| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institut de Chimie Biologique and INSERM U-575, Faculty of Medicine, Strasbourg, France
1Correspondence: Faculty of Medicine and INSERM U-575, 11, rue Humann, 67085, Strasbourg Cedex, France. E-mail: maitre{at}neurochem.u-strasbg.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxybutyrate (GHB) receptor, C12K32 and GHBh1, were isolated from a human frontal cortex cDNA library. The two clones differ by a deleted cytosine in C12K32. CHO cells transfected with either C12K32 or GHBh1 responded positively to submicromolar GHB stimulation. However, unlike C12K32, GHBh1 desensitizes rapidly on application of low concentrations of GHB. GHB receptor properties were then studied on C12K32 expressed in CHO cells. C12K32 bound GHB with a Kd of 114 nM and has no affinity for GABA or glutamate. GHB and NCS-382 displaced [3H]GHB with an IC50 of 53 ± 8 and 120 ± 18 nM, respectively. In patch-clamp experiments, GHB induced a dose-dependent response with an EC50 of 130 nM. This response was antagonized by NCS-382, was not reproduced by GABA, and was sensitive to the addition of GTP--S in the recording pipette. CHO cells transfected with C12K32 exhibited GTP-35S binding with an EC50 of 462 nM for GHB and an IC50 of 2.9 µM for NCS-382. The present data led to the conclusion that both C12K32 and GHBh1 are two closely related isoforms of a human GHB receptor, GHBh1, that is described in the databank as the GPCR 172A—Andriamampandry, C., Taleb, O., Kemmel, V., Humbert, J.-P., Aunis, D., Maitre, M. Cloning and functional characterization of a gamma-hydroxybutyrate receptor identified in the human brain.
Key Words: G-protein coupled receptors • electrophysiological characterization • functional expression • GHB
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-HYDROXYBUTYRATE (GHB) IS A NATURALLY occurring biological molecule that is present at micromolar concentrations in both brain and peripheral tissues (1
, 2)
. In the brain, GHB is derived from the transformation of GABA by GABA-transaminase into succinic semialdehyde (SSA), which is subsequently reduced in the neuronal cytosol into GHB in the presence of NADPH and succinic semialdehyde reductase (SSR) (3)
. GHB is coaccumulated with GABA/glycine into synaptic vesicles via the vesicular inhibitory amino acid transporter (VIAAT) and is then coreleased with GABA in GABAergic synapses (4)
. GHB is thought to participate in the regulation of the activity of these synapses through its receptors. The existence of GHB receptors has been suspected since the description of high affinity binding sites for GHB in the brains of various species (5
, 6)
. These sites are specific in their distribution in the brain, their ontogenesis, plasticity, kinetics, and pharmacology. The cloning of the first GHB receptor from rat brain (7)
prompted us to look for homologous in human brain.
GHB has been used as a drug since its synthesis by Laborit in 1964 (8)
because it acts as a GABA-mimetic able to freely cross the blood brain barrier. Pharmacological doses of GHB (several grams), resulting in >800 µM GHB in the rat brain, is generally thought to stimulate GABAB receptors (9)
. The specific role of GHB receptors remains elusive, but it has been proposed that high GHB concentrations desensitize GHB receptors that permit a potentiation of GABA synapse activity and a hyperstimulation of GABA receptors (mainly GABAB). GHB has been used clinically as an adjuvant to anesthesia and in the treatment of narcolepsy (10)
and has been proposed in the treatment of opiate withdrawal, alcohol dependence and in fibromyalgia (11)
. Since the early 1990s, GHB has become a substance available in health food stores for its supposed aid to body building and a recreational drug because it promotes anxiolysis, social interaction and some degree of euphoria (12)
. High doses of GHB associated with other central nervous system (CNS) depressants (such as alcohol) induce intoxication leading to respiratory depression and coma. For these reasons, severe restrictions have been imposed on the availability of this drug in several countries.
