Clustered and non-clustered GABAA receptors in cultured hippocampal neurons

www.elsevier.com/locate/ymcne

Mol. Cell. Neurosci. 31 (2006) 1 – 14

Clustered and non-clustered GABAA receptors in cultured

hippocampal neurons

S.B. Christie, R.-W. Li, C.P. Miralles, B.-Y. Yang, and A.L. De Blas*

Department of Physiology and Neurobiology, University of Connecticut, 3107 Horsebarn Hill Rd., U-4156, Storrs, CT 06269, USA

Received 7 July 2005; revised 17 August 2005; accepted 23 August 2005

Available online 21 September 2005

In cultured hippocampal neurons, ;2 subunit-containing GABAARs

form large postsynaptic clusters at GABAergic synapses and small

clusters outside GABAergic synapses. We now show that a pool of non-

clustered ;2 subunit-containing GABAARs are also present at the cell

surface. We also demonstrate that myc- or EGFP-tagged ;2, A2, B3 or

A1 subunits expressed in these neurons assemble with endogenous

subunits, forming GABAARs that target large postsynaptic clusters,

small clusters outside GABAergic synapses or a pool of non-clustered

surface GABAARs. In contrast, myc- or EGFP-tagged D subunits only

form non-clustered GABAARs, which can be induced to form clusters

by antibody capping. A myc-tagged chimeric ;2 subunit possessing the

large intracellular loop (IL) of the y-subunit IL (myc;2S/D-IL)

assembled into GABAARs, but it did not form clusters, therefore

behaving like the D subunit. Thus, the large intracellular loops of ;2

and D play an important role in determining the synaptic clustering/

non-clustering capacity of the GABAARs.

D 2005 Elsevier Inc. All rights reserved.

Introduction

The g2 GABAA receptor (GABAAR) subunit knockout and

conditional knockout mouse models demonstrate that the g2subunit is essential for the postsynaptic clustering of GABAARs

(Gunther et al., 1995; Essrich et al., 1998; Schweizer et al., 2003).

This interpretation is also supported by the results obtained after

knocking down the g2 subunit by RNA interference (Li et al., in

press). The g2 knockout mice show a loss of most postsynaptically

clustered GABAARs accompanied by severe deficits in GABAer-

gic synaptic transmission, sensorimotor deficits and neonatal

lethality. In these animals, the expression of the GABAAR a and

h subunits was not affected, and they formed functional GABAARs

just containing ah subunits that were translocated to the surface

(Gunther et al., 1995; Essrich et al., 1998). Thus, the postsynaptic

accumulation of GABAARs apposed to GABA-releasing presy-

1044-7431/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.mcn.2005.08.014

* Corresponding author. Fax: +1 860 486 5439.

E-mail address: [email protected] (A.L. De Blas).

Available online on ScienceDirect (www.sciencedirect.com).

naptic terminals is thought to be central to the phasic inhibitory

GABAergic synaptic transmission in the brain.

The g2 and y subunits do not co-assemble within the same

GABAAR (Quirk et al., 1994, 1995; Araujo et al., 1998;

Jechlinger et al., 1998). Contrary to the postsynaptic clustering

of the g2 subunit-containing GABAARs, y subunit-containing

GABAARs are non-synaptic or perisynaptic as shown by

immunoelectron microscopy in granule cells of the cerebellum

(Nusser et al., 1998) and dentate gyrus of the hippocampus (Wei

et al., 2003). These extrasynaptic GABAAR receptors mediate the

tonic GABAergic inhibition by sensing the ambient levels and/or

spillover of GABA released from synapses (Brickley et al., 1996;

De Schutter, 2002; Nusser and Mody, 2002; Stell et al., 2003;

Yeung et al., 2003).

Gephyrin, a glycine receptor (GlyR) interacting protein

present in glycinergic and GABAergic synapses, has also been

suggested to be required for the clustering of GABAARs

(Essrich et al., 1998; Kneussel et al., 1999). However, not all

GABAARs require gephyrin for clustering as shown in a

gephyrin knockout mouse (Fischer et al., 2000; Kneussel et

al., 2001; Levi et al., 2004). There is no biochemical evidence

for a direct interaction of gephyrin with any of the GABAAR

subunits (Meyer et al., 1995). Nevertheless, gephyrin plays an

important role in normal synaptic transmission because deficits

in clustering of GABAARs and GABAergic synaptic trans-

mission are apparent in gephyrin�/� mice (Levi et al., 2004).

Furthermore, a missense mutation in collybistin, a GDP–GTP

exchange factor that is important for gephyrin clustering, leads

to impaired inhibitory synapse formation (Harvey et al., 2004).

There still exists an incomplete understanding of the key

molecular mechanisms by which g2 subunit-containing GABAARs

(g2-GABAARs) are selectively retained and/or concentrated at

GABAergic synapses, while y-GABAARs do not. However, it is

likely that an interaction of the GABAAR with scaffolding proteins

at the GABAergic synapses occurs via the IL of some of the

GABAAR subunits. Studies using the yeast two-hybrid system

have identified several proteins that interact with the IL domain of

GABAAR subunits (recently reviewed or discussed by Bedford et

al., 2001; Fritschy and Brunig, 2003; Herring et al., 2003; Charych

et al., 2004a; Keller et al., 2004; Luscher and Keller, 2004).

http://www.sciencedirect.com

S.B. Christie et al. / Mol. Cell. Neurosci. 31 (2006) 1–142

Nevertheless, the identified proteins seem to be involved in

GABAAR trafficking rather than in synaptic clustering.

In this communication, we have tested the hypothesis that the

IL domains of g2 and y GABAAR subunits play a role in the

synaptic clustering capacity of the GABAARs.

Results

Before studying the expression in hippocampal neurons shown

below, all the tagged subunits and the chimera used in this study

were coexpressed in HEK293 cells with non-tagged h2 and a6

subunits. Expression, assembly and translocation to the cell surface

were demonstrated by immunofluorescence after life cell incubation

with anti-myc or anti-EGFP antibodies. These experiments showed

myc or EGFP immunofluorescence at the cell surface of the HEK293

cells that colocalized with h2 and a6 cell surface immunofluor-

escence (as determined with anti-h2/3 and anti-a6 antibodies

respectively) in triple-label experiments. Live cell incubation with

the anti-EGFP antibody of the cotransfected HEK293 cells induced

the capping of the EGFP-tagged subunit (g2 or y) and the colocalizedcapping of the h2 and a6 subunits (not shown).

Exogenous mycc2 or c2-EGFP form clusters at GABAergic synapses

that colocalize with endogenous GABAAR clusters

This is shown in Fig. 1. Hippocampal neurons transfected

with only mycg2S DNA express the exogenous subunit, which

forms large clusters (arrows, Fig. 1A) localized postsynaptically

to GAD+ (GABAergic) terminals (arrows, Fig. 1C). The mycg2S

also forms small clusters that are present in dendritic areas not

receiving GABAergic innervation (arrowheads, Figs. 1A and C).

The larger (synaptic) and smaller clusters of the endogenous

GABAAR associated protein, gephyrin, were highly colocalized

with both large (arrows, Fig. 1B) and small (arrowheads, Fig.

1B) mycg2S clusters respectively. Moreover, the size of the largemycg2S clusters at GABAergic synapses and small clusters

outside GABAergic synapses was similar to that of the

endogenous GABAAR and gephyrin clusters in non-transfected

cells (Figs. 1D–F). These larger (synaptic) and smaller clusters

have been characterized in detail elsewhere (Christie et al.,

2002a).

Table 1 shows that there is no statistically significant

difference (P = 0.58, Student’s t test) between the density ofmycg2S clusters in transfected cells (22 T 2 clusters per 100 Am2

dendrite area; n = 193 clusters in 15 dendrites from 9 cells) and

the density of endogenous g2 clusters (20 T 1 clusters per 100

Am2 dendrite area; n = 386 clusters in 18 dendrites from 9

cells) in untransfected cells. Neither is there significant differ-

ence (P = 0.74) between the density of endogenous gephyrin

clusters in transfected cells (19 T 2 clusters/100 Am2; n = 175

in 15 dendrites from 9 cells) and the density of endogenous

gephyrin clusters in untransfected cells (19 T 1 clusters/100

Am2; n = 357 in 18 dendrites from 9 cells). In transfected

neurons (Table 1) 82 T 3% of mycg2S clusters were colocalized

with gephyrin, while 91 T 3% of gephyrin clusters were

colocalized with mycg2S clusters. These values were not

significantly different than the 85 T 2% colocalization of

endogenous g2 clusters with gephyrin clusters (P = 0.19) or

the 93 T 2% colocalization of gephyrin clusters with endoge-

nous g2 clusters (P = 0.32) observed in untransfected neurons.

Thus, we find that the exogenous expression of the mycg2S subunit

in the transfected neurons integrates with the endogenous

GABAAR assembly without significantly altering the targeting of

GABAARs, density of the clusters or their colocalization with

gephyrin.

Transfection of hippocampal neurons with only g2S-EGFP

(Figs. 1G–I) or g2L-EGFP (Figs. 1J–L) also formed large and

small clusters as shown by EGFP fluorescence (Figs. 1G and J)

and with an anti-g2 antibody (Fig. 1H). These clusters

colocalized with endogenous GABAAR subunits, as shown by

immunofluorescence with anti-h2/3 GABAAR subunit (arrows,

Figs. 1I and K), at GABAergic (vGAT+) synapses (arrows, Fig

1L). These experiments strongly suggest that the mycg2S, g2S-

EGFP or g2L-EGFP co-assemble with endogenous subunits to

form GABAARs. This notion is further supported because others

have shown that some myc- and EGFP-tagged GABAAR

subunits co-assemble with endogenous subunits forming func-

tional GABAARs (Connolly et al., 1996; Kittler et al., 2000;

Alldred et al., 2005). Moreover, g2 GABAAR subunits that do

not assemble with a or h GABAAR subunits are retained within

the ER, do not translocate to the cell surface and are rapidly

degraded (Connolly et al., 1999).

