The ligand-gated chloride channels including s GABAC
and a, b, y, e, π, d and GABAAreceptors are key
elements in the tonic and synaptic inhibitory signalling in
the CNS (Cutting et al. 1991; MacDonald & Olsen, 1994;
Wang et al. 1994). Unlike GABACreceptors, GABAA
receptors are reversibly blocked by bicuculline and
modulated by barbiturates and benzodiazepines (Polenzani
et al. 1991; Shimada et al. 1992). GABAAreceptors are
widely distributed in the retina, spinal cord, hippo-
campus, cerebellum, superior colliculus, thalamus and
other brain regions (Houser et al. 1988; Zimprich et al.
1991; Feigenspan & Bormann, 1994; MacDonald & Olsen,
1994). GABAC receptors are widely expressed in the
retina, with lower levels in the brain and spinal cord
(Strata & Cherubini, 1994; Zhang et al. 1995; Enz et al.
1996; Koulen et al. 1998; Lukasiewicz & Shields, 1998).
Numerous CNS diseases such as epilepsy, hepatic
encephalopathy, spinocerebellar degeneration and dementia
may be associated with a functional abnormality of
GABAergic transmission (Cossart et al. 2001). A potential
method to treat these abnormalities is the delivery of the
DNA coding for functional GABA receptors into the
disease-affected tissue. The human adenovirus (serotypes
2, 5) is a potentially powerful gene-delivery vehicle in
that it satisfies the following stringent criteria: (i) high
level of transduction, (ii) high insert capacity, (iii) wide
variety of cell targets, (iv) amplification to very high
titres, (v) non-oncogenic, and (vi) replication deficient
(Douglas & Curiel, 1997; Krasnykh et al.2000). The prime
receptor for the human adenovirus (serotypes 2, 5) was
shown to be similar to that for coxsackie B virus and has
therefore been termed the coxsackie/adenovirus receptor
Recombinant GABACreceptors expressed in rat
hippocampal neurons after infection with an adenovirus
containing the human s1 subunit
Natalia Filippova, Anna Sedelnikova, William J. Tyler, Terri L. Whitworth,
Henry Fortinberry and David S. Weiss
Department of Neurobiology, University of Alabama at Birmingham School of
Medicine, 1719 Sixth Avenue South CIRC 410, Birmingham, AL 35294-0021, USA
(Received 20 April 2001; accepted after revision 22 June 2001)
1. A recombinant adenovirus was generated with the human s1 GABACreceptor subunit
(adeno-s). Patch-clamp and antibody staining were employed to confirm functional expression
of recombinant s1 receptors after infection of human embryonic kidney cells (HEK293 cell
line), human embryonic retinal cells (911 cell line), dissociated rat hippocampal neurons and
cultured rat hippocampal slices.
2. Standard whole-cell recording and Western blot analysis using s1 GABACreceptor antibodies
revealed that recombinant s1 receptors were expressed in HEK293 and 911 cells after adeno-s
infection and exhibited properties similar to those of s1 receptors after standard transfection.
3. Cultured rat hippocampal neurons (postnatal day (P)3–P5) did not show a native GABAC-like
current. After adeno-s infection, however, a GABAC-like current appeared in 70–90% of the
4. Five days after infection, expression of GABACreceptors in hippocampal neurons significantly
decreased native GABAAreceptor currents from 1200 ± 300 to 150 ± 70 pA (n = 10). The
native glutamate-activated current was unchanged.
5. Hippocampal slices (P8) did not show a native GABAC-like current, although recombinant s1
receptors could be expressed in cultured hippocampal slices after adeno-s infection.
6. These data indicate that an adenovirus can be used to express recombinant GABACreceptors in
hippocampal neurons. This finding could represent an important step towards the gene
therapy of CNS receptor-related diseases.
Journal of Physiology (2001), 535.1, pp.145–153
(CAR) (Roelvink et al.1998). Biochemical analysis of CAR
revealed that it is a 46 kDa glycoprotein widely
distributed in human fibroblasts, glia, and to a lesser
extent in the differentiated respiratory epithelium,
mature skeletal muscle and human lymphocytes (Zabner
et al. 1997; Walters et al. 1999; Nalbantoglu et al. 1999;
Hidaka et al.1999). Less is known about CAR distribution
in neuronal cells.
In this study we have used adenovirus serotype 5 to
deliver DNA encoding the s1 GABACreceptor subunit
into neuronal hippocampal cells. Recombinant adeno-
virus containing the s1 subunit (adeno-s) was produced
under the human cytomegalovirus (CMV) promoter.
Recombinant s1 GABAC receptors were expressed in
70–90% of cultured hippocampal neurons after adeno-s
infection. Patch-clamp analysis of GABA-activated
current revealed that the s1 receptors had similar
properties to s1 receptors expressed using standard
transfection methods in non-neuronal cells. This finding
could represent an important step towards the gene
therapy of CNS receptor-related diseases.