In an effort to isolate and characterize human brain GHB receptors, we have selected two clones from a human frontal cortex cDNA library, C12K32 and GHBh1, the latter corresponding to an already-reported sequence of unknown function, GPR172A or PERV-A (13)
. Because of the predicted structure of GPR172A and the presence in its sequence of a consensus site for G-protein coupling, this receptor has been classified as G-protein coupled receptor (GPCR) in the NCBI database. In the present work, we demonstrate that the two clones we isolated (C12K32 and GHBh1) are functional GPCRs, specifically activated by submicromolar concentrations of GHB.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, were chosen among several others because of their ability to amplify a single cDNA band of the expected size (Fig. 1
). In this cloning procedure, a PCR reaction was positive when it showed the amplification of a cDNA fragment of 
400 bp that is the expected fragment size if the PCR was run on the rat receptor clone. 48 PCR reactions were performed to reveal positive plates and visualized by 1% agarose gel electrophoresis. On each positive plate, row and column pools were screened to reveal the unique address of the positive well within each plate. Phage plating was performed at low density on NZY-agar plates with 200 µl Xl1-Blue MRF’ and 4 ml NZY top agarose and incubated overnight at 37°C. The phage plaques were resuspended in 500 µl SM buffer and PCR screened using the primers described above until homogeneity. The positive plaques were cloned as follows. Two E. coli strains (Xl1-Blue MRF’ and SOLR) were grown separately overnight in NZY medium containing 0.2% maltose. In a sterile tube, 200 µl of the Xl1-Blue MRF’ culture were inoculated with 250 µl of a positive phage suspension in the presence of ExAssist phage helper for 15 min at 37°C. Three milliliters of NZY were added and incubated at 37°C for 3 h. The tube was heated at 70°C for 20 min and centrifuged at 1,000 g for 15 min. An aliquot of the supernatant (10 µl) was added to 200 µl of the SOLR culture and incubated at 37°C for 15 min. After 300 µl NZY were added, incubation was continued for 45 min at 37°C. 200 µl of the mixture was plated on LB-agar plates containing 50 µg/ml ampicillin and incubated overnight at 37°C. A distinct colony was picked and grown overnight in LB-ampicillin medium to prepare plasmid DNA for sequencing and transfections. Plasmid DNA was prepared with the GenElute miniprep kit (Sigma-Aldrich, Saint Quentin Fallavier, France).
|
This procedure led to the isolation of the C12K32 full length cDNA of 1461 bp. When searching databases [basic local alignment search tool (BLAST) at NCBI], a G protein-coupled receptor 172A (GPR172A) also named putative G-protein coupled receptor 41 (GPCR41) and PERV-A receptor 1 (PERV-A1) (13)
was found to be identical to C12K32 with the exception of the 3' region. Indeed, the analysis of the C12K32 sequence showed a cytosine deletion at position 1374 when compared to that of these two former clones, leading to the modification of its 3' peptide sequence. We then decided to name GPR172A, GPCR 41, or PERV-A1 as GHBh1. Unexpectedly, C12K32 showed no significant homology with the rat brain GHB receptor. The subcloning and the sequencing of the 400 bp cDNA fragment were then performed as described below.
Cloning of GHBh1
To isolate GHBh1, the full-length cDNA of the nonmutated isoform of C12K32, we PCR amplified the coding sequence from another human cortex cDNA library (a gift from Dr. Garnier; IGBMC, France), using two flanking primers (AGTCTTCACTTCCCAGGAGAGCCAAAGCGT and TCAGGAGTCACAGGGGTCTGCACAGTCCTT).
3. Subcloning and sequencing of the cDNA fragment of 400 bp
PCR reactions were performed, according to the above protocol, on the cloned rat GHB receptor and a plasmid cDNA preparation representative of the human cDNA library with the primer couples designed from the rat GHB receptor sequence. After 1% agarose gel electrophoresis, bands with sizes 400 bp were collected, purified by the Montage Gel Extraction Kit (Millipore, Bedford, MA, USA), and subcloned with the TOPO TA cloning kit (Invitrogen, Cergy Pontoise, France). Briefly, 4 µl of each PCR product were mixed with 1 µl of salt solution and 1 µl of the TOPO vector and incubated 5 min at room temperature. The reaction was placed on ice, and 2 µl of the TOPO cloning reaction were added into a vial of One Shot TOP10 E. Coli competent cells and gently mixed. After 30 min incubation on ice, the reaction was heat shocked at 42°C for 30 s, placed on ice, and grown in 250 µl of SOC medium under gentle shaking at 37°C for 1 h. Aliquots were spread on LB-Ampi plates and incubated overnight at 37°C. Plasmids were prepared from selected colonies and sequenced with T7 and BGH reverse primers.
Generation of tagged GHB receptors
To generate an enhanced GFP (EGFP)-tagged N-terminal C12K32 and GHBh1 fusion protein (EGFP/C12K32 and EGFP/GHBh1), the GHBR ORFs were amplified from the original clones using the primers TA/AGATCT/ATGGCAGCACCCACGCCCGC and TA/TCTAGA/CAGGAGTCACAGGGGTCTGCAC, which contain BglII and XbaI restriction sites. The resulting fragments were then cloned into the pEGFP-c1 vector under the control of the human cytomegalovirus (CMV) immediate early promoter.
