex expressed transcripts for one or more of the γ-subunit transcripts. The relative abundance of the γ2- and γ3-subunit transcripts increased markedly in MGE-derived cells that have entered the cortex (Fig. 5F).

Figure 5.

Individual BAC-Lhx6 eGFP+ cells in the embryonic cortex, but not those in the area proximal to the CSJ, display GABA_A receptor γ-subunit pharmacology and expression profiles. (A and B) Whole-cell current responses of eGFP+ cells in the area proximal to the CSJ (A) or the cortex (B) to focal application of GABA either alone (100 μM; black traces) or in conjunction with zinc chloride (Zinc; 100 μM; A1 and B1; red traces) or diazepam (DZ; 3 μM; A2 and B2; red traces). The GABA response of the cell in the area proximal to the CSJ was abolished in the presence of zinc (A1) but was insensitive to modulation by diazepam (A2). The GABA response of the cell recorded in the cortex, however, was insensitive to zinc (B1) but potentiated in the presence of diazepam (B2). (C) Percentage change in the amplitude of GABA-induced current responses in the presence of zinc or diazepam in eGFP+ cells examined in the area proximal to the CSJ (black bars) and cortex (gray bars). Data are expressed as mean ± SEM. The dashed line denotes no change from the current response amplitude elicited by application of GABA alone. (D and E) The amplicons corresponding to the GABA_A receptor γ1–3 subunits in individual eGFP+ cells located in the CSJ (D) and the cortex (E) are electrophoresed parallel to a molecular weight (MW) ladder. The 600-bp position is indicated in the lanes containing the molecular weight ladder. The arrowheads point to the predicted sizes of the amplicons corresponding to the γ-subunit transcripts, which are also given at the top of each lane. (F) Semiquantitative results of GABA_A receptor γ-subunit transcripts (γ1–3) of single eGFP+ cells harvested from the area proximal to the CSJ (black bars) and the cortex (gray bars). Amount of transcript was normalized to the amount of β-actin in the individual cell. Data are expressed as mean ± SEM. Asterisk denotes a significant difference in the normalized GABA + modulator response amplitude between eGFP+ cells from the area proximal to the CSJ and the cortex (C) or transcript abundance (F) relative to that found in the area proximal to the CSJ (P < 0.05, Student's t-test).

In certain extrasynaptic GABA_A receptor isoforms, the γ-subunit is replaced with the δ-subunit (Nusser et al. 1998). We asked whether this might be the case in MGE-derived cells that displayed γ-less GABA_A receptors. Figure 6 illustrates that THIP (10 μM), a GABA_A receptor agonist with preference for δ-containing receptors, induced an inward current response in an eGFP+ cells recorded in the area proximal to the CSJ (Fig. 6A; 13 of 13 cases) and the cortex (Fig. 6B; 14 of 14 cases). THIP induced currents of comparable amplitude in cells regardless of location (Fig. 6C; CSJ: −41.06 ± 3.64 pA; cortex: −46.46 ± 12.43 pA; P = 0.23). Consistent with this, the δ-subunit transcript was present at a similar relative expression level in every cell profiled (Fig. 6F, P = 0.32) in the area proximal to the CSJ (Fig. 6D; 36 of 36 cells) and cortex (Fig. 6E; 44 of 44 cells). This suggested that, unlike the family of γ-subunits, the δ-subunit is not developmentally regulated in MGE-derived cells.

Figure 6.

The relative expression of the δ-subunit of the GABA_A receptor in BAC-Lhx6 eGFP+ cells is unchanged regardless of location along the tangential migratory. (A and B) Whole-cell current responses of BAC-Lhx6 eGFP+ cells recorded in the area proximal to the CSJ (A) or the cortex (B) to focal application of THIP (10 μM). (C) The mean amplitude of the current response to THIP is not significantly different between cells recorded in the area proximal to the CSJ and in the cortex (Student's t-test; p = 0.23). Data expressed as mean ± SEM. (D and E) Single-cell expression profiling of individual eGFP+ cells in the area proximal to the CSJ (D) and the cortex (E) reveals the presence of the GABA_A receptor δ-subunit. The darker bands in the molecular weight (MW) ladder lanes denote the 600-bp position. The arrowheads point to the predicted size of amplicon corresponding to the δ-subunit transcript, which is also indicated at the top of each lane. (F) Abundance of the GABA_A receptor δ-subunit transcript in single BAC-Lhx6 eGFP+ cells harvested from the area proximal to the CSJ and the cortex relative to that of β-actin in the same cell. Data are expressed as mean ± SEM.

