- Email: [email protected]
The GABAA receptor agonist THIP is neuroprotective in organotypic hippocampal slice cultures
Brain Research 973 (2003) 303–306 www.elsevier.com / locate / brainres
Short communication
The GABAA receptor agonist THIP is neuroprotective in organotypic hippocampal slice cultures Bjarne W. Kristensen a,b , *, Jens Noraberg a , Jens Zimmer a,b a
b
Anatomy and Neurobiology, Institute of Medical Biology, University of Southern Denmark, Winsløwparken 21, DK-5000 Odense C, Denmark NeuroScreen ApS, c /o Anatomy and Neurobiology, Institute of Medical Biology, University of Southern Denmark, DK-5000 Odense C, Denmark Accepted 28 February 2003
Abstract The potential neuroprotective effects of the GABAA receptor agonists THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol) and muscimol, and the selective GluR5 kainate receptor agonist ATPA ((RS)-2-amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid), which activates GABAergic interneurons, were examined in hippocampal slice cultures exposed to N-methyl-D-aspartate (NMDA). The NMDA-induced excitotoxicity was quantified by densitometric measurements of propidium iodide (PI) uptake. THIP (100–1000 mM) was neuroprotective in slice cultures co-exposed to NMDA (10 mM) for 48 h, while muscimol (100–1000 mM) and ATPA (1–3 mM) were without effect. The results demonstrate that direct GABAA agonism can mediate neuroprotection in the hippocampus in vitro as previously suggested in vivo. 2003 Elsevier Science B.V. All rights reserved. Theme: Neurotransmitters, modulators, transporters and receptors Topic: Excitatory amino acids: excitotoxicity Keywords: Excitotoxicity; Experimental ischemia; Explant culture; g-Aminobutyric acid; Neurodegeneration; Propidium iodide; GluR5
Activation of glutamate receptors evidently plays a key role in the induction of neurodegeneration in ischemia, hypoxia, trauma and epilepsia [5]. Since the inhibitory transmitter g-aminobutyric acid (GABA) counteracts glutamate receptor-mediated depolarization, enhanced GABAergic neurotransmission has been proposed as a neuroprotective strategy, in parallel with application of glutamate receptor antagonists, Ca 21 -channel blockers and NO-pathway inhibitors [5,9,17,18]. THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol) appears to act as a partial agonist at some GABAA receptor subunit combinations, and as an agonist with higher efficacy than GABA at other GABAA receptor subunit combinations [6,8]. Given that full GABAA receptor agonists may be associated with severe side effects by activating all GABAA receptors in the central nervous system, interest has focused on partial GABAA receptor agonists as potential therapeutics [7]. As both THIP and *Corresponding author. Tel.: 145-6550-3800; fax: 145-6590-6321. E-mail address: [email protected] (B.W. Kristensen).
the non-specific GABAA receptor agonist muscimol moreover can penetrate the blood–brain barrier [16], we therefore decided to investigate the neuroprotective actions of THIP and muscimol against N-methyl-D-aspartate (NMDA)-induced CA1 pyramidal cell death in hippocampal slice cultures. Interneurons in stratum oriens of CA1 in hippocampal slices have recently been shown to be activated by (RS)-2amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid (ATPA) through activation of postsynaptic GluR5 containing kainic acid (KA) receptors, resulting in increased tonic GABAergic inhibition of CA1 pyramidal cells [4]. ATPA has also recently been shown to have an anti-epileptic effect in hippocampal slices [11]. This inspired us to specifically investigate whether the proposed ATPA-induced activation of CA1 inhibitory interneurons could protect CA1 pyramidal cells against NMDA-induced cell death in the hippocampal slice cultures. A preliminary report of the present study has appeared in abstract form [12]. Details of the methodology have been described else-
0006-8993 / 03 / $ – see front matter 2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02550-2
304
B.W. Kristensen et al. / Brain Research 973 (2003) 303–306