The human s1 subunit and rat a1, b2 and y2 subunits were obtained
from cDNA libraries via the polymerase chain reaction as described
previously (Amin et al.1994; Amin & Weiss, 1994). The cDNA of the
s1 subunit was excised using BamHI and XbaI restriction enzymes
and inserted in the pShuttle CMV vector using BglII and XbaI
ligation sites. Recombinant adenovirus containing s1 under the
control of the human CMV promoter was produced using the
QuantumAdEasy kit (Quantum Biotechnologies, Quebec, Canada)
and has been termed adeno-s. Adeno-s was propagated in 109
HEK293 cells and was purified by centrifugation in a CsCl gradient
according to Quantum protocols. The titre of infectious viral particles
of adeno-s determined by plaque assay after large-scale purification
was 2 w 1011plaque-forming units (PFU) ml_1. Dialysed adeno-s was
aliquoted and stored at _80°C.
Transfection.HEK293 cells were transfected with s1 and/or a1, b2
and y2 subunits in the pCDNA3 vector using Fugene 6 (Roche,
Indianapolis, IN, USA) as described by the manufacturer. a1, b2 and
y2 were cotransfected at a cDNA ratio of 1:1:2 with a total of 4 µg of
cDNA per 35 mm dish. For the case of the cotransfection of s1, a1, b2
and y2 , the cDNA ratio was 1:1:1:2 with a total of 5 µg of cDNA per
35 mm dish. In all cases, 1 µg of green fluorescent protein (GFP) was
included for visualization of transfected cells.
Primary culture of hippocampal neurons and cell infection
For preparation of dissociated neurons, Sprague-Dawley rats at stage
P3–P5 (Harlan, Indianapolis, IN, USA) were rapidly decapitated
after cervical dislocation, and the hippocampi were removed from
the brain and dissected free of meninges in cooled (6°C), oxygenated,
phosphate-buffered saline (PBS) containing Ca2+and Mg2+. This
procedure, as well as the procedure for obtaining hippocampal slices
(described below), were carried out under the guidelines and approval
of the UAB Institutional Animal Care and Use Committee. The
hippocampi were then transferred into Ca2+, Mg2+-free PBS, cut into
small pieces and incubated with 0.3% (w/v) protease from Aspergillus
oryzae (Type XXIII; Sigma, St Louis, MO, USA) and 0.1% (w/v)
DNase (Type I, Sigma) for 20 min at 25°C. The tissue was washed
and triturated. After a brief centrifugation, the cell pellet was
resuspended in culture medium (minimal essential medium (MEM),
Gibco BRL, Gaithersburg, MD, USA), supplemented with 10% NU
serum (Fischer Scientific, Pittsburgh, PA, USA), penicillin (5 U ml_1)
and streptomycin (5 µg ml_1), and plated at a density of
8–10 (w 104) cells cm_2on glass coverslips coated with poly-L-lysine.
Cells were used after 10–14 days in culture.
Adeno-s was used at a concentration of 100 PFU cell_1for the
neuronal cultures, and in the range of 2 to 100 PFU cell_1for the
HEK293 and 911 cell lines. GABA-activated currents were recorded
from 12 h to 5 days after infection.
Organotypic hippocampus slice culture and slice infection
Stage P7 Sprague-Dawley rats (Harlan) were cervically dislocated
and rapidly decapitated. Hippocampal slices (200–400 µm thick)
were prepared with a custom-designed wire slicer and maintained in
vitro on Millicell-CM filter inserts (Millipore, Bedford, MA, USA) in a
36°C, 5% CO2, humidified (99%) incubator (Stoppini et al.1991). The
concentration of horse serum (Gibco) in the culture medium was
reduced from 20 to 10% at 6 days in vitro. Over the next 2 days,
serum was reduced to 5% and then 0%. The culture medium was
completely exchanged every 3 days. The best infection of slices was
observed in serum-free medium.
Experiments were performed at room temperature (20–24°C) using
the whole-cell recording patch-clamp technique as previously
described (Filippova et al. 1999). The holding potential was _50 mV.
The external recording solution contained (mM): NaCl, 160; KCl, 3.5;
glucose, 10; CaCl2, 2; and Hepes, 10 (pH 7.4). In some experiments
TTX (1 µM), DL-2-amino-5-phosphonopentanoic acid (DL-APV,
10 µM), 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 10 µM) or
bicuculline (30–50 µM) was added to the external bath solution to
decrease spontaneous synaptic activity. The recording pipettes
(borosilicate glass) had resistances of 3–5 MΩ when filled with
internal solution containing (mM): CsCl, 150; CaCl2, 0.25; EGTA, 1.1
(free Ca2+, 5 w 10_8M); Hepes, 10; and Mg-ATP, 4 (pH 7.2). GABA,
glycine and glutamate were applied to the cells through a double-
barrelled perfusion system. In some experiments, bicuculline
(30–50 µM) was added in the GABA-containing solution. In order to
determine the GABAAand GABACcurrent amplitudes in native
neurons infected with adeno-s, the amplitude of the GABAAcurrent
was calculated by subtraction of the GABACcurrent amplitude
(activated by 20 µMGABA in the presence of 50 µMbicuculline) from
the current activated by 300 µM GABA without bicuculline. In the
case of the s1 and s1–a1b2y2 coexpression studies in HEK293 cells,
the amplitude of the GABAAcurrent was calculated by subtraction of
the GABACcurrent amplitude (activated by 10 µM GABA in the
presence of 20 µM bicuculline) from the current activated by 200 µM
GABA without bicuculline.