Eukaryotic expression
CHO cells were transiently transfected using two gene transfer methods. For electrophysiological recording, polyethylenimine 25 kDa (Sigma Chemical) was used to transfect CHO cells according to Boussif et al. (14)
. For biochemical assays, the transfection reagent LIPOFECTAMINE 2000 (Invitrogen, Cergy Pontoise, France) was used according to the manufacturer’s protocol. Briefly, CHO cultured cells at 50–70% confluence were transfected with 12 µg of plasmid DNA and 30 µl of LIPOFECTAMINE 2000 (diluted in OPTI-MEM I Reduced Serum Medium; GIBCO Life Technologies, Inc., Invitrogen) per 100 mm petri dish, mixed, and incubated at room temperature for 30 min.
Ligand binding experiments
Transiently transfected CHO cells were harvested and resuspended in cold phosphate buffer (100 mM KH2PO4; pH 6.0) containing 5 mM EDTA and centrifuged at 18,000 g for 10 min at 4°C. The pellet was resuspended in EDTA-free phosphate buffer and centrifuged at 30,000 g for 20 min. The pelleted membrane preparation was diluted in phosphate buffer and used immediately for binding experiments.
Saturation experiments were performed using 0.3–0.4 mg protein per assay. The incubation was carried out in phosphate buffer pH 6.0 for 25 min at 0°C in the presence of various concentrations of [3H]GHB (10–10-2.10–6 M; 110 Ci/mmol, CEA, France). Nonspecific binding was determined in the presence of 5 mM unlabeled GHB. The binding reaction was terminated by adding ice-cold phosphate buffer and bound [3H]GHB was separated from the free molecule by rapid filtration through GF/B filters (Whatman International, Ltd., Maidstone, England). After being rinsed twice with ice-cold phosphate buffer, the membrane-bound [3H]GHB remaining on the filters was determined by liquid scintillation counting in 5 ml Rotiszint (Carl Roth, GmbH + Co, Karlsruhe, Germany).
Competition experiments were performed at 0°C with 0.3–0.4 mg protein per assay in 600 µl phosphate buffer pH 6.0 for 25 min in the presence of 10 µM of either GHB, GHB structural analogs, glutamate or GABA with 800 nM [3H]GHB (110 Ci/mmol, CEA, France). Nonspecific binding was determined with 5 mM GHB and results were expressed in percentage of [3H]GHB binding in the presence of the different compounds. IC50 was measured for GHB and NCS-382 with 800 nM [3H]GHB, in the presence of various concentrations of GHB or NCS-382 (10–10-10–3 M).
[35S]GTP--S binding assay
CHO cells were transiently transfected with the pEGFP/C12K32 construction as described above. After 30–48 h transfection, cells were harvested and resuspended in GTP--S buffer (50 mM Tris-HCl, pH 7.4, containing 100 mM NaCl, MgCl2 at indicated concentrations, 0.2 mM EGTA and a cocktail of protease inhibitors (Complete from Roche Diagnostics, Meylan, France). Cells were homogenized for 20 s by polytron and centrifuged at 20,000g for 20 min. The pellet was resuspended in GTP--S buffer at a concentration of 250 µg protein/ml. The membrane suspension (10–15 µg protein) and drugs were preincubated in a MultiScreen 96-well plate (Millipore, Saint-Quentin-en-Yvelines, France) at the indicated GDP concentrations for 30 min at room temperature. The reaction was started by the addition of [35S]GTP--S (Amersham Biosciences, Saclay, France) at a final concentration of 0.3 nM. After 2 h, the samples were filtered through the MultiScreen 96-well plate and washed two times with 200 µl of ice-cold incubation buffer. The membranes were removed with the MultiScreen Punch Kit (Millipore, Saint-Quentin-en-Yvelines, France) and the radioactivity was counted in 4 ml liquid scintillation fluid (Carl Roth, GmbH + Co, Karlsruhe, Germany). Non-specific binding was measured in the presence of 100 µM unlabeled GTP and was assayed in the absence of GHB and in the presence of GDP.
The stimulation by agonist was defined as a percentage increase in the binding activity above the basal level [(cpm agonist–cpm no agonist)/cpm no agonist] x 100. Data are mean ± SE of 2 experiments, each performed in triplicate. Nonlinear regression analysis of concentration-response data was performed using Prism 4.0 software (GraphPad Prism Program, GraphPad, San Diego, CA, USA).
