- Email: [email protected]
Functional expression of metabotropic GABAB receptors in primary cultures of astrocytes from rat cerebral cortex
BBRC Biochemical and Biophysical Research Communications 341 (2006) 874–881 www.elsevier.com/locate/ybbrc
Functional expression of metabotropic GABAB receptors in primary cultures of astrocytes from rat cerebral cortex q Michiko Oka a, Miyuki Wada a, Qiang Wu b, Akira Yamamoto b, Takuya Fujita a
a,b,*
Department of Biochemical Pharmacology, Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto, Japan b Department of Biopharmaceutics, Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto, Japan Received 5 January 2006 Available online 23 January 2006
Abstract GABAB receptor subunits are widely expressed on neurons throughout the central nervous system (CNS), at both pre- and postsynaptic sites, where they mediate the late and slow component of the inhibitory response to the major inhibitory neurotransmitter GABA. Recently, GABAB receptors have been reported to be expressed in astrocytes and microglia in the rat CNS by immunocytochemistry. However, there are few reports available for the functional characterization of GABAB receptors on astrocytes. In the present study, we therefore investigated the functional expression and characteristics of GABAB receptors in primary cultures of astrocytes from rat cerebral cortex. In the presence of 10 lM GTP, forskolin concentration-dependently increased adenylylcyclase (AC) activity in membranes prepared from rat astrocytes. The selective GABAB agonist (R)-baclofen concentration-dependently reduced forskolin-stimulated AC activity in the presence of 10 lM GTP. This effect was reversed by the selective GABAB antagonists, CGP-55845 and CGP-54626, and was completely abolished by treatment of astrocytic membranes with pertussis toxin. In addition, RT-PCR, Western blotting, and immunocytochemistry clearly showed that metabotropic GABAB receptor isoforms (GABABR1 and GABABR2) are expressed in rat cerebrocortical astrocytes. Taken collectively, these results demonstrate that functionally active metabotropic GABAB receptors are expressed in rat cerebrocortical astrocytes. 2006 Elsevier Inc. All rights reserved. Keywords: GABA; GABAB receptor; (R)-Baclofen; Adenylylcyclase; G-protein-coupled receptor; Astrocytes
In mammalian central nervous system, c-aminobutyric acid (GABA) is the main inhibitory neurotransmitter and plays a key role in modulating neuronal activity. GABA interacts with distinct receptor systems, ionotropic (GABAA and GABAC) and metabotropic (GABAB) receptors [1]. The GABAA receptors contain an integral action channel [2] that is gated directly by the binding of GABA [3]. On the other hand, GABAB receptor was first identified as a metabotropic receptor with a pharmacological profile distinct from that of GABAA receptor [4]. GABAB recepq Abbreviations: DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; G3PDH, glyceraldhyde 3-phosphate dehydrogenase; AC, adenylylcyclase; PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate. * Corresponding author. Fax: +81 75 595 4761. E-mail address: [email protected] (T. Fujita).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.01.039
tors belong to members of the seven transmembrane G-protein-coupled receptor superfamily [5], and which can exist as either auto- or heteroreceptors to modulate neurotransmitter release on both pre- and postsynaptic terminals [1]. Dysfunction of GABA-mediated synaptic transmission in the CNS is believed to underlie various nervous system disorders. For example, hypoactivity of the GABA system was linked to epilepsy, spasticity, anxiety, stress, sleep disorders, depression, addiction, and pain. On the contrary, hyperactivity of the GABAergic system was associated with schizophrenia [6]. In neurons, GABAB receptor activation may influence presynaptic transmitter release and cause hyperpolarized neurons postsynaptically via the inhibition of second messenger systems, mainly by adenylylcyclase and by modulation of Ca2+ and K+ channel activities [1,7]. The molecular cloning of GABAB receptors revealed that functional
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
GABAB receptors are obligate heterodimers composed of two receptor subunits GABABR1 and GABABR2 (reviewed by [5,8]). Physiological responses following activation of GABAB receptors require the co-assembly of GABABR1 and GABABR2 (reviewed by [5,9]). Although detailed analysis has indicated that GABAB heterodimer component proteins are expressed widely throughout neurons in the brain, spinal cord, and dorsal ganglia, the expression of GABAB receptors in non-neuronal cells such as astrocytes and microglia has remained to be explored. Glia including astrocytes, the most diverse population of glial cells in CNS, have been believed for a long time that they have a largely passive role as neuronal support cells and play a minor functional role in the CNS. Astrocytes were known to be responsible for the maintenance of perineuronal homeostasis, for example, astrocytes help terminate inhibitory synaptic transmission via GABA uptake mechanism [10]. Recently, Charles et al. [11] have reported that certain types of glial cells, astrocytes, and microglia, from the CNS exhibit GABAB receptor immunoreactivity, suggesting that these cells may play a much more important role in neurotransmission via GABAB receptors and may be a potential target for the GABAB receptor agonist than previously envisaged [12,13]. However, because they did not investigate the functional characterization of GABAB receptors expressed in glial cells, there is few information available for glial GABAB receptor function in the CNS. In the present study, we precisely investigated the functional expression and characteristics of GABAB receptors in primary cultures of astrocytes from rat cerebral cortex and assessed the functional profile associated with adenylylcyclase system via the pertussis toxin (PTX)-sensitive G protein (Gi). Materials and methods Materials. Dulbecco’s modified Eagle’s medium (DMEM), DMEM/ F12, antibiotic/antimycotic solution for tissue culture, carbenicillin, bovine apotransferrin, human insulin, and other culture reagents were purchased from GIBCO-Invitrogen and fetal bovine serum (FBS) was from Eqitech-Bio (Kerrville, TX, USA). Pertussis toxin was purchased from Seikagakukogyo (Tokyo, Japan). (R)-Baclofen was purchased from Sigma-RBI (Natick, MA, USA). CGP-54626 and CGP-55845 were obtained from Tocris Cookson (Bristol, UK). Other chemicals were all of guaranteed grade. Cell culture. All animal procedures were carried out in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committee of Kyoto Pharmaceutical University. Primary astrocyte cultures were prepared from newborn (0- to 2-days-old) cerebral cortices of Wistar rat (Shimizu Laboratory Supplies, Kyoto, Japan) as described previously [14– 16]. After the cerebral cortices were dissected, the meninges were carefully removed in a Ca2+-free Puck’s solution (pH 7.4). Tissues were minced and washed with Ca2+-free Puck’s solution, followed by treatment with 0.1% trypsin dissolved in Ca2+-free Puck’s solution at 37 C for 5 min under the stream of gas mixture of 95% O2/5% CO2. The trypsin digestion was terminated by the addition of ice-cold DMEM supplemented with antibiotic-antimycotic and 20% FBS, and tissues were triturated with a Pasteur pipette. The dispersed cells were collected after centrifugation at 900g at 4 C for 2 min. The resultant pellet was then re-suspended in DMEM containing 10% FBS followed by trituration, then the cell suspension was
875
passed through a nylon sieve (mesh size 70 lm) and seeded on 150-cm2 flask (BD Biosciences). After 2 days in cultures, the medium was renewed. After 1 week, the medium was changed to DMEM/F12 supplemented with 5% FBS, 25 lg/mL insulin, 50 lg/mL apotransferrin, and 1% antibiotic/ antimycotic, and further cultured for 1 week to obtain morphological differentiation of glial cells. This procedure routinely produced astrocyte cultures of >95% purity as assessed by staining with a polyclonal antibody against glial fibrillary acidic protein. In the set of experiments, astrocytes were pretreated with 100 ng/mL pertussis toxin for 15 h as described previously [17]. Membrane preparation. Astrocytic membranes were prepared as per previous reports with a slight modification [17,18]. After washing twice with PBS, primary cultured astrocytes were detached with cell scraper (Corning). The collected cells were homogenized using a Teflon homogenizer (10 up-and-down strokes by hand) in 10 volumes of ice-cold buffer containing 10 mM Tris–HCl (pH 7.4), 1 mM EDTA, and 0.32 M sucrose. The homogenate was centrifuged at 27,000g for 30 min at 4 C. The final pellet was resuspended in 10 mM Tris–HCl (pH 7.4), 1 mM EDTA buffer containing various protease inhibitors at a protein concentration of 10 mg/mL. Protein content was measured with BCA Protein Assay kit (Pierce) using bovine serum albumin as a standard. Cyclic AMP assay in membrane preparations. Adenylylcyclase (AC) activity was measured according to previous reports with a slight modification [17–22]. AC was stimulated by various concentrations of forskolin in the absence or presence of GTP. The reaction buffer contained 2 mM MgCl2, 1 mM dithiothreitol, 0.3 mM EDTA, 10 lM GTP, 10 lM ATP, 1 mM IBMX, 50 mM phosphocreatine, and 50 U/mL creatine phosphokinase in 50 mM Hepes–NaOH (pH 7.4). Unless otherwise indicated, adenylylcyclase was stimulated by 1 lM forskolin. The reaction was initiated by addition of membrane preparations (20 lg protein) to prewarmed reaction buffer (500 lL) in the absence or presence of GABAB ligands, and allowed to proceed for 10 min at 37 C, then terminated by addition of an equal volume of ice-cold 0.1 M HCl. The cyclic AMP content was assayed by enzyme immunoassay using cAMP assay kit (Amersham Biosciences). RT-PCR. Total RNA was isolated from rat cerebrocortical astrocytes using Sepazol RNA I reagent (Nacalai Tesque, Kyoto, Japan) according to the manufacturer’s instructions. Total RNA (5 lg) was used for a reverse transcription reaction (20 lL). First-strand cDNA was synthesized using Superscript II (Invitrogen, Carlsbad, CA, USA) with an oligo(dT) primer. Specific primers (Invitrogen) for AC1-9, GABAB receptors, and glyceraldehyde-3-phosphate dehydrogenase (G3PDH) are summarized in Table 1. All primers used have been shown previously to amplify selectively the specific AC and GABAB receptor isoform from rat tissues [24,28]. PCR was performed according to the following protocol: 94 C for 45 s, adequate annealing temperature (see Table 1) for 90 s, and 72 C for 90 s (35 cycles); 94 C for 45 s, 58 C for 45 s, and 72 C for 90 s (25 cycles) for G3PDH. PCR products were separated on a 1% agarose gel, and were visualized with ethidium bromide under ultraviolet light. Immunocytochemistry. Immunocytochemistry for GABAB receptors was performed according to previous reports [15,23]. Rat cerebrocortical astrocytes were plated onto glass-bottomed dishes. Seven to 10 days after plating, the cells were washed twice with PBS and fixed with 2% paraformaldehyde for 30 min at 4 C. Following fixation of rat cortical astrocytes, the cells were washed three times with PBS, incubated in methanol (20 C) for 5 min, and washed again three times with PBS. The cells were blocked in 10% calf serum/PBS containing 0.3% Tween 20 for 3 h, washed three times in PBS, and incubated in the primary antibodies, rabbit anti-GABAB receptor 1 or GABAB receptor 2 antibody (1:100, Chemicon), overnight at 4 C. The next day, the cells were rinsed three times in PBS for 5 min, incubated with biotinylated goat anti-rabbit IgG (1:200 dilution: Histofine, Nichirei, Osaka, Japan) for 1 h at room temperature, and then were washed again three times with PBS for 5 min. The cells were then incubated with ABC solution (1:100 dilution: Vectastain ABC kit, Vector Laboratories, Burlingame, CA, USA) for 30 min at room temperature and were visualized with 3,3-diaminobenzidine tetrahydrochloride and 0.01% H2O2, and were examined under an Olympus AX80 microscope for observation and photography.