Dose–response relationships were fitted with the following Hill
equation using a non-linear least-squares method:
I = Imax/(1 + (EC50/[A])nH),
where I is the peak current response at a given concentration of
agonist (A), Imaxis the maximum current response, EC50is the
concentration of agonist with half-maximal activation, and nHis the
Hill coefficient. Data were compared statistically by Student’s t test.
Statistical significance was determined at the 5% level. All results
are presented as means ± S.E.M.
N-terminal GABACreceptor antibodies
N-terminal s1 GABACreceptor antibodies were raised against the
GABAC receptor s1 subunit by synthesizing a fusion protein
N. Filippova and others
J. Physiol. 535.1
corresponding to the s1 N-terminal region (positions 14–191) with a
6His tag on the N-terminus. The specific oligonucleotide primers
used for the N-terminal fusion protein were as follows.
Forward primer (position 42–67): 5fi-CCACGCGGATCCGGCCAC
Reverse primer (position 573–538): 5fi-GACTGAGCCCAAGCTTCTA
The PCR product was subcloned, using BamHI and HindIII sites
added to the primers, into the bacterial expression vector pQE-30
(Qiagen Inc., Valencia, CA, USA). The 6His fusion protein was
expressed in Epicurian Coli BL21-Gold(DE3) pLysS cells (Stratagene,
La Jolla, CA, USA) and was purified from urea-solubilized inclusion
bodies by Ni-NTA chromatography (Qiagen) and then refolded. Mice
were injected with the antigen and the serum was purified using an
antigen-coupled, cyanogen bromide-activated column (Sigma). The
mice were humanely killed at the end of the procedure, which was
carried out under the guidelines and approval of the UAB
Institutional Animal Care and Use Committee.
Gel electrophoresis and Western blot analysis
HEK293 cells expressing s1 were lysed in cell culture lysis reagent
(Promega, Madison, WI, USA). The concentration of total cell protein
was determined using a DC protein assay kit (Bio-Rad, Hercules, CA,
USA) with bovine serum albumin as a standard. Total cell protein
from HEK293 cells and bacterial fusion proteins were separated by
SDS-PAGE, transferred to Hybond-P membrane (Amersham
Pharmacia Biotech, Piscataway, NJ, USA) and detected using the
ECL+Plus Western blotting detection system (Amersham Pharmacia
Biotech). The following dilutions ofantibody were used: anti-
N-terminal s1 GABACreceptor antibody, 1:500 for HEK293 cell
lysates and 1:2000 for E. coli fusion proteins; secondary sheep anti-
mouse Ig horseradish peroxidase-linked antibody (Amersham Life
Immunocytochemistry and laser scanning confocal
Slices or primary hippocampal cells were fixed overnight in 4 and 2%
paraformaldehyde in PBS solution, respectively, rinsed in PBS,
incubated in blocking solution (10% horse serum, 2% bovine serum
albumen in PBS), and then incubated overnight at 4°C in primary
antibody (anti-N-terminal s1 GABAC receptor antibody, 1:200).
Following washout of the primary antibody with PBS solution, slices
or cells were incubated overnight with the secondary antibody (Texas
Red-conjugated antibody, Amersham Life Sciences) and mounted.
Imaging was performed with a laser scanning confocal microscope
(LSCM; Olympus Fluoview, Mellville, NY, USA). In some cases, to
aid cell visualization, the membrane-permeable red fluorescent dye
Ro31-8222 (Roche Molecular Biochemicals) was added to the external
solution. Appropriate controls lacking primary and secondary
antibodies were performed, and background fluorescence was
adjusted for each experiment.
The following drugs were used for the experiments: bicuculline,
GABA, glutamate, glycine (all from Sigma), CNQX, DL-APV5,
3-aminopropylphosphonic acid (3-APA), (2S)(+)-5,5-dimethyl-2-
morpholineacetic acid (SCH 50911) and trans-4-aminocrotonic acid
(TACA) (all from Tocris, Ballwin, MO, USA).
Characterization of recombinant s1 receptors in
HEK293 cells after adeno-s infection
HEK293 cells were infected with adeno-s at a
concentration of 2–10 PFU cell_1. To confirm expression
of recombinant s1 receptors, N-terminal s1 GABAC
receptor antibodies were used in a Western blot analysis
Virus-mediated expression of GABACreceptors
J. Physiol. 535.1
Figure 1. Properties of recombinant
s1 receptors expressed in HEK293
cells after adeno-s infection
A, Western blot with N-terminal s1
GABACreceptor antibodies, which
recognized a bacterially synthesized
N-terminal fusion protein. The band
indicated by the asterisk probably
represents non-specific staining of an
unidentified protein. B, Western blot
from uninfected HEK293 cells (lane 1)
and HEK293 cells infected with adeno-s
(2 and 10 PFU cell_1, lanes 2 and 3,
respectively). Note the specific band of
the s1 subunit of the GABACreceptor
(50 kDa) in lanes 2 and 3. C, whole-cell
current evoked at a holding potential of
_50 mV with GABA (10 µM) in the
presence of bicuculline (30 µM) from
HEK293 cells after adeno-s infection.