where [13–15,19]. In brief, 5–7-day-old rats of Wistar strain (Møllegaard, Denmark) were killed by instant decapitation, the two hippocampi isolated and their dorsal halves sectioned transversely at 300–350 mm by a McIlwain tissue chopper. The resulting tissue slices were placed on insert membranes (Millipore Corp., Bedford, MA, USA) and the inserts transferred to culture trays (Corning Costar, Corning, NY, USA), containing culture medium composed of 50% Opti-MEM, 25% horse serum, 25% Hank’s BSS (HBSS) (all from Gibco BRL), supplemented by D-glucose to a final concentration of 25 mM. The trays were kept in an incubator with 5% CO 2 at 36 8C. After 3–4 days of incubation the culture medium was replaced with chemically defined, serum-free Neurobasal medium (Gibco BRL) with 25 mM D-glucose, 1 mM L-glutamine and 0.6 mM of MgSO 4 (Sigma, Vallensbæk Strand, Denmark), and 2% B27 supplement (Gibco BRL). The medium was thereafter changed twice per week. Propidium iodide (PI, Sigma) uptake has been used to monitor neuronal death in organotypic brain slice cultures exposed to excitotoxins [1,14,15,19,26] and oxygen-glucose deprivation [2,21]. PI is a very stable fluorescent dye and is basically nontoxic to neurons [22]. As a polar compound it only enters dead or dying cells with a damaged cell membrane. Once inside the cell PI interacts with DNA to yield a brightly red fluorescence (630 nm), when excited by blue-green light (493 nm). After 3–4 weeks of culturing and at all subsequent medium changes, PI was added to the culture medium to achieve a final concentration of 2 mM. In order to monitor the basic levels of PI uptake, revealing the level of spontaneous cell death in the cultures, PI was added at least 3 h before the NMDA exposures and the PI uptake recorded by fluorescence microscopy (Olympus IMT-2, 4X (Splan FL2)) and a digital camera (Sensys KAF 1400 G2, Photometrics, Tucson, AZ, USA). After recording of the basic PI uptake, the cultures were exposed to drugs as described below and the PI uptake recorded again after 48 h. For densitometric measurements, the digital photos were analyzed in NIH Image 1.62 image analysis program (National Institutes of Health, USA) after outlining the CA1 pyramidal cell layer. Cultures were exposed to 100–1000 mM THIP or 100– 1000 mM muscimol (Tocris, cat. no. TOC0289) together with 10 mM NMDA (Sigma, cat. no. M-3262) for 48 h. Other cultures were exposed to low GluR5 selective concentrations of ATPA (1 or 3 mM) [3], or to 0.3 mM KA as control for kainate receptor activation (Sigma, cat. no. K-025), together with 10 mM NMDA. The concentrations of ATPA and KA used are from a recent study known to be non-toxic [15]. All experiments included control cultures not subjected to drug treatment. To monitor a possible toxic effect of 10 mM NMDA on GABAergic hippocampal interneurons, glutamic acid decarboxylase (GAD) activity was assessed in tissue samples pooled from three hippocampal cultures taken from the same culture well (n51 corresponds to three cultures). The
tissue was sonicated and the GAD activity was assessed on the same day according to a previously published protocol [14]. GAD activity was expressed as nmol glutamate / mg protein / h at 37 8C. All data were expressed as means6standard error of mean (S.E.M.). Statistical significance was assessed in GraphPad Instat (GraphPad Software, San Diego, CA,
Fig. 1. Protection against 10 mM NMDA-induced excitotoxicity and PI uptake in CA1 pyramidal cells (A) by the GABAA receptor agonist THIP in 100–1000 mM concentrations (B,C). Data in C are shown as means1 S.E.M. with n536–87, **P,0.01, ***P,0.001, ANOVA with Bonferroni correction for comparison with cultures treated with NMDA. Cultures treated with NMDA and 100 or 1000 mM THIP were also compared without finding statistical significance. AU, arbitrary units; FD, fascia dentata.
B.W. Kristensen et al. / Brain Research 973 (2003) 303–306
USA) using single-factor analysis of variance (ANOVA) with Bonferroni correction for comparison of the groups of interest. Differences were considered significant at P, 0.05. A sigmoid shaped curve was fitted to the NMDA– GAD concentration–response data set using Microcal Origin 3.5 (Microcal Software, Northampton, MA, USA). From this curve the EC 50 value corresponding to 50% of the plateau level (maximum GAD activity) in the concentration–response curve was obtained. In co-exposure with 10 mM NMDA, 100–1000 mM THIP protected CA1 pyramidal cells (Fig. 1) (n560–87) reducing NMDA-induced PI uptake. A lower concentration of 30 mM THIP was not neuroprotective (n536), just as 100–1000 mM muscimol failed to be protective (n511– 24) (data not shown). Other cultures exposed to 10 mM NMDA were cotreated with 1 or 3 mM ATPA (n59–23) or 0.3 mM KA (n511), but without neuroprotective effect (Fig. 2A). The GAD activity measurements (based on 6–8 samples of sets of three cultures per sample) gave an EC 50 value of 19 mM NMDA corresponding to 50% of the plateau level (maximum GAD activity) in the concentration–response curve. For 10 mM NMDA, there was a 30% reduction on GAD activity (Fig. 2B). Our results demonstrate that application of 100, 300 and 1000 mM of the GABAA receptor agonist THIP to hippocampal slice cultures reduced the neuronal cell death induced by 10 mM NMDA by 25–35%. For most subunit combinations of GABAA receptors expressed in oocytes the electrophysiological EC 50 values of THIP were within the 100–300 mM range [6]. Application of concentrations higher than 100 mM provided a tendency towards decreased protection, probably due to rapid desensitization of the GABA receptors by higher concentrations. In vivo, intraperitoneal injections of THIP offered no significant neuroprotection of CA1 in the rat hippocampus after global