Electrophysiological recordings
The CHO cells in culture were transiently transfected with the human GHB receptors as described above. The day before transfection, cells were plated at low density (2.104 cells/ml). GHB-induced responses were recorded in the whole-cell configuration of the patch-clamp technique (15)
using an Axopatch-B200 amplifier (Axon Instruments, CA, USA). Solutions for recording were defined such that the equilibrium potentials for potassium, sodium, cationic and chloride ions were imposed far away from each other, allowing us to identify the charge carrier of the current response. The recording pipette had 2 to 5 M
access resistance when filled with recording solution (in mM): K-gluconate 120, HEPES 10, EGTA 5.5, CaCl2 1, MgCl2 2, NaCl 3 and KCl 5. In all experiments (if not specified) pipette medium was added with 1 mM ATP just before use. Isolated CHO cells were selected and recorded under continuous perfusion with control medium containing (in mM): NaCl 132, KCl 5, CaCl2 0.5, MgCl2 2, HEPES 10 and D-glucose (Glc) 10. The pH of the pipette and bath medium was adjusted to 7.2 with KOH and 7.4 with NaOH, respectively. Voltage command and current trace digitization were achieved using the Digidata 1322A card interface (Axon Instruments, Foster City, CA, USA) and Pclamp software (Axon Instruments). Current traces were low-pass filtered at 1 kHz before digitization with a period of 0.5 ms. To determine the I-V relationship, test potential values were corrected for the liquid junction potential according to the calculation of junction potential given by the clampex-8 routine of the Pclamp package.
Drugs were applied to the recorded cell through a multibarrel perfusion system, the rapid solution exchanger RCS-160 (Bio-Logic, Grenoble, France). Each barrel had a 1 mm inner diameter, and the selected tube was placed 50 µm from the recorded cell. The application starts when the selected tube moves into place in front of the cell. To reduce the application delay, the corresponding electrovalve was opened 50 ms before its arrival in front of the recorded cell. The kinetics of solution exchange around the recorded cell was estimated using the change in external solution potassium concentration test. Each tube was filled with the same solution containing 20 mM higher K concentration than the control medium. Under these conditions, the solution exchange induced a cellular current change whose kinetics are dependent on the kinetics of solution exchange. No significant differences between tubes were observed, and the overall data gave a time for complete solution exchange around the recorded cell of 1.9 ± 0.3 s.
The membrane of the recorded cell was electrically stimulated by a protocol as described below. From a holding potential of –40 mV, a hyperpolarization to –50 incremented in each episode by –10 mV up to –140 mV and depolarization to –20 mV with increment of 10 mV were separated by a constant test potential of –80 mV. Episodes of stimulation were applied periodically with a frequency of 0.2 Hz. The amplitude of the current obtained at a given test potential corresponded to the mean steady state current measured at the end of the test potential. Data were analyzed using the clampfit routine of the Pclamp software package and graphs were realized with Sigmaplot software (SSP).
Confocal microscopy
CHO cells were transfected and labeled with EGFP-GHBh1 or EGFP-C12K32 as described above. The cells were plated onto 35-mm glass-bottomed culture dishes at a density of 3.104 cells/ml and incubated at 37°C. Forty-eight hours after transfection (4 µg plasmid), cells expressing EGFP-GHBh1 or EGFP-C12K32 were observed by confocal microscopy at 37°C in culture medium containing 20 mM HEPES, pH 7.4. A Leica TCS-2 confocal inverted microscope was used to measure the fluorescence of EGFP-tagged proteins. Excitation was obtained with an Argon-Krypton laser, with lines set at 488 nm. The emitted light (from 510 to 600 nm) was obtained through appropriate filters (BF 530/30). Images were taken with a x40 objective, numerical apertune (NA) 1.2, with an electronic zoom of 2- to 4-fold. A look-up table (glowoverglowunder, Leica) ensured that the full dynamic range of the photomultipliers was used. Image processing and fluorescence intensity measurements were performed on the LCS (Leica Confocal System, Paris, France) software. The image size was 512 x 512 pixels, and before each measurement a section series was acquired in the vertical axis to choose the equatorial section.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The sequencing of this fragment showed no significant homology to the rat brain clone, except for the two primers (Fig. 2
). When the rat GHB receptor cDNA was used as a template, the PCR reaction amplified a 411 bp cDNA fragment with 100% homology, as expected. When run with the plasmid cDNAs preparation from the human cortex library, the PCR generated a chimeric cDNA of 341 bp composed of a 291 bp fragment that is 100% homologue to C12K32 sequence and the two primers designed from the rat GHB receptor sequence flanked on its 5' and 3' ends, showing only 50–60% similarity (Fig. 3
). The amplification probably occurred owing to the very low stringency used for the four first PCR cycles.