876
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
Table 1 Primers used for adenylylcyclase and GABAB receptor RT-PCR cDNA
Primer sequence (5 0 -to-3 0 )
Product size (bp)
Annealing temperature (C)
Accession No.
AC1
Forward: 5 0 -CCA CGT CCT ACA TCC TCG TT-3 0 Reverse: 5 0 -AAG TGG TAG GGG CAC CTT CT-3 0
521
58
AF 053980
AC2
Forward: 5 0 -CGT GTC ACT CTC CAT ATT C-3 0 Reverse: 5 0 -CCT TGT TCA CAT CTG ACT C-3 0
302
53
M80550
AC3
Forward: 5 0 -CAT CGA GTG TCT ACG CTT C-3 0 Reverse: 5 0 -GGA TGA CCT GTG TCT CTT CT-3 0
442
56
M55075
AC4
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
455
64
M80633
AC5
Forward: 5 0 -ACC ATT GTG CCC CAC TCC CTG TT-3 0 Reverse: 5 0 -TCG TCG CCC AGG CTG TAG TTG AA-3 0
498
62
M96159
AC6
Forward: 5 0 -CAA AGG AAG GGA CGC CGA GAG G-3 0 Reverse: 5 0 -TGG GGA GAG ATC ACG GGA CTA GGA-3 0
417
64
L01115
AC7
Forward: 5 0 -CCA GTT ATT TAG AGA GAG ACC TG-3 0 Reverse: 5 0 -CTT GCT CAT CAG GGC CAT GCT AA-3 0
560
58
U12919
AC8
Forward: 5 0 -TTC ACT TGA GCC TAG CCT CG-3 0 Reverse: 5 0 -GGA TGT AGA TGC GGT GGA AC-3 0
628
58
L26986
AC9
Forward: 5 0 -AGC TTA TCC TCA CCT TCT TCT TCC TC-3 0 Reverse: 5 0 -AGG ACA CGG TAG CAC TCC TTG CC-3 0
312
64
U30602
GABABR1a
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
197
60
Y10369
GABABR1b
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
349
55
Y10370
GABABR1c
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
337 470
58
AB016160
GABABR1
Forward: 5 0 -CGT CTT CTT CTG CTG GTG AT-3 0 Reverse: 5 0 -AAG TCC CAC GAT GAT TCG AG-3 0
693
58
NM_019439
GABABR2
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
705
58
X02231
AC, adenylylcyclase. Data analysis. Each experiment was determined with duplicate or triplicate monolayers. Results are expressed as means ± SEM. Statistical significance was determined by Student’s t test for unpaired sample for assuming equal variance; p < 0.05 was considered significant. Statistical differences among multiple different groups were determined by one-way analysis of variance and p < 0.05 was considered by Dunnett’s test using commercial software (Instat, GraphPad software, San Diego, CA, USA).
Results Identification of adenylylcyclase in membrane of rat astrocytes We first examined the effect of forskolin on adenylylcyclase activity (AC) in membrane of rat astrocytes. As shown in Fig. 1, the basal adenylylcyclase activity in the membrane fraction was quite responsive to forskolin. So far, molecular cloning techniques have identified nine mammalian genes of AC isoforms (AC1–AC9) [24,25] and one gene cluster of a soluble isoform [26]. Among them, inhibitory trimeric G-protein Gi can inhibit AC 5 and 6. Therefore, we investigated the mRNA expression of these AC isoforms in primary cultures of rat cerebrocor-
Fig. 1. Effect of forskolin on adenylylcyclase activity in membranes of rat astrocytes. Membranes (20 lg of membrane protein) were incubated at 37 C for 10 min in the absence or presence of various concentrations of forskolin (1–100 lM) and GTP (10 lM). The reaction buffer contained 10 mM MgCl2, 1 mM dithiothreitol, 1 mM EDTA, 1 mM ATP, and 1 mM IBMX in 200 lL of 5 · 102 M Hepes–NaOH (pH 7.4). Data are expressed as the percentage of the basal cAMP value without forskolin stimulation. Each column represents the mean ± SE of three experiments. *p < 0.05, ***p < 0.001 vs. non-treated (none) group (Dunnett’s test).
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
tical astrocytes by RT-PCR. Using primer pairs specific for each AC isoform, we confirmed the presence of AC2–AC9 in cerebral cortex RNA (Fig. 2B). As shown in Fig. 2A, PCR products corresponding to AC5 and AC6 were clearly detected in primary cultures of rat cerebrocortical astrocytes. These results indicate that AC5 and 6 were functionally expressed in primary cultures of rat cerebrocortical astrocytes. Effects of (R)-baclofen on forskolin-stimulated adenylylcyclase activity in membrane of rat astrocytes GABAB receptor-mediated inhibition of AC in neurons is well known to be mediated by inhibitory G-protein (Gi), and this action requires GTP [27]. To examine whether (R)baclofen inhibits the AC activity via an inhibitory G-protein, we investigated the concentration-dependent effect of GTP. Fig. 3 shows the requirement of GTP for (R)-baclofen-induced inhibition of AC activity. Although forskolin alone markedly stimulated the basal AC activity with or without 10 lM GTP, (R)-baclofen attenuated the forskolin-stimulated AC activity in the presence of 10 lM GTP in a concentration-dependent manner and a significant effect was observed at more than 100 lM of (R)-baclofen (Fig. 3B). On the other hand, (R)-baclofen had no influence even at the highest concentration of 300 lM on forskolin-stimulated AC in the absence of GTP (Fig. 3A). In addition, (R)-baclofen (100 lM) suppressed the forskolin-stimulated AC activity in a GTP-dependent manner (Fig. 4). Inhibition of forskolin-stimulated AC activity by
877
(R)-baclofen (100 lM) was antagonized by specific GABAB receptor antagonists, CGP-54626 (0.1–100 lM) and CGP-55845 (0.1–100 lM), in concentration-dependent manners (Figs. 5A and B), although the potency of inhibitory effect by CGP-54626 was more marked than that by CGP-55845. When the membranes were prepared from PTX-pretreated astrocytes, (R)-baclofen (300 lM) did not affect the forskolin-stimulated AC activity at all, suggesting that the (R)-baclofen-induced inhibition of forskolin-stimulated AC activity is mediated by PTX-sensitive G-protein (Gi). Taken together, these results indicate that functional GABAB receptors are expressed in rat primary cultured cerebrocortical astrocytes. Identification of GABAB receptor expression in primary cultures of astrocytes from rat cerebral cortex To confirm that GABAB receptors were expressed in primary cultures of astrocytes from rat cerebral cortex, we evaluated, by means of RT-PCR, whether the mRNA is coding for GABAB receptors (Figs. 7A and B). Analysis in the cerebral cortex was performed as a positive control (Fig. 7A, lanes 3 and 4). The RT-PCR demonstrated that GABAB receptor mRNAs were expressed in primary cultured of astrocytes from rat cerebral cortex (Fig. 7A, lanes 1 and 2). These findings were also supported by immunocytochemistry. As shown in Fig. 7C, immunoreactive GABABR1 and GABABR2 were detected in primary cultured astrocytes, consistent with the results of RT-PCR. Furthermore, the data obtained from Fig. 7B indicated
A
B
Fig. 2. Expression of adenylylcyclase mRNA in cerebrocortical astrocytes (A) and cerebral cortex in rat. The reactions using specific primer sets were described in text. PCR products were separated by electrophoresis on 1.2% agarose gel and detected with ethidium bromide to visualize bands. M corresponds to 1-kb plus DNA ladder (Invitrogen).
