| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Subunits of GABAA Receptor Regulates Inhibitory Synaptic Strength
Interdisciplinary Program in Neuroscience and Department of Physiology and Biophysics, Georgetown University School of Medicine, Washington, DC 20007
Submitted 10 March 2004; accepted in final form 15 April 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
subunit subtypes in the molecular assembly of GABAA receptors are a critical determinant of the functional properties of inhibitory synapses and their modulation by a range of pharmacological agents. We investigated the contribution of these subunits to the developmental changes of inhibitory synapses in cerebellar granule neurons in primary cultures from wild-type and 1 subunit –/– mice. The decay time of miniature inhibitory postsynaptic currents (mIPSCs) halved between 6 days in vitro (DIV6) and DIV12. This was paralleled by the decrease of 2 and 3 subunits, the increase of 1 and 6 subunits expression at synapses, and changes in the action of selective subunit modulators. A small but significant shortening of mIPSCs was observed with development in cells from –/– mice together with a decrease in the expression of 3 subunit. In contrast, the expression of 2 subunit at inhibitory synapses in –/– cells was significantly higher than in +/+ cells at DIV11-12. 5 subunit was not detected, and increased sensitivity to a selective 4/6 subunit agonist suggests increased expression of extrasynaptic receptors in –/– mice. 
2/3 subunit expression and loreclezole sensitivity increased with development in +/+ but not in –/– cells, supporting the preferential association of the 1 with the 2 subunit. Synaptic charge transfer strongly decreased with development but was not different between cells in the +/+ and –/– groups until DIV11-12. Our results uncover a pattern of sequential expression of subunits underlying the changes in functional efficacy of GABAergic networks with development. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
1 subunit, for example, produces fast current decay and confers sensitivity to the imidazopyridine, zolpidem (Vicini et al. 2001
), 2 and 3 prolong deactivation (Gingrich et al. 1995; Lavoie et al. 1997; McClellan and Twyman 1999; Verdoorn 1994), and 6 confers sensitivity to furosemide (Korpi et al. 1995) and is responsible for tonic inhibition (Hamann et al. 2002; Kaneda et al. 1995; Wall and Usowicz 1997). In addition, a GABA agonist 4,5,6,7-tetrahydroisothiazolo-[5,4-c]pyridin-3-ol (THIP), has been shown to act as a superagonist at extrasynaptic 4 and 
subunit-containing receptors (Brown et al. 2002) that are preferentially enhanced by ethanol (Wallner et al. 2003).
In cerebellar neurons of mice and rats, expression of 1 and 6 subunit mRNA increases with age, whereas 2 and 3 subunit mRNA is downregulated (Laurie et al. 1992a,b; Poulter et al. 1992). Consequently, the kinetics of spontaneous and miniature IPSCs (sIPSCs and mIPSCs) in cerebellar granule cells (CGCs) become faster with development as demonstrated by electrophysiological recordings in slices (Brickley et al. 1996; Tia et al. 1996). This is due to increased expression of 1 but not 6 subunits as shown in mutant mice lacking these subunits (Farrant et al. 1999; Vicini et al. 2001). Development of an in vitro system allows direct correlation between anatomical (Gao and Fritschy 1995) and functional approaches to study inhibitory synapses, but it has not yet been established whether developmental changes in inhibitory synaptic function observed in vivo can be reliably replicated in culture conditions. Primary cultures of CGCs are best maintained in a depolarizing medium containing 25 mM KCl (Gallo et al. 1987), which, however, prevents detection of synaptic activity with electrophysiological methods (Mellor et al. 1998). In this culture condition, immunocytochemical studies have shown some clustering of 1 but not 3 or 6 subunits (Gao and Fritschy 1995; Studler et al. 2002). Low (5 mM) concentration of K+ in a medium supplemented with specific additives favors formation of both inhibitory (Mellor et al. 1998; Ueno et al. 1996; Virginio et al. 1995) and excitatory (Chen et al. 2000; Losi et al. 2002) networks between CGCs and GABAergic interneurons in culture. This allows electrophysiological recordings from such cultures. sIPSCs and mIPSCs under these conditions can be recorded several days after plating, but have only been extensively characterized late in development [>12 days in vitro (DIV12)] (Mellor et al. 2000). In these cultures, whole cell recordings revealed tonic GABA-mediated current that is TTX sensitive and therefore mostly mediated by spillover of GABA from high-density active inhibitory synaptic terminals (Leao et al. 2000). This tonic current is absent in CGCs in slices from mice with deletion of the 6 or subunits (Brickley et al. 2001; Stell et al. 2003).