Our profiling of GABA_A receptor subunit expression in the MGE-derived cells revealed heterogeneity not only among but also within the family of α-, β-, and γ-subunits. A striking trend was for individual migrating MGE-derived cells to incorporate more GABA_A subunit isomers that can potentially assemble into pharmacologically distinct receptor isoforms. This was particularly evident in the expression profile of the α1-subunit. Specifically, assuming a hypothetical α/β/γ subunit stoichiometric motif, only 26% of the MGE-derived cells proximal to the CSJ could potentially express the combination of α1/βx/γx subunits, while it could be expressed in 75% of those located in the cortex (Fig. 7A). An increased potential was also observed for expressing the α3/βx/γx and α5/βx/γx subunit combinations but not for expressing the α2/βx/γx or α4/βx/γx subunit combinations (Fig. 7A) nor those containing the β1–3 or γ1–2 subunits (Fig. 7B and C, respectively). Overall, although the putative GABA_A receptor subunit combinations expressed in any given MGE-derived cortical interneuronal subpopulation await elucidation, our results underscore the potential for dynamic expression of select members of the α-subunit family in tangentially migrating MGE-derived cells.

Figure 7.

Percentage of eGFP+ MGE-derived cells expressing hypothetical α1–5/βxγx, αx/β1–3γx, and αx/βxγ1–3 GABA_A receptor subunit combinations. Summary of the potential for individual eGFP+ cells expressing any one of the α1-5 GABA_A receptor subunit transcripts to form α/β/γ receptor isoforms in combination with βxγx subunits profiled in the same cells (A) for β1–3 subunit transcripts in combination with αxγx (B) and for γ1–3 in combination with αxβx (C). The percentages reflect the potential for a given cell to express multiple GABA_A receptor isoforms since each cell profiled expressed multiple α, β, and γ isomers.

Discussion

In this study, we demonstrate that primordial GABAergic cortical interneurons migrating from the region of the MGE to the embryonic cortical anlage display regionally selective differences in pharmacological GABA response properties as well as profiles of GABA_A receptor subunit transcripts. The use of acute telencephalic slices from E14.5 BAC-Lhx6 embryos permitted us to target corticopetal eGFP+ cells of MGE origin when tangential migration is at its height. We reasoned that their location and distribution in any given E14.5 telencephalic slice should reflect the progression of the tangential migratory process. Specifically, those situated in the cortex presumably would be more mature than those found in the region of the MGE since they were ahead of the migratory exodus and, thus, were born earlier. In this light, we propose that, as part of their maturational process, primordial GABAergic cortical interneurons acquire increasing responsiveness to GABA and switch to expressing more mature GABA_A receptor isoforms with advancing migration into the cortex. Indeed, while the literature is replete with observations of either changes in neuronal responsiveness to GABA (Shen et al. 1988; LoTurco and Kriegstein, 1991; Owens et al. 1999; Hutcheon et al. 2000) or expression of GABA_A receptor subunits in the immature brain (Gambarana et al. 1991; Araki et al. 1992; Laurie et al. 1992; Cheng et al. 2006; Peden et al. 2008; Yu et al. 2009), our study integrated these developmental processes into perspective with the embryonic development of an identified neuronal population, notably, that of the GABAergic cortical interneurons.