878
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
A
B
Fig. 3. Effect of (R)-baclofen on forskolin-stimulated adenylylcyclase activity in the absence (A) or presence (B) of GTP in membranes of rat astrocytes. Membranes were incubated at 37 C for 10 min in the presence of indicated concentrations of (R)-baclofen (10–300 lM) with or without 10 lM GTP. Each column represents the mean ± SE of three experiments. **p < 0.01 vs. without (R)-baclofen (Dunnett’s test).
Fig. 4. Effect of GTP on inhibition of forskolin-stimulated adenylylcyclase activity by (R)-baclofen in membranes of rat astrocytes. Membranes were incubated at 37 C for 10 min in the presence of indicated concentrations of GTP (108–105 M) with 100 lM (R)-baclofen. Each column represents the mean ± SE of three experiments. *p < 0.05, **p < 0.01 vs. non-treated (none) group (Dunnett’s test).
that GABAB receptor isoforms 1a, 1b, and 1c are expressed in cerebrocortical astrocytes. These results strongly supported that GABAB receptors in primary cultures astrocytes from rat cerebral cortex are coupled to the AC system via an inhibitory G-protein and that it suppresses cyclic AMP formation in rat astrocytes. Discussion The major feature of this report is to demonstrate the functional characterization of GABAB receptors coupled to AC system via an inhibitory GTP-binding protein in astrocytes from rat cerebral cortex. RT-PCR revealed that both GABABR1 and GABABR2 are expressed in astro-
cytes (Fig. 7). The presence of GABABR1 and GABABR2 subunits in GFAP-positive rat astrocytes is further supported by our immunocytochemical analysis. Although, in our experiment, we do not know the precise intracellular localization of GABABR1 and GABABR2 subunits, some literature have reported that the staining of GABABR1 and GABABR2 subunits is widely distributed preferentially in the cell body rather than their process in rat hippocampal astrocytes [11] and appears in the perinuclear space and in the cellular processes in other glial cells, rat peripheral Schwann cells [28]. At present, the precise physiological role of GABAB receptors in astrocytes is not understood. Recently, Magnaghi et al. [28] demonstrated that GABAB receptors are expressed in the glial cells of a peripheral nerve, e.g., the rat sciatic nerve, and influence their proliferation and myelin protein expression. In our experiment, GABAB receptor subunits were observed not only in normal brain but also in primary cultured astrocytes. Astrocytes in normal brain would be thought to be in a resting state compared to those in culture that would be an ‘‘activated’’ state [29]. Our results therefore show that GABAB receptors may play a role in the tonic modulation of astrocyte function including astrocyte-to-neuron signaling and uptake or release of several neurotransmitters [30] since GABAB receptors are expressed both in ‘‘activated’’ and ‘‘resting’’ state astrocytes. In the present study, we showed that GABAB receptors on the rat cerebrocortical astrocytes are expressed functionally active. The features of these receptors in neurons are to be coupled to adenylylcyclase, Ca2+ and K+ channels [1,7]. Among these signals, we measured the inhibition of adenylylcyclase activity as the index of GABAB receptors. In agreement with enormous previous reports using neurons, we provided that (R)-baclofen, a specific agonist
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
A
879
B
Fig. 5. Antagonism of (R)-baclofen-induced inhibition of forskolin-stimulated adenylylcyclase activity by CGP-54626 (A) and CGP-55845 (B) in membranes of rat astrocytes. Membranes were incubated at 37 C for 10 min in the presence of indicated concentrations of CGP-54626 (0.1–100 lM) with 1 lM forskolin and 100 lM (R)-baclofen. Each column represents the mean ± SE of three experiments. *p < 0.05, **p < 0.01 vs. Control (Cont) group (Dunnett’s test).
of GABAB receptor, suppressed the forskolin-stimulated adenylylcyclase activity. Selective antagonists of GABAB receptors such as CGP-54626 and CGP-55845 reversed these effects in a concentration-dependent manner (Fig. 5). Green et al. [31] reported that the rank order of the binding to the GABAB receptors stably expressed in CHO-K1 cells was CGP-54626 > CGP-55845. However, we do not know the precise reason that the potency of CGP-54626 for antagonizing action in our experiment using astrocytes was more marked than that of CGP55845, the differences in indices of GABAB receptor function, cell types or GABAB receptor subunit combination may result in such difference between our data and those reported by others. In astrocytes, (R)-baclofen inhibited GTP-dependently the forskolin-stimulated adenylylcyclase activity (Fig. 4), suggesting that GABAB receptors are coupled to adenylylcyclase system via an inhibitory G-protein (Gi) similar to GABAB receptors in neurons. These results were further supported by the result that (R)-baclofen-mediated inhibition of adenylylcyclase activity was PTX-sensitive (Fig. 6). So far nine mammalian genes of AC isoforms (AC1–AC9) and one gene cluster a soluble isoform were identified [26]. Among ACs, inhibitory trimeric G-protein Gi is reported to be able to inhibit AC5 and AC6 [26]. RT-PCR analysis revealed that both AC isoforms are expressed in rat primary cultured cerebrocortical astrocytes (Fig. 2B), suggesting that AC5 and/or AC6 are responsible for astrocytic GABAB signaling. Glutamate is an important neurotoxin as well as the major excitatory neurotransmitter. Elevated glutamate levels stimulate postsynaptic N-methyl-D-aspartate (NMDA) receptors, which induce a massive Ca2+ entry [32,33]. The resultant activation of a variety of intracellular Ca2+-dependent enzymes such as proteases, nucleases, phospholip-
Fig. 6. Effect of (R)-baclofen on forskolin-stimulated adenylylcyclase activity in PTX-treated membrane of rat astrocytes. Membranes treated with or without PTX were incubated at 37 C for 10 min in the presence of 1 lM forskolin and 300 lM (R)-baclofen. Each column represents the mean ± SE of three experiments.
ases, and nitric oxide synthase (NOS) causes an irreversible neuronal damage [34,35]. Astrocytes were shown to be involved in glutamate uptake by glutamate transporters EAAT1 (GLAST) and EAAT2 (GLT-1) under pathogenic conditions such as acute ischemic brain injury, indicating a neuroprotective role for these cells [36]. A recent study indicates that GABA opposes the effect of glutamate, not only as a neurotransmitter, but also by acting in an opposite fashion on neuronal function. It is reported that GABA protects against anoxic injury in a model of white matter dysfunction, which action was inhibited by GABAB receptor antagonist but not GABAA receptor antagonist, indicating GABA acts at GABAB receptors [37]. Astrocytes
880
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
A
B
C
Fig. 7. RT-PCR analysis and immunocytochemical detection of GABAB receptors. (A) Total RNA was isolated from rat cerebrocortical astrocytes or cerebral cortex, and RT-PCR was performed using specific primer sets of GABABR1 (lanes 1 and 3) and GABABR2 (lanes 2 and 4). Lanes 1 and 2, cerebrocortical astrocytes; lanes 3 and 4, cerebral cortex. The expected size of GABABR1 and GABABR2 was 693 and 705 bp, respectively. (B) Expression of GABABR1 subunits in cerebrocortical astrocytes. RT-PCR showed the presence of three kinds of GABABR1 subunits, GABABR1a (product size 197 bp), GABABR1b (349 bp), and GABABR1c (337 and 470 bp), in rat cerebrocortical astrocytes. M corresponds to 1-kb plus DNA ladder (Invitrogen). (C) Immunocytochemical detection of GABABR1 and GABABR2 in rat cerebrocortical astrocytes. Original magnification: 200·.