305
cerebral ischemia when given alone, but when administered together with diazepam, it significantly reduced the CA1 pyramidal cell loss [10]. Intraventricular infusion of muscimol has been found to attenuate neurodegeneration after transient forebrain ischemia in the gerbil hippocampus and neocortex in vivo [24]. In the study of Yoshikawa et al. [27], muscimol, however, did not protect against NMDA in cortical slice cultures, corresponding to that muscimol did not protect against NMDA in hippocampal slice cultures in the present study. However, in dispersed cortical cell cultures muscimol reduced NMDA-induced release of lactate dehydrogenase [20]. The reason for this difference in observed effects of muscimol may be due to differences in GABAA receptor subunit combinations among brain regions and among strains, as well as between different developmental stages of the neurons. According to the ‘overinhibition’ hypothesis proposed by Cossart et al. [4], glutamate activation of postsynaptic KA receptors containing the GluR5 subunit should stimulate CA1 interneurons, leading to an increase of GABAergic inhibition on pyramidal cells. One or 3 mM ATPA or 0.3 mM KA did not, however, prevent or ameliorate the excitotoxic effects of 10 mM NMDA. This might be due to a limited, too short lasting increased release of GABA on pyramidal cells. Another reason might be that the reduction in CA1 pyramidal cell excitability was limited due to concomitant weak activation of AMPA receptors by 1 or 3 mM ATPA [25]. The lack of neuroprotective effect by ATPA might also be due to partial, NMDA-induced damage to the GABAergic interneurons, as suggested by the observed reduction in GAD activity. The GABAergic interneurons in the hippocampus are, however, in general thought to be much more resistant to ischemia than CA1 pyramidal cells [23], possibly not preventing ATPA from protecting against ischemia in vivo.
Fig. 2. (A) Lack of protective effect of 1 or 3 mM ATPA and of 0.3 mM KA on CA1 pyramidal cells in slice cultures exposed to 10 mM NMDA. (B) By acting on GluR5 containing KA receptors, ATPA and KA were supposed to activate inhibitory GABAergic CA1 interneurons in stratum oriens. The lack of protection might be because NMDA also damaged the GABAergic interneurons, supported by the finding that 10 mM NMDA caused a decrease in GAD activity (nmol glutamate (glu) / mg protein per hour). An EC 50 value of 19 mM corresponding to 50% of the plateau level (maximum GAD activity) in the concentration–response curve was obtained. Data are shown as means6S.E.M. with n59–24 for the PI uptake and n56–8 (n51 corresponds to three cultures) for the GAD activity measurements, **P,0.01, ***P,0.001, ANOVA with Bonferroni correction for comparison with control. AU, arbitrary units.
306
B.W. Kristensen et al. / Brain Research 973 (2003) 303–306
This study demonstrates that a GABAA receptor agonist (THIP) can protect against excitotoxicity (NMDA) in organotypic hippocampal slice cultures. The results thereby suggest that slice cultures can by used for testing the hypothesis that enhancement of GABAergic activation is neuroprotective. Unfortunately, supposed (ATPA) stimulation of CA1 interneurons leading to an increase of GABAergic inhibition on pyramidal cells did not mediate neuroprotection, thereby questioning this indirect strategy of increased GABAergic stimulation.
Acknowledgements The technical help of Inge Holst, Randi Godskesen and Dorte Bramsen is gratefully acknowledged. Thanks are also due to Professor Povl Krogsgaard-Larsen for providing the ATPA and THIP compounds, and to Professor Bjarke Ebert for fruitful discussions and help with the manuscript. Associate Professor Jan Bert Gramsbergen is thanked for help with the GAD measurements. The study was supported by Director Jacob Madsens and wife Olga Madsens Foundation, a Danish MRC grant (9902660) to J.N., Neuroscience Pharmabiotec Research Center and the FP5-EU grant (QLK3-CT-2001-00407) to J.Z. (Anatomy and Neurobiology) and to J.N. (NeuroScreen ApS).