Decay of the current upon GABA
removal was well described by a single
exponential component with a time
constant (r) of 7 s. D, mean
dose–response relationship for GABA-
activated current fitted with a Hill
equation (n = 4).
performed on HEK293 cells. Figure 1A demonstrates
that the N-terminal s1 antibody recognized a bacterially
synthesized N-terminal fusion protein of the human s1
subunit at a concentration of 20 and 2 ng. N-terminal s1
GABACreceptor antibodies did not recognize any specific
proteins from untransfected HEK293 cells (Fig. 1B, lane
1). However, we observed specific signals of the expected
size for the s1 subunit of GABACreceptors in HEK293
cells 24 h after adeno-sinfection at two different titres, 2
and 10 PFU cell_1(Fig. 1B, lanes 2 and 3, respectively).
Figure 1C illustrates current activated by GABA
application (10 µM) in the presence of bicuculline (30 µM)
from HEK293 cells 24 h after adeno-s infection. The
GABA-evoked current showed no desensitization, had a
linear current–voltage relationship, and was insensitive
to bicuculline (not shown). The time constant of decay
upon GABA removal (deactivation) was 8 ± 1 s (Fig. 1C,
n = 6). The GABACreceptor antagonist 3-APA (300 µM)
completely and reversibly blocked this current (n = 3).
Figure 1D is the mean dose–response relationship best
fitted with the Hill equation yielding an EC50 of
1 ± 0.3 µM, and a Hill coefficient of 2.6 ± 0.4 (n = 4).
These properties of recombinant s1 GABAC receptors
after adeno-s infection are indistinguishable from those
of s1 receptors in HEK293 cells after standard
transfection protocols (Filippova et al. 1999).
Pharmacological properties of current recorded from
hippocampal neurons in culture before and after
Before expressing s1 subunits, we first characterized the
native ligand-activated current in uninfected neurons.
Based on these studies, we divided uninfected neurons
into three groups according to the properties of the
ligand-activated current (Fig. 2Ai–iii). The first group
(Fig. 2Ai) represented about 14% (6/44 cells) of the
cultured neurons tested and had GABAAreceptors; that
is, a GABA-activated current which was inhibited by
bicuculline and had strong desensitization at high GABA
concentrations (> 100 µM). The second group (Fig. 2Aii)
represented 63% (28/44 cells) of the neurons and had
GABAAreceptors similar to group i, but also exhibited a
glutamate-activated current. Note that in both groups,
GABA (10 µM) in the presence of bicuculline (30 µM) did
not activate a current, confirming the absence of native
N. Filippova and others
J. Physiol. 535.1
Figure 2. Characteristics of ligand-activated current from hippocampal neurons in culture before
and after adeno-s infection
Ai–iii, examples of ligand-activated currents from uninfected neurons at a holding potential of _50 mV.
Note that GABA (10 µM) in the presence of bicuculline (Bicuc; 30 µM) did not activate current from group
i and ii neurons, but evoked a small current from group iii neurons. B, 3-APA (300 µM) did not block the
bicuculline-insensitive current from neurons in group iii. SCH 50911 (SCH; 50 µM) was present to block
GABAB receptors. C, the bicuculline-insensitive current from group iii neurons had a linear
current–voltage relationship. D, examples of ligand-activated currents from neurons after adeno-s
infection. Note that GABA (10 µM) in the presence of bicuculline (30 µM) evoked a GABAC-like current.
E, 3-APA (300 µM) completely and reversibly blocked the bicuculline-insensitive current that appeared
after adeno-s infection. F, dose–response relationship of bicuculline-insensitive current after adeno-s
infection. Glyc, glycine; Glut, glutamate.
GABAC-like receptors. The third group of cells
represented 23% of the population (10/44 cells). In
addition to the GABAAand glutamate currents as in
group ii, these neurons contained a glycine-activated
current (Fig. 2Aiii). Moreover, GABA (10 µM) in the
presence of bicuculline (30 µM) induced a current with an
amplitude of 40–100 pA, a linear current–voltage
relationship and fast deactivation (r < 1 s; Fig. 2C).
Unlike a GABACreceptor current, this current was not
blocked by the GABACcompetitive antagonist 3-APA
(300 µM; Fig. 2B). In fact, we observed a slight
potentiation of this current in the presence of 3-APA and
the GABABantagonist SCH50911. A similar potentiation
was observed with 3-APA alone (data not shown). Our
results suggest that rat hippocampal neurons do not
express a classic GABAC-like current, although we have
not yet identified the receptors responsible for the
bicuculline-insensitive, GABA-activated current in
neurons from group iii.