are potential candidates mediating this protection via GABAB receptors, since we have now demonstrated that astrocytes express GABAB receptors. More recently, we provided that human astrocytes functionally express voltage-gated Na+ channel as well as L-type Ca2+ channel [23]. Moreover, hypoxic injury induced in cerebrocortical slices by oxygen deprivation in the absence of glucose was inhibited by Na+ channel blockers and N-type or P/ Q-type Ca2+ channel blockers [19–22]. Taken together, under pathologic conditions such as cerebral ischemia, astrocytes are potential targets for protective drugs acting on these receptors and channels. In conclusion, rat cerebrocortical astrocytes express GABAB receptors negatively coupled to adenylylcyclase. Their pharmacological profiles are similar to those of GABAB expressed in neurons. Further studies are required to fully understand the physiological role that these receptors play in cerebrocortical astrocytes and how this complements the well-established roles of GABAB receptors in generating inhibitory synaptic inputs and inhibition of neurotransmitter release. Acknowledgments This work was supported in part by Scientific Research in Priority Areas (16015305, 16047229), Academic Frontier Project ‘‘Research on Medicinal Chemistry Based on Molecular Recognition: Aiming of Development of New Drugs to Overcome Difficult Diseases’’, and the 21st Century Center of Excellent Program ‘‘Development of Drug Discovery Frontier Integrated from Tradition to Proteome’’ from the Ministry of Education, Science, Sports
and Culture of Japan, and by grant from Takeda Science Foundation and Uehara Memorial Foundation. References [1] N.G. Bowery, S.J. Enna, c-Aminobutyric acidB receptors: first of the functional metabotropic heterodimers, J. Pharmacol. Exp. Ther. 292 (2000) 2–7. [2] C. Czajkowski, D.H. Farb, Transmembrane topology and subcellular distribution of the benzodiazepine receptor, J. Neurosci. 6 (1986) 2857–2863. [3] J. Bormann, O.P. Hamill, B. Sakmann, Mechanism of anion permeation through channels gated by glycine and c-aminobutyric acid in mouse cultured spinal neurons, J. Physiol. 385 (1987) 243–286. [4] D.R. Hill, N.G. Bowery, [3H]Baclofen and [3H]GABA bind to bicuculline-sensitive GABAB sites in rat brain, Nature 290 (1981) 149–151. [5] N.G. Bowery, B. Bettler, W. Froestl, J.P. Gallagher, F. Marshall, M. Raiteri, T.I. Bonner, S.J. Enna, International Union of Pharmacology. XXXIII. Mammalian c-aminobutyric acidB receptors: structure and function, Pharmacol. Rev. 54 (2002) 247–264. [6] B.P. Blum, J.J. Mann, The GABAergic system in schizophrenia, Int. J. Neuropsychopharmacol. 5 (2002) 159–179. [7] F.H. Marshall, K.A. Jones, K. Kaupmann, B. Bettler, GABAB receptors—the first 7TM heterodimers, Trends Pharmacol. Sci. 20 (1999) 396–399. [8] A. Billinton, A.O. Ige, J.P. Bolam, J.H. White, F.H. Marshall, P.C. Emson, Advances in the molecular understanding of GABAB receptors, Trends Neurosci. 24 (2001) 277–282. [9] A. Couve, S.J. Moss, M.N. Pangalos, GABAB receptors: a new paradigm in G protein signaling, Mol. Cell. Neurosci. 16 (2000) 296–312. [10] A. Vernadakis, Glia-neuron intercommunications and synaptic plasticity, Prog. Neurobiol. 49 (1996) 185–214. [11] K.J. Charles, J. Deuchars, C.H. Davies, M.N. Pangalos, GABAB receptor subunit expression in glia, Mol. Cell. Neurosci. 24 (2003) 214–223.
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881 [12] J. Kang, L. Jiang, S.A. Goldman, M. Nedergaard, Astrocytemediated potentiation of inhibitory synaptic transmission, Nat. Neurosci. 1 (1998) 683–692. [13] K.J. Charles, A.R. Calver, S. Jourdain, M.N. Pangalos, Distribution of a GABAB-like receptor protein in the rat central nervous system, Brain Res. 989 (2003) 135–146. [14] T. Fujita, H. Katsukawa, E. Yodoya, M. Wada, A. Shimada, N. Okada, A. Yamamoto, V. Ganapathy, Transport characteristics of N-acetyl-L-aspartate in rat astrocytes: involvement of sodium-coupled high-affinity carboxylate transporter NaC3/ NaDC3-mediated transport system, J. Neurochem. 93 (2005) 706–714. [15] M. Wada, S. Miyakawa, A. Shimada, N. Okada, A. Yamamoto, T. Fujita, Functional linkage of H+/peptide transporter PEPT2 and Na+/H+ exchanger in primary cultures of astrocytes from mouse cerebral cortex, Brain Res. 1044 (2005) 33–41. [16] E. Yodoya, M. Wada, A. Shimada, H. Katsukawa, N. Okada, A. Yamamoto, V. Ganapathy, T. Fujita, Functional and molecular identification of sodium-coupled dicarboxylate transporters in rat primary cultured cerebrocortical astrocytes and neurons, J. Neurochem. (2006) (in press). [17] M. Oka, Y. Itoh, T. Shimidzu, Y. Ukai, Y. Yoshikuni, K. Kimura, Involvement of metabotropic glutamate receptors in Gi- and Gsdependent modulation of adenylate cyclase activity induced by a novel cognition enhancer NS-105 in rat brain, Brain Res. 754 (1997) 121–130. [18] T. Shimidzu, Y. Itoh, M. Oka, T. Ishima, Y. Ukai, Y. Yoshikuni, K. Kimura, Effect of a novel cognition enhancer NS-105 on learned helplessness in rats: Possible involvement of GABAB receptor upregulation after repeated treatment, Eur. J. Pharmacol. 338 (1997) 225–232. [19] M. Oka, Y. Itoh, T. Fujita, Halothane attenuates the cerebroprotective action of several Na+ and Ca2+ channel blockers via reversal of their ion channel blockade, Eur. J. Pharmacol. 452 (2002) 175–181. [20] M. Oka, Y. Itoh, Y. Ukai, Involvement of Na+ and Ca2+ channel activation and resultant nitric oxide synthesis in glutamate-mediated hypoxic injury in rat cerebrocortical slices, Life Sci. 67 (2000) 2331– 2343. [21] M. Oka, Y. Itoh, Y. Ukai, Preferential inhibition by a novel Na+/ Ca2+ channel blocker NS-7 of severe to mild hypoxic injury in rat cerebrocortical slices: a possible involvement of a highly voltagedependent blockade of Ca2+ channel, J. Pharmacol. Exp. Ther. 293 (2000) 522–529. [22] M. Oka, Y. Itoh, M. Wada, A. Yamamoto, T. Fujita, A comparison of Ca2+ channel blocking mode between gabapentin and verapamil:
[23]
[24]
[25]
[26] [27]
[28]
[29]
[30]
[31]
[32]
[33]
[34] [35] [36] [37]
881
implication for protection against hypoxic injury in rat cerebrocortical slices, Br. J. Pharmacol. 139 (2003) 435–443. M. Oka, M. Wada, A. Yamamoto, Y. Itoh, T. Fujita, Functional expression of constitutive nitric oxide synthases regulated by voltagegated Na+ and Ca2+ channels in cultured human astrocytes, Glia 46 (2004) 53–62. M.J. Bek, S. Zheng, J. Xu, I. Yamaguchi, L.D. Asico, X.-G. Sun, P.A. Jose, Differential expression of adenylyl cyclases in the rat nephron, Kidney Int. 60 (2001) 890–899. T. Abaffy, K.R. Trubey, N. Chaudhari, Adenylyl cyclase expression and modulation of cAMP in rat taste cells, Am. J. Physiol. Cell Physiol. 284 (2003) C1420–C1428. R.K. Sunahara, R. Taussig, Isoforms of mammalian adenylyl cyclase: multiplicities of signaling, Mol. Interv. 2 (2002) 168–184. W.J. Wojcik, I. Holopainen, Role of central GABAB receptors in physiology and pathology, Neuropsychopharmacology 6 (1992) 201– 214. V. Magnaghi, M. Ballabio, I.T.R. Cavarretta, W. Froestl, J.J. Lambert, I. Zucchi, R.C. Melcang, GABAB receptors in Schwann cells influence proliferation and myelin protein expression, Eur. J. Neurosci. 19 (2004) 2641–2649. J.T. Porter, K.D. McCarthy, Hippocampal astrocytes in situ respond to glutamate released from synaptic terminals, J. Neurosci. 16 (1996) 5073–5081. A. Araque, V. Parpura, R.P. Sanzgiri, P.G. Haydon, Tripartite synapses: glia, the unacknowledged partner, Trends Neurosci. 22 (1999) 208–215. A. Green, S. Walls, A. Wise, R.H. Green, A.K. Martin, F.H. Marshall, Characterization of [3H]-CGP54626A binding to heterodimeric GABAB receptors stably expressed in mammalian cells, Br. J. Pharmacol. 131 (2000) 1766–1774. A. Schousboe, A. Frandsen, P. Krogsgaard-Larsen, Pharmacological and functional characterization of excitatory amino acid mediated cytotoxicity in cerebral cortical neurons, Cell. Biol. Toxicol. 8 (1992) 93–100. E.L. Gleason, N.C. Spitzer, AMPA and NMDA receptors expressed by differentiating Xenopus spinal neurons, J. Neurophysiol. 79 (1998) 2986–2998. S. Orrenius, D.J. McConkey, G. Bellomo, P. Nicotera, Role of Ca2+ in toxic cell killing, Trends Pharmacol. Sci. 10 (1989) 281–285. D.W. Choi, Excitotoxic cell death, J. Neurobiol. 23 (1992) 1261–1276. Y. Chen, R.A. Swanson, Astrocytes and brain injury, J. Cereb. Blood Flow Metab. 23 (2003) 137–149. R. Fern, S.G. Waxman, B.R. Ransom, Endogenous GABA attenuates CNS white matter dysfunction following anoxia, J. Neurosci. 15 (1995) 699–708.
