Download Molecular and Electrophysiological Evidence for a GABAC Receptor

0013-7227/00/$03.00/0
Endocrinology
Copyright © 2000 by The Endocrine Society
Vol. 141, No. 5
Printed in U.S.A.
Molecular and Electrophysiological Evidence for a
GABAC Receptor in Thyrotropin-Secreting Cells*
ERIC BOUE-GRABOT†, ANNE TAUPIGNON, GÉRARD TRAMU,
AND
MAURICE GARRET
Laboratoire de Neurophysiologie Centre National de la Recherche Scientifique-Unité Mixte de
Recherche 5543 (E.B.-G., A.T., M.G.), Université Victor Segalen Bordeaux 2, 33076 Bordeaux
Cedex, France and Laboratoire de Neurocytochimie Fonctionnelle (G.T.), Centre National
de la Recherche Scientifique-Unité Mixte de Recherche 5807, Université de Bordeaux 1,
33405 Talence Cedex, France
ABSTRACT
In the pituitary, GABA regulates the release of several hormones
via different receptors. GABAC receptors are heterooligomers that
differ from GABAA receptors in that they contain ␳-subunits and are
insensitive to bicuculline. However, molecular and functional evidence for the presence of GABAC receptors outside the retina has yet
to be established. The present work was performed on guinea pig and
rat pituitaries. Both Northern blot and RT-PCR analysis showed that,
although ␳1- and ␳2-subunits were expressed at similar levels in the
rat retina, ␳1 messenger RNA (mRNA) was enriched, relative to ␳2
mRNA in the rat pituitary. Northern blot experiments also showed
that, in the pituitary, ␳1 and ␳2 mRNAs are shorter in size than those
expressed in the retina. The use of a subunit-specific antibody revealed colocalization of ␳1-subunit and anti-TSH labeling on rat pituitary sections. TSH guinea pig pituitary cells were also labeled with
a ␳-subunit antiserum. Moreover, whole-cell patch clamp on single
guinea pig TSH cells showed that GABA induced a bicucullineinsensitive Cl⫺ current. In contrast to the Cl⫺ current generated by
GABAC receptors in the retina, the bicuculline-insensitive Cl⫺ currents in TSH cells quickly desensitized. These results suggest that a
novel GABAC receptor may regulate TSH secretion and that the
structure and/or biochemical regulation of this pituitary receptor is
different from that found in the retina. (Endocrinology 141: 1627–
1632, 2000)
T
when recombinant rat ␳1- and ␳2-subunits were coexpressed in oocytes (11). These results, combined with studies demonstrating expression of ␳-subunits in rat bipolar
cells (12, 13), suggested that retinal GABAC receptors in
rats were heteromeric and composed of at least ␳1- and
␳2-subunits. Bicuculline-insensitive GABA effects have
also been reported in various parts of vertebrate brains (14,
15), and we have previously reported the expression of ␳1and ␳2-subunits in restricted brain domains that may contain functional GABAC receptors (16). So far, however, a
clear correlation between the expression of ␳-subunits and
functional bicuculline-insensitive GABA-gated Cl⫺ channels has only been demonstrated in retina cells (6, 12, 13).
GABA is also active in the pituitary, where it modulates
the release of several hormones via A-, B-, and C-type
GABA receptors (17–20). We and others have previously
demonstrated the expression of different GABAA receptor
subunits in the anterior lobe of the rat pituitary (21, 22).
Our RT-PCR experiments have also shown the expression
of the ␳1-subunit in this tissue (22). Because GABAC receptors in the rat retina are believed to be composed of ␳1and ␳2-subunits, in the present study we have further
examined the expression of ␳-subunits in the anterior pituitary by combining RT-PCR and Northern blot. We also
report the immunodetection of the ␳-subunits in rat and
guinea pig TSH cells. For technical reasons, the electrophysiological characterization of the GABA-induced current was conducted on guinea-pig pituitary cells. This
analysis revealed the presence of a bicuculline-insensitive
GABA-gated chloride current in individual thyrotroph
cells. Our results suggest that this current is mediated by
a novel GABAC receptor.
HE MAJOR INHIBITORY neurotransmitter, ␥-aminobutyric acid (GABA), activates three pharmacologically and structurally different classes of GABA receptors: GABAA, GABAB, and GABAC receptors. GABAB
receptors are coupled to G protein, activated selectively by
baclofen, inhibited by saclofen, and insensitive to bicuculline. GABAA and GABAC receptors are both ligandgated Cl⫺ channels selectively activated by muscimol.