It is also worth mentioning that it has been proposed (Kittler

et al., 2000) that: (I) a g2S when tagged at the N-terminus with

EGFP (EGFP-g2S), but not when tagged at the C-terminus

(g2S-EGFP), can form clusters that colocalize with other

GABAAR subunits, (II) exogenous a and h subunits need to

be cotransfected together with the EGFP-g2S for the latter to

assemble into GABAARs and to form clusters. Our results

instead show that (I) the g2S and g2L GABAAR subunits

tagged at the C-terminus with EGFP can also form synaptic

GABAAR clusters and (II) cotransfection of exogenous a and hsubunits is not required since g2S-EGFP and g2L-EGFP can co-

assemble with endogenous GABAAR subunits and form clusters

of similar density and size to the endogenous GABAAR

clusters, probably because the endogenous subunits are rate-

limiting for receptor assembly.

Various GABAAR subunits tagged at the N-terminus or the

C-terminus form clusters that colocalize with endogenous GABAAR

clusters

We also tagged other GABAAR subunits with EGFP (a1-EGFP,

a2-EGFP and h3-EGFP) and made a double-tagged mycg2S-EGFP

subunit. Hippocampal neurons transfected with each of these

tagged subunits displayed large EGFP clusters (arrows, Figs. 2A, D

and G, respectively) that were colocalized with endogenous

gephyrin clusters (arrows, Figs. 2B, E and H) in areas postsynaptic

to GAD+ boutons (arrows, Figs. 2C, F and I). Similar results were

obtained using an a2-EGFP made by us (not shown) or another a1-

EGFP construct produced in Dr. Stefano Vicini’s laboratory (not

shown). Smaller clusters of EGFP-tagged subunit that colocalized

with smaller gephyrin clusters were observed in dendritic areas not

receiving GABAergic innervation (arrowheads, Fig. 2).

These experiments strongly suggest that the tagged a1, a2 or h3

subunits co-assemble with the endogenous subunit partners (e.g.

exogenous-tagged a assembles with endogenous h and g subunits

or exogenous-tagged h assembles with endogenous a or g

subunits) since only GABAAR pentamers that contain the g

subunit (plus have a and h) target to GABAergic synapses (Essrichet al., 1998; Schweizer et al., 2003; Li et al., in press).

Fig. 1. A similar pattern of GABAAR cluster organization is seen for the endogenous g2 and exogenous myc-tagged g2 subunits expressed in cultured

hippocampal neurons. Triple-label immunofluorescence. Neurons transfected with mycg2S (A–C), untransfected (D–F) or transfected with g2S-EGFP (G–I) or

g2L-EGFP (J–L) were fixed with paraformaldehyde, permeabilized and incubated with rabbit anti-myc (A) or rabbit anti-GABAAR g2 subunit (D, and H),

mouse anti-gephyrin (B and E), sheep anti-GAD (C and F), guinea pig anti-vGAT (L) or mouse anti-h2/3 subunit (I and K). Large clusters of exogenousmycg2S

(arrows A) or endogenous g2 (arrows, D) colocalized with large clusters of gephyrin (arrows, B and E, respectively) that were apposed to GAD+ terminals

(arrows, C and F, respectively). Additionally, there were small clusters of exogenous mycg2S subunit (arrowheads, A) or endogenous g2GABAAR (arrowheads,

D) that colocalized with small clusters of gephyrin (arrowheads, B and E, respectively) in areas not receiving GABAergic innervation (arrowheads, C and F,

respectively). Neurons transfected with the g2S-EGFP or g2L-EGFP also displayed a clustered organization of EGFP fluorescence (arrows, G and J,

respectively) that was extensively colocalized with large clusters of GABAAR subunit immunoreactivity for the g2 (arrows, H) and/or endogenous h2/3 (arrows,

I and K) subunits or vGAT+ terminals (arrows, L). Scale bar in panel A is 5 Am.

S.B. Christie et al. / Mol. Cell. Neurosci. 31 (2006) 1–14 3

Antibody-induced capping of the c2 subunit in living hippocampal

cultures reveals the existence of a pool of non-clustered c2subunit-containing GABAARs at the cell surface

Figs. 3A–C show the cluster distribution of endogenous g2-

containing GABAARs (g2-GABAARs) on the cell surface of

untransfected neurons after incubating the living pyramidal

cells with an antibody recognizing the extracellular N-terminus

of the g2 subunit (rabbit anti-g2NH2) followed by fixation. We

call this procedure ‘‘one-step capping’’ to differentiate from

‘‘two-step capping’’, in which the living cells are incubated

sequentially with the primary and secondary antibodies before

fixation (see below). The one-step capping produced results

similar in many respects to the distribution of g2-GABAARs

observed in fixed cells (compare Figs. 3A–C with Figs. 1D–

F). Thus, we observed the colocalization of large clusters of

g2-GABAARs (arrows, Fig. 3A) and gephyrin (arrows, Fig.

3B) that were apposed to presynaptic GABAergic terminals

(arrows, Fig. 3C), as well as a high degree of colocalization

between the smaller g2-GABAAR clusters and gephyrin in

areas not innervated by GABAergic terminals (filled arrow-

heads, Figs. 3A–C).

Table 1

Antibody-induced capping of g2- andmycg2S-containing GABAARs in cultured hippocampal neurons

g2 ormycg2S clustersa,b Gephyrin clustersa,b g2 or

mycg2S clusters

colocalized with gephyrina,cGephyrin clusters colocalized

with g2 ormycg2S

a,d

No capping:

Endogenous g2 20 T 1 (n = 386) 19 T 1 (n = 355) 17 T 1 (85 T 2%) 17 T 1 (93 T 2%)mycg2S 22 T 2 (n = 193) 19 T 2 (n = 175) 17 T 2 (82 T 3%) 17 T 2 (91 T 3%)

One-step capping:

Endogenous g2 34 T 3e (n = 510) 24 T 2 (n = 365) 22 T 2e (65 T 3%) 22 T 2 (91 T 3%)mycg2S 35 T 2e (n = 339) 25 T 2f (n = 234) 22 T 2e (65 T 3%) 22 T 2 (90 T 3%)

Two-step capping:

Endogenous g2 76 T 7e,g (n = 1153) 23 T 3 (n = 349) 20 T 2e,g (26 T 2%) 20 T 2 (89 T 3%)mycg2S 74 T 7e,g (n = 990) 23 T 3 (n = 294) 20 T 3e,g (26 T 3%) 20 T 3 (88 T 3%)

a Values are mean T SEM of cluster density given as number of clusters per 100 Am2.b (n) is the number of clusters counted. Samples were collected from 6 to 10 neurons in 1–2 coverslips, 2–3 dendrites/neuron (15–18 dendrites) for each

immunolabeling condition and antibody used. The cluster density values correspond to all clusters regardless of the size.c Percent values represent g2 or

mycg2S clusters colocalizing with gephyrin clusters.d Percent values represent gephyrin clusters colocalizing with g2 or

mycg2S clusters.e Difference is significant compared to cells without antibody capping; P < 0.001, Student’s t test.f Difference is significant compared to cells without antibody capping; P < 0.05, Student’s t test.g Difference is significant compared to one-step antibody capping; P < 0.001, Student’s t test.


Similar results were obtained after one-step capping of

exogenous mycg2S-GABAARs in transfected hippocampal neu-

rons after incubating the living neurons with the anti-myc

antibody to the external myc epitope (Figs. 3D–F). The surface

labeling of the living cells indicates that the mycg2S-GABAARs

were properly assembled and trafficked to the surface. We

observed large clusters of surface mycg2S-GABAARs (arrows,

Fig. 3D) that colocalized with large clusters of endogenous

gephyrin (arrows, Fig. 3E) and GAD+ synaptic boutons (arrows,

Fig. 3F). Smaller clusters that colocalized with gephyrin in the

absence of GAD+ boutons were also present (filled arrowheads,

Figs. 3D–F). Thus, in many respects, the one-step capping of

transfected cells gave results similar to those obtained with

primary antibody labeling after fixation of the transfected cells

(compare Figs. 3D–F with Figs. 1A–C).

However, a noticeable difference in the results (see Table 1 for

quantification) was that transfected (Figs. 3D–F) or untransfected

(Figs. 3A–C) neurons subjected to one-step capping had a higher

density of clusters of g2-GABAARs ormycg2S-GABAARs (34–35

vs. 20–22 clusters/100 Am2) and that many more of these clusters

had no gephyrin (empty arrowheads; 12–13 vs. 3–4 clusters/100

Am2) when compared to transfected (Figs. 1A–C) or untransfected

(Figs. 1D–F) neurons in which the incubation with the primary

antibody was done after cell fixation. A two-step capping of

untransfected (Figs. 3G–I) or transfected neurons (Figs. 3J–L)

produced an even larger increase in the cluster density of g2-

GABAARs ormycg2S-GABAARs when compared to untransfected

or transfected neurons in which the incubation with the primary

antibody was done after cell fixation (74–76 vs. 20–22 clusters/

100 Am2). Many of these new clusters did not have gephyrin (54–

56 after capping vs. 3–5 clusters/100 Am2 before capping, Table

1). Others have shown that the two-step capping efficiently

induced clustering of GABAARs (Levi et al., 2004).