Functional expression of metabotropic GABAB receptors in primary cultures of astrocytes from rat cerebral cortex q Michiko Oka a, Miyuki Wada a, Qiang Wu b, Akira Yamamoto b, Takuya Fujita a
a,b,*
Department of Biochemical Pharmacology, Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto, Japan b Department of Biopharmaceutics, Kyoto Pharmaceutical University, Misasagi, Yamashina-ku, Kyoto, Japan Received 5 January 2006 Available online 23 January 2006
Abstract GABAB receptor subunits are widely expressed on neurons throughout the central nervous system (CNS), at both pre- and postsynaptic sites, where they mediate the late and slow component of the inhibitory response to the major inhibitory neurotransmitter GABA. Recently, GABAB receptors have been reported to be expressed in astrocytes and microglia in the rat CNS by immunocytochemistry. However, there are few reports available for the functional characterization of GABAB receptors on astrocytes. In the present study, we therefore investigated the functional expression and characteristics of GABAB receptors in primary cultures of astrocytes from rat cerebral cortex. In the presence of 10 lM GTP, forskolin concentration-dependently increased adenylylcyclase (AC) activity in membranes prepared from rat astrocytes. The selective GABAB agonist (R)-baclofen concentration-dependently reduced forskolin-stimulated AC activity in the presence of 10 lM GTP. This effect was reversed by the selective GABAB antagonists, CGP-55845 and CGP-54626, and was completely abolished by treatment of astrocytic membranes with pertussis toxin. In addition, RT-PCR, Western blotting, and immunocytochemistry clearly showed that metabotropic GABAB receptor isoforms (GABABR1 and GABABR2) are expressed in rat cerebrocortical astrocytes. Taken collectively, these results demonstrate that functionally active metabotropic GABAB receptors are expressed in rat cerebrocortical astrocytes. 2006 Elsevier Inc. All rights reserved. Keywords: GABA; GABAB receptor; (R)-Baclofen; Adenylylcyclase; G-protein-coupled receptor; Astrocytes
In mammalian central nervous system, c-aminobutyric acid (GABA) is the main inhibitory neurotransmitter and plays a key role in modulating neuronal activity. GABA interacts with distinct receptor systems, ionotropic (GABAA and GABAC) and metabotropic (GABAB) receptors [1]. The GABAA receptors contain an integral action channel [2] that is gated directly by the binding of GABA [3]. On the other hand, GABAB receptor was first identified as a metabotropic receptor with a pharmacological profile distinct from that of GABAA receptor [4]. GABAB recepq Abbreviations: DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; G3PDH, glyceraldhyde 3-phosphate dehydrogenase; AC, adenylylcyclase; PAGE, polyacrylamide gel electrophoresis; SDS, sodium dodecyl sulfate. * Corresponding author. Fax: +81 75 595 4761. E-mail address: [email protected] (T. Fujita).
0006-291X/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.01.039
tors belong to members of the seven transmembrane G-protein-coupled receptor superfamily [5], and which can exist as either auto- or heteroreceptors to modulate neurotransmitter release on both pre- and postsynaptic terminals [1]. Dysfunction of GABA-mediated synaptic transmission in the CNS is believed to underlie various nervous system disorders. For example, hypoactivity of the GABA system was linked to epilepsy, spasticity, anxiety, stress, sleep disorders, depression, addiction, and pain. On the contrary, hyperactivity of the GABAergic system was associated with schizophrenia [6]. In neurons, GABAB receptor activation may influence presynaptic transmitter release and cause hyperpolarized neurons postsynaptically via the inhibition of second messenger systems, mainly by adenylylcyclase and by modulation of Ca2+ and K+ channel activities [1,7]. The molecular cloning of GABAB receptors revealed that functional
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
GABAB receptors are obligate heterodimers composed of two receptor subunits GABABR1 and GABABR2 (reviewed by [5,8]). Physiological responses following activation of GABAB receptors require the co-assembly of GABABR1 and GABABR2 (reviewed by [5,9]). Although detailed analysis has indicated that GABAB heterodimer component proteins are expressed widely throughout neurons in the brain, spinal cord, and dorsal ganglia, the expression of GABAB receptors in non-neuronal cells such as astrocytes and microglia has remained to be explored. Glia including astrocytes, the most diverse population of glial cells in CNS, have been believed for a long time that they have a largely passive role as neuronal support cells and play a minor functional role in the CNS. Astrocytes were known to be responsible for the maintenance of perineuronal homeostasis, for example, astrocytes help terminate inhibitory synaptic transmission via GABA uptake mechanism [10]. Recently, Charles et al. [11] have reported that certain types of glial cells, astrocytes, and microglia, from the CNS exhibit GABAB receptor immunoreactivity, suggesting that these cells may play a much more important role in neurotransmission via GABAB receptors and may be a potential target for the GABAB receptor agonist than previously envisaged [12,13]. However, because they did not investigate the functional characterization of GABAB receptors expressed in glial cells, there is few information available for glial GABAB receptor function in the CNS. In the present study, we precisely investigated the functional expression and characteristics of GABAB receptors in primary cultures of astrocytes from rat cerebral cortex and assessed the functional profile associated with adenylylcyclase system via the pertussis toxin (PTX)-sensitive G protein (Gi). Materials and methods Materials. Dulbecco’s modified Eagle’s medium (DMEM), DMEM/ F12, antibiotic/antimycotic solution for tissue culture, carbenicillin, bovine apotransferrin, human insulin, and other culture reagents were purchased from GIBCO-Invitrogen and fetal bovine serum (FBS) was from Eqitech-Bio (Kerrville, TX, USA). Pertussis toxin was purchased from Seikagakukogyo (Tokyo, Japan). (R)-Baclofen was purchased from Sigma-RBI (Natick, MA, USA). CGP-54626 and CGP-55845 were obtained from Tocris Cookson (Bristol, UK). Other chemicals were all of guaranteed grade. Cell culture. All animal procedures were carried out in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committee of Kyoto Pharmaceutical University. Primary astrocyte cultures were prepared from newborn (0- to 2-days-old) cerebral cortices of Wistar rat (Shimizu Laboratory Supplies, Kyoto, Japan) as described previously [14– 16]. After the cerebral cortices were dissected, the meninges were carefully removed in a Ca2+-free Puck’s solution (pH 7.4). Tissues were minced and washed with Ca2+-free Puck’s solution, followed by treatment with 0.1% trypsin dissolved in Ca2+-free Puck’s solution at 37 C for 5 min under the stream of gas mixture of 95% O2/5% CO2. The trypsin digestion was terminated by the addition of ice-cold DMEM supplemented with antibiotic-antimycotic and 20% FBS, and tissues were triturated with a Pasteur pipette. The dispersed cells were collected after centrifugation at 900g at 4 C for 2 min. The resultant pellet was then re-suspended in DMEM containing 10% FBS followed by trituration, then the cell suspension was
875
passed through a nylon sieve (mesh size 70 lm) and seeded on 150-cm2 flask (BD Biosciences). After 2 days in cultures, the medium was renewed. After 1 week, the medium was changed to DMEM/F12 supplemented with 5% FBS, 25 lg/mL insulin, 50 lg/mL apotransferrin, and 1% antibiotic/ antimycotic, and further cultured for 1 week to obtain morphological differentiation of glial cells. This procedure routinely produced astrocyte cultures of >95% purity as assessed by staining with a polyclonal antibody against glial fibrillary acidic protein. In the set of experiments, astrocytes were pretreated with 100 ng/mL pertussis toxin for 15 h as described previously [17]. Membrane preparation. Astrocytic membranes were prepared as per previous reports with a slight modification [17,18]. After washing twice with PBS, primary cultured astrocytes were detached with cell scraper (Corning). The collected cells were homogenized using a Teflon homogenizer (10 up-and-down strokes by hand) in 10 volumes of ice-cold buffer containing 10 mM Tris–HCl (pH 7.4), 1 mM EDTA, and 0.32 M sucrose. The homogenate was centrifuged at 27,000g for 30 min at 4 C. The final pellet was resuspended in 10 mM Tris–HCl (pH 7.4), 1 mM EDTA buffer containing various protease inhibitors at a protein concentration of 10 mg/mL. Protein content was measured with BCA Protein Assay kit (Pierce) using bovine serum albumin as a standard. Cyclic AMP assay in membrane preparations. Adenylylcyclase (AC) activity was measured according to previous reports with a slight modification [17–22]. AC was stimulated by various concentrations of forskolin in the absence or presence of GTP. The reaction buffer contained 2 mM MgCl2, 1 mM dithiothreitol, 0.3 mM EDTA, 10 lM GTP, 10 lM ATP, 1 mM IBMX, 50 mM phosphocreatine, and 50 U/mL creatine phosphokinase in 50 mM Hepes–NaOH (pH 7.4). Unless otherwise indicated, adenylylcyclase was stimulated by 1 lM forskolin. The reaction was initiated by addition of membrane preparations (20 lg protein) to prewarmed reaction buffer (500 lL) in the absence or presence of GABAB ligands, and allowed to proceed for 10 min at 37 C, then terminated by addition of an equal volume of ice-cold 0.1 M HCl. The cyclic AMP content was assayed by enzyme immunoassay using cAMP assay kit (Amersham Biosciences). RT-PCR. Total RNA was isolated from rat cerebrocortical astrocytes using Sepazol RNA I reagent (Nacalai Tesque, Kyoto, Japan) according to the manufacturer’s instructions. Total RNA (5 lg) was used for a reverse transcription reaction (20 lL). First-strand cDNA was synthesized using Superscript II (Invitrogen, Carlsbad, CA, USA) with an oligo(dT) primer. Specific primers (Invitrogen) for AC1-9, GABAB receptors, and glyceraldehyde-3-phosphate dehydrogenase (G3PDH) are summarized in Table 1. All primers used have been shown previously to amplify selectively the specific AC and GABAB receptor isoform from rat tissues [24,28]. PCR was performed according to the following protocol: 94 C for 45 s, adequate annealing temperature (see Table 1) for 90 s, and 72 C for 90 s (35 cycles); 94 C for 45 s, 58 C for 45 s, and 72 C for 90 s (25 cycles) for G3PDH. PCR products were separated on a 1% agarose gel, and were visualized with ethidium bromide under ultraviolet light. Immunocytochemistry. Immunocytochemistry for GABAB receptors was performed according to previous reports [15,23]. Rat cerebrocortical astrocytes were plated onto glass-bottomed dishes. Seven to 10 days after plating, the cells were washed twice with PBS and fixed with 2% paraformaldehyde for 30 min at 4 C. Following fixation of rat cortical astrocytes, the cells were washed three times with PBS, incubated in methanol (20 C) for 5 min, and washed again three times with PBS. The cells were blocked in 10% calf serum/PBS containing 0.3% Tween 20 for 3 h, washed three times in PBS, and incubated in the primary antibodies, rabbit anti-GABAB receptor 1 or GABAB receptor 2 antibody (1:100, Chemicon), overnight at 4 C. The next day, the cells were rinsed three times in PBS for 5 min, incubated with biotinylated goat anti-rabbit IgG (1:200 dilution: Histofine, Nichirei, Osaka, Japan) for 1 h at room temperature, and then were washed again three times with PBS for 5 min. The cells were then incubated with ABC solution (1:100 dilution: Vectastain ABC kit, Vector Laboratories, Burlingame, CA, USA) for 30 min at room temperature and were visualized with 3,3-diaminobenzidine tetrahydrochloride and 0.01% H2O2, and were examined under an Olympus AX80 microscope for observation and photography.