|
|
As reported above, the analysis of the C12K32 nucleotide sequence showed a cytosine deletion at position 1374 compared to the GHBh1 gene (located on chromosome 8) and the cDNA clone GHBh1. This deletion resulted in ORF changes and thus a modification of the C-terminal tail. Figure 4
A shows the amino acids sequence comparison between GHBh1 (GPCR 172A) and C12K32. Previous studies on hydrophobicity profiles indicated that they are 10 or possibly 11 transmembrane regions in GPCR 172A (PERV.A) receptor (13)
.
|
Structure of C12K32 and GHBh1 receptors
SMART program (16
, 17)
was used to analyze the GHBh receptors structural organization. GHBh receptor belongs to proteins with multiple characters and functions that are performed by one or more component domains. SMART has been designed to allow easy and rapid annotation of signaling multidomain proteins. The tool contains several unique aspects, including automatic seed alignment generation, automatic detection of repeated motifs or domains, and a protocol for combining domain predictions from homologous subfamilies. SMART uses Hidden Markov Models (profile HMMs) to carry out sensitive database searching and applied to the problems of statistical modeling and multiple sequence alignment of protein families and protein domains. TMHMM2 predicts the location and topology of transmembrane helices. As shown by this diagram (Fig. 4B
), six putative transmembrane domains were described at positions 46–68, 81–103, 113–135, 147–69, 196–218, and 405–424. SEG is a program of Wootton and Federhen (18)
that detects regions of sequence that have low compositional complexity. SignalP program predicts the presence and location of signal peptide cleavage sites at position 1–25. A Pfam (protein family database) domain, DUF1011 (Accession number: PF06237), a collection of protein families and domains that contains multiple protein alignments and profile-HMMs was detected at position 265–318 (19)
. DUF1011 functions are not known yet but it is a structure that is shared by several putative GPCRs.
As shown in Fig. 4A
, GHBh1 and C12K32 are identical except in their C terminal regions (445 AA in total for GHBh1, 487 AA for C12K32 because of an additional C terminal region with 42 AA). Putative phosphorylation sites are present in GHBh1 structure (underlined italic letters).
Functional expression of C12K32 and GHBh1 clones
Sensitivity of control CHO cells to GHB
The CHO cells in control conditions (nontransfected) were tested in whole-cell recording experiments for their GHB sensitivity between 24 and 72 h after plating. Isolated cells were chosen and tested for GHB responses at concentrations from 0.25 to 10 µM. No response to GHB was seen at 48 to 72 h (n=70). However, at 24 h, in 5 out of 32 cells tested, very small responses could be recorded in the presence of GHB. Indeed, when the agonist was applied for example at a concentration of 0.25 µM, the response amplitude varied from –4 to –22 pA at –80 mV with an average value of –10 ± 8 pA (n=5). Consequently, transfected cells were tested at 48 to 72 h.
Pharmacological profile and functional coupling of GHBh1 and C12K32
Whole-cell recording experiments carried out on CHO cells transiently transfected with C12K32 (Fig. 5
A, B, D) or GHBh1 (Fig. 5C
) led to specific cell responses which are blocked by the GHB receptor antagonist NCS-382 (Fig. 5B
). The cell response observed in the presence of 0.25 µM GHB was inhibited by 84.8 ± 1.2% (n=5) in the presence of 1 µM NCS-382.
|
The GHB-induced response corresponded to an inward current that was delayed from the onset of agonist application by a period that had a duration value in between 5 s and 2 min. For example, with 0.25 µM GHB a mean delay value of 16.8 ± 2.5 s (mean±SE; n=11) was obtained. The amplitude of the current strongly varied from one cell to the other, and at test potential of –80 mV a mean value of –830 ± 245 pA (mean±SE; n=13) was obtained with 0.25 µM GHB. However, GHBh1 and C12K32 differed markedly in the reproducibility of the response. GHBh1 was strongly desensitized after the first GHB application and no further response was obtained (Fig. 5C
) even after 1 h of washout (data not shown). In contrast, when the cells were transfected with C12K32, reproducible responses could be recorded from the same cell (Fig. 5D
). Thus, C12K32 was used to study the pharmacological characteristics of the human GHB receptor.