References [1] C. Bonde, B.W. Kristensen, M. Blaabjerg, T.E. Johansen, J. Zimmer, M. Meyer, GDNF and neublastin protect against NMDA-induced excitotoxicity in hippocampal slice cultures, Neuroreport 11 (2000) 4069–4073. [2] C. Bonde, J. Noraberg, J. Zimmer, Nuclear shrinkage and other markers of neuronal cell death after oxygen-glucose deprivation in rat hippocampal slice cultures, Neurosci. Lett. 327 (2002) 49–52. [3] V.R. Clarke, B.A. Ballyk, K.H. Hoo, A. Mandelzys, A. Pellizzari, C.P. Bath, J. Thomas, E.F. Sharpe, C.H. Davies, P.L. Ornstein, D.D. Schoepp, R.K. Kamboj, G.L. Collingridge, D. Lodge, D. Bleakman, A hippocampal GluR5 kainate receptor regulating inhibitory synaptic transmission, Nature 389 (1997) 599–603. [4] R. Cossart, M. Esclapez, J.C. Hirsch, C. Bernard, Y. Ben-Ari, GluR5 kainate receptor activation in interneurons increases tonic inhibition of pyramidal cells, Nat. Neurosci. 1 (1999) 579–586. [5] J. De Keyser, G. Sulter, P.G. Luiten, Clinical trials with neuroprotective drugs in acute ischaemic stroke: are we doing the right thing?, Trends Neurosci. 22 (1999) 535–540. [6] B. Ebert, K.A. Wafford, P.J. Whiting, P. Krogsgaard-Larsen, J.A. Kemp, Molecular pharmacology of g-aminobutyric acid type A receptor agonists and partial agonists in oocytes injected with different a, b, and g receptor subunit combinations, Mol. Pharmacol. 46 (1994) 957–963. [7] B. Frølund, B. Ebert, U. Kristiansen, T. Liljefors, P. KrogsgaardLarsen, GABA(A) Receptor ligands and their therapeutic potentials, Curr. Top. Med. Chem. 2 (2002) 817–832. [8] B. Frølund, U. Kristiansen, L. Brehm, A.B. Hansen, P. KrogsgaardLarsen, E. Falch, Partial GABAA receptor agonists. Synthesis and in vitro pharmacology of a series of nonannulated analogs of 4,5,6,7tetrahydroisoxazolo[5,4-c]pyridin-3-ol, J. Med. Chem. 38 (1995) 3287–3296.
[9] A.R. Green, Clomethiazole (Zendra) in acute ischemic stroke: basic pharmacology and biochemistry and clinical efficacy, Pharmacol. Ther. 80 (1998) 123–147. [10] F.F. Johansen, N.H. Diemer, Enhancement of GABA neurotransmission after cerebral ischemia in the rat reduces loss of hippocampal CA1 pyramidal cells, Acta Neurol. Scand. 84 (1991) 1–6. [11] I. Khalilov, J. Hirsch, R. Cossart, Y. Ben Ari, Paradoxical antiepileptic effects of a GluR5 agonist of kainate receptors, J. Neurophysiol. 88 (2002) 523–527. [12] B.W. Kristensen, B. Ebert, J. Zimmer, Excitotoxic AMPA-like effects of ATPA in hippocampal slice cultures, Soc. Neurosci. Abstr. (1999) 890.3. [13] B.W. Kristensen, H. Noer, J.B. Gramsbergen, J. Zimmer, J. Noraberg, Colchicine induces apoptosis in organotypic hippocampal slice cultures, Brain Res. 964 (2003) 264–278. [14] B.W. Kristensen, J. Noraberg, B. Jakobsen, J.B. Gramsbergen, B. Ebert, J. Zimmer, Excitotoxic effects of non-NMDA receptor agonists in organotypic corticostriatal slice cultures, Brain Res. 841 (1999) 143–159. [15] B.W. Kristensen, J. Noraberg, J. Zimmer, Comparison of excitotoxic profiles of ATPA, AMPA, KA and NMDA in organotypic hippocampal slice cultures, Brain Res. 917 (2001) 21–44. [16] P. Krogsgaard-Larsen, E. Falch, H. Hjeds, Heterocyclic analogues of GABA: chemistry, molecular pharmacology and therapeutic aspects, Prog. Med. Chem. 22 (1985) 67–120. [17] P.D. Lyden, GABA and neuroprotection, Int. Rev. Neurobiol. 40 (1997) 233–258. [18] B. Meldrum, Protection against ischaemic neuronal damage by drugs acting on excitatory neurotransmission, Cerebrovasc. Brain Metab. Rev. 2 (1990) 27–57. [19] J. Noraberg, B.W. Kristensen, J. Zimmer, Markers for neuronal degeneration in organotypic slice cultures, Brain Res. Protoc. 3 (1999) 278–299. [20] S. Ohkuma, S.H. Chen, M. Katsura, D.Z. Chen, K. Kuriyama, Muscimol prevents neuronal injury induced by NMDA, Jpn. J. Pharmacol. 64 (1994) 125–128. [21] D.E. Pellegrini-Giampietro, F. Peruginelli, E. Meli, A. Cozzi, S. Albani-Torregrossa, R. Pellicciari, F. Moroni, Protection with metabotropic glutamate 1 receptor antagonists in models of ischemic neuronal death: time-course and mechanisms, Neuropharmacology 38 (1999) 1607–1619. [22] L.D. Pozzo Miller, N.K. Mahanty, J.A. Connor, D.M. Landis, Spontaneous pyramidal cell death in organotypic slice cultures from rat hippocampus is prevented by glutamate receptor antagonists, Neuroscience 63 (1994) 471–487. [23] M. Schlander, S. Hoyer, M. Frotscher, Glutamate decarboxylaseimmunoreactive neurons in the aging rat hippocampus are more resistant to ischemia than CA1 pyramidal cells, Neurosci. Lett. 91 (1988) 241–246. [24] A. Shuaib, R. Mazagri, S. Ijaz, GABA agonist ‘muscimol’ is neuroprotective in repetitive transient forebrain ischemia in gerbils, Exp. Neurol. 123 (1993) 284–288. [25] T.B. Stensbøl, L. Borre, T.N. Johansen, J. Egebjerg, U. Madsen, B. Ebert, P. Krogsgaard-Larsen, Resolution, absolute stereochemistry and molecular pharmacology of the enantiomers of ATPA, Eur. J. Pharmacol. 380 (1999) 153–162. [26] J.J. Vornov, R.C. Tasker, J.T. Coyle, Direct observation of the agonist-specific regional vulnerability to glutamate, NMDA, and kainate neurotoxicity in organotypic hippocampal cultures, Exp. Neurol. 114 (1991) 11–22. [27] M. Yoshikawa, S. Goto, A. Okamura, T. Hamasaki, Y. Ushio, Neurotoxicity evoked by N-methyl-D-aspartate in the organotypic static slice cultures of rat cerebral cortices: effect of GABA(A) receptor activation, Acta Neuropathol. (Berl) 95 (1998) 592–596.