We investigated age-dependent changes of GABAA-mediated IPSCs and tonic current in CGC cultures. The data gathered with electrophysiological, immunocytochemical, and pharmacological techniques applied to CGCs cultured from wild-type (+/+) and 1 knock-out (–/–) mice demonstrate specific contributions of distinct subunits underlying developmental changes of IPSC kinetics in vitro.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
-subunit-deficient mice were produced at the University of Pittsburgh as described (Vicini et al. 2001) and shipped to Georgetown University for breeding and all experimental procedures. Heterozygous mice on a mixed genetic background (C57BL/6J, strain 129/Sv/SvJ, and FVB/N) (Vicini et al. 2001) were interbred to produce wild-type (+/+), heterozygous (±), and homozygous (–/–) knockout mice. Polymerase chain reaction (PCR) was used to genotype mutant mice. Total genomic DNA was isolated at the third postnatal day from tail snips with DNeasy Tissue Kit (Qiagen, Valencia, CA), according to the manufacturer's instructions. DNA was amplified with PCR using the primers GAB1 (tctgcatgtgggacaaagac), GAB2 (acgcataccctctcttggtg), and GAB3 (tgattgcttttctgagataggg). GAB1 and GAB2 amplify across the deleted region of 1 DNA sequence and allow identification of the wild-type allele; GAB1 and GAB3 amplify outside the deleted sequence and allow identification of the knock-out allele. Primers (2 µl each) and DNA (5 µl) were mixed with Ready-to-Go PCR Beads (Amersham Pharmacia Biotech, Piscataway, NJ) and diluted to 25 µl with distilled H2O. Twenty-seven PCR cycles were run after initial denaturation at 95°C for 5 min. Each cycle consisted of: denaturation at 95°C for 1 min, primer annealing at 57°C for 1.5 min, and primer elongation at 72°C for 2 min. A final elongation step was performed at 72°C for 3 min. PCR amplification products were loaded at 10 µl/lane into a 2% agarose gel, separated by electrophoresis, and visualized with ethidium bromide staining.
CGC cell culture and immunocytochemistry
Primary cultures of mouse CGCs were prepared from postnatal day 5–7 mice. Mouse pups were killed by decapitation in agreement with the guidelines of the Georgetown University Animal Care and Use Committee. The cerebella were dissociated with trypsin (0.25 mg/ml, Sigma, St. Louis, MO) and plated in 35-mm Nunc dishes at a density of 1.1 x 106 cells/ml on glass coverslips (Fisher Scientific, Pittsburgh, PA) coated with poly-L-lysine (10 µg/ml; Sigma). The cells were cultured in basal Eagle's medium supplemented with 10% bovine calf serum, 2 mM glutamine, and 100 µg/ml gentamycin (all from Invitrogen, Carlsbad, CA) and maintained at 37°C in 5% CO2. The final concentration of KCl in the culture medium was adjusted to 25 mM (high K+). At DIV5, the medium was replaced with low (5 mM) K+ medium [MEM supplemented with 5 mg/ml glucose, 0.1 mg/ml transferrin, 0.025 mg/ml insulin, 2 mM glutamine, 20 µg/ml gentamicin (Invitrogen) and 10 µM cytosine arabinofuranoside, Sigma] as previously described (Chen et al. 2000; Losi et al. 2002).
All immunostaining experiments were carried out at room temperature. Live cultured neurons were fixed by incubation with 4% paraformaldehyde, 4% sucrose in PBS for 10 min and washed three times in PBS. Fixed neurons were permeabilized with 0.3% Triton X-100/PBS for 5 min, washed several times with PBS, and incubated in 10% bovine serum albumin in PBS for 1 h to block nonspecific staining. Cells were then incubated for 1 h with one of the following primary antibodies: glutamic acid decarboxylase (GAD65) monoclonal antibody (GAD6; Developmental Studies Hybridoma Bank, University of Iowa, Iowa City, IA; 1:1000), rabbit anti-1 (Upstate, Waltham, MA; 4 µg/ml), guinea pig anti-2 (1:2,000), guinea pig anti-3 (1:1,000), guinea pig anti-5 (1:1,000), kind gifts from Dr. Jean Marc Fritschy (University of Zurich), or rabbit anti-6 (1:1,000), a kind gift from Dr. Angel DeBlas (University of Connecticut). After washing with PBS, cells were incubated with indocarbocyanine (Cy3)-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, West Grove, PA; 1:1,000) and, in double-staining experiments, with Alexa488-conjugated secondary antibodies (Molecular Probe; 1:1,000).
Neurons were imaged on a Nikon E600 microscope (Nikon) equipped with a x60, 1.0 N.A. objective. Nikon band-pass filter cubes were used for Cy3 or Alexa488 fluorescence. Digital images were acquired with a CFW-1310 (Scion, Frederick, MD), 10-bit (1,024 gray scale intensity level) CCD digital camera, 1,360 x 1,024 pixel array. Images were analyzed with MetaMorph (Universal Imaging, Downingtown, PA), and pseudocolored for presentation with Adobe Photoshop 7.0 (Adobe, San Jose, CA). Antibody staining was considered positive when staining intensity was at least twice the fluorescence intensity of the background. For quantification of developmental changes of synaptic density, the number of GAD-positive puncta was counted along 100-µm segments of randomly selected dendrites. At least 12 segments were analyzed at each age. For the analysis of synaptic localization of subunits, the percent of GAD-positive puncta containing subunit-specific staining along 100-µm segments of randomly selected dendrites was counted (>10 segments for each subunit). Synaptic localization was assumed when the center pixels of subunit clusters and GAD puncta were <1 µm apart. Statistical comparisons were done using a two-tailed Student's t-test assuming homogeneity of variances of the samples with Bonferroni corrections. Data values are expressed as means ± SE.