Two to three days after adeno-sinfection, GABA (10 µM),
in the presence of bicuculline (30 µM), induced a current
with a linear current–voltage relationship, no
desensitization and a slow deactivation rate (r = 6 ± 2 s,
n = 24) in all three types of cell. Figure2Dshows currents
from a type ii neuron after infection. Figure 2E
demonstrates that 3-APA (300 µM) completely and
reversibly blocked the bicuculline-insensitive GABA-
activated current (n = 8), and the GABA agonist TACA
(10 µM) evoked a current with the same amplitude as
GABA (10 µM) with or without bicuculline (30 µM). To
estimate the dose–response relationship, we applied
GABA at different concentrations in the presence of
bicuculline (30 µM; Fig.2F). The dose–response relationship
was best fitted with a single Hill equation yielding an
EC50of 1 ± 0.4 µM, and a Hill coefficient of 2.2 ± 0.4
(n = 4). These results confirm that recombinant s1
receptors were expressed in hippocampal neurons after
adeno-s infection, and had similar properties to s1
receptors after infection of HEK293 cells (Fig. 1C and D).
Level and time course of expression of recombinant
To estimate the percentage of cells that can be infected
with adenovirus, we infected HEK293 cells, 911 cells, and
hippocampal neurons with an adenovirus containing GFP
Virus-mediated expression of GABACreceptors
J. Physiol. 535.1
Figure 3. Level and time course of expression of s1 GABACreceptors in neurons and in HEK293
and 911 cell lines after adeno-s infection
A, cultured neurons were infected with adeno-GFP and visualized with the red fluorescent dye Ro31-
8222. Images of a single neuron using a confocal scanning microscope. The bottom image shows the
merging of the green and red channels. B, same as in A but a group of neurons is shown. C, images of
HEK293 cells after adeno-GFP infection and treatment with Ro31-8222. Note the 100% co-localization
of red and green fluorescence. Scale bars: 50 µm in A and C,100 µm in B. D,percentage of cells expressing
recombinant s1 receptors after adeno-sinfection of HEK293 and 911 cell lines and hippocampal neurons
over time. E,examples of GABA-activated currents on the second and fourth days after adeno-sinfection
of hippocampal cells. Note that on the fourth day, the GABAAcurrent was greatly diminished. Only a
GABAC-like current with the characteristic slow decay time was present.
(adeno-GFP). For cell visualization, the membrane-
permeable red fluorescent dye Ro31-8222 was added to
the external solution 5 min prior to examination.
Figure 3A (left) shows a dissociated hippocampal neuron
infected with adeno-GFP (top, green fluorescence) and
treated with Ro31-8222 dye (middle, red fluorescence).
The two images are merged in the bottom panel of
Fig. 3A. Confocal scanning microscopy revealed that
70–90% of the neurons in the hippocampal culture had
green fluorescence and, thus, this percentage of cells could
be potentially targeted by adenovirus (Fig. 3B). In the
case of the HEK293 and 911 cell lines, all cells had green
fluorescence and showed a cytopathic effect (Fig. 3C).
To estimate the percentage of infectible neurons that
express recombinant s1 GABACreceptors, we co-infected
neuronal cells with both adeno-GFP and adeno-s. From
12 h to 4 days after infection, we analysed the current
amplitude of recombinant s1 GABACreceptors activated
by GABA (20 µM) in the presence of bicuculline (50 µM)
and found that the amplitude of the GABACcurrent
increased over this time course. Four days after infection,
100% of the infected cells (about 90% of all neurons)
contained a GABACcurrent (Fig. 3D). In the HEK293
and 911 cell lines, 100% of the cells expressed
recombinant s1 GABACreceptors 12 h after infection
Figure 3E presents examples of GABAC- and GABAA-
activated currents in the hippocampal neurons 2 and
4 days after infection. The amplitude of the recombinant
GABAC current on the second day of infection was
100–400 pA, which was 4- to 10-fold less than that of the
typical GABAA current (1–2 nA). However, 4 days
post-infection, the GABACcurrent amplitude reached
0.8–1.5 nA, while the GABAA current decreased to
< 300 pA. The mean ratio between the GABAC and
GABAAcurrent amplitudes was 0.16 ± 0.05 (n = 10) and
8 ± 1 (n = 10) on the second and fourth days post-
infection, respectively. We did not observe a significant
change in the amplitude of the glutamate-activated
current after s1 receptor expression (290 ± 70 and
210 ± 40 pA before and at 4 days post-adeno-sinfection,
respectively; n = 5).
We also observed a decrease in the expression level of
a1b2y2 GABAAreceptors transfected into HEK293 cells
following infection with adeno-s. Seventy-two hours
after transfection and 12 h after adeno-s infection, the
amplitude of the a1b2y2 GABAA current was
335 ± 150 pA (n = 5) as compared to 1486 ± 600 pA
(n = 5) without adeno-s infection. A qualitatively similar
finding was observed when, rather than adeno-s
infection, HEK293 cells were transfected with both s1
GABACand a1b2y2 GABAAreceptors. In this case, the
GABAAcurrent was 100 ± 80 pA (n = 4) as compared to
1486 ± 600 pA (n = 5) with a1b2y2 expression alone.
N. Filippova and others
J. Physiol. 535.1
Figure 4. Recombinant s1 receptors expressed in
cultured hippocampal slices after adeno-s
A–C, images of cultured hippocampal slices after
co-infection with adeno-GFP and adeno-s using a
confocal scanning microscope. A, GFP fluorescence.