The one-step and two-step GABAAR capping only marginally

increased the density of gephyrin clusters over primary antibody

incubation after fixation (23–25 vs. 19 clusters/100 Am2, Table 1)

compared to the large increase in g2-GABAARs or mycg2S-

GABAARs clusters (74–76 vs. 20–22 clusters/100 Am2, Table

1). This resulted in a decreased colocalization of g2-GABAARs ormycg2S-GABAARs clusters with gephyrin, from fixed cells (82–

85%) to one-step capping (65%) to two-step capping (26%), while

there was little effect on the proportion of gephyrin clusters that

colocalized with g2-GABAARs ormycg2S-GABAARs clusters (91–

93% vs. 90–91% vs. 88–89% respectively). Thus, the results

indicate that: (I) antibody-induced capping in living cells induces

the rapid formation de novo of small clusters from a laterally

diffusible pool of non-clustered g2-GABAARs or mycg2S-

GABAARs that are present at the surface of neurons and (II) the

majority of the new GABAAR clusters induced by antibody capping

had no gephyrin associated to them. Thus, gephyrin does not co-

cluster simultaneously with the GABAARs. Instead, there is a delay

in gephyrin co-clustering with GABAARs (see Discussion).

The exogenous expression of mycd or d-EGFP subunits in cultured

hippocampal neurons produces non-clustered and non-synaptic

GABAARs

The y-GABAARs are non-synaptic and are diffusely distributed

on the neuronal surface (Nusser et al., 1998; Jechlinger et al., 1998).

Therefore, the endogenous y-GABAARs are very difficult to

visualize in culture, even in neurons that express them in relatively

large quantities, because they do not form synaptic or non-synaptic

clusters. Moreover, in the hippocampal cultures, the y subunit is

expressed at very low levels (Killisch et al., 1991). This is

consistent with the low level of y subunit expression observed in

CA1 and CA3 regions of the hippocampus in the intact brain

(Persohn et al., 1992; Wisden et al., 1992; Sperk et al., 1997; Pirker

et al., 2000). Therefore, we have examined the synaptic targeting of

the exogenous myc and EGFP-tagged y subunit in these cultures.

The expression of the exogenous mycy subunit in transfected

hippocampal neurons was confirmed by showing robust immuno-

labeling with an anti-y-IL antibody (Fig. 4A). The mycy immuno-

labeling was diffusely distributed throughout the soma and

dendrites but failed to cluster or colocalize specifically with

Fig. 2. Various exogenous GABAAR-EGFP subunits form GABAAR clusters that colocalize with gephyrin clusters at GABAergic synapses. Triple-label

immunofluorescence. Hippocampal neurons transfected with either a1-EGFP (A), h3-EGFP (D) or mycg2S-EGFP (G) were fixed, permeabilized and labeled

with mouse anti-gephyrin (B, E and H) and sheep anti-GAD (C, F and I). For all EGFP fusion subunits examined, the larger clusters of EGFP fluorescence

(arrows, A, D and G) colocalized with the larger clusters of gephyrin (arrows, B, E and H) that were apposed to GAD+ synapses (arrows, C, F and I). In

addition, extensive colocalization was observed between the small EGFP clusters (arrowheads, A, D and G) and gephyrin (arrowheads, B, E and H) in the

absence of GABAergic terminals (arrowheads, C, F and I). Scale bar in panel A is 5 Am.


gephyrin clusters or GAD at GABAergic synapses (arrows, Figs.

4B and C, respectively) or with any of the small gephyrin clusters

outside of GABAergic synapses (arrowheads, Fig. 4B). Similar

subcellular distribution of anti-y-IL immunolabeling was observed

in fixed neurons that had been transfected with either an untagged

y subunit (not shown) or a novel y-EGFP subunit generated in our

laboratory (Figs. 4D–F).

In the cerebellar granule cells, the a6 subunit co-assembles with

the y subunit (and h) and forms extrasynaptic y-GABAARs or with

the g2 subunit (and h) and forms synaptic g2-GABAARs (Nusser et

al., 1998; Jechlinger et al., 1998). When we transfected hippo-

campal neurons with a plasmid encoding the exogenous non-tagged

a6 subunit (a subunit not expressed in the hippocampus or in these

cultures), they produced a6-GABAAR clusters that colocalized with

gephyrin (arrows, Figs. 5A and B, respectively) at GABAergic

GAD+ synapses (not shown). The observed clustering and synaptic

targeting were presumably mediated by the assembly of exogenous

a6 with endogenous g2 and h subunits since the synaptic targeting

of the a subunits is severely impaired unless it co-assembles with g2(Gunther et al., 1995; Essrich et al., 1998; Li et al., in press).

When we cotransfected hippocampal neurons with a combina-

tion of y-EGFP and a6, we found a diffuse distribution of the y-EGFP fluorescence (Fig. 5C), similar to that observed in cells

expressing only the y-EGFP (Fig. 4D), thereby discounting a role

for a6 in significantly recruiting y-EGFP to postsynaptic clusters

since there was no colocalization with gephyrin clusters (Fig. 5D).

These results are consistent with the notion that in the brain a6 co-

assembles with either g2 or y (or y-EGFP) but not with both

subunits in the same receptor pentamer (Quirk et al., 1994, 1995;

Araujo et al., 1998; Jechlinger et al., 1998).

Further evidence for the non-synaptic distribution of the y-EGFP containing GABAARs that are present at the cell surface was

obtained by antibody-induced capping. One-step capping of living

neurons with polyclonal anti-GFP antibody revealed numerous

small and disorganized EGFP clusters that decorate the processes

(arrowheads, Fig. 5E) of cells transfected with the y-EGFP subunit.

These small EGFP clusters are not readily observed in areas of

contact by GAD+ terminals (arrows, Fig. 5F).

In order to confirm that in cotransfected neurons the exogenous

a6 co-assembled with y-EGFP to form surface expressed but non-

clustered GABAARs, we performed a two-step capping of living

neurons using a polyclonal anti-GFP and a secondary antibody

(arrowheads, Fig. 5G). We observed that the y-EGFP clusters

induced by two-step capping frequently colocalized with a subset of

small a6-GABAAR clusters (arrowheads, Figs. 5G and H,

respectively). However, none of the induced y-EGFP clusters

colocalized with the larger and synaptic clusters of a6 (arrows, Figs.

5G and H, respectively), which are postsynaptic to GAD+ terminals

Fig. 3. Antibody-induced capping of endogenous g2- ormycg2-GABAARs that are present on the cell surface. Triple-label immunofluorescence. One-step

antibody capping (A–F) of live neurons with an affinity-purified rabbit antibody to the N-terminus of the g2 (A) or a rabbit anti-myc (D) induced the formation

of GABAAR clusters of g2-GABAARs in untransfected cells (A) or mycg2S-GABAARs in transfected cells (D). Two-step antibody-induced capping (G–L)

induced the formation of additional clusters of endogenous g2-GABAARs in untransfected cells (G) ormycg2S-GABAARs in transfected cells (J). After capping,

cells were fixed, permeabilized and incubated with a mixture of mouse anti-gephyrin (B, E, H and K) and sheep anti-GAD (C, F, I and L) followed by a mixture

of the secondary antibodies. Larger clusters of the endogenous g2-GABAARs (arrows, A and G) and the exogenous mycg2S-GABAARs (arrows, D and J)

colocalized with large gephyrin clusters (arrows, B, E, H and K) at GAD+ synapses (arrows, C, F, I and L). Smaller clusters of the endogenous g2-GABAARs

(filled arrowheads, A and G) and the exogenous mycg2S-GABAARs (filled arrowheads, D and J) were also colocalized with gephyrin (filled arrowheads, B, E,

H and K) in areas not innervated by GABAergic terminals (filled arrowheads, C, F, I and L). However, aggregates of endogenous g2-GABAARs (empty

arrowheads, A and G) or exogenous mycg2S-GABAARs (empty arrowheads, D and J) that did not colocalize with gephyrin (empty arrowheads, B, E, H and K)

or GAD+ terminals (empty arrowheads, C, F, I and L) were abundant throughout all processes. These clusters were greater in number after two-step capping

than one-step capping. Scale bar in panel A is 5 Am.


(not shown). These and the aforementioned results indicate that a6

co-assembled with either the endogenous g2 or exogenous y-EGFPsubunits (but not with both), forming GABAARs that were

differentially targeted to synaptic or non-synaptic areas, respec-

tively. This is consistent with the behavior of endogenous a6 in

cerebellar granule cells, where it assembles with either y or g2(Quirk et al., 1995; Jechlinger et al., 1998) to form extrasynaptic or

synaptic receptors, respectively, in the same cell (Nusser et al.,

1998).

The large intracellular loop of the c2 and d subunits plays a role in

the synaptic clustering of the GABAA receptors

Studies with other receptors of the same family have shown that

the large IL domain of the GlyR h subunit (Kirsch et al., 1995) and

the nAChR a3 subunit (Williams et al., 1998) is critical to the

synaptic targeting and clustering of these receptors. We have now

tested the hypothesis that the IL of the g2 and y subunits plays an

important role in determining the clustering of the GABAARs. For

Fig. 4. The mycy or y-EGFP GABAARs do not form clusters or target to synapses in cultured hippocampal neurons. Triple-label immunofluorescence. Neurons

transfected with mycy (A–C) or y-EGFP (D–F) were fixed, permeabilized and incubated with rabbit anti-y (A), mouse anti-gephyrin (B and E) and sheep anti-

GAD (C and F) or by EGFP fluorescence (D). The mycy or y-EGFP GABAARs did not form clusters (A and D) and failed to colocalize with gephyrin clusters at

GABAergic synapses (arrows, A–F) or outside GABAergic synapses (arrowheads, A–F). Instead, diffuse immunolabeling was present that concentrated in the

perikaryon and dendrites. Scale bar in panel A is 10 Am.


this purpose, we have constructed a myc-tagged chimeric subunit in

which the IL domain of the y subunit has been substituted for the g2-IL of the mycg2S to produce a mycg2S/y-IL chimera.