Electrophysiology
Coverslips with CGCs were placed on the stage of an inverted microscope (TM2000, Nikon) equipped with fluorescent and phase contrast optics. All recordings were performed at room temperature (24–26°C) from neurons maintained for 6–14 DIV. Continuously perfused extracellular solution contained (in mM) 145 NaCl, 5 KCl, 1 MgCl2, 1 CaCl2, 5 HEPES, 5 glucose, and 15 sucrose and 0.25 mg/l phenol red and 10 µM D-serine (all from Sigma) adjusted to pH 7.4 with NaOH. Electrodes were pulled in two stages on a vertical pipette puller from borosilicate glass capillaries (Wiretrol II, Drummond, Broomall, PA) and filled with recording solution containing (in mM) 145 KCl, 10 HEPES, 5 ATP.Mg, 0.2 GTP.Na, and 10 BAPTA, adjusted to pH 7.2 with KOH. Pipette resistance was 5–8 M
. Whole cell voltage-clamp recordings were made at –60 mV with an Axopatch-1D amplifier (Axon Instruments, Union City, CA), and access resistance was monitored throughout the recordings. Capacitance was calculated from a transient current response to a hyperpolarizing 10-mV pulse. Currents were filtered at 1 kHz with an 8-pole low-pass Bessel filter (Frequency Devices, Haverhill, MA), digitized at 5–10 kHz using an IBM-compatible microcomputer equipped with Digidata 1322A data-acquisition board and pCLAMP9 software (both from Axon Instruments). Off-line data analysis, curve fitting, and figure preparation were performed with Clampfit 9 software. GABA, THIP, flunitrazepam, furosemide, zolpidem, bretazenil, bicuculline metabromide (BMR, Sigma), and loreclezole (a gift from Janssen Research Foundation, Beerse, Belgium) were dissolved either in water or in dimethylsulfoxide (DMSO, <0.001% final concentration, Sigma) and diluted in the extracellular medium. Drugs were locally applied by means of a Y tube (Murase et al. 1989). mIPSCs were isolated by application of 0.5 µM tetrodotoxin (TTX, Alamone Labs, Israel). Infrequent AMPA-mediated miniature excitatory postsynaptic currents (mEPSCs) were easily identified by the very rapid decay (<2 ms) and excluded from the analysis. sIPSC and mIPSC averages were based on 
50 events in each cell studied. The decay phase of currents was fitted using a simplex algorithm for least-squares exponential fitting routines with triple exponential equation of the form I(t) = I1 * exp(–t/
1) + I2 * exp(–t/2) + I3 * exp(–t/3), where Ix is a peak amplitude of a decay component and x is the corresponding decay time constant. To allow for easier comparison of decay times between experimental conditions, the three decay time components were combined into a weighted time constant w = [I1/(I1 + I2 + I3)] * 1 + [I2/(I1 + I2 + I3)] * 2+[I3/(I1 + I2 + I3)] * 3. Mean current changes with BMR and TTX were performed by computing the differences between 2 averages of baseline current that were devoid of sIPSCs before and during applications of each drug. All data are expressed as mean ± SE, unless otherwise indicated; P values represent the results of 2-tailed Student's t-test with Bonferroni corrections.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
sIPSCs were studied in mouse CGCs in primary cultures with whole cell voltage-clamp recordings. Granule neurons were identified as neurons with a cell body axis of 
7 µm that had relatively simple processes and lacked sustained high-frequency action potential firing, a hallmark of GABAergic interneurons. Cells were voltage-clamped at –60 mV using an intracellular pipette solution containing symmetrical chloride to the bath solution and were observed as inward currents (Fig. 1). As reported previously, culture medium containing 5 mM KCl allows the formation of functional inhibitory and excitatory synapses (Chen et al. 2000; Mellor et al. 1998, 2000; Virginio et al. 1995). We began our studies in cells cultured for 6 days in vitro (DIV6), 24 h after lowering K+ concentration in the medium. sIPSCs occurred in the majority of cells investigated at this age with highly variable frequency of occurrence and amplitude. Examples of sIPSCs recorded from control cells at DIV6 and DIV9 are reported in Fig. 1A. The decay of sIPSCs in these cells was best described by three exponential curves (Fig. 1A), similar to sIPSCs recorded from CGCs in cerebellar slices (Farrant and Brickley 2003). The three component decay was maintained when current flow was reverted by clamping the cell at +60 mV (not shown).
|
w) based on a triple-exponential fit of decay (see METHODS) was 35.5 ± 1.7 ms (n = 33) at DIV6-7 and decreased to 18.2 ± 1 ms (49%) by DIV11-12 (n = 34, P < 0.01). To investigate the role of the 1 subunit in developmental changes of sIPSCs w, we cultured CGCs from 1 –/– mice. Developmental speeding of sIPSCs in these cells could be observed but was less pronounced (Fig. 1B): sIPSCs averaged 36.7 ± 1.7 ms (n = 36) at DIV6-7 and decreased to 30.6 ± 1.9 ms (17%) by DIV11-12 (n = 24, P < 0.01). The w of sIPSC averages in the +/+ group were significantly faster than those of sIPSCs in the –/– group for all ages after DIV7 (P < 0.01). These data support the critical role of 1 subunit in mediating the developmental decrease of sIPSCs in CGC cultures first reported in cerebellar slices (Vicini et al. 2001).