876
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
Table 1 Primers used for adenylylcyclase and GABAB receptor RT-PCR cDNA
Primer sequence (5 0 -to-3 0 )
Product size (bp)
Annealing temperature (C)
Accession No.
AC1
Forward: 5 0 -CCA CGT CCT ACA TCC TCG TT-3 0 Reverse: 5 0 -AAG TGG TAG GGG CAC CTT CT-3 0
521
58
AF 053980
AC2
Forward: 5 0 -CGT GTC ACT CTC CAT ATT C-3 0 Reverse: 5 0 -CCT TGT TCA CAT CTG ACT C-3 0
302
53
M80550
AC3
Forward: 5 0 -CAT CGA GTG TCT ACG CTT C-3 0 Reverse: 5 0 -GGA TGA CCT GTG TCT CTT CT-3 0
442
56
M55075
AC4
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
455
64
M80633
AC5
Forward: 5 0 -ACC ATT GTG CCC CAC TCC CTG TT-3 0 Reverse: 5 0 -TCG TCG CCC AGG CTG TAG TTG AA-3 0
498
62
M96159
AC6
Forward: 5 0 -CAA AGG AAG GGA CGC CGA GAG G-3 0 Reverse: 5 0 -TGG GGA GAG ATC ACG GGA CTA GGA-3 0
417
64
L01115
AC7
Forward: 5 0 -CCA GTT ATT TAG AGA GAG ACC TG-3 0 Reverse: 5 0 -CTT GCT CAT CAG GGC CAT GCT AA-3 0
560
58
U12919
AC8
Forward: 5 0 -TTC ACT TGA GCC TAG CCT CG-3 0 Reverse: 5 0 -GGA TGT AGA TGC GGT GGA AC-3 0
628
58
L26986
AC9
Forward: 5 0 -AGC TTA TCC TCA CCT TCT TCT TCC TC-3 0 Reverse: 5 0 -AGG ACA CGG TAG CAC TCC TTG CC-3 0
312
64
U30602
GABABR1a
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
197
60
Y10369
GABABR1b
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
349
55
Y10370
GABABR1c
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
337 470
58
AB016160
GABABR1
Forward: 5 0 -CGT CTT CTT CTG CTG GTG AT-3 0 Reverse: 5 0 -AAG TCC CAC GAT GAT TCG AG-3 0
693
58
NM_019439
GABABR2
Forward: 5 0 -GGA AGA CGA GAA GGG CAC CGA GAG-3 0 Reverse: 5 0 -GAG CTG GGG TGG TGG TCA C-3 0
705
58
X02231
AC, adenylylcyclase. Data analysis. Each experiment was determined with duplicate or triplicate monolayers. Results are expressed as means ± SEM. Statistical significance was determined by Student’s t test for unpaired sample for assuming equal variance; p < 0.05 was considered significant. Statistical differences among multiple different groups were determined by one-way analysis of variance and p < 0.05 was considered by Dunnett’s test using commercial software (Instat, GraphPad software, San Diego, CA, USA).
Results Identification of adenylylcyclase in membrane of rat astrocytes We first examined the effect of forskolin on adenylylcyclase activity (AC) in membrane of rat astrocytes. As shown in Fig. 1, the basal adenylylcyclase activity in the membrane fraction was quite responsive to forskolin. So far, molecular cloning techniques have identified nine mammalian genes of AC isoforms (AC1–AC9) [24,25] and one gene cluster of a soluble isoform [26]. Among them, inhibitory trimeric G-protein Gi can inhibit AC 5 and 6. Therefore, we investigated the mRNA expression of these AC isoforms in primary cultures of rat cerebrocor-
Fig. 1. Effect of forskolin on adenylylcyclase activity in membranes of rat astrocytes. Membranes (20 lg of membrane protein) were incubated at 37 C for 10 min in the absence or presence of various concentrations of forskolin (1–100 lM) and GTP (10 lM). The reaction buffer contained 10 mM MgCl2, 1 mM dithiothreitol, 1 mM EDTA, 1 mM ATP, and 1 mM IBMX in 200 lL of 5 · 102 M Hepes–NaOH (pH 7.4). Data are expressed as the percentage of the basal cAMP value without forskolin stimulation. Each column represents the mean ± SE of three experiments. *p < 0.05, ***p < 0.001 vs. non-treated (none) group (Dunnett’s test).
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
tical astrocytes by RT-PCR. Using primer pairs specific for each AC isoform, we confirmed the presence of AC2–AC9 in cerebral cortex RNA (Fig. 2B). As shown in Fig. 2A, PCR products corresponding to AC5 and AC6 were clearly detected in primary cultures of rat cerebrocortical astrocytes. These results indicate that AC5 and 6 were functionally expressed in primary cultures of rat cerebrocortical astrocytes. Effects of (R)-baclofen on forskolin-stimulated adenylylcyclase activity in membrane of rat astrocytes GABAB receptor-mediated inhibition of AC in neurons is well known to be mediated by inhibitory G-protein (Gi), and this action requires GTP [27]. To examine whether (R)baclofen inhibits the AC activity via an inhibitory G-protein, we investigated the concentration-dependent effect of GTP. Fig. 3 shows the requirement of GTP for (R)-baclofen-induced inhibition of AC activity. Although forskolin alone markedly stimulated the basal AC activity with or without 10 lM GTP, (R)-baclofen attenuated the forskolin-stimulated AC activity in the presence of 10 lM GTP in a concentration-dependent manner and a significant effect was observed at more than 100 lM of (R)-baclofen (Fig. 3B). On the other hand, (R)-baclofen had no influence even at the highest concentration of 300 lM on forskolin-stimulated AC in the absence of GTP (Fig. 3A). In addition, (R)-baclofen (100 lM) suppressed the forskolin-stimulated AC activity in a GTP-dependent manner (Fig. 4). Inhibition of forskolin-stimulated AC activity by
877
(R)-baclofen (100 lM) was antagonized by specific GABAB receptor antagonists, CGP-54626 (0.1–100 lM) and CGP-55845 (0.1–100 lM), in concentration-dependent manners (Figs. 5A and B), although the potency of inhibitory effect by CGP-54626 was more marked than that by CGP-55845. When the membranes were prepared from PTX-pretreated astrocytes, (R)-baclofen (300 lM) did not affect the forskolin-stimulated AC activity at all, suggesting that the (R)-baclofen-induced inhibition of forskolin-stimulated AC activity is mediated by PTX-sensitive G-protein (Gi). Taken together, these results indicate that functional GABAB receptors are expressed in rat primary cultured cerebrocortical astrocytes. Identification of GABAB receptor expression in primary cultures of astrocytes from rat cerebral cortex To confirm that GABAB receptors were expressed in primary cultures of astrocytes from rat cerebral cortex, we evaluated, by means of RT-PCR, whether the mRNA is coding for GABAB receptors (Figs. 7A and B). Analysis in the cerebral cortex was performed as a positive control (Fig. 7A, lanes 3 and 4). The RT-PCR demonstrated that GABAB receptor mRNAs were expressed in primary cultured of astrocytes from rat cerebral cortex (Fig. 7A, lanes 1 and 2). These findings were also supported by immunocytochemistry. As shown in Fig. 7C, immunoreactive GABABR1 and GABABR2 were detected in primary cultured astrocytes, consistent with the results of RT-PCR. Furthermore, the data obtained from Fig. 7B indicated
A
B
Fig. 2. Expression of adenylylcyclase mRNA in cerebrocortical astrocytes (A) and cerebral cortex in rat. The reactions using specific primer sets were described in text. PCR products were separated by electrophoresis on 1.2% agarose gel and detected with ethidium bromide to visualize bands. M corresponds to 1-kb plus DNA ladder (Invitrogen).