GABAA receptors are antagonized by bicuculline and have
allosteric binding sites for pharmacologically important
drugs (for review, see Ref. 1). They are formed by differential assembly of multiple subunits (␣1– 6, ␤1–3, ␥1–3, ␦,
and ⑀) (2– 4). By contrast, GABAC receptors, identified in
bovine, rat, and perch retinas (5–7), are neither sensitive
to bicuculline nor modulated by benzodiazepines, barbiturates, and steroids. Human ␳-subunits, cloned from
human retina libraries and expressed in oocytes, form
homooligomeric chloride channels that share many pharmacological properties with retinal GABAC receptors (8 –
10). In the rat retina, bicuculline-insensitive GABA-gated
Cl⫺ channels, localized in bipolar cells, are not inhibited
by picrotoxin (6). The native response was mimicked only
Received October 12, 1999.
Address all correspondence and requests for reprints to: M. Garret,
Laboratoire de Neurophysiologie, Unité Mixte de Recherche 5543,
Université de Bordeaux 2, 146 rue Léo-Saignat 33076 Bordeaux cedex,
France. E-mail: [email protected].
* This work has been supported by Centre National de la Recherche
Scientifique, the University of Bordeaux 2, and the Conseil Regional
Aquitaine.
† Present address: Institut neurologique de Montréal, Universite
McGill, Montréal H3A 2B4, Canada.
1627
1628
GABAC RECEPTOR IN THYROTROPIN-SECRETING CELLS
Materials and Methods
RNA isolation, complementary DNA (cDNA) probes, and
Northern blot analysis
In the absence of any information on guinea pig ␳-subunit gene
sequences, these experiments were done on rats. These methods, using
tissues from adult Wistar rats, have been reported extensively (16). The
␳1⫹2 cDNA probe derived from a conserved sequence in the presumed
extracellular domain. The ␳1 and ␳2 cDNA probes were chosen from the
␳1 and ␳2 variable intracellular domain, respectively. High-stringency
hybridization (Saline-Sodium-Phosphate-EDTA 0.1⫻, 65 C) was carried
out; then the filters were exposed to x-ray films for 1–10 days.
PCR analysis
The primers used for gene expression analysis were selected from
domains conserved among the ␳1–3-subunits. PCR experiments and
amplification product analyses were conducted as described (16); 1 ␮l
35
S deoxycycidine triphosphate (800 Ci/mmol, 10 mCi/ml, Amersham
Pharmacia Biotech, Saclay, France) was also added to label the PCR
product. Negative control experiments were run with water instead of
cDNA, and with RNA samples treated like RT-PCR template (except that
reverse transcriptase was omitted). No amplification products were
found in the control experiments (not shown). The 35S-labeled PCR
products were separated on an 8% acrylamide gel and vacuum dried.
Radiolabeling was detected by exposure to x-ray films or analyzed on
a Molecular Dynamics, Inc. Phosphorimager system (Saclay, France).
Immunohistochemistry
Adult Wistar rats (300 g) and guinea pigs (600 g) were overdosed with
pentobarbital and immediately perfused transcardially with 150 ml saline solution. Pituitaries were rapidly dissected out, fixed for 1 h by
immersion in a solution containing 4% paraformaldehyde and 0.2%
picric acid in 0.1 m phosphate buffer, and soaked overnight at 4 C in
phosphate buffer containing 20% sucrose. Then the pituitary was frozen
and cut on a cryostat. Regularly spaced, 10-␮m, horizontal sections were
collected. Polyclonal antisera directed against ␳1-subunit-specific peptide (16), ␳-subunits (Enz et al., 1996), and pituitary hormones (TSH, FSH,
ACTH, and PRL) raised in rabbits were used as primary antibodies. The
antihormone antisera are extensively used, and their specificity has been
evaluated in previous studies: anti-hTSH and anti-hFSH (23), antiACTH (24), and anti-rPRL (25) (h, human; r, rat).
Pituitary sections were incubated first for 18 h at room temperature
with purified ␳1 antiserum (diluted 1:200) or with ␳ antiserum (1 : 100)
in 0.01 m veronal buffer (VB) containing 0.2% Triton X-100. Sections were
then incubated for 2 h at room temperature in goat antirabbit peroxidaselinked IgGs (1:200; Jackson ImmunoResearch Laboratories, Inc.). The
peroxidatic activity was revealed using 4 chloro-1 naphthol chromogen.
It was not possible to identify rat GH- and guinea pig TSH-producing
cells after immunodetection of ␳-subunits, probably because of alteration of the antigens during the elution procedure.
The first staining was photographed, then the blue reaction products
were removed by immersion in acetone. Antibodies were then eluted by
gentle stirring in a mixture made of 1 vol of 2.5% KMnO4, 1 vol of 5%
H2SO4, and 200 vol of distilled H2O for 1 min (26). Sections were washed
in VB before immunodetection of the second antigen: sections were
incubated for 2 h with anti-rPRL (1:400), anti-hTSH (1:1000), anti-ACTH
(1:500), or anti-hFSH (1:500). After washing with VB, goat antirabbit
peroxidase-linked IgGs (1:150, Jackson ImmunoResearch Laboratories,
Inc., Asnières, France) was applied for 1 h. Reaction products were
formed with 3,3⬘ diaminobenzidine tetra HCL, then photographed.