sIPSCs in CGCs are generated by presynaptic action potentials in GABAergic interneurons and can be due to simultaneous activation of multiple synaptic sites. Analysis of mIPSCs, which are generated at single synaptic sites, allows for more robust conclusions about changes at individual synapses. mIPSCs were isolated by application of TTX (0.5 µM) and mIPSC averages fitted with three exponential curves. Data from consecutive days were grouped to increase statistical power (Fig. 2). Individual time constant values and their relative contributions to peak mIPSCs are reported in Table 1. The w of mIPSC averages in both the +/+ and the –/– groups showed progressive developmental decrease (Fig. 2A). mIPSCs from the +/+ group were faster than those from the –/– group at all ages after DIV6-7 (P < 0.01) similar to what was observed for sIPSCs. The decrease in w was accompanied by a decrease in peak amplitude in both genotypes (Fig. 2B, Table 1). w changes were significant for the wild-type group at DIV11-12 and only reached significance between the two groups at DIV8-9. Charge transfer was significantly reduced in both the +/+ and the –/– groups as early as DIV8-9 (Fig. 2C), indicating that the decreased mIPSCs amplitude compensates for the slower decay in –/– CGCs. However, by DIV11-12, charge transfer in +/+ CGCs was smaller than in –/– CGCs (P < 0.05). The frequency of mIPSCs increased in both the –/– and the +/+ groups at DIV11-12 (P < 0.01 for –/–; P < 0.05 for +/+) and as early as DIV8-9 for the +/+ group (Fig. 2D). The differences were only significant between groups at DIV8-9.
|
|
1 subunit in developmental changes of IPSCs. However, changes in IPSCs in cells from 1 –/– mice suggest additional and/or compensatory mechanisms underlying GABA channel changes in developing CGCs. Pharmacology of mIPSCs in developing CGS
We further assessed the involvement of other GABAA receptor subunits in developmental changes of mIPSCs in CGCs from +/+ and –/– mice using pharmacological tools. We first studied prolongation of current decay by the imidazopyridine, zolpidem (200 nM), shown to preferentially enhance Cl- currents at 1-subunit-containing receptor subtypes (Pritchett and Seeburg 1990; Sanna et al. 2002). We limited our analysis to mIPSC duration because effects on mIPSCs peak amplitude relate to the degree of occupancy of postsynaptic receptors (Perrais and Ropert 1999). Y tube application of zolpidem at DIV11 resulted in a significantly greater prolongation of the w in the +/+ group than in the –/– group (P < 0.01, Fig. 3, A and B ), confirming the developmental increase of 1 expression in wild-type CGCs (Fig. 3B).
|
6 subunit has been previously shown to be developmentally regulated in cerebellar cultures grown in 5 mM KCl (Mellor et al. 1998). Participation of the 6 subunit in inhibitory synaptic transmission has likewise been demonstrated (Mellor et al. 2000; Tia et al. 1996). To assess if 1 subunit deletion affects the expression of 6 subunit at inhibitory synapses, we studied inhibition of sIPSCs in the +/+ and the –/– groups by furosemide (100 µM), a diuretic selective antagonist of GABAA receptors containing 6 subunits (Korpi et al. 1995). Application of furosemide at DIV11 minimally affected decay kinetics in both groups and reduced peak current to a similar extent in the +/+ and the –/– groups (P > 0.39, Fig. 3, C and D). Compared with DIV8, the percent reduction of the peak sIPSC amplitude was significantly greater at DIV11 in both genotypes (Fig. 3D). These findings agree with previous reports of developmental increase of 6 subunit in inhibitory synapses to CGCs in vivo (Tia et al. 1996) and show, in addition, that such increase is not markedly affected by 1 subunit deletion.
We also investigated tonic current recorded in CGCs by measuring mean baseline current in the presence and absence of BMR, TTX, and BMR + TTX (Leao et al. 2000). The average BMR-sensitive current was very small (<7 pA) among all ages and genotypes tested and was not statistically different from the TTX-sensitive current (P > 0.29). Thus in primary CGC cultures, tonic GABAA receptor activation is minimal in contrast to observations in slices (Kaneda et al. 1995; Rossi and Hamann 1998; Wall and Usowicz 1997), likely due to effective local perfusion that facilitates GABA removal. We, therefore, directly activated GABAA receptor with THIP (Fig. 4A ), a superagonist at 43 receptors but with considerable potency and efficacy also for 6 subunit-containing receptors (Brown et al. 2002; Ebert et al. 1997). The response to THIP (at 2.5, 10, and 50 µM) in every CGC studied was compared with the maximal current elicited by 1 mM GABA. As shown in Fig. 4A, the ratio of THIP to maximal GABA did not increase with development in +/+ CGCs but was significantly larger in –/– CGCs, suggesting an increase in expression of 4- and/or 6-containing receptors. The maximal GABA current density was also compared between DIVs and genotypes. The results of these comparisons revealed that GABA current density significantly increased with age in +/+ CGCs (from 223 ± 32 pA/pF, n = 11 at DIV6, to 492 ± 91 pA/pF, n = 12 at DIV12, P < 0.01) but not in –/– CGCs (from 310 ± 68 pA/pF, n = 12 at DIV6, to 182 ± 30 pA/pF, n = 10 at DIV12). Consequently, at DIV12 maximal GABA current density was significantly larger for +/+ than for –/– CGCs (P < 0.01).