Short communication
The GABAA receptor agonist THIP is neuroprotective in organotypic hippocampal slice cultures Bjarne W. Kristensen a,b , *, Jens Noraberg a , Jens Zimmer a,b a
b
Anatomy and Neurobiology, Institute of Medical Biology, University of Southern Denmark, Winsløwparken 21, DK-5000 Odense C, Denmark NeuroScreen ApS, c /o Anatomy and Neurobiology, Institute of Medical Biology, University of Southern Denmark, DK-5000 Odense C, Denmark Accepted 28 February 2003
Abstract The potential neuroprotective effects of the GABAA receptor agonists THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol) and muscimol, and the selective GluR5 kainate receptor agonist ATPA ((RS)-2-amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid), which activates GABAergic interneurons, were examined in hippocampal slice cultures exposed to N-methyl-D-aspartate (NMDA). The NMDA-induced excitotoxicity was quantified by densitometric measurements of propidium iodide (PI) uptake. THIP (100–1000 mM) was neuroprotective in slice cultures co-exposed to NMDA (10 mM) for 48 h, while muscimol (100–1000 mM) and ATPA (1–3 mM) were without effect. The results demonstrate that direct GABAA agonism can mediate neuroprotection in the hippocampus in vitro as previously suggested in vivo. 2003 Elsevier Science B.V. All rights reserved. Theme: Neurotransmitters, modulators, transporters and receptors Topic: Excitatory amino acids: excitotoxicity Keywords: Excitotoxicity; Experimental ischemia; Explant culture; g-Aminobutyric acid; Neurodegeneration; Propidium iodide; GluR5
Activation of glutamate receptors evidently plays a key role in the induction of neurodegeneration in ischemia, hypoxia, trauma and epilepsia [5]. Since the inhibitory transmitter g-aminobutyric acid (GABA) counteracts glutamate receptor-mediated depolarization, enhanced GABAergic neurotransmission has been proposed as a neuroprotective strategy, in parallel with application of glutamate receptor antagonists, Ca 21 -channel blockers and NO-pathway inhibitors [5,9,17,18]. THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol) appears to act as a partial agonist at some GABAA receptor subunit combinations, and as an agonist with higher efficacy than GABA at other GABAA receptor subunit combinations [6,8]. Given that full GABAA receptor agonists may be associated with severe side effects by activating all GABAA receptors in the central nervous system, interest has focused on partial GABAA receptor agonists as potential therapeutics [7]. As both THIP and *Corresponding author. Tel.: 145-6550-3800; fax: 145-6590-6321. E-mail address: [email protected] (B.W. Kristensen).