No immunoreactivity was observed when elution efficiency was controlled by incubation of sections with normal rabbit serum instead of
hormone antibodies or when specificity controls were performed by
preincubation of the purified ␳1 antiserum (50 ␮g/ml) with the peptide
antigen ␳1N (100 ␮g/ml, not shown).
Guinea-pig pituitary cell culture and patch clamp recording
For studies on isolated thyrotrophs, pituitaries from 250 –350-g female guinea pigs were used. The anterior pituitary was first separated
from the posterior pituitary. It was further dissected to isolate the rostral
Endo • 2000
Vol 141 • No 5
(anteromedian and lateral) part of the pars distalis known, in guinea pigs,
to contain most thyrotrophs (27). The cell dissociation method was
adapted from that described (19). Triturating took place in calcium- and
magnesium-free Hanks’ medium containing 2 mm EGTA. Yield was
5– 8 ⫻ 105 cells per gland. Cells were harvested by gentle centrifugation
and plated onto glass coverslips in 35-mm Petri dishes at a density of
2.5 ⫻ 105 cells/dish. Experiments were performed on days 1 and 2.
Guinea pig thyrotrophs have been described as voluminous cells,
always much bigger than any other cells (27). Indeed, coverslips viewed
at ⫻1000, exhibited conspicuously large round cells (diameter ⱖ 25 ␮m).
We selected such cells for our whole-cell patch clamp experiments. In
some experiments, cell type identification was confirmed, after recording on scored coverslips, by immunocytochemical labeling using our
anti-hTSH antibody (not shown). The patch amplifier was a RK-300 unit
(Biologic, Claix, France). The bath medium (38 C) contained Hanks’
saline buffered solution with 10 mm HEPES. To test the dependency of
the reversal potential of GABA-induced currents on external Cl⫺, a low
Cl⫺ Hanks’ saline was used. It contained 45 mm Cl⫺ instead of 149.7 mm
Cl⫺, and NaCl was replaced by Na methanesulfonate. Pipettes, coated
with sylgard and fire-polished, had an average resistance of 3 M⍀ when
filled with (in mm): KCl, 120; HEPES, 10; EGTA, 11; ATP-Mg, 2; GTP-Na,
0.4; 280 –300 mosmol/liter; pH 7.25. The expected ECL was ⫺2 mV or
(with the low Cl⫺ Hanks’ medium) ⫹29 mV. Drugs were applied to
isolated cells from double-barreled pipettes by pneumatic ejection.
Results
Expression of ␳-subunit messenger RNAs (mRNAs) in
pituitary by Northern blot and RT-PCR analysis
To examine the expression of ␳-subunit mRNAs, Northern
blots with poly(A)⫹-selected RNA extracted from retina and
pituitary were hybridized with probes derived from ␳-subunit cDNAs. The ␳1⫹2 probe (Fig. 1A), derived from a conserved sequence among ␳-subunits, demonstrated hybridization to mRNAs in retina (5 and 2.4 kb) and pituitary (4.5,
2, and 1 kb). When the same membrane was hybridized with
a ␳1-specific probe (Fig. 1B), only the 5- and 4.5-kb mRNAs
were detected in eye and pituitary, respectively. In the same
way, a ␳2-specific probe (Fig. 1C) revealed the 2.4-kb mRNA
in the retina, and the 2- and 1-kb mRNAs in the pituitary. The
products of the ␳1-subunit gene (5 and 4.5 kb), as of the ␳2
FIG. 1. Northern blot analysis of rat ␳1 and ␳2-subunit expression.
Poly A⫹ RNA was extracted from the retina (R) and pituitary (P). Ten
micrograms of RNAs were loaded on the gel, electrophoresed, transferred, and subjected to RNA hybridization analysis. Northern blot
membranes were hybridized, stripped, and reprobed with successive
probes derived from a region common to ␳1 and ␳2 (A), or exclusively
from ␳1 (B) and ␳2 (C) GABA receptor subunits, respectively. A glyceraldehyde 3-phosphate dehydrogenase probe was used to control for
variations in sample loading (D). Estimated sizes for each band shown
on the right were determined using an RNA ladder.
GABAC RECEPTOR IN THYROTROPIN-SECRETING CELLS
gene (2.4 and 2 kb), may be attributable to alternative splicing
or alternative polyadenylation between retina and pituitary.
It should be noted that the 2-kb ␳2 signal in the pituitary was
very faint, suggesting a low expression level. The 1-kb band,
revealed by the ␳1⫹2 and ␳2 probes in the pituitary, may
reflect an alternative splicing of the ␳2 gene product unlikely
to encode a functional GABA receptor subunit (approximately 450 amino acids). Correction for the variation in sample loading, using the ubiquitous G3PDH probe (Fig. 1D),
indicated that the expression levels of ␳-subunit genes were
within the same range in the retina and the pituitary.