|
w in the +/+ and the –/– groups (Fig. 4B). This effect was not significantly different at DIV14 for either genotype. Given that the presence of 
2 and 6 subunits has opposite actions on benzodiazepine sensitivity (Mohler et al. 2002; Olsen and MacDonald 2002), our data suggest that a developmental increase in the expression of both subunits but in different receptor subtypes occurs at CGC's inhibitory synapses. Furthermore, no compensatory action in the expression of either subunit is produced by the 1 subunit deletion.
We then investigated two additional allosteric modulators with subunit selectivity (Fig. 4, C and D). We compared prolongation of mIPSC decay (w) by the 4 subunit selective modulator, bretazenil (Knoflach et al. 1996). Bretazenil (10 µM) failed to significantly prolong the w in either genotype at all ages (Fig. 4C). This suggests that the expression of 4 is minimal at inhibitory synapses in CGCs, it does not increase with development, and it does not change between genotypes. We also compared the prolongation of mIPSCs decay by the 2/3-selective drug, loreclezole (Fisher et al. 2000; Wafford et al. 1994). Bath perfusion with 10 µM loreclezole similarly prolonged the w in neurons of both genotypes at DIV7 (Fig. 4D). However, by DIV13 mIPSC prolongation of w by loreclezole was significantly greater in +/+ than in –/– CGCs. This result indicates that with development in culture, the expression of the 2/3 subunit increases or the expression of 1 subunit decreases in CGCs from wild-type but not 1 knock out mice.
Immunohistochemistry
Immuno-fluorescence staining was used to correlate the electrophysiological results with directly visualized pattern of expression of subunits at synaptic sites during development. Cultures of CGCs were immunostained using subunit specific antibodies and antibodies to glutamic acid decarboxylase (GAD)—a marker of GABAergic synapses. As illustrated in Fig. 5, CGCs, often clustered in small groups, form extensive neuritic networks. 1 subunit staining of CGCs at DIV6-7 and -11-12 from +/+ mice revealed a strong increase of 1 immunoreactivity. Such staining was absent in CGCs from 1 –/– mice. In +/+ cultures, a small population of larger neurons, possibly a subtype of GABAergic interneurons, displayed intense staining for the 1 subunits at all ages (not shown). Quantification of 1 subunit immunoreactivity at GABAergic synapses is described in the following text.
|
subunit clusters. Smaller clusters and more diffused immunofluorescence similar to that previously reported in CGC cultures grown in high potassium were also present (Gao and Fritschy 1995; Studler et al. 2002). As can be observed in the high-magnification micrographs in Fig. 6, 1 and 2 subunit clusters were most dense along the dendrites but were observed also on cell bodies. We speculated that the punctate expression of 1 and 2 subunit clusters along the dendrites and cell bodies might correspond to areas of synaptic contact between the cells. Indeed, double staining of CGCs with antibodies against 1 subunit and GAD confirmed that a number of GAD-positive puncta were opposed to a postsynaptic 1 cluster at DIV8 in +/+ CGCs (Fig. 6). Quantitative assessment of GAD/ subunit co-localization data supports these observations (see following text). Double staining with antibodies against 2 subunit and GAD showed that GAD-positive puncta were opposed to postsynaptic 2 subunit clusters in both +/+ and –/– CGCs (Fig. 6).
|
subunits in developmental changes of inhibitory synapses in CGCs, we extended our study to 3, 5, and 6 subunits in developing +/+ and –/– CGCs. Immunolabeling with the 5 subunit at all ages was very faint as compared with all other subunits tested and without well-defined clustering, and it was not investigated further. As seen in Fig. 7, double staining with antibodies against distinct subunits and GAD showed remarkable changes in punctate 1, 2, 3, and 6 subunit immunoreactivity with development. The density of GAD puncta per 16-µm dendritic length did not change with development, being on average 3.9 ± 0.4. The punctate expression of 3 and 2 subunits at GAD-positive sites was pronounced early in development. By DIV12, 2 subunit immunofluorescence became more diffuse, and well-defined clusters were preserved only at select few synapses (Fig. 7) with large differences in staining intensity from field to field (not shown). Similarly to 2, 3 subunit in wild-type CGC cultures appears to undergo developmental downregulation. However, the decrease in GAD colocalization was stronger for 3 than for 2 subunit, with more than threefold decrease of 3 expression at GAD-positive sites from DIV6 to -12 (Fig. 7, A and B). In contrast, GAD-positive puncta gradually became colocalized with 1 and 6 subunit clusters as development in vitro progressed. We quantified the percent of GAD puncta colocalized with each subunit's clusters along a set length of randomly selected dendrites (Fig. 7B). These results taken together indicate that 3 subunit first and 2 subunit later are progressively replaced by 1 and 6 subunits at inhibitory synaptic sites in developing CGCs, supporting the critical role of these subunits in developmental regulation of synaptic efficacy seen with electrophysiological recordings. In addition, the number of synapses facing 1, 2, 3, and 6 subunits was always >100%, indicating that some subunits must be colocalized at the same synaptic sites.