the non-specific GABAA receptor agonist muscimol moreover can penetrate the blood–brain barrier [16], we therefore decided to investigate the neuroprotective actions of THIP and muscimol against N-methyl-D-aspartate (NMDA)-induced CA1 pyramidal cell death in hippocampal slice cultures. Interneurons in stratum oriens of CA1 in hippocampal slices have recently been shown to be activated by (RS)-2amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid (ATPA) through activation of postsynaptic GluR5 containing kainic acid (KA) receptors, resulting in increased tonic GABAergic inhibition of CA1 pyramidal cells [4]. ATPA has also recently been shown to have an anti-epileptic effect in hippocampal slices [11]. This inspired us to specifically investigate whether the proposed ATPA-induced activation of CA1 inhibitory interneurons could protect CA1 pyramidal cells against NMDA-induced cell death in the hippocampal slice cultures. A preliminary report of the present study has appeared in abstract form [12]. Details of the methodology have been described else-
0006-8993 / 03 / $ – see front matter 2003 Elsevier Science B.V. All rights reserved. doi:10.1016 / S0006-8993(03)02550-2
304
B.W. Kristensen et al. / Brain Research 973 (2003) 303–306
where [13–15,19]. In brief, 5–7-day-old rats of Wistar strain (Møllegaard, Denmark) were killed by instant decapitation, the two hippocampi isolated and their dorsal halves sectioned transversely at 300–350 mm by a McIlwain tissue chopper. The resulting tissue slices were placed on insert membranes (Millipore Corp., Bedford, MA, USA) and the inserts transferred to culture trays (Corning Costar, Corning, NY, USA), containing culture medium composed of 50% Opti-MEM, 25% horse serum, 25% Hank’s BSS (HBSS) (all from Gibco BRL), supplemented by D-glucose to a final concentration of 25 mM. The trays were kept in an incubator with 5% CO 2 at 36 8C. After 3–4 days of incubation the culture medium was replaced with chemically defined, serum-free Neurobasal medium (Gibco BRL) with 25 mM D-glucose, 1 mM L-glutamine and 0.6 mM of MgSO 4 (Sigma, Vallensbæk Strand, Denmark), and 2% B27 supplement (Gibco BRL). The medium was thereafter changed twice per week. Propidium iodide (PI, Sigma) uptake has been used to monitor neuronal death in organotypic brain slice cultures exposed to excitotoxins [1,14,15,19,26] and oxygen-glucose deprivation [2,21]. PI is a very stable fluorescent dye and is basically nontoxic to neurons [22]. As a polar compound it only enters dead or dying cells with a damaged cell membrane. Once inside the cell PI interacts with DNA to yield a brightly red fluorescence (630 nm), when excited by blue-green light (493 nm). After 3–4 weeks of culturing and at all subsequent medium changes, PI was added to the culture medium to achieve a final concentration of 2 mM. In order to monitor the basic levels of PI uptake, revealing the level of spontaneous cell death in the cultures, PI was added at least 3 h before the NMDA exposures and the PI uptake recorded by fluorescence microscopy (Olympus IMT-2, 4X (Splan FL2)) and a digital camera (Sensys KAF 1400 G2, Photometrics, Tucson, AZ, USA). After recording of the basic PI uptake, the cultures were exposed to drugs as described below and the PI uptake recorded again after 48 h. For densitometric measurements, the digital photos were analyzed in NIH Image 1.62 image analysis program (National Institutes of Health, USA) after outlining the CA1 pyramidal cell layer. Cultures were exposed to 100–1000 mM THIP or 100– 1000 mM muscimol (Tocris, cat. no. TOC0289) together with 10 mM NMDA (Sigma, cat. no. M-3262) for 48 h. Other cultures were exposed to low GluR5 selective concentrations of ATPA (1 or 3 mM) [3], or to 0.3 mM KA as control for kainate receptor activation (Sigma, cat. no. K-025), together with 10 mM NMDA. The concentrations of ATPA and KA used are from a recent study known to be non-toxic [15]. All experiments included control cultures not subjected to drug treatment. To monitor a possible toxic effect of 10 mM NMDA on GABAergic hippocampal interneurons, glutamic acid decarboxylase (GAD) activity was assessed in tissue samples pooled from three hippocampal cultures taken from the same culture well (n51 corresponds to three cultures). The
tissue was sonicated and the GAD activity was assessed on the same day according to a previously published protocol [14]. GAD activity was expressed as nmol glutamate / mg protein / h at 37 8C. All data were expressed as means6standard error of mean (S.E.M.). Statistical significance was assessed in GraphPad Instat (GraphPad Software, San Diego, CA,
Fig. 1. Protection against 10 mM NMDA-induced excitotoxicity and PI uptake in CA1 pyramidal cells (A) by the GABAA receptor agonist THIP in 100–1000 mM concentrations (B,C). Data in C are shown as means1 S.E.M. with n536–87, **P,0.01, ***P,0.001, ANOVA with Bonferroni correction for comparison with cultures treated with NMDA. Cultures treated with NMDA and 100 or 1000 mM THIP were also compared without finding statistical significance. AU, arbitrary units; FD, fascia dentata.