The C12K32 receptor expressed in CHO cells showed a pharmacological profile closely resembling the characteristics obtained from binding data (see below). Indeed, the response was specific to GHB and did not respond to GABA since the latter, at a concentration up to 100 µM, was unable to reproduce the GHB effect. Figure 6
A illustrates a cell where a large current was obtained after application of 0.1 µM GHB and where no current response was seen when GABA was applied at a 103 times higher concentration.
|
The amplitude of the response increased with GHB concentration (50 nM to 5 µM) according to a sigmoid model and reached an apparent maximum that differed from one cell to another (0.1 to several nA). Hence, the construction of a mean dose-response curve was made on relative responses with respect to this maximum. Figure 6B
illustrates the relative statistical dose-response distribution obtained from eight cells. A nonlinear fit of the Hill equation to the data points gave the value of half-efficient concentration EC50 = 0.13 µM and a slope h = 2.9.
To identify the ion charge carrier involved in the GHB-induced response, I-V relations were obtained and the reversal potential (Er) of the responses was calculated. (Fig. 6C
), illustrates the I-V relationship of the response to 0.25 µM GHB. The I-V curves were linear, and Er values were –45.6, –21.7, and –44.3 in control, during GHB application, and wash-out conditions respectively. The specific GHB-induced current corresponding to the difference between steady-state current in the control and under GHB perfusion was also linear and had an Er value of 12.5 mV in this case. A mean value of 9.2 ± 1.8 (mean±SE; n=26) mV was obtained. This value was close to the monovalent cationic ion (Na and K) equilibrium potential (Ecat=1.7 mV), suggesting that cationic membrane permeability was activated after GHB application. Nevertheless, on average, our value was 7.5 mV more depolarized, suggesting that the GHB activated cationic channels maybe also be permeable to calcium ions or that Na ions is more permeable than K ions.
The functional coupling of GHBh1 to G-protein was investigated by using GTP--S. For GHBh1 and C12K32, when 1 mM GTP--S was added to the pipette medium, the response irreversibly evolved to a maximum steady-state current in five and six cells out of six and eight responding cells respectively. Figure 6D
illustrates the response elicited on application of 5 µM GHB, reaching a steady-state inward current of about –4.5 nA at –80 mV. To verify whether this irreversible current was due to a seal breakdown or increased membrane permeability, outside-out patch was pulled (arrow in Fig. 6D
) and recorded. The amplitude (–17 pA at –80 mV) of the current obtained in this configuration corresponded to a patch resistance of 4.7 G, which confirms an increased membrane conductance as the origin of the irreversible current observed in whole-cell. This result strongly suggests that GHBh1 and C12K32 are G-protein coupled receptors, an idea that is reinforced by the following GTP--S binding data.
GHB Binding assay
Saturation experiments were carried out as described above in phosphate buffer pH 6.0 for 25 min at 0°C in the presence of increasing concentrations of [3H]GHB from 10–10 M to 2.10–6 M. Nonspecific binding was determined in the presence of 5 mM unlabeled GHB. Non linear curve fitting (Graphpad Prism Programm) gave a Kd of 114 ± 24 nM for GHB binding and a Bmax of 295 ± 96 fmol/mg proteins (Fig. 7
A).
|
Competition experiments carried out with 10 µM of the tested substances revealed no affinity of the cloned receptor for GABA or glutamate. Various GHB structural analogs that exhibit affinity for the GHB binding site(s) ex vivo have been screened for their ability to displace GHB binding at 10 µM concentration (Fig. 7B
). Trans-hydroxycrotonate (T-HCA) revealed no affinity for the cloned receptor. The other substances tested (GHB itself, NCS-382, 4-phenylbutyrate) possess significant affinity for the C12K32 receptor.
IC50 values for displacement of radioactive GHB (800 nM) have been determined for GHB itself and for NCS-382. This last compound showed an IC50 of 120 ± 18 nM (Fig. 7C
) while GHB possesses higher potency (IC50 of 53±8 nM; Fig. 7D
). These results appear to indicate that the C12K32 receptor belongs to a subclass of GHB receptors showing affinity for NCS-382, which possesses antagonistic properties for some of the described effects of GHB in vivo and in vitro (20
21
22)
.