We further determined, by RT-PCR, which ␳-subunits
were present in the anterior pituitary, using primers previously validated in experiments with RNA extracted from the
retina and brain (16). We used a set of primers, rhoA and rhoB
(Fig. 2A), from conserved regions among ␳1–3-subunits. An
aliquot of each PCR amplification product (330 bp) was digested with the appropriate restriction enzyme: EcoRI restriction hydrolysis of the PCR product obtained with ␳1
transcript should produce 164- and 166-bp fragments. In the
same way, BglII restriction enzyme should produce 137- and
193-bp fragments from ␳2 cDNA, and XbaI restriction enzyme should produce 51- and 279-bp fragments from ␳3
cDNA (Fig. 2A). Amplification, using retina and antepituitary cDNA, yielded a band of the expected size, 330 bp. The
PCR product from the retina, verified by restriction pattern
analysis, contained ␳1, ␳2, and ␳3-subunits, as reported (8, 28,
29). The PCR products obtained from the anterior pituitary
were digested with EcoRI and BglII restriction enzymes but
not with the ␳3-specific restriction enzyme XbaI. These results indicate that, in the anterior pituitary, the ␳1 mRNA
FIG. 2. Autoradiogram showing restriction analysis of RT-PCR products of GABA receptor ␳1–3-subunits. A schematic representation of
the PCR primers and restriction enzyme sites used in the study is
shown at the top of each autoradiogram. Experiments were performed
on rat retina (Rt) and rat anterior pituitary tissue (AP). A, Restriction
analysis of GABA receptor ␳1–3-subunit RT-PCR products were digested with EcoRI (E), BglII (B), and XbaI (X) restriction enzymes
specific for ␳1-, ␳2-, and ␳3-subunits, respectively. Restriction DNA
fragments from ␳1-, ␳2-, and ␳3-subunits are indicated on the left of
the gel, and the sizes of fragments are shown on the right. B, Restriction analysis of GABA receptor ␳1- and ␳2-subunit RT-PCR products were digested with EcoRI (E) and BglII (B) restriction enzymes
specific for ␳1- and ␳2-subunits, respectively. DNA restriction fragments from ␳1- and ␳2-subunits are indicated on the left of the gel, and
the sizes of fragments are shown on the right. The ratio of ␳2 to ␳1
cDNA products from four independent reactions is shown at the bottom of B.
1629
level was higher than that of ␳2 mRNA and that the ␳3
subunit gene was not expressed (Fig. 2).
Because rhoB primer is degenerate and we wished to examine more closely the relative levels of ␳1 and ␳2 mRNAs
in the anterior pituitary, we used a rhoC primer (Fig. 2B) from
the ␳1 and ␳2 nucleotidic sequences (rhoC does not amplify
␳3-subunit mRNA). The ␳1- and ␳2-subunit cDNA sequences
have a 100% match to rhoA and rhoC oligonucleotide primers
and compete for primer binding and amplification (for review, see Ref. 30). PCR fragments were identified by restriction analysis: EcoRI restriction hydrolysis of the PCR product
obtained with ␳1 transcript should produce 166- and 101-bp
fragments, and BglII restriction hydrolysis of the PCR product from ␳2 cDNA should produce 137- and 130-bp fragments. Figure 2B showed that, in the retina and anterior
pituitary, the ratios of ␳2- vs. ␳1-mRNA were different. In
conclusion, the comparison of both tissues, by Northern blot
and RT-PCR analyses, indicated that, in the rat pituitary, the
␳1 mRNA level was considerably higher than that of ␳2
mRNA.
Localization of GABAC receptor ␳1-subunit in the pituitary
Horizontal sections of rat pituitary were immunostained
with an antibody against the ␳1-subunit (16). Figure 3 shows
micrographs of the rat pituitary. As shown in Fig. 3A, immunoreactivity was found on anterior lobe cells. No significant immunolabeling was observed in the neurointermediate lobe.
To identify ␳1-expressing cells, among the various types of
adenohypophyseal cells, successive staining experiments
were performed using first the antiserum to the ␳1-subunit
and then antisera to the various pituitary hormones (see
Materials and Methods). In sections examined, all cells that
displayed ␳1-subunit immunoreactivity (Fig. 3B1) were also
stained with anti-TSH antibodies after the elution of ␳1antibodies (Fig. 3B2). However, 20% of anti-TSH-labeled cells
showed little or no immunoreaction to the ␳1-subunit (37/
200), indicating that some thyrotrophs may not express the
␳1-subunit. On the other hand, ␳1-subunit immunoreactivity
(Fig. 3C1) was never colocalized with FSH (Fig. 3C2), ACTH,
and PRL (not shown) staining, demonstrating that gonadotrophs, corticotrophs, and lactotrophs are devoid of ␳1-subunit immunoreactivity. Because the ␳1-antiserum is directed
against the N-terminal peptide (15 amino acids) that varies
from one species to another, we used an antiserum directed
against the N-terminal region (positions 16 –171) of the ␳1subunit that recognizes the ␳1-, ␳2-, and ␳3-subunits (13). The
pattern of ␳ labeling observed (Fig. 4B) was consistent with
TSH cells (Fig. 4A) within the guinea pig pituitary, strongly
suggesting that, in both species, ␳1 is expressed in
thyrotrophs.