B.W. Kristensen et al. / Brain Research 973 (2003) 303–306
USA) using single-factor analysis of variance (ANOVA) with Bonferroni correction for comparison of the groups of interest. Differences were considered significant at P, 0.05. A sigmoid shaped curve was fitted to the NMDA– GAD concentration–response data set using Microcal Origin 3.5 (Microcal Software, Northampton, MA, USA). From this curve the EC 50 value corresponding to 50% of the plateau level (maximum GAD activity) in the concentration–response curve was obtained. In co-exposure with 10 mM NMDA, 100–1000 mM THIP protected CA1 pyramidal cells (Fig. 1) (n560–87) reducing NMDA-induced PI uptake. A lower concentration of 30 mM THIP was not neuroprotective (n536), just as 100–1000 mM muscimol failed to be protective (n511– 24) (data not shown). Other cultures exposed to 10 mM NMDA were cotreated with 1 or 3 mM ATPA (n59–23) or 0.3 mM KA (n511), but without neuroprotective effect (Fig. 2A). The GAD activity measurements (based on 6–8 samples of sets of three cultures per sample) gave an EC 50 value of 19 mM NMDA corresponding to 50% of the plateau level (maximum GAD activity) in the concentration–response curve. For 10 mM NMDA, there was a 30% reduction on GAD activity (Fig. 2B). Our results demonstrate that application of 100, 300 and 1000 mM of the GABAA receptor agonist THIP to hippocampal slice cultures reduced the neuronal cell death induced by 10 mM NMDA by 25–35%. For most subunit combinations of GABAA receptors expressed in oocytes the electrophysiological EC 50 values of THIP were within the 100–300 mM range [6]. Application of concentrations higher than 100 mM provided a tendency towards decreased protection, probably due to rapid desensitization of the GABA receptors by higher concentrations. In vivo, intraperitoneal injections of THIP offered no significant neuroprotection of CA1 in the rat hippocampus after global
305
cerebral ischemia when given alone, but when administered together with diazepam, it significantly reduced the CA1 pyramidal cell loss [10]. Intraventricular infusion of muscimol has been found to attenuate neurodegeneration after transient forebrain ischemia in the gerbil hippocampus and neocortex in vivo [24]. In the study of Yoshikawa et al. [27], muscimol, however, did not protect against NMDA in cortical slice cultures, corresponding to that muscimol did not protect against NMDA in hippocampal slice cultures in the present study. However, in dispersed cortical cell cultures muscimol reduced NMDA-induced release of lactate dehydrogenase [20]. The reason for this difference in observed effects of muscimol may be due to differences in GABAA receptor subunit combinations among brain regions and among strains, as well as between different developmental stages of the neurons. According to the ‘overinhibition’ hypothesis proposed by Cossart et al. [4], glutamate activation of postsynaptic KA receptors containing the GluR5 subunit should stimulate CA1 interneurons, leading to an increase of GABAergic inhibition on pyramidal cells. One or 3 mM ATPA or 0.3 mM KA did not, however, prevent or ameliorate the excitotoxic effects of 10 mM NMDA. This might be due to a limited, too short lasting increased release of GABA on pyramidal cells. Another reason might be that the reduction in CA1 pyramidal cell excitability was limited due to concomitant weak activation of AMPA receptors by 1 or 3 mM ATPA [25]. The lack of neuroprotective effect by ATPA might also be due to partial, NMDA-induced damage to the GABAergic interneurons, as suggested by the observed reduction in GAD activity. The GABAergic interneurons in the hippocampus are, however, in general thought to be much more resistant to ischemia than CA1 pyramidal cells [23], possibly not preventing ATPA from protecting against ischemia in vivo.
Fig. 2. (A) Lack of protective effect of 1 or 3 mM ATPA and of 0.3 mM KA on CA1 pyramidal cells in slice cultures exposed to 10 mM NMDA. (B) By acting on GluR5 containing KA receptors, ATPA and KA were supposed to activate inhibitory GABAergic CA1 interneurons in stratum oriens. The lack of protection might be because NMDA also damaged the GABAergic interneurons, supported by the finding that 10 mM NMDA caused a decrease in GAD activity (nmol glutamate (glu) / mg protein per hour). An EC 50 value of 19 mM corresponding to 50% of the plateau level (maximum GAD activity) in the concentration–response curve was obtained. Data are shown as means6S.E.M. with n59–24 for the PI uptake and n56–8 (n51 corresponds to three cultures) for the GAD activity measurements, **P,0.01, ***P,0.001, ANOVA with Bonferroni correction for comparison with control. AU, arbitrary units.
306
B.W. Kristensen et al. / Brain Research 973 (2003) 303–306
This study demonstrates that a GABAA receptor agonist (THIP) can protect against excitotoxicity (NMDA) in organotypic hippocampal slice cultures. The results thereby suggest that slice cultures can by used for testing the hypothesis that enhancement of GABAergic activation is neuroprotective. Unfortunately, supposed (ATPA) stimulation of CA1 interneurons leading to an increase of GABAergic inhibition on pyramidal cells did not mediate neuroprotection, thereby questioning this indirect strategy of increased GABAergic stimulation.