[35S]GTP--S binding studies
[35S]GTP--S binding assay
Effects of increasing concentrations of GHB—
Thirty hours after transfection, cells were harvested and membranes prepared as described above. [35S]GTP--S binding was performed on membrane preparations (10–15 µg protein/well) in a 50 mM Tris-HCl buffer (pH 7.4) containing 5 mM MgCl2, 100 mM NaCl, 0.2 mM EGTA and a tablet of protease inhibitor. Membranes were preincubated with 5 µM GDP for 30 min at room temperature. [35S]GTP--S was then added and binding carried out at room temperature for 1 h in the presence of increasing concentrations of GHB from 5.10–8 M to 2.10–6 M. The reaction was terminated with two rinses with ice-cold buffer and the radioactivity was determined by scintillation spectrometry. As shown in Fig. 8
A, [35S]GTP--S specifically bound to C12K32 on GHB stimulation with an EC50 of 462 nM.
|
Stimulation of [35S]GTP--S binding to membranes expressing C12K32 by 10 µM of GHB, as a function of concentration of GDP and MgCl2—
The binding assay was performed as before in the presence of 2.5 mM to 50 mM MgCl2 and 0.1 µM to 100 µM GDP. As shown in Fig. 8B, C
, maximum [35S]GTP--S binding was obtained at 10 mM MgCl2 and 20 µM GDP.
Effect of increasing concentrations of NCS-382 on the stimulation of [35S]GTP--S binding to membrane—
[35S]GTP--S binding was carried out in the presence of 20 µM GDP, 10 mM MgCl2, 10 µM GHB and various concentrations of the antagonist NCS-382 (10–8 to 10–3 M). As shown in Fig. 8D
, the antagonist NCS-382 inhibited [35S]GTP--S binding with an apparent IC50 of 2.9 µM.
Effect of pertussis toxin on the stimulation of [35S]GTP--S binding to membrane—
Transfection was performed as described above. Sixteen hours before cell harvesting, 100 ng/ml pertussis toxin were added to the culture medium. Cell membranes were prepared in a 50 mM Tris-HCl buffer (pH 7.4) containing 10 mM MgCl2, 100 mM NaCl, 0.2 mM EGTA, and a tablet of protease inhibitor and frozen at –80°C for 4 days. The binding assay was performed (10–15 µg protein/well) after a preincubation with 20 µM GDP for 30 min at room temperature followed by a 1 h incubation in the presence of 0.3 nM [35S]GTP--S and increasing concentrations of GHB at room temperature. As shown in Fig. 8E
, pertussis toxin strongly inhibits the GHB-induced stimulation of [35S]GTP--S binding, reducing the affinity of GTP by 1,400-fold compared to control conditions. These results suggest the implication of Gi/Go in the mechanism of coupling to the receptor. However, in these experiments carried out with frozen/thawed transfected cell membranes, we observed a shift in the EC50 for GHB (control experiments without pertussis toxin, 13 µM vs. 462 nM with fresh membranes). Other experiments performed with frozen membranes confirmed the decrease with time of GTP binding to transfected cell membranes, even though the membranes were stored at low temperature.
Confocal studies of C12K32 and GHBh1
To analyze the spatial organization of the GHB receptors, we used GFP N terminus tagged GHBh1 or C12K32 protein. As shown in Fig. 9
A, B, the fluorescence was mainly located at the cell surface of transiently transfected CHO cells. An intracellular staining of the cells was also observed, which probably corresponds to the trafficking process. Indeed, the precursor proteins may accumulate in the endoplasmic reticulum. However, receptors could also accumulate inside the cell, suggesting that some GHB receptor molecules may be trapped. To confirm this hypothesis, live cells were treated with the protein inhibitor cycloheximide (10 µM). After 2 h, we found that the expression of EGFP-GHBh1 or EGFP-C12K32 was only observed on membranes including the cell surface and cytoplasmic exo- and endocytotic vesicles (data not shown). The EGFP labeling appeared 12 h after transfection and lasted for 72 h. The EGFP without the targeted protein was separately transfected into CHO cells (Fig. 9C
). The staining was only present in the cytoplasm and the nucleus as aggregated or light diffused fluorescence.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, led us to PCR-screen a human brain cDNA library with primer couples designed from the nucleotide sequence of GHBR. One of these primer couple (ADO228/ADU34) amplified a unique cDNA band of 400 bp (exactly 411 bp). We opted for this primer couple because this would be the size of the amplified fragment if the rat cDNA library was used as template. Surprisingly, we did not find any significant homology between the PCR-amplified 341 bp from the human cortex library and GHBR except for the two primers. The low stringency of the first four cycles of the PCR protocol probably allowed the amplification of this chimeric cDNA. Out of several positive clones isolated, only C12K32 exhibited both pharmacological and functional responses in accordance with a GHB receptor.