|
2 and 3 subunits but not for 6 subunit (Fig. 7B). Colocalization of GAD terminals with 2 or 3 subunit staining was not significantly decreased in –/– CGCs between DIV-6-7 and DIV-8-9. However, at DIV-11-12, GAD/3 subunit colocalization decreased in –/– CGCs, but GAD/2 subunit colocalization did not and it was significantly higher than in +/+ CGCs. (Fig. 7B). In parallel experiments, we investigated the expression of 2/3 as compared with 2 subunits in DIV12 CGCs from +/+ and –/– mice. The density of 2/3 subunits puncta per 16 µm dendritic length was higher for +/+ CGCs (7.0 ± 0.7) than for –/– CGCs (4.5 ± 0.8), reflecting the differing action of loreclezole on mIPSC decay between genotypes. The extent of cluster colocalization between 2/3 and 2 subunits was significantly smaller for +/+ (39 ± 10%) than –/– CGCs (74 ± 10%) as an expected consequence of the decrease of 2 subunit expression in +/+ but not in –/– CGCs. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
1subunit knock-out mice. The specific culture condition we used facilitates formation of functional inhibitory synapses allowing us to study them with whole cell voltage-clamp recordings and immunohistochemical staining. sIPSCs occur at high frequency as expected by the large number of GABAergic terminals seen with immunohistochemical methods and the high level of spontaneous activity of cerebellar GABAergic neurons (Farrant and Brickley 2003; Leao et al. 2000; Mellor et al. 2000). Our results indicate that the developmental changes of GABAA-mediated current kinetics and amplitude reported in cerebellar granule neurons in slices (Brickley et al. 1996; Tia et al. 1996; Vicini et al. 2001) can be observed in primary cultures of these neurons despite differences in synaptic connectivity. This provides a powerful model for investigation of the mechanisms underlying changes of inhibitory synaptic efficacy. An overview of the correlated immunohistochemical and electrophysiological findings is presented in Table 2.
|
), sIPSC and mIPSC decay is best fit by three exponential components. The fastest component of GABA responses can be seen with rapid agonist applications to excised patches but is generally lost when synaptic currents are small or distorted by dendritic filtering. The relatively large currents and high-quality voltage clamp that can be achieved with CGCs (Silver et al. 1992) allows effective resolution of all three components of the synaptic decay. If an exponential component were related to the presence of receptor subtypes with distinct subunit composition, this may have resulted in its loss with development. Because the triple-exponential decay of mIPSCs was preserved during development, our data suggest rather that multiple decay components are characteristic of each of the distinct receptor subtypes as seen in responses obtained with ultra-rapid GABA applications from distinct recombinant receptors (Barberis et al. 2002; Boileau et al. 2003).
Supplementing the findings with cerebellar slices (Vicini et al. 2001), we observed that 1 deletion greatly attenuated the developmental acceleration of sIPSCs decay in cultures but did not completely prevent it. This suggests a role for additional factors in normal development or involvement of compensatory mechanisms in the 1 –/– cultures. To perform a more reliable assessment of the developmental changes at inhibitory synapses containing and lacking the 1 subunit, we focused on mIPSCs. The results of our analyses revealed that, in parallel to the decrease of the decay time of mIPSCs, there are also changes in mIPSCs amplitude and frequency of occurrence. A significant decrease of mIPSC amplitude was observed first in CGCs cultured from the –/– mice and only several days later in CGCs cultured from the +/+ animals. The weighted time constant declined, albeit to a different extent, in both the –/– and the +/+ groups. Consequently, the synaptic charge transfer dropped by a half in both the wild-type and the knock-out CGCs over a period of 2 days, suggesting that developmental changes in decay time and amplitude are related and might act to compensate each other. It is tempting to speculate that such compensatory action might underlie the lack of behavioral alterations in 1 –/– mice (Kralic et al. 2002; Sur et al. 2001), although differences in charge transfer between genotypes were observed later in development. An additional consideration when evaluating developmental changes of mIPSCs amplitude are possible changes in the expression of potassium/chloride cotransporters (Payne et al. 2003; Staley and Smith 2001). These changes would not be apparent in our experimental conditions due to symmetrical chloride, yet they may have an important physiological role concomitant to the changes in GABAA receptor subunit expression.