878
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
A
B
Fig. 3. Effect of (R)-baclofen on forskolin-stimulated adenylylcyclase activity in the absence (A) or presence (B) of GTP in membranes of rat astrocytes. Membranes were incubated at 37 C for 10 min in the presence of indicated concentrations of (R)-baclofen (10–300 lM) with or without 10 lM GTP. Each column represents the mean ± SE of three experiments. **p < 0.01 vs. without (R)-baclofen (Dunnett’s test).
Fig. 4. Effect of GTP on inhibition of forskolin-stimulated adenylylcyclase activity by (R)-baclofen in membranes of rat astrocytes. Membranes were incubated at 37 C for 10 min in the presence of indicated concentrations of GTP (108–105 M) with 100 lM (R)-baclofen. Each column represents the mean ± SE of three experiments. *p < 0.05, **p < 0.01 vs. non-treated (none) group (Dunnett’s test).
that GABAB receptor isoforms 1a, 1b, and 1c are expressed in cerebrocortical astrocytes. These results strongly supported that GABAB receptors in primary cultures astrocytes from rat cerebral cortex are coupled to the AC system via an inhibitory G-protein and that it suppresses cyclic AMP formation in rat astrocytes. Discussion The major feature of this report is to demonstrate the functional characterization of GABAB receptors coupled to AC system via an inhibitory GTP-binding protein in astrocytes from rat cerebral cortex. RT-PCR revealed that both GABABR1 and GABABR2 are expressed in astro-
cytes (Fig. 7). The presence of GABABR1 and GABABR2 subunits in GFAP-positive rat astrocytes is further supported by our immunocytochemical analysis. Although, in our experiment, we do not know the precise intracellular localization of GABABR1 and GABABR2 subunits, some literature have reported that the staining of GABABR1 and GABABR2 subunits is widely distributed preferentially in the cell body rather than their process in rat hippocampal astrocytes [11] and appears in the perinuclear space and in the cellular processes in other glial cells, rat peripheral Schwann cells [28]. At present, the precise physiological role of GABAB receptors in astrocytes is not understood. Recently, Magnaghi et al. [28] demonstrated that GABAB receptors are expressed in the glial cells of a peripheral nerve, e.g., the rat sciatic nerve, and influence their proliferation and myelin protein expression. In our experiment, GABAB receptor subunits were observed not only in normal brain but also in primary cultured astrocytes. Astrocytes in normal brain would be thought to be in a resting state compared to those in culture that would be an ‘‘activated’’ state [29]. Our results therefore show that GABAB receptors may play a role in the tonic modulation of astrocyte function including astrocyte-to-neuron signaling and uptake or release of several neurotransmitters [30] since GABAB receptors are expressed both in ‘‘activated’’ and ‘‘resting’’ state astrocytes. In the present study, we showed that GABAB receptors on the rat cerebrocortical astrocytes are expressed functionally active. The features of these receptors in neurons are to be coupled to adenylylcyclase, Ca2+ and K+ channels [1,7]. Among these signals, we measured the inhibition of adenylylcyclase activity as the index of GABAB receptors. In agreement with enormous previous reports using neurons, we provided that (R)-baclofen, a specific agonist
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
A
879
B
Fig. 5. Antagonism of (R)-baclofen-induced inhibition of forskolin-stimulated adenylylcyclase activity by CGP-54626 (A) and CGP-55845 (B) in membranes of rat astrocytes. Membranes were incubated at 37 C for 10 min in the presence of indicated concentrations of CGP-54626 (0.1–100 lM) with 1 lM forskolin and 100 lM (R)-baclofen. Each column represents the mean ± SE of three experiments. *p < 0.05, **p < 0.01 vs. Control (Cont) group (Dunnett’s test).
of GABAB receptor, suppressed the forskolin-stimulated adenylylcyclase activity. Selective antagonists of GABAB receptors such as CGP-54626 and CGP-55845 reversed these effects in a concentration-dependent manner (Fig. 5). Green et al. [31] reported that the rank order of the binding to the GABAB receptors stably expressed in CHO-K1 cells was CGP-54626 > CGP-55845. However, we do not know the precise reason that the potency of CGP-54626 for antagonizing action in our experiment using astrocytes was more marked than that of CGP55845, the differences in indices of GABAB receptor function, cell types or GABAB receptor subunit combination may result in such difference between our data and those reported by others. In astrocytes, (R)-baclofen inhibited GTP-dependently the forskolin-stimulated adenylylcyclase activity (Fig. 4), suggesting that GABAB receptors are coupled to adenylylcyclase system via an inhibitory G-protein (Gi) similar to GABAB receptors in neurons. These results were further supported by the result that (R)-baclofen-mediated inhibition of adenylylcyclase activity was PTX-sensitive (Fig. 6). So far nine mammalian genes of AC isoforms (AC1–AC9) and one gene cluster a soluble isoform were identified [26]. Among ACs, inhibitory trimeric G-protein Gi is reported to be able to inhibit AC5 and AC6 [26]. RT-PCR analysis revealed that both AC isoforms are expressed in rat primary cultured cerebrocortical astrocytes (Fig. 2B), suggesting that AC5 and/or AC6 are responsible for astrocytic GABAB signaling. Glutamate is an important neurotoxin as well as the major excitatory neurotransmitter. Elevated glutamate levels stimulate postsynaptic N-methyl-D-aspartate (NMDA) receptors, which induce a massive Ca2+ entry [32,33]. The resultant activation of a variety of intracellular Ca2+-dependent enzymes such as proteases, nucleases, phospholip-
Fig. 6. Effect of (R)-baclofen on forskolin-stimulated adenylylcyclase activity in PTX-treated membrane of rat astrocytes. Membranes treated with or without PTX were incubated at 37 C for 10 min in the presence of 1 lM forskolin and 300 lM (R)-baclofen. Each column represents the mean ± SE of three experiments.
ases, and nitric oxide synthase (NOS) causes an irreversible neuronal damage [34,35]. Astrocytes were shown to be involved in glutamate uptake by glutamate transporters EAAT1 (GLAST) and EAAT2 (GLT-1) under pathogenic conditions such as acute ischemic brain injury, indicating a neuroprotective role for these cells [36]. A recent study indicates that GABA opposes the effect of glutamate, not only as a neurotransmitter, but also by acting in an opposite fashion on neuronal function. It is reported that GABA protects against anoxic injury in a model of white matter dysfunction, which action was inhibited by GABAB receptor antagonist but not GABAA receptor antagonist, indicating GABA acts at GABAB receptors [37]. Astrocytes
880
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881
A
B
C
Fig. 7. RT-PCR analysis and immunocytochemical detection of GABAB receptors. (A) Total RNA was isolated from rat cerebrocortical astrocytes or cerebral cortex, and RT-PCR was performed using specific primer sets of GABABR1 (lanes 1 and 3) and GABABR2 (lanes 2 and 4). Lanes 1 and 2, cerebrocortical astrocytes; lanes 3 and 4, cerebral cortex. The expected size of GABABR1 and GABABR2 was 693 and 705 bp, respectively. (B) Expression of GABABR1 subunits in cerebrocortical astrocytes. RT-PCR showed the presence of three kinds of GABABR1 subunits, GABABR1a (product size 197 bp), GABABR1b (349 bp), and GABABR1c (337 and 470 bp), in rat cerebrocortical astrocytes. M corresponds to 1-kb plus DNA ladder (Invitrogen). (C) Immunocytochemical detection of GABABR1 and GABABR2 in rat cerebrocortical astrocytes. Original magnification: 200·.