The anti-y-IL antibody immunolabeling (Fig. 6A) of fixed

neurons that had been cotransfected with the mycg2S/y-IL chimeric

subunit and EGFP (used as a marker of transfection, Fig. 6C)

produced diffuse staining with no indication of specific colocaliza-

tion with gephyrin clusters (Fig. 6B). A similar diffuse distribution

of the mycg2S/y-IL chimeric subunit was also observed with the anti-

myc antibody (not shown). Thus, the diffuse distribution of

immunolabeling for the chimeric subunit was similar to that ofmycyor y-EGFP (Fig. 4) rather than to the clustered distribution of

themycg2S or g2S-EGFP (Fig 1). The mycg2S/y-IL chimeric subunit

was present at the cell surface as shown by the numerous small

clusters distributed evenly over the surface of transfected neurons

induced by two-step capping with the anti-myc antibody (Fig. 6D).

The chimeric GABAAR clusters induced by two-step capping did

not colocalize with gephyrin clusters (arrows, Fig. 6E) or GAD+

GABAergic synapses (arrows, Fig. 6F). Only 3 T 1% (mean T SEM)

of all the mycg2S/y-IL-induced clusters colocalized with endogenousgephyrin clusters. This value was not significantly different than the

5 T 1% random colocalization observed when the mycg2S/y-ILchannel was rotated 180- and superimposed on the gephyrin

channel for quantitation (P > 0.05). These results show that (I) the

chimeric mycg2S/y-IL subunit is transported to the cell surface,

indicating that co-assembly with endogenous subunits has occurred,

as discussed above; (II) the clusters induced by the two-step

antibody capping were derived from a pool of laterally diffusible

surface mycg2S/y-IL-containing GABAARs; and (III) the clusters

induced by the two-step antibody capping of the mycg2S/y-IL-containing GABAARs are excluded from GABAergic synapses.

See below also.

We next cotransfected neurons with a combination of themycg2S/y-IL, a6 and h3-EGFP subunits to induce co-assembly of the

exogenous subunits and produce GABAARs that could be simulta-

neously monitored by anti-myc immunolabeling and EGFP

fluorescence (Figs. 6G–I). In these experiments, anti-myc immu-

nolabeling and h3-EGFP fluorescence were used to identify

transfected neurons. Two-step antibody capping of mycg2S/y-IL-GABAARs (using the anti-myc antibody) induced the formation ofmycg2S/y-IL-GABAAR clusters (Fig. 6G) and h3-EGFP clusters

(Fig. 6H) that colocalized with each other (filled arrowheads). This

indicates that the chimeric mycg2S/y-IL and h3-EGFP subunits co-

assemble into the same receptor. However, the mycg2S/y-IL clusters

induced by the two-step capping were absent from the GAD+

GABAergic synapses (arrows, Figs. 6G–I). The h3-EGFP clusters

were found both postsynaptic to GABAergic terminals (arrows

Figs. 6H and I) and outside of GABAergic synapses (arrowheads,

Figs. 6H and I), many of which colocalizing with mycg2S/y-ILclusters (filled arrowheads, Fig. 6G). Furthermore, there were a

number of small h3-EGFP clusters outside GABAergic synapses

that were not colocalized with clusters of mycg2S/y-IL GABAARs

(empty arrowheads, Figs. 6G and H). These results indicate that the

h3-EGFP is able to assemble (I) with the endogenous g2 and other

subunits into GABAARs that form clusters at GABAergic synapses

and outside GABAergic synapses, as we have also shown in Figs.

2D–F, and (II) with the exogenous mycg2S/y-IL and other subunits

forming non-clustered GABAARs at the cell surface that do not

target GABAergic synapses.

Thus, the mycg2S/y-IL, like the y subunit, shows mutual

exclusivity with the g2 subunit (Quirk et al., 1994, 1995; Araujo

et al., 1998; Jechlinger et al., 1998), and, like the y subunit, it

competes with g2 for assembly with the a and h subunits (Korpi et

al., 2002; Peng et al., 2002).

Discussion

Like the endogenous g2 subunit, the exogenous mycg2S,mycg2L, g2S-EGFP, mycg2S-EGFP or g2L-EGFP subunits

Fig. 5. Exogenous a6 subunit assembles into GABAARs that form clusters or co-assembles with exogenous y-EGFP into GABAARs that do not form clusters.

Double-label immunofluorescence. Hippocampal live neurons transfected with only a6 were surface-labeled with a rabbit anti-a6 antibody (A) prior to being

fixed, permeabilized and immunolabeled with anti-gephyrin (B) followed by incubation with fluorescently labeled secondary antibodies. Contrary to

untransfected hippocampal neurons, which did not express the a6 nor did they show a6 clusters, the transfected neurons showed a high degree of colocalization

between clusters of exogenous a6 (arrows, A) and endogenous gephyrin (arrows, B). Cotransfection of y-EGFP with a6 subunit failed to recruit y-EGFPfluorescence (C) to the gephyrin clusters (arrows, D) apposed to GABAergic terminals (not shown) or to the smaller gephyrin clusters (arrowheads, D) that

formed outside of GABAergic synapses. In cells transfected with only y-EGFP, one-step antibody-induced capping of surface y-EGFP GABAARs (E) induced

clusters that were exclusively extrasynaptic (arrowheads, E) and that were absent from areas contacted by GAD+ terminals (arrows, H). Two-step antibody

induced capping of the y-EGFP (G) in cells cotransfected with y-EGFP, and a6 produced y-EGFP clusters that colocalized with a subpopulation of smaller a6

clusters (arrowheads, H), while larger clusters of a6 were devoid of y-EGFP (arrows, H). Scale bar in panel A is 5 Am for panels A, B, E–H. Scale bar in panel

C is 10 Am for panels C and D.


expressed in transfected cultured hippocampal neurons assembled

with endogenous subunits to form receptors. This is supported by

our observations that GABAARs containing the exogenous

subunits are translocated to the cell surface, participate in

postsynaptic GABAAR clusters that are apposed to presynaptic

GABAergic terminals and/or colocalize with gephyrin clusters.

Both g2S-EGFP and g2L-EGFP target to GABAergic synapses

(Fig. 1). This is consistent with the notion that, in the intact brain,

GABAARs containing either of the g2 splice forms can target to

GABAergic synapses (Baer et al., 2000). The two splice forms

only differ in an octapeptide insert that is present in the large

intracellular loop of g2L (Whiting et al., 1990). This peptide can

be phosphorylated in serine 343 (Moss et al., 1992). Antibody-

induced capping also revealed the existence of a pool of non-

Fig. 6. Surface GABAARs containing the chimeric mycg2S/y-IL subunit do not form clusters unless capping of the GABAARs in the living neuron is induced by

antibodies. Triple-label immunofluorescence. Hippocampal neurons cotransfected with the chimeric mycg2S/y-IL subunit and EGFP (A–C) were fixed,

permeabilized and incubated with rabbit anti-y-IL (A) and mouse anti-gephyrin (B). Transfected cells were identified by EGFP fluorescence (C). Most EGFP-

positive neurons also displayed y-IL immunoreactivity (used to indicate proper expression and monitor subcellular localization of the chimeric subunit) that

was diffusely distributed throughout the soma and proximal dendrites (A) and did not colocalize with clustered gephyrin (B). Two-step antibody-induced

capping of living cells with the rabbit anti-myc antibody and the FITC-conjugated donkey anti-rabbit IgG secondary antibody (D–I) of hippocampal neurons

transfected with only mycg2S/y-IL (D–F) or cotransfected with a mixture of mycg2S/y-IL, a6 and h3-EGFP (G–I) induces the formation of mycg2S/y-ILGABAAR clusters (D and G). The antibody-induced clusters (filled arrowheads, D and G) did not colocalized with gephyrin clusters (E) or were present at

GABAergic synapses containing GAD+ terminals (arrows, F and I) and postsynaptic gephyrin (arrows, E). In the triple-transfected neurons (G–I), antibody-

induced clustering of mycg2S/y-IL (filled arrowheads, G) induces the co-clustering of h3-EGFP (filled arrowheads, H). These neurons also have clusters of h3-

EGFP at GABAergic synapses (arrows, H) and outside GABAergic synapses (empty arrowheads, H) that do not colocalize with the antibody-induced mycg2S/y-IL clusters. Scale bar in panel A is 10 Am for panels A–C; scale bar in panel D is 5 Am for panels D–I.


clustered GABAARs containing the endogenous g2 or the tagged

g2 subunit that is present at the cell surface of the non-transfected

and transfected hippocampal neurons, respectively. Quantification

of the cluster density indicates that the pool of non-clustered

endogenous or tagged g2-GABAARs at the cell surface is

significant.

A different result was obtained with the y subunit since the y-GABAARs were diffusely localized at the cell surface, not

forming any clusters at GABAergic synapses or outside

GABAergic synapses. We have also found that the chimericmycg2S/y-IL subunit, a g2 subunit with the IL of the y subunit, co-

assembles with other subunits to produce GABAARs that are

diffusely distributed on the cell surface and do not participate in

synaptic or non-synaptic GABAAR clusters, thus behaving like

the y subunit. These results support the notion that the large

intracellular loops of the native g2 and y subunits are domains

critical to the clustering of the GABAARs. The results also

suggest that the IL is more relevant to the clustering of GABAARs

than to their targeting to GABAergic synapses since y or mycg2S/

y-IL-containing GABAARs do not cluster at GABAergic synapses

or outside GABAergic synapses.

We and others have shown that clustered GABAARs are not

only localized to GABAergic synapses, forming large clusters, but

are capable of forming small clusters outside of GABAergic

synapses, often being mismatched to glutamatergic presynaptic

terminals (Rao et al., 2000; Christie et al., 2002a; Christie and de

Blas, 2003, Studler et al., 2005). Both large and small clusters in

these mature cultures have a high degree of colocalization with

gephyrin (Christie et al., 2002a). The present study, and that of

Levi et al. (2004), demonstrates that GABAAR clusters formed

rapidly after antibody-induced capping do not initially colocalize

with gephyrin. However, given enough time, gephyrin clusters

colocalize with the antibody-induced a2-GABAAR clusters,

suggesting that gephyrin is recruited to GABAAR clusters after

the latter are formed (Levi et al., 2004). Moreover, most

extrasynaptic clusters of a2 subunit-containing GABAARs in

immature cultures form independently of gephyrin (Studler et al.,

2005). Thus, gephyrin clustering follows GABAAR clustering.