B, staining with N-terminal s1 GABACreceptor
antibodies (red). C, co-localization of GFP and s1.
Scale bar, 50 µm. D, example of a recombinant
GABACcurrent in the slice after adeno-s infection.
The current was blocked by 3-APA, an antagonist of
Recombinant s1 GABACreceptors expressed in
cultured hippocampal slices after adeno-s infection
Cultured hippocampal slices were co-infected with both
adeno-GFP and adeno-s. Three days after infection,
slices were fixed and primary N-terminal s1 GABAC
receptor and secondary Texas Red-conjugated antibodies
antibodies were used for visualization of s1 GABAC
receptors. Control non-infected slices did not exhibit
specific antibody staining (not shown). However, we
observed bright red fluorescence confirming the presence
of s1 receptors on the cell surface 3 days after adeno-s
infection (Fig. 4B). Moreover, we observed a strong
correlation between green and red fluorescence
(Fig. 4A–C) suggesting that adeno-GFP and adeno-s
infected the same neurons.
To confirm the functional expression of s1 receptors, we
employed the patch-clamp technique in the whole-cell
recording configuration. In uninfected slices, we did not
observe a native GABAC-like current (5/5 cells).
However, in adeno-s1-infected neurons, GABA (10 µM) in
the presence of bicuculline (30 µM) induced a current with
a linear current–voltage relationship, and slow decay
time (r > 10 s; 4 of 4 cells). As expected for GABAC
receptors, the current was reversibly blocked by 3-APA
(300 µM; n = 4; Fig. 4D). Thus, recombinant s1 receptors
were expressed in hippocampal slices after adeno-s
We have demonstrated that recombinant s1 GABAC
receptors could be expressed in hippocampal neurons
after adeno-s infection and these receptors exhibited
properties similar to those of recombinant s1 GABAC
receptors previously described in HEK293 cells (Filippova
et al. 1999). Furthermore, the hippocampal neurons
expressed s1 receptors regardless of their existing
complement of ligand-activated receptors.
The distribution of s1 subunits has been identified by
RT-PCR and in situ hybridization in the retina, superior
colliculus, dorsal lateral geniculate nucleus and visual
cortex (Boue-Grabot et al. 1998). In our experiments, we
did not observe specific immunostaining of hippocampal
neurons using N-terminal s1 GABACreceptor antibodies,
suggesting either the absence, or very low levels, of
native GABACreceptors in the hippocampus. Overall,
based on patch-clamp experiments and immunostaining,
we conclude that s1 GABACreceptors are not evident in
the rat hippocampus at P4–P8. The presence of other
subunit combinations (s2, s3) of GABACreceptors cannot
be confirmed or eliminated. Considering the native
bicuculline-insensitive current that was not blocked by
3-APA (Fig. 2Aiii), it is possible that GABACreceptor
subunits interact with GABAAor glycine subunits, and
form GABA-activated receptors with unexpected
An interesting observation in our study was that
expression of s1 GABACreceptors after adeno-sinfection
of hippocampal neurons diminished functional expression
of GABAA receptors. Functional coassembly of the
s1(T314A) GABACsubunit with the y2 GABAAreceptor
subunit was recently confirmed (Pan et al. 2000).
Moreover, the authors observed a dramatic decrease in
the functional expression of GABAA and glycine
receptors after co-expression with s1 GABACreceptors in
Xenopus oocytes. It is possible that in native neurons, an
interaction of recombinant s1 subunits with native y2
GABAAsubunits could replace GABAAreceptors with
s1–y2 chimeric receptors. If this occurs, the interaction
would not be evident functionally as the proposed wild-
type s1–y2 receptor has properties indistinguishable
from those of s1 alone (Pan et al. 2000). Another
possibility for the decrease in the GABAA receptor
current after expression of GABACreceptors is that the
GABACreceptors monopolize the translational machinery
of the cells. This explanation seems unlikely, since the
amplitude of the glutamate-activated current was
unchanged after GABACreceptor expression.
During the review of our manuscript, a report appeared
that also demonstrated expression of GABACreceptors by
infection with an adenovirus containing the s1 subunit
(Cheng et al. 2001) In addition to documenting functional
expression, the authors demonstrated that expression of
GABACreceptors eliminated neuronal hyperactivity and
delayed the neuronal death induced by chronic blockade
of glutamate receptors. One interesting difference
between the two studies was that these authors noted an
increased expression of GABAAreceptors after infection
with the s1 subunit as opposed to our observed decrease
in GABAAexpression. A possible explanation for this
difference could be the choice of promoter. While we
employed a CMV promoter, Cheng et al. (2001) used a
promoter from Rous sarcoma virus (RSV). The promoter
affects cell-type specificity, temporal patterns of
expression and absolute expression levels (Smith et al.
2000). Comparison of b-galactosidase or GFP expression
under the control of different viral promoters in
hippocampal neurons demonstrated a high expression
level in pyramidal neurons and low expression in granule
cells with CMV with the opposite pattern with the RSV
promoter (Smith et al. 2000). Furthermore, expression
under the CMV promoter peaked rapidly and remained
high, whereas the RSV promoter produced lower levels of
b-galactosidase that began to decrease after several days
in culture. These findings are confirmed with s1 as the
highest level of expression Chen et al. (2001) could obtain
with the RSV promoter without observing gross
morphological abnormalities was an infection of 10–20%
of the cells, whereas we estimated ~70–90% of the cells
expressed GABAC-like currents with the CMV promoter.