Whole-cell patch clamp recording.
In our primary cultures of normal male rat, TSH-positive
cells after immunocytochemical staining accounted for only
a very small proportion of cells (ⱕ2%), in agreement with
published data (31). Moreover, these cells had no specific
morphological features, so they were difficult to identify
among other cells under a light microscope. We therefore
1630
GABAC RECEPTOR IN THYROTROPIN-SECRETING CELLS
FIG. 3. Localization of ␳1-subunit and TSH or FSH by successive
immunohistochemical staining on horizontal sections of rat pituitary.
A, Immunodetection of ␳1-subunit protein. B1, B2 and C1, C2, Successive localization on the same section of 2 antigens. B, Section
incubated first with ␳1-specific antibodies (B1) and then with antiTSH (B2), after decolorizing and elution of ␳1-antibodies. Arrows
mark cells stained by both antibodies, displaying similar labeling
pattern. C, Section incubated successively with ␳1-specific antibodies
(C1) and then with FSH hormone antibodies (C2). Arrows show a
nonoverlapping distribution of ␳1-subunit and FSH. IL, intermediate
lobe. Calibration bars: A, 100 ␮m; B and C, 50 ␮m.
used guinea pig pituitaries, because: 1) the rostral part of the
gland is especially rich in thyrotrophs (see Materials and
Methods), thus making it possible to enrich the culture in TSH
cells; and 2) these cells were easily identified by their size,
much larger than that of any other cell type (27).
As shown in the example in Fig. 5A, application of GABA
(10 ␮m) to such a large cell evoked an inward desensitizing
whole-cell current that was insensitive to bicuculline (100
␮m). It ran down quickly, as a further application of GABA
evoked no current. In some experiments, cells were plated
onto scored coverslips, and the coordinates of each recorded
cell were registered. Cell types were then identified by immunocytochemical labeling using the anti-hTSH antibody.
All the bicuculline-insensitive cells recorded were positive
Endo • 2000
Vol 141 • No 5
FIG. 4. Immunoreactivity for TSH and ␳-subunits of the GABAC receptor on sagittal sections through guinea pig pituitary. A, Low magnification showing TSH-specific immunostaining of cells essentially
localized in the rostroventral part of the anterior pituitary; B, neighboring section labeled with ␳ antibody; C, higher magnification of the
microscopic field boxed in B, showing typical cytoplasmic staining of
large cells; PS, pituitary stalk; calibration bars: A and B, 250 ␮m; C,
50 ␮m.
after anti-TSH labeling (not shown). In the same culture
plates, smaller cells displayed a GABA-induced current that
was completely and reversibly abolished by bicuculline (Fig.
5B). Unlike the bicuculline-sensitive currents that did not run
down quickly, the bicuculline-insensitive currents ran down
in less than 3 min (Fig. 5C). In both cases, however, the
GABA-activated current was not inhibited by the GABAB
inhibitor saclofen (10 ␮m; not shown). Furthermore, the current-voltage relationship of the bicuculline-insensitive current (Fig. 5D) revealed a mean measured reversal potential
(⫺0.56 ⫾ 1.7 mV; n ⫽ 5) close to the chloride Nernst potential.
GABAC RECEPTOR IN THYROTROPIN-SECRETING CELLS
1631
Discussion
FIG. 5. Bicuculline-insensitive GABA-activated membrane currents
are found in isolated guinea pig pituitary cells. Membrane currents,
activated by 10 sec application of GABA or GABA plus bicuculline,
were recorded from cells that were presumably thyrotrophs (A) or
lactotrophs (B), at a holding potential of ⫺60 mV, at various times (45,
90, and 135 sec for the first, second, and third traces, respectively),
after formation of the whole-cell recording mode. Mean initial responses from bicuculline-insensitive cells were ⫺125 ⫾ 32 pA (mean
cell capacitance: 24 ⫾ 4 pF, n ⫽ 10). Those of bicuculline-sensitive cells
were ⫺213 ⫾ 54 pA (mean cell capacitance: 8 ⫾1 pF, n ⫽ 10). Bars
show the application periods of the drugs. C, Time-dependence of
bicuculline-insensitive (squares) and bicuculline-sensitive (circles)
GABA-activated membrane currents. To collate data from separate
cells, currents were normalized to the initial response to 10 ␮M GABA,
recorded in each cell after formation of the whole-cell recording mode,
defined as 100%. All points are mean ⫹ SEM. D, Current voltage curve
of a bicuculline-insensitive current obtained with a fast voltage ramp.