The antibody-induced capping supports a model of lateral g2-

GABAARs diffusion across the surface membrane of neurons

(Barnes, 2001; Triller and Choquet, 2005) that may also mediate

a dynamic exchange between surface clusters. This model is also

consistent with: (I) at the EM level, g2-GABAARs were not only

observed to be concentrated in postsynaptic membranes but were

also seen in non-synaptic membranes of granule cells of the intact

cerebellum (Nusser et al., 1998); (II) several reports describe the

presence of diffuse GABAAR subunit immunoreactivity that is

present outside of the synaptic clusters (Essrich et al., 1998;

Brunig et al., 2002b; Christie et al., 2002a,b; Christie and de

Blas, 2003; Levi et al., 2004; Studler et al., 2005); and (III) the

surface distribution of a5-GABAARs in cultured hippocampal

neurons is found in both GABAergic synapses (Christie and De

Blas, 2002) and extrasynaptically (Brunig et al., 2002a; Christie

and De Blas, 2002). Since the hippocampal a5-containing

GABAARs have a predominant a5h3g2 composition (Fritschy et

al., 1997; Sur et al., 1998), the enumerated data above suggest

that the presence of the g2 subunit within GABAARs does not

necessarily targets all the surface g2-GABAARs to the post-

synaptic region. Therefore, the g2 subunit may be essential for

the postsynaptic clustering of GABAARs (Gunther et al., 1995;

Essrich et al., 1998; Schweizer et al., 2003; Li et al., in press),

but it might not be sufficient since other subunits seem to also

play a role (i.e. a5 and a2).

For glycine receptors, it has been shown that their postsynaptic

accumulation in clusters is a highly dynamic process involving

lateral diffusion of the GlyRs (Meier et al., 2000, 2001; Rosenberg

et al., 2001; Meier, 2003; Triller and Choquet, 2003). The direct

interaction of the GlyR h-IL subunit with gephyrin seems to be

involved in the postsynaptic accumulation of the GlyR (Kirsch et

al., 1993; Meyer et al., 1995). The GlyR–gephyrin interaction is

important for the rate of surface accumulation, postsynaptic

accumulation, postsynaptic stabilization and internalization of

GlyRs from synapses to intracellular organelles (Kirsch et al.,

1993; Feng et al., 1998; Rasmussen et al., 2002; Hanus et al., 2004;

Levi et al., 2004). Furthermore, GlyR and Ca+ channel activity has

also been shown to be important for GlyR surface stability and the

postsynaptic accumulation of the glycine receptor (Kirsch and

Betz, 1998; Levi et al., 1998; Kneussel and Betz, 2000).

However, in the case of GABAARs clusters, the colocalizing

gephyrin has not been shown to directly interact with any of the

GABAAR subunit isoforms and appears to be dispensable for the

postsynaptic formation of some GABAergic postsynaptic receptor

clusters (Levi et al., 2004; Studler et al., 2005). Moreover, unlike

GlyR clustering, the clustering of GABAARs occurs independently

of receptor activation (Craig et al., 1994). Recent reports suggest

that the mechanism by which the GABAARs normally cluster and

are postsynaptically localized are dissociable events, with extra-

synaptic GABAAR clustering preceding colocalization with

gephyrin and/or synaptic targeting (Levi et al., 2004; Graf et al.,

2004; Studler et al., 2005). This is also supported by our antibody-

induced capping experiments as discussed above.

In the hippocampus, expression of the y subunit is primarily

localized to granule cells of the dentate gyrus (Sperk et al., 1997;

Pirker et al., 2000). Within these cells, electrophysiological and EM

immunocytochemical evidence supports exclusion of these sub-

types from GABAergic synapses (Stell and Mody, 2002; Wei et al.,

2003). The in situ hybridization studies (Persohn et al., 1992,

Wisden et al., 1992) also reveal low levels of y subunit expression

within the pyramidal cells of CA1 field. Our results showing that

the tagged y subunit distributed diffusely at the cell surface and did

not form synaptic or non-synaptic clusters were consistent with the

reported extrasynaptic distribution of the y subunit-containing

receptors in all cells expressing this subunit.

Like the myc- or EGFP-tagged y subunit, the mycg2S/y-ILchimeric subunit formed GABAARs that translocated to the surface

but did not form or participate in synaptic or non-synaptic GABAAR

clusters. Non-synaptic clusters of these receptors were observed

only after antibody-induced capping. Thus, the loss of clustering

capacity observed for the mycg2S/y-IL-GABAARs, compared to themycg2S-GABAARs, is consistent with the hypothesis that the IL

domain of the g2 subunit is necessary for clustering to occur.

However, this interpretation contrasts a recent report in which

analysis of various chimeric g2–a2 subunits mapped the TM4

region of g2 as the domain that was ‘‘required and sufficient’’ for

synaptic clustering (Alldred et al., 2005). Because our mycg2S/y-ILchimeric subunit contains the TM4 domain of g2, we concluded that

the TM4 domain of g2 is not sufficient for clustering activity when

the y-IL is present, despite the ability of receptors containing these

chimeric subunits to access the surface. We were unable to test

whether the g2-IL was sufficient to induce the clustering of the mycy/g2S-IL-GABAARs. The chimeric mycy/g2S-IL subunit did not

express in HEK293 cells or in hippocampal neurons.

The apparent discrepancy between Alldred et al.’s (2005)

results and our results may be due to the fact that the various g2chimeras used by Alldred et al. (2005) were produced by

exchanging equivalent domains between g2 and a2, two subunits

that cluster at GABAergic synapses. In our case, we exchanged

equivalent domains between a subunit that clusters at GABAergic

synapses (g2) and a subunit that does not cluster at GABAergic

synapses (y). Thus, the a2 subunit may be neutral or weakly

permissive with regard to clustering and/or synaptic targeting,

while the y-IL may prevent the clustering of GABAARs.

Furthermore, post-translational palmitoylation of the g2-IL domain

has been shown to occur (Keller et al., 2004; Rathenberg et al.,

2004) and is essential in regulating the clustering of GABAARs at

postsynaptic sites (Rathenberg et al., 2004) which is also consistent

with our results. Perhaps in the mycg2S/y-IL GABAARs, the

presence of the y-IL disrupts or supersedes the g2 TM4 clustering

signal identified by Alldred et al. (2005). Therefore, GABAAR

clustering might depend on the presence of permissive domains

(e.g. possessing the g2-IL and the g2 TM4 domains but not the y-ILdomain) or restrictive (e.g. possessing the y-IL domain) signals that

are present within the pentameric receptor.

Experimental methods

Construction of c-myc tagged GABAAR subunits

Coding sequences for the GABAAR subunits rat g2S, rat y and

human g2L, contained within the pcDNA3.1 (+) mammalian

expression vector (Invitrogen), were tagged with the 10-amino-acid

epitope (EQKLISEEDL) from c-myc (Evan et al., 1985) between the

fourth and fifth amino acids of the mature subunit peptide sequence.

Briefly, a site-directed mutagenesis strategy described by Connolly

et al. (1996) was performed using the Gene Editor in vitro Site-

Directed Mutagenesis System (Promega) with PAGE purified 5Vphosphorylated antisense oligonucleotides (5V-GC ATA ATC TTC

ATA GTC ATC TAG GTC TTC TTC TGA TAT TAG TTT TTG

TTC ATC TGA CTT TTG GCTAGT G-3V, 5V-GCATAATCT TCA


TAG TCA TCT AGG TCT TCT TCT GAT ATT AGT TTT TGT

TCATCAGATTTCTGGCTAGTG 3Vand 5V-GTCATTCATTGC

TCT GGC GCC TAG GTC TTC TTC TGA TAT TAG TTT TTG

TTC ATG GTG CGG CTG CGT GCA CAG-3V) to create the

ratmycg2S, ratmycy and human mycg2L subunits, respectively. The

correct nucleotide sequences for the full-lengthmyc-tagged subunits

were confirmed by direct sequencing. Correct expression of themyc-

tagged subunit proteins was confirmed in transfected HEK293 cells

by immunolabeling with the monoclonal antibody (mAb) 9E10

(DSHB, University of Iowa) or a rabbit anti-myc (Upstate, Waltham

MA) antibody. A rabbit anti y-IL, produced in our laboratory to a

synthetic peptide (amino acids, a.a., 390–402) of y, was also used tostudy the expression of the mycy subunit and the mAb KC5-5E1,

raised in our laboratory against the g2-IL region (Fernando et al.,

1995), to study the expression of the mycg2S and mycg2L subunits.

Construction of chimeric subunits

To produce the chimeric mycg2S/y-IL subunit, BamHI and

EcoRI restriction sites flanking the large IL region were introduced

into the rat mycg2S and rat mycy subunits by site-directed

mutagenesis, as described above. Thus, the BamHI site spanning

the junction of the TM3/IL and the EcoRI site, six nucleotide bases

into the TM4 region, were created using the 5V phosphorylatedantisense oligonucleotides in the rat mycg2S (5V-TGG TTT CCG

GTT GCT CAC AAA ATA GGATTC GGTACC ATA CTC CAC

CAA AGC TGA-3V, 5V-GGC GGT AGG GAA GAA GAT GAA

TTC ATA GAA GTC CAT TTT GGC-3V) and in the mycy (5V-CCGTTT CTT CCT GTA GTC AGC ATT GAA GGATCC AAATGC

ATA CTC CAC CAG GGC AGC-3V, 5V-GGC TGC CGG GAA

CAC AGC GAA TTC ATA GAT GTC GAT GGT GTC-3V).The mutated clones were then double-digested with BamHI

and EcoRI enzymes, and the resultant fragments were gel-

extracted, purified and cross-ligated to produce the chimericmycg2S/y-IL subunit containing BamHI and EcoRI restriction sites.