MGE-Derived Cells Increase Receptivity to GABA With Tangential Migration

In the present study, GABA concentration–response profiles, coupled with assessing the effectiveness of GABA_A receptor subunit–selective modulators, facilitated a pharmacological analysis of the GABA response monitored in tangentially migrating MGE-derived cells. We found that the concentration–response curve derived from eGFP+ neurons in the region of the MGE shifted leftward relative to that obtained from their cohorts in the embryonic cortex. This suggested that the MGE-derived neurons en route to the embryonic cortex acquire GABA_A receptors with increasing apparent efficacy and affinity to GABA. While apparent efficacy provides an indication of receptor number, apparent affinity reflects in part the subunit makeup of functional GABA_A receptor isoforms. Thus, we propose that MGE-derived neurons upregulate and incorporate functional GABA_A receptors of the same isoforms, isoforms with higher affinity to GABA, or both as they migrate from the MGE to the embryonic cortex. MGE-derived cells located in the MGE region also displayed GABA-activated currents characterized by slow kinetics and decay time with little to no desensitization compared with those recorded in the cortex. In addition, profiling of GABA_A receptor subunit transcripts in individual migrating cells revealed differential expression of a number of α- and γ-subunit transcripts. These observations reinforce the notion that different subunits contribute to the makeup of GABA_A receptors in MGE-derived cells at different locations along the tangential migratory route.

MGE-Derived Cells in the Cortex Express More Mature GABA_A Receptor Isoforms

The expression of many GABA_A receptor subunits is developmentally regulated (Gambarana et al. 1991; Araki et al. 1992; Laurie et al. 1992; Cheng et al. 2006; Peden et al. 2008). We focused on analyzing GABA_A receptor subunits reported to be expressed in the developing cortex, notably the α1–α5, β1–β3, γ1–γ3, and δ-subunits (Gambarana et al. 1991; Araki et al. 1992; Laurie et al. 1992; Cheng et al. 2006; Peden et al. 2008). Although not an exhaustive analysis, the present data illustrate the dynamic expression and potential for functional diversity of GABA_A receptors in migrating MGE-derived neurons.

In individual primordial GABAergic cortical interneurons located proximal to the CSJ, there was robust expression of the α2- and α3-subunit transcripts vis-a-vis the α1-subunit, reminiscent of type II benzodiazepine GABA_A receptor isoforms. As these neurons enter the developing cortex, the α2- and α3-subunit transcripts continue to be expressed, but a dramatic upregulation in the α1- and γ1–3 subunit transcripts occurs, suggesting a switch to a predominantly type I benzodiazepine pharmacology. Corroborating these findings was electrophysiological results, indicating that zolpidem and DMCM were ineffective in modulating GABA-activated current responses in eGFP+ neurons in the area proximal to the CSJ but potentiated and suppressed, respectively, the responses elicited in the same population of neurons situated in the developing cortex. These results are consistent with previous studies reporting the predominant expression of the α2- and α3-subunits early during development, which subsequently is replaced by the α1-subunit with advancing neuronal migration (Laurie et al. 1992; Fritschy et al. 1994; Hornung and Fritschy 1996; Bosman et al. 2002; Liu and Wong-Riley 2004; Takayama and Inoue 2004).

Along a similar developmental trend, our electrophysiological results employing diazepam and zinc, combined with transcript profiling, indicate that MGE-derived neurons first express γ-subunit–less GABA receptor isoforms as they migrate in the subpallium and then express γ-subunit–containing isoforms once they enter and traverse the developing cortex. This scheme of a γ-less to γ-containing switch in GABA_A receptor isoform is reminiscent of our previous study that demonstrated expression of γ-less GABA_A receptors in embryonic but not in postnatal Cajal–Retzius cells during corticogenesis (Cheng et al. 2006) and is consistent with the notion that MGE-derived neurons switch from expressing immature to more mature GABA_A receptor isoforms as they migrate. However, our expression profiling revealed that virtually all MGE-derived cells, regardless of location, expressed at least one of the γ-subunit transcripts. It may be that MGE-derived cells express γ-subunit transcripts early in the course of migration, but the encoded proteins are not translated and assembled into functionally demonstrable GABA_A receptor isoforms until they have reached the cortical anlage. Such a developmental relationship between transcript and protein expression may also apply to other GABA_A receptor subunits. The present study contributes to the groundwork for a comprehensive immunohistochemical analysis to resolve these and other issues related to the assembly of functionally distinct GABA_A receptor isoforms in migrating and mature MGE-derived cortical interneurons.