The probability of GABA release is also important for inhibitory synaptic efficacy and may change both with development and as a compensatory effect of the 1 subunit deletion. Developmental increase of mIPSC frequency was not correlated with an increase in inhibitory synaptic terminal density as seen with GAD staining. This indicates, most likely, that changes in the probability of release take place with development. Indeed, Studler et al. (2002) reported a significant enlargement of GABAergic terminals in developing CGC cultures. The delayed increase of mIPSCs frequency in 1 subunit –/– CGCs could be due to a smaller number of postsynaptic sites. The immunostaining data, however, show compensatory changes at postsynaptic sites with 2 and 3 subunits (see following text). Therefore the lower mIPSCs frequency seen in –/– CGCs is likely due to a compensatory alteration in release probability.
In previous immunocytochemical studies in CGCs cultures, GABAA receptor subunit clusters were described for , , and subunits (Balduzzi et al. 2002; Gao and Fritschy 1995; Studler et al. 2002). However, although individual well-defined clusters were sometime identified, staining was mostly dispersed with microclusters seen all over cell body and dendrites. In contrast, we observed well-defined clusters similar to those reported in hippocampal neurons in culture (Brunig et al. 2002; Christie et al. 2002; Levi et al. 2004). It is likely that the reason for this difference lies in the formation of functional synapses as seen with electrophysiological recordings. In addition, the use of insulin in the culture medium facilitates surface expression of GABAA receptor subunits (Wan et al. 1997).
Inhibition of sIPSCs by furosemide increased with development but was comparable between cells in the –/– and +/+ groups at all ages tested. Immunostaining with 6 subunit specific antibodies support the increase of 6 subunit expression at synapses. These data are consistent with the results in cerebellar slices suggesting that 6 subunits increasingly participate in inhibitory synaptic transmission in developing GCGs (Tia et al. 1996). We were not able to detect a significant staining with antibodies against the 5 subunit in contrast to the reported presence of this subunit in hippocampal cultures with the same antibody (Brunig et al. 2002). This result is consistent with the lack of 5 subunits mRNA in cerebellum (Laurie et al. 1992a,b).
The 4 subunit is mostly expressed in granule neurons of the hippocampus dentate gyrus, although there are reports indicating its presence in CGCs (Mascia et al. 2002; Poltl et al. 2003). Although we did not directly assess the expression of 4 subunit with immunocytochemical studies, we investigated its presence pharmacologically. The 4-selective allosteric modulator, bretazenil, (Knoflach et al. 1996) failed to change mIPSCs across ages or genotypes, indicating that 4 subunit is not expressed at synaptic sites. 4 and 6 subunits in combination with subunit are responsible for tonic activation of perisynaptic and extrasynaptic receptor in hippocampal and cerebellar granule neurons, respectively (Stell et al. 2003; Sundstrom-Poromaa et al. 2002; Wei et al. 2003). The whole cell current elicited by THIP, a superagonist at 43 receptors, although with considerable potency and efficacy for containing receptors (Brown et al. 2002; Ebert et al. 1997), increased with development and was significantly higher in –/– CGCs. This suggests an increased expression of either 42 or 62 extrasynaptic receptors in the knock-outs.
Additional regulation of the efficacy of GABAergic synapses in CGCs with development may also be contributed by the isoforms of or subunits. A similar action of flunitrazepam on mIPSCs at different DIV suggests that an increase of 2 subunit may occur in parallel with the observed increase in benzodiazepine-insensitive 6 subunit at inhibitory synapses. In addition, greater prolongation of mIPSCs with development of +/+ but not –/– CGCs by loreclezole demonstrates a role for 2/3 subunits. Because the sensitivity to loreclezole of the 2/3 subunit is similar to that of 1 subunit when coexpressed with 2 subunits (Wafford et al. 1994), our results indicate that the expression of the 2/3 subunits increased with development in wild-type but not in knock-out CGS. This is further supported by the greater density of 2/3 subunit puncta in +/+ than in –/– CGCs. Studies of mIPSCs in neurons from 3 –/– mice, suggested that l-containing receptors were selectively retained by co-assembled 2 subunit (Ramadan et al. 2003). In addition, anatomical and biochemical data show preferential association of 2 with 3 subunit and of 1 with 2 subunit (Benke et al. 1994; Bollan et al. 2003; Fritschy et al. 1994). CGCs from 1 –/– mice might thus be expected to have lower expression levels of functional 2 subunit-containing receptors.
In summary, the sequential expression of 3 followed by 2 and later by 1 subunits is the most likely regulator of inhibitory synaptic efficacy with development, although additional subunits may also be involved. Our results supplement previous findings in cerebellar slices and elucidate the role of distinct subunits in regulation of functional efficacy of GABAergic networks.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
1 –/– mice, Dr. Jean Marc Fritschy, University of Zurich, and Dr. Angel DeBlas, University of Connecticut, Storrs, for the gift of the various subunit antibodies and to Drs. Stephen Logan and Andrea Barberis for helpful discussions. |
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: S. Vicini, Dept. of Physiology and Biophysics, BSB225, Georgetown University School of Medicine, 3900 Reservoir Rd., Washington, DC 20007 (E-mail: svicin01{at}georgetown.edu).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Barberis A, Petrini EM, Cherubini E, and Mozrzymas JW. Allosteric interaction of zinc with recombinant 122 and 12 GABAA receptors. Neuropharmacology 43: 607–618, 2002.[CrossRef][ISI][Medline]
Benke D, Fritschy JM, Trzeciak A, Bannwarth W, and Mohler H. Distribution, prevalence, and drug binding profile of GABAA receptor subtypes differing in the subunit variant. J Biol Chem 269: 27100–27107, 1994.