The choice of the expression vector could be an important
consideration in the control of the levels of GABAC
Virus-mediated expression of GABACreceptors
J. Physiol. 535.1
receptors as well as the desired spatial and temporal
expression pattern within the central nervous system.
The successful example of the employment of adenovirus
in the treatment of cystic fibrosis as well as carcinomas of
various organs, including the lung, bladder, ovary and
liver, is already well documented (Wilson, 1995; Eck et al.
1996). The high infectional capabilities of adeno-s with
respect to hippocampal neurons opens the way for gene
therapy in the treatment of CNS-related diseases such as
epilepsy. If the goal is to decrease neuronal excitability,
the delivery of GABACreceptors composed of s1 subunits
may be an optimal choice for several reasons. First, they
have a higher sensitivity for GABA compared to GABAA
receptors. Second, s1 receptors do not desensitize and
they demonstrate a slow rate of deactivation upon
agonist removal. Both of these factors could increase the
efficiency of tonic inhibition of neurons in the presence of
low concentrations of extracellular GABA. Finally, the
structure–function relationship of s1 receptors has been
well characterized allowing the use of custom-made s1
mutants with preferred properties.
AMIN, J., DICKERSON, I. M. & WEISS, D. S. (1994). The agonist
binding site of the y-aminobutyric acid type A channel is not
formed by the extracellular cysteine loop. Molecular Pharmacology
AMIN, J. & WEISS, D. S. (1994). Homomeric rho 1 GABA channels:
activation properties and domains. Receptors and Channels 2,
BOUE-GRABOT, E., ROUDBARAKI, M., BASCLES, L., TRAMU, G., BLOCH,
B. & GARRET, M. (1998). Expression of GABA receptor rho
subunits in rat brain. Journal of Neurochemistry 70, 899–907.
CHENG, Q., KULLI, J. C. & YANG, J. (2001). Suppression of neuronal
hyperexcitability and associated delayed neuronal death by
adenoviral expression of GABAC receptors. Journal of
Neuroscience 21, 3419–3428.
COSSART, R., DINOCOURT, C., HIRSCH, J., MERCHAN-PEREZ, A.,
FELIPE, J., ESCLAPEZ, M., BERNARD, C. & BEN-ARI, Y. (2001).
Dendritic but not somatic GABAergic inhibition is decreased in
experimental epilepsy. Nature Neuroscience 4, 52–62.
CUTTING, G. R., LU, L., O’HARA, B. F., KASCH, L. M., MONTROSE-
RAFIZADEH, C., DONOVAN, D. M., SHIMADA, S., ANTONARAKIS,
S. E., GUGGINO, W. B. & UHL, G. R. (1991). Cloning of the gamma-
aminobutyric acid (GABA) rho 1 cDNA: a GABA receptor subunit
highly expressed in the retina. Proceedings of the National
Academy of Sciences of the USA 88, 2673–2677.
DOUGLAS, J. T. & CURIEL, D. T. (1997). Adenoviruses as vectors for
gene therapy. Science and Medicine 4, 44–53.
ECK, S., ALAVI, J., ALAVI, A., DAVIS, A., HACKNEY, D., JUDY, K.,
MOLLMAN, J., PHILLIPS, P., WHEELDON, E. & WILSON, J. (1996)
Treatment of advanced CNS malignancies with the recombinant
adenovirus H5.101RSVTK: a phase I trial. Human Gene Therapy
ENZ, R., BRANDSTATTER, J. H., WASSLE, H. & BORMANN, J. (1996).
Immunocytochemical localization of the GABAC receptor rho
subunits in the mammalian retina. Journal of Neuroscience 16,
FEIGENSPAN, A. & BORMANN, J. (1994). Differential pharmacology of
GABAA and GABAC receptors on rat retinal bipolar cells.
European Journal of Pharmacology 288, 97–104.
FILIPPOVA, N., DUDLEY, R. & WEISS, D. S. (1999). Evidence for
phosphorylation-dependent internalization of recombinant
human s1 GABACreceptors. Journal of Physiology 518, 385–399.
HIDAKA, C., MILANO, E., LEOPOLD, P. L., BERGELSON, J. M.,
HACKETT, N. R., FINBERG, R. W., WICKHAM, T. J., KOVESDI, I.,
ROELVINK, P. & CRYSTAL, R. G. (1999). CAR-dependent and CAR-
independent pathways of adenovirus vector-mediated gene
transfer and expression in human fibroblasts. Journal of Clinical
Investigation 103, 579–587.
HOUSER, C. R., OLSEN, R. W., RICHARDS, J. G. & MOHLER, H. (1988).
Immunohistochemical localization of benzodiazepine/GABAA
receptors in the human hippocampal formation. Journal of
Neuroscience 8, 1370–1383.
KOULEN, P., BRANDSTATTER, J. H., ENZ, R., BORMANN, J. & WASSLE,
H. (1998). Synaptic clustering of GABA(C) receptor rho-subunits
in the retina. European Journal of Neuroscience 10, 115–127.