Background current was subtracted. E, Membrane current activated
by GABA and bicuculline in an HEK-293 cell transfected with ␳1subunit cDNA at a holding potential of ⫺60 mV.
Furthermore, when the external Cl⫺ was changed, the reversal potential changed as expected for a current carried by
chloride ions. It shifted by ⫹ 26 ⫾ 2 mV (n ⫽ 2) when external
Cl⫺ was lowered from 149.7 to 45 mm (not shown). This
strongly suggested that GABA opened bicuculline-insensitive Cl⫺ selective channels. Using the same recording
method, this rapid desensitization of the bicuculline-insensitive current was not found in HEK-293 cells (Fig. 5E) transfected with rat ␳1-subunit cDNA (16).
We demonstrate here, for the first time outside the retina,
the presence of a ␳1-subunit polypeptide restricted to one cell
type in the endocrine pituitary. Moreover, we show that
guinea pig TSH cells express a functional bicuculline-insensitive GABA-gated Cl⫺ channel. Because it is now evident
that the subunit composition of GABA-gated Cl⫺ channels
influences their pharmacological and electrophysiological
properties (1), it is of interest that the expression profile of
␳-subunit mRNAs and GABAC functional behavior suggest
different GABAC receptors in the pituitary and retina.
Northern blot analyses revealed that ␳-subunit mRNAs
were expressed in the pituitary in sufficiently high amounts
to be detected by Northern blot. The slight difference in
mRNA sizes of ␳1 and ␳2 mRNAs in the retina and pituitary
suggested tissue-specific RNA splicing. It would be worthwhile to screen a pituitary library to determine whether these
alternative splicings result in alternative coding sequences,
as reported for the retinal ␳1-subunit (32). The functional
alternatively spliced ␳1-subunits described in that report did
not show any apparent difference in channel behavior. The
presence of a 1-kb mRNA, revealed by the ␳2 probe in the
pituitary, suggests an aberrantly spliced form, as reported for
this subunit in the human retina, as well as for other GABAA
receptor subunits (4, 8). However, an alternative spliced form
of the ␳2-subunit containing features common to members of
the GABA-gated chloride channel family (i.e. a signal peptide, a Cys-Cys loop, and four transmembrane domains)
cannot be completely ruled out. RT-PCR experiments used to
compare the ratio of ␳-subunit mRNAs in the rat pituitary
and retina confirmed that ␳1 was the main GABAC receptor
subunit expressed in the pituitary, whereas ␳1 and ␳2 were
equally expressed in the retina. Taken together, our experiments clearly showed a tissue-specific expression of ␳subunit genes in the retina and pituitary and suggest that the
molecular composition of the GABAC receptor is different in
these two tissues.
Effects of GABA on TSH and LH release via so-called
bicuculline-insensitive GABAA receptors, have been reported (18, 20). We showed that the anti-FSH antibody well
known as a gonadotroph marker (33) did not label cells
stained by the ␳1-antibody. We also showed that all the rat
pituitary cells labeled by the ␳1-antibody were labeled by an
anti-TSH antibody, demonstrating cell-specific expression of
the ␳1 protein. Moreover, in single identified guinea-pig
TSH-producing cells, GABA induced a bicuculline-insensitive Cl⫺ current. The whole-cell GABA-activated membrane
current displayed a rapid rising phase followed by a decrease
(despite continued application of GABA) consistent with
desensitization. Although GABAC currents in rat bipolar
cells do not desensitize, desensitizing GABAC receptormediated currents have been reported in carp bipolar cells
(34). The bicuculline-insensitive GABA-gated Cl⫺ current in
thyrotrophs also showed a rapid down-regulation of the
whole-cell current. It is known that the rat retinal GABAC
receptor is down-regulated by protein kinase C (35). It is also
well established that the amplitude and desensitization kinetics response of GABAA receptor, like other ligand-gated
channels, depend both on their phosphorylation states and
1632
GABAC RECEPTOR IN THYROTROPIN-SECRETING CELLS
their subunit composition (36, 37). It can be hypothesized that
pituitary GABAC receptors contain an alternative ␳1-subunit.
In that case, the alternative sequence would be in the intracellular domain that is the target of protein kinases (36). It is
also possible that TSH cells contain protein kinases not
present in mammalian retinal bipolar cells. The rapid rundown of the bicuculline-insensitive GABA-gated Cl⫺ current
expressed in thyrotrophs also suggest that such channels
may have been overlooked in the brain. Because ␳-subunits
may be differentially expressed in rats and guinea pigs, it will
be necessary to find a method for enriching rat primary
cultures in TSH cells. This will make it possible to fully
characterize the pharmacology and correlate the electrophysiological properties of the pituitary GABAC receptor to its
molecular structure.