C12K32 was found to be identical to GPR or GPCR 172A (GHBh1) with the exception of a deleted cytosine at the position 1374 in the C12K32 nucleotide sequence. This deletion resulted in a shift in the ORF that led to a different C-terminal tail and a longer protein length for C12K32 (487 amino acids vs. 445 for GHBh1). Both receptors seem to be membrane proteins as shown by using EGFP/C12K32 and EGFP/GHBh1. Indeed the GFP-tagged fusion protein expressed in CHO cells was clearly located at the plasma membrane, suggesting that the receptor was effectively associated with the cell membrane. Moreover, CHO cells expressing C12K32 or GHBh1 responded positively to GHB stimulation in a sensitive manner to the GHB receptor antagonist NCS-382 and these responses were not reproduced by GABA. These results demonstrate that C12K32 and GHBh1 encode membrane proteins that are receptors activated by GHB. Nevertheless, GHBh1 and C12K32 differed in their responsiveness to GHB. GHBh1 showed strong desensitization at concentrations as low as 0.1 µM, and only one response was obtained under our conditions, suggesting very slow recovery kinetics from desensitization and/or internalization. Consequently, the specific effects of GHB on GHBh1 may be masked in native cells if GHB is applied at millimolar concentrations (23
, 24)
.
Rapid agonist-induced desensitization is a common mechanism to interrupt the intracellular signal propagation of many GPCRs, including those coupled to Gi (25
, 26)
. Agonist-induced GPCR desensitization has been shown to be mediated by receptor phosphorylation through either second messenger-dependent kinases (PKA and PKC) or G protein-coupled receptor kinases (GRK). The Ser or Thr residues present in the intracellular C-terminal domain of GHBh1 are probably involved in these mechanisms. It would be worthwhile to further investigate this hypothesis.
C12K32 and GHBh1 structures do not correspond to the typical 7TM model even though GHBh1 is described as a GPCR in the database. Both have 10 potential TM domains as predicted by SOSUI (TM prediction algorithm). A consensus site for putative G protein coupled receptors (75% identity) was found in their third intracellular loops (PROSITE). With the use of the Simple Modular Architecture Research Tools (SMART) (16
, 27)
, a structure with six TM domains and a seventh PFAM domain of 107 amino acids called DUF1011 (accession number: PF06237) of unknown function was found. A number of eukaryotic proteins share this conserved domain, including human putative GPCRs. These receptors constitute a vast protein family that encompasses a wide range of functions. They show considerable diversity at the sequence level, on the basis of which they can be separated into distinct groups. The term "clan" has been used to describe these GPCRs, as they are a group of different families with no significant sequence similarity between them. C12K32 and GHBh1 receptors probably constitute members of a new "clan" and are believed to transduce extracellular signals through interaction with guanine nucleotide-binding proteins regardless of the absence of significant homology with any known GPCR and, in particular, with GABAB receptors. Results from [35S]GTP--S binding as well as electrophysiological experiments strongly support the coupling of C12K32 and GHBh1 to G protein. Taken together, our data strongly support the concept that C12K32, like GHBh1, belongs functionally to a GPCR group. First, micromolar amounts of GHB potentiate the binding of radioactive GTP to cell membranes transfected with C12K32. An EC50 of 462 nM was obtained. Second, GTP binding was inhibited by the antagonist NCS-382 with an apparent IC50 of 2.9 µM. This observation was correlated with the inhibitory effect obtained with micromolar amounts of NCS-382 in the GHB-induced electrophysiological response in transfected cells. Third, GTP binding was inhibited when the transfected cells were preincubated with pertussis toxin. This indicates that most probably Gi or Go is involved in the coupling of the receptor to the cellular response. The precise nature of this response remains to be clarified.
Similarly to the rat GHBR (7)
, the human GHB receptors are probably G-protein coupled receptors the activation of which in our heterologous expression conditions results in cationic channel opening. However, the channels involved in GHBh1 and C12K32 responses seem to be different than those activated by rat GHBR as revealed by the calculated Er values. In the present data, Er had more depolarized values suggesting either involvement of channels with a higher selectivity for Na+ ions than for K+ ions and/or channels that are permeable also to Ca2+ ions, suggesting that the human and the rat GHB receptors are coupled to different signal transduction pathways.
The high affinity of C12K32 for GHB and the rapid desensitization of GHBh1 suggest that these receptors are not involved in the response to high doses of GHB reported in the literature (28
, 29)
. The disappearance of the pharmacological effects of GHB in GABAB(1)-deficient mice while the GHB binding remains unchanged supports such a contention (30)
. A possible explanation of the poorly specific GHB cellular effect reported in the literature may be in part attributed to the strongly desensitizing feature associated with these receptors.
|
ACKNOWLEDGMENTS |
|---|
Received for publication August 17, 2006. Accepted for publication October 11, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