The restriction sites flanking the IL region were then removed by

site-directed mutagenesis using the 5V phosphorylated oligos formycg2S/y-IL (5V-CCG TTT CTT CCT GTA GTC AGC ATT GAA

GTG CAG GGT ACC ATA CTC CAC CAA AGC TGA-3V, 5V-GCA GAA GGC GGT AGG GAA GAA GAT CCG AGC ATA

GAT GTC GAT GGT GTC TGC ATC-3V) to restore the original

nucleotide coding sequences appropriate to the respective subunit

isoform domains. The resulting chimeric subunit clones were then

sequenced to confirm that the mutagenesis reactions did not alter

the predicted primary amino acid sequence of the translated

chimeras. Correct expression of the mycg2S/y-IL chimera was

confirmed in transfected HEK293 cells, by separate immunolab-

eling experiments using rabbit anti-myc or rabbit anti y-ILantibodies.

Construction of EGFP-tagged GABAAR subunits

EGFP-tagged subunits (rat a1, a2, h3, g2S, g2L and human ytagged at the C-terminal) were fusion proteins constructed using

the pEGFP-N1 vector (Invitrogen, Carlsbad CA). Briefly, the

coding region for each subunit was amplified by PCR using

primers that created a restriction site before the 5V start codon

(NheI, for a1, h3, g2S, g2L and y; XhoI for a2) and replaced the 3Vstop codon with a restriction site (XhoI for a1, PstI for a2, h3, g2S,

g2L and y). The replacement of the 3V subunit insert’s stop codon

with either PstI or XhoI restriction sites ensured that the coding

region for the subunit was in frame with the start codon of the

EGFP coding region, thus creating a linker sequence between the

subunit and EGFP that was based on the sequence in the plasmid

vector (13 a.a. for PstI site and 20 a.a. for the XhoI site). EGFP-

tagged subunit clones were sequenced to confirm correctness and

expressed in HEK cells to confirm expression and detection by

both EGFP fluorescence and anti-subunit antibodies.

Transfection of primary cultures of hippocampal neurons

Hippocampal cultures were prepared according to Goslin et al.

(1998), as described elsewhere (Christie et al., 2002a; Christie and

de Blas, 2003). Briefly, dissociated cells were prepared from

embryonic day 18 Sprague–Dawley rat pup hippocampi and

plated at a density of 20,000–30,000 cells per 18 mm diameter

circular coverslip treated with poly-l-lysine (Sigma). The cover-

slips with cell suspension media were then placed in 5% CO2 at

37-C for 4–6 h to allow settling and attachment of cells. The

coverslips with attached cells were then placed upside down in 60

mm diameter petri dishes containing glial feeder layers (Goslin et

al., 1998) to maintain conditioning of the support medium (DMEM

containing 0.6% glucose, 26 mM NaHCO3, N-2 supplement and

0.1% ovalbumin) and placed in a 5% CO2 atmosphere at 37-C.After 2–3 days, a final concentration of 5 � 10�6 M cytosine

arabinoside was added for 24 h. Cultures were maintained by

replacement of one-third volume of fresh N2 supplemented

DMEM medium every 3–5 days.

Hippocampal neurons were maintained in culture for 12 days

prior to transfection by using the CalPhos Mammalian Transfection

Kit (BD Biosciences, Palo Alto, CA) following manufacturer’s

protocol using 1–2 Ag DNA of each plasmid for each coverslip

used in the transfection. Immunocytochemistry was done 7 days

after transfection, unless otherwise noted.

Antibodies

The primary antibodies, rabbit anti-g2 (1–15 a.a.) and rabbit

anti-y-IL (390–402 a.a.) were raised and affinity-purified in our

laboratory against synthetic peptides made to a unique extracellular

epitope (N-terminus for g2) or the large IL domain (for y) of ratGABAAR subunits. The specificity of the affinity-purified anti-

bodies has been determined by Western blots and immunopreci-

pitation of rat brain extracts, rat brain immunohistochemistry and

immunofluorescence after transfection of GABAAR subunits into

HEK293 cells. The rabbit anti-a6 and rabbit anti-g2 antibodies

have been used in previous studies (Gutierrez et al., 1996; Christie

et al., 2002a,b; Christie and De Blas, 2002, 2003; Charych et al.,

2004b). For antibody identification of myc- or EGFP-tagged

receptor subunits, rabbit affinity-purified anti-myc antibody from

Upstate (Charlottesville, VA) and either chicken anti-GFP or rabbit

anti-GFP from Chemicon (Temecula, CA) was used. The mono-

clonal mouse anti-gephyrin (mAb 7a) was purchased from

Cedarlane (Accurate Chemical and Scientific Corp., Westbury,

NY). Sheep anti-GAD (gift of I. Kopin, NINDS, Bethesda) and

guinea pig anti-synaptic vesicle GABA transporter (VGAT) from

Chemicon were used to identify interneurons and/or GABAergic

presynaptic processes. The secondary antibodies purchased from

Jackson Immunoreagents (West Grove, PA) were all raised in

donkey (crossreactivity eliminated by adsorption with IgG from

other species) and were conjugated with Texas Red, FITC or

AMCA fluorophores.


Live cell fluorescence immunocytochemistry by incubation with

primary or combination of primary and secondary antibodies

Triple-label immunofluorescence detection of fixed and per-

meabilized cells was done as described elsewhere (Christie et al.,

2002a). For antibody-induced capping of surface receptors by

primary antibody (one-step capping), 19 DIV live neurons were

incubated in support medium containing anti-g2-NH2, rabbit anti-

myc or rabbit anti-GFP antibodies at 37-C for 30 min in a 5% CO2

atmosphere. Coverslips were then twice washed with pH- and

temperature-equilibrated DMEM medium at 37-C for 5 min. In the

two-step capping, there was a live cell incubation with the primary

antibody (done as above) followed by washing with temperature-

and pH-equilibrated DMEM and by incubation for 30 min at 37-Cwith DMEM containing FITC- or Texas-Red-conjugated donkey

anti-rabbit antibodies. Others have shown that antibody-induced

capping with this protocol induces the clustering of GABAARs at

the cell surface with little or no internalization (Levi et al., 2004).

Following live cell incubations with primary or primary and

secondary antibodies, samples were briefly rinsed in phosphate-

buffered saline (PBS) and fixed by immersion in 4% paraformal-

dehyde, 4% sucrose in PBS for 15 min at RT followed by

permeabilization with 0.25% Triton X-100 in PBS for 5 min. Non-

specific antibody labeling was minimized by treatment with 5%

donkey serum in PBS for 30 min at room temperature. For labeling

cytoplasmic antigens (i.e. gephyrin and GAD), the fixed cultures

were incubated with a mixture of primary antibodies to the

intracellular antigens diluted in 0.25% Triton X-100 PBS for 2 h at

RT. Coverslips were washed with PBS and incubated for 1 h at

room temperature with a mixture of species-specific secondary

anti-IgG antibodies raised in donkey and conjugated to either

Texas Red, FITC or AMCA diluted in 0.25% Triton X-100 PBS.

The coverslips were washed with PBS and mounted using Prolong

anti-fade mounting solution (Molecular Probes; Eugene, Oregon).

Image acquisition and analysis

Images were collected using a 60� pan-fluor objective on a

Nikon Eclipse T200 microscope with a Spot 2e cooled CCD camera

(with a Kodak 1401E chip) driven by Spot image acquisition

software. Image files (1315 � 1035 pixel resolution) were then

sharpened using the unsharpen mask tool (settings: amount = 120%,

radius = 1.5 pixels, threshold = 0 levels) and merged for color

colocalization and figure preparation using PhotoShop 4.01

(Adobe). Control slides in which one or more primary antibodies

were omitted showed no spill over in the other two fluorescence

channels. Random drift of the sample’s fluorescence signal between

channels was controlled by alignment of all channels using triple-

labeled fluorescent microspheres (0.1 Am and 0.4 Am diameter;

Molecular Probes, Eugene, OR). Prior to processing images for

quantitation, two or three 50 Am dendritic lengths to be examined

were collected in 6–12 cells from each coverslip and condition.

Areas of the selected regions were measured directly for each

dendrite by outlining and dividing the total number of pixels

contained within the region by the constant value 81 pixels/Am2.

This value was determined by calibration of the Nikon 60�objective and CCD camera field (9 pixels/Am) using a Leica

micrometer. Quantification of signal was performed by normalizing

maximum intensity data between fluorophore channels followed by

the subtraction of the diffuse non-clustered fluorescence signal seen

in the dendrites. The two- or three-color channel images to be

compared were merged and clusters counted over 50 Am dendrite

lengths. A cluster in a fluorescence channel was considered

colocalizing with a cluster in the other fluorescence channel when

>25% of the surface of one of the two clusters overlapped with the

cluster in the other channel. Values were then normalized to 100 Am2

and used for statistical comparison by Student’s t test and reported as

mean T standard error of the mean (SEM).

Acknowledgments

This work was supported by National Institute of Neurological

Disorders and Stroke grants NS38752 and NS39287. We would

also like to thank Dr. Peter Seeburg for kindly providing the cDNA

clones for the GABAAR a1, a2, h3, g2S and g2L subunits, Dr.

Robert Macdonald for the GABAAR y subunit cDNA clone and

Dr. Stephano Vicini for one of the a1-EGFP plasmids.

References

Alldred, M.J., Mulder-Rosi, J., Lingenfelter, S.E., Chen, G., Luscher, B.,

2005. Distinct gamma2 subunit domains mediate clustering and

synaptic function of postsynaptic GABAA receptors and gephyrin.