Boileau AJ, Li T, Benkwitz C, Czajkowski C, and Pearce RA. Effects of 2S subunit incorporation on GABAA receptor macroscopic kinetics. Neuropharmacology 44: 1003–1012, 2003.[CrossRef][ISI][Medline]
Bollan K, King D, Robertson LA, Brown K, Taylor PM, Moss SJ, and Connolly CN. GABAA receptor composition is determined by distinct assembly signals within and Subunits. J Biol Chem 278: 4747–4755, 2003.
Brickley SG, Cull-Candy SG, and Farrant M. Development of a tonic form of synaptic inhibition in rat cerebellar granule cells resulting from persistent activation of GABAA receptors. J Physiol 497: 753–759, 1996.
Brickley SG, Revilla V, Cull-Candy SG, Wisden W, and Farrant M. Adaptive regulation of neuronal excitability by a voltage-independent potassium conductance. Nature 409: 88–92, 2001.[CrossRef][Medline]
Brown N, Kerby J, Bonnert TP, Whiting PJ, and Wafford KA. Pharmacological characterization of a novel cell line expressing human 43 GABAA receptors. Br J Pharmacol 136: 965–974, 2002.[CrossRef][ISI][Medline]
Brunig I, Scotti E, Sidler C, and Fritschy JM. Intact sorting, targeting, and clustering of GABAA receptor subtypes in hippocampal neurons in vitro. J Comp Neurol 443: 43–55, 2002.[CrossRef][ISI][Medline]
Chen L, Chetkovich DM, Petralia RS, Sweeney NT, Kawasaki Y, Wenthold RJ, Bredt DS, and Nicoll RA. Stargazing regulates synaptic targeting of AMPA receptors by two distinct mechanisms. Nature 408: 936–943, 2000.[CrossRef][Medline]
Christie SB, Miralles CP, and De Blas AL GAB Aergic innervation organizes synaptic and extrasynaptic GABAA receptor clustering in cultured hippocampal neurons. J Neurosci 22: 684–697, 2002.
Ebert B, Thompson SA, Saounatsou K, McKernan R, Krogsgaard-Larsen P, and Wafford KA. Differences in agonist/antagonist binding affinity and receptor transduction using recombinant human GABAA receptors. Mol Pharmacol 52: 1150–1156, 1997.
Farrant M and Brickley SG. Properties of GABAA receptor-mediated transmission at newly formed Golgi-granule cell synapses in the cerebellum. Neuropharmacology 44: 181–189, 2003.[CrossRef][ISI][Medline]
Farrant M, Wisden W, Cull-Candy S, and Brickley SG. Absence of tonic GABAA mediated conductance in cerebellar granule cells of 6 –/– mice. Soc Neurosci Abstr 25: 246, 1999.
Fisher JL, Hinkle DJ, and Macdonald RL. Loreclezole inhibition of recombinant 112L GABAA receptor single channel currents. Neuropharmacology 39: 235–245, 2000.[CrossRef][ISI][Medline]
Fritschy JM, Paysan J, and Mohler H. Switch in the expression of rat GABAA receptor subtypes during postnatal development: an immunohistochemical study. J Neurosci 9: 5302–5324, 1994.
Gallo V, Kingsbury A, Balazs R, and Jorgensen OS. The role of depolarization in the survival and differentiation of cerebellar granule cells in culture. J Neurosci 7: 2203–2213, 1987.[Abstract]
Gao B and Fritschy JM. Cerebellar granule cells in vitro recapitulate the in vivo pattern of GABAA-receptor subunit expression. Brain Res Dev Brain Res 88: 1–16, 1995.[CrossRef][Medline]
Gingrich KJ, Roberts WA, and Kass RS. Dependence of the GABAA receptor gating kinetics on the -subunit isoform: implications for structure-function relations and synaptic transmission. J Physiol 489: 529–543, 1995.
Hamann M, Rossi DJ, and Attwell D. Tonic and spillover inhibition of granule cells control information flow through cerebellar cortex. Neuron 33: 625–633, 2002.[CrossRef][ISI][Medline]
) and 
) mice (*P < 0.01 vs. +/+; +P < 0.01 vs. DIV6-7 within group). DIV6-7, -8-9, and -11-12 values derive from 
), –/– CGCs at DIV12 (
). B–D: bar graphs summarizing the average percent increases of mIPSC decay (%
) or 
) subunits (% synapses) in CGCs from +/+ (top) and –/– (bottom) mice at 3 developmental age groups. *, P < 0.01 vs. 
, P < 0.05 vs. wild-type; +, P < 0.01 vs. DIV6-7.