are potential candidates mediating this protection via GABAB receptors, since we have now demonstrated that astrocytes express GABAB receptors. More recently, we provided that human astrocytes functionally express voltage-gated Na+ channel as well as L-type Ca2+ channel [23]. Moreover, hypoxic injury induced in cerebrocortical slices by oxygen deprivation in the absence of glucose was inhibited by Na+ channel blockers and N-type or P/ Q-type Ca2+ channel blockers [19–22]. Taken together, under pathologic conditions such as cerebral ischemia, astrocytes are potential targets for protective drugs acting on these receptors and channels. In conclusion, rat cerebrocortical astrocytes express GABAB receptors negatively coupled to adenylylcyclase. Their pharmacological profiles are similar to those of GABAB expressed in neurons. Further studies are required to fully understand the physiological role that these receptors play in cerebrocortical astrocytes and how this complements the well-established roles of GABAB receptors in generating inhibitory synaptic inputs and inhibition of neurotransmitter release. Acknowledgments This work was supported in part by Scientific Research in Priority Areas (16015305, 16047229), Academic Frontier Project ‘‘Research on Medicinal Chemistry Based on Molecular Recognition: Aiming of Development of New Drugs to Overcome Difficult Diseases’’, and the 21st Century Center of Excellent Program ‘‘Development of Drug Discovery Frontier Integrated from Tradition to Proteome’’ from the Ministry of Education, Science, Sports
and Culture of Japan, and by grant from Takeda Science Foundation and Uehara Memorial Foundation. References [1] N.G. Bowery, S.J. Enna, c-Aminobutyric acidB receptors: first of the functional metabotropic heterodimers, J. Pharmacol. Exp. Ther. 292 (2000) 2–7. [2] C. Czajkowski, D.H. Farb, Transmembrane topology and subcellular distribution of the benzodiazepine receptor, J. Neurosci. 6 (1986) 2857–2863. [3] J. Bormann, O.P. Hamill, B. Sakmann, Mechanism of anion permeation through channels gated by glycine and c-aminobutyric acid in mouse cultured spinal neurons, J. Physiol. 385 (1987) 243–286. [4] D.R. Hill, N.G. Bowery, [3H]Baclofen and [3H]GABA bind to bicuculline-sensitive GABAB sites in rat brain, Nature 290 (1981) 149–151. [5] N.G. Bowery, B. Bettler, W. Froestl, J.P. Gallagher, F. Marshall, M. Raiteri, T.I. Bonner, S.J. Enna, International Union of Pharmacology. XXXIII. Mammalian c-aminobutyric acidB receptors: structure and function, Pharmacol. Rev. 54 (2002) 247–264. [6] B.P. Blum, J.J. Mann, The GABAergic system in schizophrenia, Int. J. Neuropsychopharmacol. 5 (2002) 159–179. [7] F.H. Marshall, K.A. Jones, K. Kaupmann, B. Bettler, GABAB receptors—the first 7TM heterodimers, Trends Pharmacol. Sci. 20 (1999) 396–399. [8] A. Billinton, A.O. Ige, J.P. Bolam, J.H. White, F.H. Marshall, P.C. Emson, Advances in the molecular understanding of GABAB receptors, Trends Neurosci. 24 (2001) 277–282. [9] A. Couve, S.J. Moss, M.N. Pangalos, GABAB receptors: a new paradigm in G protein signaling, Mol. Cell. Neurosci. 16 (2000) 296–312. [10] A. Vernadakis, Glia-neuron intercommunications and synaptic plasticity, Prog. Neurobiol. 49 (1996) 185–214. [11] K.J. Charles, J. Deuchars, C.H. Davies, M.N. Pangalos, GABAB receptor subunit expression in glia, Mol. Cell. Neurosci. 24 (2003) 214–223.
M. Oka et al. / Biochemical and Biophysical Research Communications 341 (2006) 874–881 [12] J. Kang, L. Jiang, S.A. Goldman, M. Nedergaard, Astrocytemediated potentiation of inhibitory synaptic transmission, Nat. Neurosci. 1 (1998) 683–692. [13] K.J. Charles, A.R. Calver, S. Jourdain, M.N. Pangalos, Distribution of a GABAB-like receptor protein in the rat central nervous system, Brain Res. 989 (2003) 135–146. [14] T. Fujita, H. Katsukawa, E. Yodoya, M. Wada, A. Shimada, N. Okada, A. Yamamoto, V. Ganapathy, Transport characteristics of N-acetyl-L-aspartate in rat astrocytes: involvement of sodium-coupled high-affinity carboxylate transporter NaC3/ NaDC3-mediated transport system, J. Neurochem. 93 (2005) 706–714. [15] M. Wada, S. Miyakawa, A. Shimada, N. Okada, A. Yamamoto, T. Fujita, Functional linkage of H+/peptide transporter PEPT2 and Na+/H+ exchanger in primary cultures of astrocytes from mouse cerebral cortex, Brain Res. 1044 (2005) 33–41. [16] E. Yodoya, M. Wada, A. Shimada, H. Katsukawa, N. Okada, A. Yamamoto, V. Ganapathy, T. Fujita, Functional and molecular identification of sodium-coupled dicarboxylate transporters in rat primary cultured cerebrocortical astrocytes and neurons, J. Neurochem. (2006) (in press). [17] M. Oka, Y. Itoh, T. Shimidzu, Y. Ukai, Y. Yoshikuni, K. Kimura, Involvement of metabotropic glutamate receptors in Gi- and Gsdependent modulation of adenylate cyclase activity induced by a novel cognition enhancer NS-105 in rat brain, Brain Res. 754 (1997) 121–130. [18] T. Shimidzu, Y. Itoh, M. Oka, T. Ishima, Y. Ukai, Y. Yoshikuni, K. Kimura, Effect of a novel cognition enhancer NS-105 on learned helplessness in rats: Possible involvement of GABAB receptor upregulation after repeated treatment, Eur. J. Pharmacol. 338 (1997) 225–232. [19] M. Oka, Y. Itoh, T. Fujita, Halothane attenuates the cerebroprotective action of several Na+ and Ca2+ channel blockers via reversal of their ion channel blockade, Eur. J. Pharmacol. 452 (2002) 175–181. [20] M. Oka, Y. Itoh, Y. Ukai, Involvement of Na+ and Ca2+ channel activation and resultant nitric oxide synthesis in glutamate-mediated hypoxic injury in rat cerebrocortical slices, Life Sci. 67 (2000) 2331– 2343. [21] M. Oka, Y. Itoh, Y. Ukai, Preferential inhibition by a novel Na+/ Ca2+ channel blocker NS-7 of severe to mild hypoxic injury in rat cerebrocortical slices: a possible involvement of a highly voltagedependent blockade of Ca2+ channel, J. Pharmacol. Exp. Ther. 293 (2000) 522–529. [22] M. Oka, Y. Itoh, M. Wada, A. Yamamoto, T. Fujita, A comparison of Ca2+ channel blocking mode between gabapentin and verapamil:
[23]
[24]
[25]
[26] [27]
[28]
[29]
[30]
[31]
[32]
[33]
[34] [35] [36] [37]
881
implication for protection against hypoxic injury in rat cerebrocortical slices, Br. J. Pharmacol. 139 (2003) 435–443. M. Oka, M. Wada, A. Yamamoto, Y. Itoh, T. Fujita, Functional expression of constitutive nitric oxide synthases regulated by voltagegated Na+ and Ca2+ channels in cultured human astrocytes, Glia 46 (2004) 53–62. M.J. Bek, S. Zheng, J. Xu, I. Yamaguchi, L.D. Asico, X.-G. Sun, P.A. Jose, Differential expression of adenylyl cyclases in the rat nephron, Kidney Int. 60 (2001) 890–899. T. Abaffy, K.R. Trubey, N. Chaudhari, Adenylyl cyclase expression and modulation of cAMP in rat taste cells, Am. J. Physiol. Cell Physiol. 284 (2003) C1420–C1428. R.K. Sunahara, R. Taussig, Isoforms of mammalian adenylyl cyclase: multiplicities of signaling, Mol. Interv. 2 (2002) 168–184. W.J. Wojcik, I. Holopainen, Role of central GABAB receptors in physiology and pathology, Neuropsychopharmacology 6 (1992) 201– 214. V. Magnaghi, M. Ballabio, I.T.R. Cavarretta, W. Froestl, J.J. Lambert, I. Zucchi, R.C. Melcang, GABAB receptors in Schwann cells influence proliferation and myelin protein expression, Eur. J. Neurosci. 19 (2004) 2641–2649. J.T. Porter, K.D. McCarthy, Hippocampal astrocytes in situ respond to glutamate released from synaptic terminals, J. Neurosci. 16 (1996) 5073–5081. A. Araque, V. Parpura, R.P. Sanzgiri, P.G. Haydon, Tripartite synapses: glia, the unacknowledged partner, Trends Neurosci. 22 (1999) 208–215. A. Green, S. Walls, A. Wise, R.H. Green, A.K. Martin, F.H. Marshall, Characterization of [3H]-CGP54626A binding to heterodimeric GABAB receptors stably expressed in mammalian cells, Br. J. Pharmacol. 131 (2000) 1766–1774. A. Schousboe, A. Frandsen, P. Krogsgaard-Larsen, Pharmacological and functional characterization of excitatory amino acid mediated cytotoxicity in cerebral cortical neurons, Cell. Biol. Toxicol. 8 (1992) 93–100. E.L. Gleason, N.C. Spitzer, AMPA and NMDA receptors expressed by differentiating Xenopus spinal neurons, J. Neurophysiol. 79 (1998) 2986–2998. S. Orrenius, D.J. McConkey, G. Bellomo, P. Nicotera, Role of Ca2+ in toxic cell killing, Trends Pharmacol. Sci. 10 (1989) 281–285. D.W. Choi, Excitotoxic cell death, J. Neurobiol. 23 (1992) 1261–1276. Y. Chen, R.A. Swanson, Astrocytes and brain injury, J. Cereb. Blood Flow Metab. 23 (2003) 137–149. R. Fern, S.G. Waxman, B.R. Ransom, Endogenous GABA attenuates CNS white matter dysfunction following anoxia, J. Neurosci. 15 (1995) 699–708.