KRASNYKH, N. V., DOUGLAS, J. T. & BEUSECHEM, V. W. (2000)
Genetic targeting of adenoviral vectors. Molecular Therapy 1,
LUKASIEWICZ, P. D. & SHIELDS, C. R. (1998). Different combinations
of GABAA and GABAC receptors confer distinct temporal
properties to retinal synaptic responses. Journal
Neurophysiology 79, 3157–3167.
MACDONALD, R. L. & OLSEN, R. W. (1994). GABAA receptor
channels. Annual Review of Neuroscience 17, 569–602.
NALBANTOGLU, J., PARI, G., KARPATI, G. & HOLLAND, P. C. (1999).
Expression of the primary coxsakie and adenovirus receptor is
downregulated during skeletal muscle maturation and limits the
efficacy of adenovirus-mediated gene delivery to muscle cells.
Human Gene Therapy 10, 1009–1019.
PAN, Z., ZHANG, D., ZHANG, X. & LIPTON, S. (2000). Evidence for
coassembly of mutant GABACs1 with GABAAy2S, glycine a1
and glycine a2 receptor subunits in vitro. European Journal of
Neuroscience 12, 3137–3145.
POLENZANI, L., WOODWARD, R. & MILEDI, R. (1991). Expression of
mammalian y-aminobutyric acid receptors with distinct
pharmacology in Xenopus oocytes. Proceedings of the National
Academy of Sciences of the USA 88, 4318–4322.
ROELVINK, P. W., LIZONOVA, A., LEE, J. G., LI, Y., BERGELSON,
J. M., FINBERG, R. W., BROUGH, D. E., KOVESDI, I. & WICKHAM,
T. J. (1998) The coxsackievirus-adenovirus receptor protein can
function as cellular attachment protein for adenovirus serotypes
from subgroup A, C, D, E, and F. Journal of Virology 72,
SHIMADA, S., CUTTING, G. & UHL, G. R. (1992). y-Aminobutyric acid
A or C receptor? y-Aminobutyric acid rho 1 receptor RNA induces
bicuculline-, barbiturate-, and benzodiazepine-insensitive
y-aminobutyric acid responses in Xenopus oocytes. Molecular
Pharmacology 41, 683–687.
SMITH, R. L., TRAUL, D. L., SCHAACK, J., CLAYTON, G. H., STALEY,
K. J. & WILCOX, C. L. (2000). Characterization of promoter
function and cell-type-specific expression from viral vectors in
the nervous system. Journal of Virology 74, 11254–11261.
STRATA, F. & CHERUBINI, E. (1994). Transient expression of a novel
type of GABA response in rat CA3 hippocampal neurones during
development. Journal of Physiology 480, 493–503.
STOPPINI, L., BUCHS, L.-A. & MULLER, D. (1991). A simple method
for organotypic cultures of nervous tissue. Journal of Neuroscience
Methods 37, 173–182.
N. Filippova and others
J. Physiol. 535.1
WALTERS, R. W., GRUNST, T., BERGELSON, J. M., FINBERG, R. W.,
WELSH, M. J. & ZABNER, J. (1999). Basolateral localization of fiber
receptors limits adenovirus infection from the apical surface of
airway epithelia. Journal of Biological Chemistry 274,
WANG, T. L., GUGGINO, W. B. & CUTTING, G. R. (1994). A novel
gamma-aminobutyric acid receptor subunit (rho 2) cloned from
human retina forms bicuculline-insensitive homooligomeric
receptors in Xenopus oocytes. Journal of Neuroscience 14,
WILSON, J. (1995). Gene therapy for cystic fibrosis: Challenges and
future directions. Journal of Clinical Investigation 96, 2547–2554.
ZABNER, J., FREIMUTH, P., PUGA, A., FABREGA, A. & WELSH, M. J.
(1997). Lack of high affinity fiber receptor activity explains the
resistance of ciliated airway epithelia to adenovirus infection.
Journal of Clinical Investigation 100, 1144–1149.
ZHANG, D., PAN, Z.-H., ZHANG, X., BRIDEAU, A. D. & LIPTON, S. A.
(1995). Cloning of a y-aminobuturic acid type C receptor subunit
in rat retina with a methionine residue critical for picrotoxinin
channel block. Proceedings of the National Academy of Sciences of
the USA 92, 11756–11760.
ZIMPRICH, F., ZEZULA, J., SIEGHART, W. & LASSMANN, H. (1991).
Immunohistochemical localization of the a1-, a2- and a3-subunit
of the GABAAreceptor in the rat brain Neuroscience Letters 127,
The authors would like to thank the UAB High Resolution Imaging
Facility for the use of the confocal microscope. The work was
supported by NS40027.
Natalia Filippova and Anna Sedelnikova contributed equally to this
D. S. Weiss: Department of Neurobiology, University of Alabama at
Birmingham School of Medicine, 1719 Sixth Avenue South CIRC 410,
Birmingham, AL 35294-0021, USA.
Virus-mediated expression of GABACreceptors
J. Physiol. 535.1