Tangentially Migrating MGE-Derived Cells Express functional “Synaptic” and “Extrasynaptic” GABA_A Receptor Isoforms

Subdomains of the MGE give rise to Nkx2.1/Lhx6–expressing precursor GABAergic cortical interneurons, including basket and Martinotti interneurons that either express or coexpress parvalbumin, somatostatin, calbindin, and NPY (Xu et al. 2004; Fogarty et al. 2007). Our experimental approach precluded ascribing the different GABA_A receptor pharmacological and subunit transcript profiles of eGFP+ cells to their origin within subdomains of the MGE (e.g., dorsal vs. lateral MGE) and to their specification into specific GABAergic interneuronal subtypes later in development. Future studies would need to incorporate genetic fate mapping strategies to refine the identification of tangentially migrating MGE-derived cells and to establish GABA_A receptor subunit expression profiles of GABAergic interneuronal subtypes in the adult cortex.

The above caveats notwithstanding, our results of expression profiling implicate migrating MGE-derived cells expressing more subunit transcripts than necessary to constitute one functional isoform of GABA_A receptor. In this light, our pharmacological analyses of whole-cell GABA-activated current responses would be expected to reflect a net outcome of simultaneous activation of multiple GABA_A receptor isoforms. Although the subunit combinations of GABA_A receptor isoforms expressed in adult GABAergic cortical interneurons await elucidation, it is reasonable to postulate that, in addition to α4-, α5- and δ-containing GABA_A receptor isoforms (e.g., α4βxδ and α5βxγ2/3), which are relatively evenly expressed in MGE-derived cells regardless of location, there is upregulation in the cortex of isoforms that include the α1- and γ-subunits, such as those of the α1βxγ2/3 combination. These GABA_A receptor isoforms have been implicated to mediate either “synaptic” or “extrasynaptic” actions of GABA (reviewed by Sieghart et al. 1999; Barnes 2001; Kittler et al. 2002). At E14.5, synapses have not yet formed on tangentially migrating MGE-derived cells, but the results of our study indicate that synaptic and extrasynaptic types of GABA_A receptors are already present and functional. Extrasynaptic GABA_A receptor isoforms, those of the α5βxγ2/3 subunit combination in particular, have been shown to be highly sensitive to GABA (Burgard et al. 1996; Nusser and Mody 2002; Caraiscos et al. 2004; Bonin et al. 2007). In this light, they are in a favorable position to mediate GABA tone and, thus, play an important role in regulating the migration of MGE-derived neurons.

Functional Implications

GABA is present in the embryonic brain, and immature neurons express GABA_A receptors well before the first signs of synaptogenesis (van Eden et al. 1989; Imamoto et al. 1994; Zecevic and Milosevic 1997; Cuzon et al. 2006). As a “developmental neurotransmitter” (Redburn and Rowe-Rendleman 1996; Redburn-Johnson 1998; Owens and Kriegstein 2002; Represa and Ben-Ari 2005; Heng et al., 2007; Manent and Represa 2007), GABA maintained at an ambient level is trophic for migrating neurons. This trophism has been reported to be mediated by GABA_A receptors (Behar et al. 1996, 1998, 2000; Bolteus and Bordey 2004; Manent et al. 2005; Cuzon et al. 2006), but whether subunit composition comes into play has not been addressed. Here, we provide evidence that MGE-derived neurons express GABA_A receptors of different subunit compositions and functional properties in the course of their migration. We propose that these differences may impart attributes to embryonic MGE-derived neurons that are important in conferring the role of GABA as a trophic factor. Specifically, the upregulation of receptor isoforms with higher affinity to GABA may be important in signaling cortical entry of MGE-derived cells and beyond. At the time point examined in this study (E14.5), GABA_A receptor activation induces membrane depolarization that activates voltage-gated calcium channels (Cherubini et al. 1991; Walton et al. 1993; Lin et al. 1994; LoTurco et al. 1995; Owens et al. 1996). Since rises in intracellular calcium have been shown to be involved in numerous gene regulatory processes, including those leading to cellular differentiation (Holliday et al. 1991; Gu and Spitzer 1995), it may be that activation of GABA_A receptors in MGE-derived cells triggers calcium influx to trigger the transcription of genes encoding GABA_A receptor subunits, leading to cellular differentiation, maturation, and synaptogenesis.

Funding

Public Heath Service Grants RO1 MH069826 (H.H.Y.) and F31 AA014698 (V.C.C.C.).