If GABAC receptors can be defined by a bicucullineinsensitive GABA-gated Cl⫺ channel and the presence of
␳-subunits, it would be reasonable to propose the presence
of a ␳1-subunit-containing GABAC receptor in mammalian
thyrotrophs. GABA modulates the pituitary hormone secretion by acting directly on the endocrine tissue that expresses
several receptor subunits (21, 22). However, so far, the identity of the GABA receptor types expressed by any individual
pituitary cell type is not known. The identification of a
GABAC receptor in thyrotrophs may be physiologically relevant, because it may account for the modulation of TSH
secretion (19). In this respect, it will be of interest to characterize the transduction pathway(s) linking the activation of
GABAC receptors to the control of hormonal secretion.
In conclusion, we propose that, besides the retina, the
pituitary also expresses a functional ␳1-subunit-containing
GABAC receptor. Our data provide the first evidence for the
hypothesis (32, 38) that GABAC receptors may be more complex than previously thought, in terms of heterogeneity and
modulation.
Acknowledgments
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
We would like to thank B. Dufy for his support, P. Séguéla and N.
Guerineau for comments on the manuscript, and Heinz Wässle for
providing the antibody directed against ␳-subunits.
27.
References
28.
1. Sieghart W 1995 Structure and pharmacology of gamma-aminobutyric acid A
receptor subtypes. Pharmacol Rev 47:181–234
2. Macdonald RL, Olsen RW 1994 GABAA receptor channels. Annu Rev Neurosci 17:569 – 602
3. Davies PA, Hanna MC, Hales TG, Kirkness EF 1997 Insensitivity to anaesthetic agents conferred by a class of GABA(A) receptor subunit. Nature
385:820 – 823
4. Garret M, Bascles L, Boue-Grabot E, Sartor P, Charron G, Bloch B, Margolskee
RF 1997 An mRNA encoding a putative GABA-gated chloride channel is expressed in the human cardiac conduction system. J Neurochem 68:1382–1389
5. Polenzani L, Woodward RM, Miledi R 1991 Expression of mammalian gamma-aminobutyric acid receptors with distinct pharmacology in Xenopus oocytes. Proc Natl Acad Sci USA 88:4318 – 4322
6. Feigenspan A, Wassle H, Bormann J 1993 Pharmacology of GABA receptor
Cl- channels in rat retinal bipolar cells. Nature 361:159 –162
7. Qian H, Dowling JE 1993 Novel GABA responses from rod-driven retinal
horizontal cells. Nature 361:162–164
8. Cutting GR, Lu L, O’Hara BF, Kasch LM, Montrose-Rafizadeh C, Donovan
DM, Shimada S, Antonarakis SE, Guggino WB, Uhl GR 1991 Cloning of the
gamma-aminobutyric acid (GABA) ␳1 cDNA: a GABA receptor subunit highly
expressed in the retina. Proc Natl Acad Sci USA 88:2673–2677
9. Cutting GR, Curristin S, Zoghbi H, O’Hara B, Seldin MF, Uhl GR 1992
Identification of a putative gamma-aminobutyric acid (GABA) receptor subunit ␳2 cDNA and colocalization of the genes encoding ␳2 (GABRR2) and ␳1
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
Endo • 2000
Vol 141 • No 5
(GABRR1) to human chromosome 6q14 – q21 and mouse chromosome 4.