J. Neurosci. 25, 594–603.

Araujo, F., Ruano, D., Vitorica, J., 1998. Absence of association between

delta and gamma2 subunits in native GABA(A) receptors from rat

brain. Eur. J. Pharmacol. 347, 347–353.

Baer, K., Essrich, C., Balsiger, S., Wick, M.J., Harris, R.A., Fritschy, J.M.,

Luscher, B., 2000. Rescue of gamma2 subunit-deficient mice by

transgenic overexpression of the GABAA receptor gamma2S or

gamma2L subunit isoforms. Eur. J. Neurosci. 12, 2639–2643.

Barnes, E.M., 2001. Assembly and intracellular trafficking of GABAA

receptors. Int. Rev. Neurobiol. 48, 1–29.

Bedford, F.K., Kittler, J.T., Muller, E., Thomas, P., Uren, J.M., Merlo, D.,

Wisden, W., Triller, A., Smart, T.G., Moss, S.J., 2001. GABA(A)

receptor cell surface number and subunit stability are regulated by the

ubiquitin-like protein Plic-1. Nat. Neurosci. 4, 908–916.

Brickley, S.G., Cull-Candy, S.G., Farrant, M., 1996. 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.

Brunig, I., Scotti, E., Sidler, C., Fritschy, J.-M., 2002a. Intact

sorting, targeting, and clustering of gamma-aminobutyric acid A

receptor subtypes in hippocampal neurons in vitro. J. Comp.

Neurol. 443, 43–55.

Brunig, I., Suter, A., Knuesel, I., Luscher, B., Fritschy, J.-M., 2002b.

GABAergic terminals are required for postsynaptic clustering of

dystrophin but not of GABA(A) receptors and gephyrin. J. Neurosci.

22, 4805–4813.

Charych, E.I., Yu, W., Miralles, C.P., Serwanski, D.R., Li, X., Rubio,

M., de Blas, A.L., 2004a. The brefeldin A-inhibited GDP/GTP

exchange factor 2, a protein involved in vesicular trafficking, interacts

with the beta subunits of the GABA-A receptors. J. Neurochem. 90,

173–189.

Charych, E.I., Yu, W., Li, R., Serwanski, D.R., Miralles, C.P., Li, X., Yang,

B.Y., Pinal, N., Walikonis, R., De Blas, A.L., 2004b. A four PDZ

domain-containing splice variant form of GRIP1 is localized in

GABAergic and glutamatergic synapses in the brain. J. Biol. Chem.

279, 38978–38990.

Christie, S.B., De Blas, A.L., 2002. a5 subunit-containing GABAARs form

clusters at GABAergic synapses in hippocampal cultures. NeuroReport

13, 2355–2358.

Christie, S.B., de Blas, A.L., 2003. GABAergic and glutamatergic axons

innervate the axon initial segment and organize GABAA receptor


clusters of cultured hippocampal pyramidal cells. J. Comp. Neurol. 456,

361–374.

Christie, S.B., Miralles, C.P., de Blas, A.L., 2002a. GABAergic innervation

organizes synaptic and extrasynaptic GABAAR clustering in cultured

hippocampal neurons. J. Neurosci. 22, 684–697.

Christie, S.B., Li, R., Miralles, C.P., Riquelme, R., Yang, B.Y., Charych, E.,

Yu, W., Daniels, S.B., Cantino, M.E., De Blas, A.L., 2002b. Synaptic

and Extrasynaptic GABAA receptor and gephyrin clusters. In: Azmitia,

E.C., DeFelipe, J., Jones, E.G., Rakic, P., Ribak, C.E. (Eds.), Progress in

Brain Research, vol. 136. Elsevier, New York, pp. 157–179.

Connolly, C.N., Krishek, B.J., McDonald, B.J., Smart, T.G., Moss, S.J.,

1996. Assembly and cell surface expression of heteromeric and

homomeric gamma-aminobutyric acid type A receptors. J. Biol. Chem.

271, 89–96.

Connolly, C.N., Uren, J.M., Thomas, P., Gorrie, G.H., Gibson, A., Smart,

T.G., Moss, S.J., 1999. Subcellular localization and endocytosis of

homomeric gamma2 subunit splice variants of gamma-aminobutyric

acid type A receptors. Mol. Cell. Neurosci. 13, 259–271.

Craig, A.M., Blackstone, C.D., Huganir, R.L., Banker, G., 1994. Selective

clustering of glutamate and gamma-aminobutyric acid receptors

opposite terminals releasing the corresponding neurotransmitters. Proc.

Natl. Acad. Sci. U. S. A. 91, 12373–12377.

De Schutter, E., 2002. Cerebellar cortex: computation by extrasynaptic

inhibition? Curr. Biol. 12, R363–R365.

Essrich, C., Lorez, M., Benson, J.A., Fritschy, J.-M., Luscher, B., 1998.

Postsynaptic clustering of major GABAA receptor subtypes requires the

gamma 2 subunit and gephyrin. Nat. Neurosci. 1, 563–571.

Evan, G.I., Lewis, G.K., Ramsay, G., Bishop, J.M., 1985. Isolation of

monoclonal antibodies specific for human c-myc proto-oncogene

product. Mol. Cell. Biol. 12, 3610–3616.

Feng, G., Tintrup, H., Kirsch, J., Nichol, M.C., Kuhse, J., Betz, H., Sanes,

J.R., 1998. Dual requirement for gephyrin in glycine receptor clustering

and molybdoenzyme activity. Science 282, 1321–1324.

Fernando, L.P., Khan, Z.U., McKernan, R.M., De Blas, A.L., 1995.

Monoclonal antibodies to the human gamma 2 subunit of the

GABAA/benzodiazepine receptors. J. Neurochem. 64, 1305–1311.

Fischer, F., Kneussel, M., Tintrup, H., Haverkamp, S., Rauen, T., Betz, H.,

Wassle, H., 2000. Reduced synaptic clustering of GABA and glycine

receptors in the retina of the gephyrin null mutant mouse. J. Comp.

Neurol. 427, 634–648.

Fritschy, J.-M., Brunig, I., 2003. Formation and plasticity of GABAergic

synapses: physiological mechanisms and pathophysiological implica-

tions. Pharmacol. Ther. 98, 299–323.

Fritschy, J.-M., Benke, D., Johnson, D.K., Mohler, H., Rudolph, U., 1997.

GABAA-receptor alpha-subunit is an essential prerequisite for receptor

formation in vivo. Neuroscience 81, 1043–1053.

Goslin, K., Asmussen, H., Banker, G., 1998. Rat hippocampal neurons in

low-density culture. In: Banker, G., Goslin, K. (Eds.), Culturing Nerve

Cells, 2nd. edR MIT Press, Cambridge, MA, pp. 339–370.

Graf, E.R., Zhang, X., Jin. S, X., Linhoff, M.W., Craig, A.M., 2004.

Neurexins induce differentiation of GABA and glutamate postsynaptic

specializations via neuroligins. Cell 119, 1013–1026.

Gunther, U., Benson, J., Benke, D., Fritschy, J-M., Reyes, G.,

Knoflach, F., Crestani, F., Aguzzi, A., Arigoni, M., Lang, Y.,

Bluethmann, H., Mohler, H., Luscher, B., 1995. Benzodiazepine-

insensitive mice generated by targeted disruption of the gamma 2

subunit gene of gamma-aminobutyric acid type A receptors. Proc.

Natl. Acad. Sci. U. S. A. 92, 7749–7753.

Gutierrez, A., Khan, Z.U., De Blas, A.L., 1996. Immunocytochemical

localization of the alpha 6 subunit of the gamma-aminobutyric acidA

receptor in the rat nervous system. J. Comp. Neurol. 365, 504–510.

Hanus, C., Vannier, C., Triller, A., 2004. Intracellular association of glycine

receptor with gephyrin increases its plasma membrane accumulation

rate. J. Neurosci. 24, 1119–1128.

Harvey, K., Duguid, I.C., Alldred, M.J., Beatty, S.E., Ward, H., Keep, N.H.,

Lingenfelter, S.E., Pearce, B.R., Lundgren, J., Owen, M.J., Smart, T.G.,

Luscher, B., Rees, M.I., Harvey, R.J., 2004. The GDP–GTP exchange

factor collybistin: an essential determinant of neuronal gephyrin

clustering. J. Neurosci. 24, 5816–5826.

Herring, D., Huang, R., Singh, M., Robinson, L.C., Dillon, G.H.,

Leidenheimer, N.J., 2003. Constitutive GABAA receptor endocytosis

is dynamin-mediated and dependent on a dileucine AP2 adaptin-binding

motif within the beta 2 subunit of the receptor. J. Biol. Chem. 278,

24046–24052.

Jechlinger, M., Pelz, R., Tretter, V., Klausberger, T., Sieghart, W., 1998.

Subunit composition and quantitative importance of hetero-oligomeric

receptors: GABAA receptors containing alpha6 subunits. J. Neurosci.

18, 2449–2457.

Keller, C.A., Yuan, X., Panzanelli, P., Martin, M.L., Alldred, M., Sassoe-

Pognetto, M., Luscher, B., 2004. The gamma2 subunit of GABA(A)

receptors is a substrate for palmitoylation by GODZ. J. Neurosci. 24,

5881–5891.

Killisch, I., Dotti, C.G., Laurie, D.J., Luddens, H., Seeburg, P.H., 1991.

Expression patterns of GABAA receptor subtypes in developing

hippocampal neurons. Neuron 7, 927–936.

Kirsch, J., Betz, H., 1998. Glycine-receptor activation is required for

receptor clustering in spinal neurons. Nature 392, 717–720.

Kirsch, J., Wolters, I., Triller, A., Betz, H., 1993. Gephyrin antisense

oligonucleotides prevent glycine receptor clustering in spinal neurons.

Nature 366, 745–748.

Kirsch, J., Kuhse, J., Betz, H., 1995. Targeting of glycine receptor subunits

to gephyrin-rich domains in transfected human embryonic kidney cells.

Mol. Cell. Neurosci. 6, 450–461.

Kittler, J.T., Wang, J., Connolly, C.N., Vicini, S., Smart, T.G., Moss, S.J.,

2000. Analysis of GABAA receptor assembly in mammalian cell lines

and hippocampal neurons using gamma 2 subunit green fluorescent

protein chimeras. Mol. Cell. Neurosci. 16, 440–452.

Kneussel, M., Betz, H., 2000. Clustering of inhibitory neurotransmitter

receptors at developing postsynaptic sites: the membrane activation

model. Trends Neurosci. 23, 429–435.

Kneussel, M., Bransdtatter, J.H., Laube, B., Stahl, S., Muller, U., Betz, H.,

1999. Loss of postsynaptic GABA(A) receptor clustering in gephyrin-

deficient mice. J. Neurosci. 19, 9289–9297.

Kneussel, M., Brandstatter, J.H., Gasnier, B., Feng, G., Sanes, J.R., Betz,

H., 2001. Gephyrin-independent clustering of postsynaptic gaba(a)

receptor subtypes. Mol. Cell. Neurosci. 17, 973–982.

Korpi, E.R., Mihalek, R.M., Sinkkonen, S.T., Hauer, B., Hevers, W.,

Homanics, G.E., Sieghart, W., Luddens, H., 2002. Altered receptor

subtypes in the forebrain of GABA(A) receptor delta subunit-

deficient mice: recruitment of gamma 2 subunits. Neuroscience 109,

733–743.

Levi, S., Vannier, C., Triller, A., 1998. Strychnine-sensitive stabilization of

postsynaptic glycine receptor clusters. J. Cell. Sci. 111, 335–345.

Levi, S., Logan, S.M., Tovar, K.R., Craig, A.M., 2004. Gephyrin is critical for

glycine receptor clustering but not for the formation of functional

GABAergic synapses in hippocampal neurons. J. Neurosci. 24, 207–217.

Li, R.-W., Yu, W., Christie, S., Miralles, C.P., Bai, J., LoTurco, J.J., De

Blas, A.L., in press. Disruption of postsynaptic GABAA receptor

clusters leads to decreased GABAergic innervation of pyramidal

neurons. J. Neurochem.

Luscher, B., Keller, C.A., 2004. Regulation of GABAA receptor trafficking,

channel activity, and functional plasticity of inhibitory synapses.

Pharmacol. Ther. 102, 195–221.

Meier, J., 2003. The enigma of transmitter-selective receptor accumulation

at developing inhibitory synapses. Cell Tissue Res. 311, 271–276.

Meier, J., Meunier-Durmort, C., Forest, C., Triller, A., Vannier, C., 2000.

Formation of glycine receptor clusters and their accumulation at

synapses. J. Cell. Sci. 113, 2783–2795.

Meier, J., Vannier, C., Serge, A., Triller, A., Choquet, D., 2001. Fast and

reversible trapping of surface glycine receptors by gephyrin. Nat.

Neurosci. 4, 253–260.

Meyer, G., Kirsch, J., Betz, H., Langosch, D., 1995. Identification of a

gephyrin binding motif on the glycine receptor beta subunit. Neuron 15,

563–572.


Moss, S.J., Doherty, C.A., Huganir, R.L., 1992. Identification of the cAMP-

dependent protein kinase and protein kinase C phosphorylation sites

within the major intracellular domains of the beta 1, gamma 2S, and

gamma 2L subunits of the gamma-aminobutyric acid type A receptor.

J. Biol. Chem. 267, 14470–14476.

Nusser, Z., Mody, I., 2002. Selective modulation of tonic and phasic inhibitions

in dentate gyrus granule cells. J. Neurophysiol. 87, 2624–2628.

Nusser, Z., Sieghart, W., Somogyi, P., 1998. Segregation of different

GABAA receptors to synaptic and extrasynaptic membranes of

cerebellar granule cells. J. Neurosci. 18, 1693–1703.

Peng, Z., Hauer, B., Mihalek, R.M., Homanics, G.E., Sieghart, W., Olsen,

R.W., Houser, C.R., 2002. GABA(A) receptor changes in delta subunit-

deficient mice: altered expression of alpha4 and gamma2 subunits in the

forebrain. J. Comp. Neurol. 446, 179–197.

Persohn, E., Malherbe, P., Richards, J.G., 1992. Comparative molecular

neuroatomy of cloned GABAA receptor subunits in the rat CNS.

J. Comp. Neurol. 326, 193–216.

Pirker, S., Schwarzer, C., Wieselthaler, A., Sieghart, W., Sperk, G., 2000.

GABA(A) receptors: immunocytochemical distribution of 13 subunits

in the adult rat. Neuroscience 101, 815–850.

Quirk, K., Gillard, N.P., Ragan, C.I., Whiting, P.J., McKernan, R.M., 1994.

Model of subunit composition of gamma-aminobutyric acid A receptor

subtypes expressed in rat cerebellum with respect to their alpha and

gamma/delta subunits. J. Biol. Chem. 269, 16020–16028.

Quirk, K., Whiting, P.J., Ragan, C.I., McKernan, R.M., 1995. Character-

isation of delta-subunit containing GABAA receptors from rat brain.

Eur. J. Pharmacol. 290, 175–181.

Rao, A., Cha, E.M., Craig, A.M., 2000. Mismatched appositions of

presynaptic and postsynaptic components in isolated hippocampal

neurons. J. Neurosci. 20, 8344–8353.

Rasmussen, H., Rasmussen, T., Triller, A., Vannier, C., 2002. Strychnine-

blocked glycine receptor is removed from synapses by a shift in

insertion/degradation equilibrium. Mol. Cell. Neurosci. 19, 201–215.

Rathenberg, J., Kittler, J.T., Moss, S.J., 2004. Palmitoylation regulates the

clustering and cell surface stability of GABAA receptors. Mol. Cell.

Neurosci. 26, 217–251.

Rosenberg, M., Meier, J., Triller, A., Vannier, C., 2001. Dynamics of glycine

receptor insertion in the neuronal plasma membrane. J. Neurosci. 21,

5036–5044.

Schweizer, C., Balsiger, S., Bluethmann, H., Mansuy, I.M., Fritschy, J.-M.,

Mohler, H., Luscher, B., 2003. The gamma 2 subunit of GABA(A)

receptors is required for maintenance of receptors at mature synapses.

Mol. Cell. Neurosci. 24, 442–450.

Sperk, G., Schwarzer, C., Tsunashima, K., Fuch, K., Sieghart, W., 1997.

GABAA receptor subunits in the rat hippocampus immunocytochemical

distribution of 13 subunits. Neuroscience 80, 987–1000.

Stell, B.M., Mody, I., 2002. Receptors with different affinities mediate

phasic and tonic GABA(A) conductances in hippocampal neurons.

J. Neurosci. 22, RC223.

Stell, B.M., Brickley, S.G., Tang, C.Y., Farrant, M., Mody, I., 2003.

Neuroactive steroids reduce neuronal excitability by selectively enhanc-

ing tonic inhibition mediated by delta subunit-containing GABAA

receptors. Proc. Natl. Acad. Sci. U. S. A. 100, 14439–14444.

Studler, B., Sidler, C., Fritschy, J.M., 2005. Differential regulation of

GABA(A) receptor and gephyrin postsynaptic clustering in

immature hippocampal neuronal cultures. J. Comp. Neurol. 484,

344–355.

Sur, C., Quirk, K., Dewar, D., Atack, J., McKernan, R., 1998. Rat and

human hippocampal alpha5 subunit-containing gamma-aminobutyric

AcidA receptors have alpha5 beta3 gamma2 pharmacological character-

istics. Mol. Pharmacol. 54, 928–933.

Triller, A., Choquet, D., 2003. Synaptic structure and diffusion dynamics of

synaptic receptors. Biol. Cell. 95, 465–476.

Triller, A., Choquet, D., 2005. Surface trafficking of receptors between

synaptic and extrasynaptic membranes: and yet they do move!. Trends

Neurosci. 28, 133–139.

Wei, W., Zhang, N., Peng, Z., Houser, C.R., Mody, I., 2003. Perisynaptic

localization of delta subunit-containing GABA(A) receptors and their

activation by GABA spillover in the mouse dentate gyrus. J. Neurosci.

23, 10650–10661.

Whiting, P., McKernan, R.M., Iversen, L.L., 1990. Another mechanism for

creating diversity in gamma-aminobutyrate type A receptors: RNA

splicing directs expression of two forms of gamma 2 phosphorylation

site. Proc. Natl. Acad. Sci. U. S. A. 87, 9966–9970.

Williams, B.M., Temburni, M.K., Levey, M.S., Bertrand, S., Bertrand, D.,

Jacob, M.H., 1998. The long internal loop of the alpha 3 subunit targets

nAChRs to subdomains within individual synapses on neurons in vivo.

Nat. Neurosci. 1, 557–562.

Wisden, W., Laurie, D.J., Monyer, H., Seeburg, P.H., 1992. The

distribution of 13 GABAA receptor subunit mRNAs in the rat

brain: I. Telencephalon, diencephalon, mesencephalon. J. Neurosci.

12, 1040–1062.

Yeung, J.Y., Canning, K.J., Zhu, G., Pennefather, P., MacDonald, J.F.,

Orser, B.A., 2003. Tonically activated GABAA receptors in hippo-

campal neurons are high-affinity, low-conductance sensors for extrac-

ellular GABA. Mol. Pharmacol. 63, 2–8.

Clustered and non-clustered GABAA receptors in cultured hippocampal neurons

Documents