Genomics 12:801– 806
Shimada S, Cutting G, Uhl GR 1992 Gamma-aminobutyric acid A or C
receptor? Gamma-aminobutyric acid ␳1 receptor RNA induces bicuculline-,
barbiturate-, and benzodiazepine-insensitive gamma-aminobutyric acid responses in Xenopus oocytes. Mol Pharmacol 41:683– 687
Zhang D, Pan ZH, Zhang X, Brideau AD, Lipton SA 1995 Cloning of a
gamma-aminobutyric acid type C receptor subunit in rat retina with a methionine residue critical for picrotoxinin channel block. Proc Natl Acad Sci
USA 92:11756 –11760
Enz R, Brandstatter JH, Hartveit E, Wassle H, Bormann J 1995 Expression of
GABA receptor ␳1- and rho2-subunits in the retina and brain of the rat. Eur
J Neurosci 7:1495–1501
Enz R, Brandstatter JH, Wassle H, Bormann J 1996 Immunocytochemical
localization of the GABAc receptor ␳-subunits in the mammalian retina. J Neurosci 16:4479 – 4490
Arakawa T, Okada Y 1988 Excitatory and inhibitory action of GABA on
synaptic transmission in slices of guinea pig superior colliculus. Eur J Pharmacol 158:217–224
Drew CA, Johnston GA 1992 Bicuculline- and baclofen-insensitive gammaaminobutyric acid binding to rat cerebellar membranes. J Neurochem
58:1087–1092
Boue-Grabot E, Roudbaraki M, Bascles L, Tramu G, Bloch B, Garret M 1998
Expression of GABA receptor rho-subunits in rat brain. J Neurochem 70:899 –907
Anderson R, Mitchell R 1986 Biphasic effect of GABAA receptor agonists on
prolactin secretion: evidence for two types of GABAA receptor complex on
lactotrophes. Eur J Pharmacol 124:1–9
Tapia-Arancibia L, Roussel JP, Astier H 1987 Evidence for a dual effect of
gamma-aminobutyric acid on thyrotropin (TSH)-releasing hormone-induced
TSH release from perifused rat pituitaries. Endocrinology 121:980 –986
Lorsignol A, Taupignon A, Dufy B 1994 Short applications of gammaaminobutyric acid increase intracellular calcium concentrations in single identified rat lactotrophs. Neuroendocrinology 60:389 –399
Virmani MA, Stojilkovic SS, Catt KJ 1990 Stimulation of luteinizing hormone
release by gamma-aminobutyric acid (GABA) agonists: mediation by GABAAtype receptors and activation of chloride and voltage-sensitive calcium channels. Endocrinology 126:2499 –2505
Berman JA, Roberts JL, Pritchett DB 1994 Molecular and pharmacological characterization of GABAA receptors in the rat pituitary. J Neurochem 63:1948 –1954
Boue-Grabot E, Dufy B, Garret M 1995 Molecular diversity of GABA-gated
chloride channels in the rat anterior pituitary. Brain Res 704:125–129
Bello AR, Dubourg P, Kah O, Tramu G 1992 Identification of neurotensinimmunoreactive cells in the anterior pituitary of normal and castrated rats. A
double immunocytochemical investigation at the light- and electron-microscopic levels. Neuroendocrinology 55:714 –723
Covenas R, de Leon M, Narvaez JA, Aguirre JA, Tramu G, Gonzalez-Baron S
1997 ACTH/CLIP immunoreactivity in the cat brain stem. Peptides 18:965–970
Hemming FJ, Dubois MP, Dubois PM 1986 Somatotrophs and lactotrophs in
the anterior pituitary of fetal and neonatal rats. Electron-microscopic immunocytochemical identification. Cell Tissue Res 245:457– 460
Tramu G, Pillez A, Leonardelli J 1978 An efficient method of antibody elution
for the successive or simultaneous localization of two antigens by immunocytochemistry. J Histochem Cytochem 26:322–324
Beauvillain JC, Mazzuca M, Dubois MP 1976 The thyrotropic cells of the
guinea pig pituitary. Electron microscopic study after characterization by
immunocytochemical means. Cell Tissue Res 174:233–244
Ogurusu T, Taira H, Shingai R 1995 Identification of GABAA receptor subunits in rat retina: cloning of the rat GABAA receptor ␳2-subunit cDNA.
J Neurochem 65:964 –968
Ogurusu T, Shingai R 1996 Cloning of a putative gamma-aminobutyric acid
(GABA) receptor subunit ␳3 cDNA. Biochim Biophys Acta 1305:15–18
Siebert PD, Larrick JW 1992 Competitive PCR. Nature 359:557–558
Ashworth R, Hinkle PM 1996 Thyrotropin-releasing hormone-induced intracellular calcium responses in individual rat lactotrophs and thyrotrophs. Endocrinology 137:5205–5212
Martinez-Torres A, Vazquez AE, Panicker MM, Miledi R 1998 Cloning and
functional expression of alternative spliced variants of the ␳1 gammaaminobutyrate receptor. Proc Natl Acad Sci USA 95:4019 – 4022
Inoue K, Kurosumi K 1984 Ultrastructural immunocytochemical localization
of LH and FSH in the pituitary of the untreated male rat. Cell Tissue Res
235:77– 83
Han MH, Li Y, Yang XL 1997 Desensitizing GABAC receptors on carp retinal
bipolar cells. Neuroreport 8:1331–1335
Feigenspan A, Bormann J 1994 Modulation of GABAC receptors in rat retinal
bipolar cells by protein kinase C. J Physiol (Lond) 481:325–330
Smart TG 1997 Regulation of excitatory and inhibitory neurotransmitter-gated
ion channels by protein phosphorylation. Curr Opin Neurobiol 7:358 –367
McDonald BJ, Amato A, Connolly CN, Benke D, Moss SJ, Smart TG 1998
Adjacent phosphorylation sites on GABAA receptor beta-subunits determine
regulation by cAMP-dependent protein kinase. Nat Neurosci 1:23–28
Johnston GA 1996 GABAc receptors: relatively simple transmitter-gated ion
channels? Trends Pharmacol Sci 17:319 –323