Acknowledgements The technical help of Inge Holst, Randi Godskesen and Dorte Bramsen is gratefully acknowledged. Thanks are also due to Professor Povl Krogsgaard-Larsen for providing the ATPA and THIP compounds, and to Professor Bjarke Ebert for fruitful discussions and help with the manuscript. Associate Professor Jan Bert Gramsbergen is thanked for help with the GAD measurements. The study was supported by Director Jacob Madsens and wife Olga Madsens Foundation, a Danish MRC grant (9902660) to J.N., Neuroscience Pharmabiotec Research Center and the FP5-EU grant (QLK3-CT-2001-00407) to J.Z. (Anatomy and Neurobiology) and to J.N. (NeuroScreen ApS).
References [1] C. Bonde, B.W. Kristensen, M. Blaabjerg, T.E. Johansen, J. Zimmer, M. Meyer, GDNF and neublastin protect against NMDA-induced excitotoxicity in hippocampal slice cultures, Neuroreport 11 (2000) 4069–4073. [2] C. Bonde, J. Noraberg, J. Zimmer, Nuclear shrinkage and other markers of neuronal cell death after oxygen-glucose deprivation in rat hippocampal slice cultures, Neurosci. Lett. 327 (2002) 49–52. [3] V.R. Clarke, B.A. Ballyk, K.H. Hoo, A. Mandelzys, A. Pellizzari, C.P. Bath, J. Thomas, E.F. Sharpe, C.H. Davies, P.L. Ornstein, D.D. Schoepp, R.K. Kamboj, G.L. Collingridge, D. Lodge, D. Bleakman, A hippocampal GluR5 kainate receptor regulating inhibitory synaptic transmission, Nature 389 (1997) 599–603. [4] R. Cossart, M. Esclapez, J.C. Hirsch, C. Bernard, Y. Ben-Ari, GluR5 kainate receptor activation in interneurons increases tonic inhibition of pyramidal cells, Nat. Neurosci. 1 (1999) 579–586. [5] J. De Keyser, G. Sulter, P.G. Luiten, Clinical trials with neuroprotective drugs in acute ischaemic stroke: are we doing the right thing?, Trends Neurosci. 22 (1999) 535–540. [6] B. Ebert, K.A. Wafford, P.J. Whiting, P. Krogsgaard-Larsen, J.A. Kemp, Molecular pharmacology of g-aminobutyric acid type A receptor agonists and partial agonists in oocytes injected with different a, b, and g receptor subunit combinations, Mol. Pharmacol. 46 (1994) 957–963. [7] B. Frølund, B. Ebert, U. Kristiansen, T. Liljefors, P. KrogsgaardLarsen, GABA(A) Receptor ligands and their therapeutic potentials, Curr. Top. Med. Chem. 2 (2002) 817–832. [8] B. Frølund, U. Kristiansen, L. Brehm, A.B. Hansen, P. KrogsgaardLarsen, E. Falch, Partial GABAA receptor agonists. Synthesis and in vitro pharmacology of a series of nonannulated analogs of 4,5,6,7tetrahydroisoxazolo[5,4-c]pyridin-3-ol, J. Med. Chem. 38 (1995) 3287–3296.
[9] A.R. Green, Clomethiazole (Zendra) in acute ischemic stroke: basic pharmacology and biochemistry and clinical efficacy, Pharmacol. Ther. 80 (1998) 123–147. [10] F.F. Johansen, N.H. Diemer, Enhancement of GABA neurotransmission after cerebral ischemia in the rat reduces loss of hippocampal CA1 pyramidal cells, Acta Neurol. Scand. 84 (1991) 1–6. [11] I. Khalilov, J. Hirsch, R. Cossart, Y. Ben Ari, Paradoxical antiepileptic effects of a GluR5 agonist of kainate receptors, J. Neurophysiol. 88 (2002) 523–527. [12] B.W. Kristensen, B. Ebert, J. Zimmer, Excitotoxic AMPA-like effects of ATPA in hippocampal slice cultures, Soc. Neurosci. Abstr. (1999) 890.3. [13] B.W. Kristensen, H. Noer, J.B. Gramsbergen, J. Zimmer, J. Noraberg, Colchicine induces apoptosis in organotypic hippocampal slice cultures, Brain Res. 964 (2003) 264–278. [14] B.W. Kristensen, J. Noraberg, B. Jakobsen, J.B. Gramsbergen, B. Ebert, J. Zimmer, Excitotoxic effects of non-NMDA receptor agonists in organotypic corticostriatal slice cultures, Brain Res. 841 (1999) 143–159. [15] B.W. Kristensen, J. Noraberg, J. Zimmer, Comparison of excitotoxic profiles of ATPA, AMPA, KA and NMDA in organotypic hippocampal slice cultures, Brain Res. 917 (2001) 21–44. [16] P. Krogsgaard-Larsen, E. Falch, H. Hjeds, Heterocyclic analogues of GABA: chemistry, molecular pharmacology and therapeutic aspects, Prog. Med. Chem. 22 (1985) 67–120. [17] P.D. Lyden, GABA and neuroprotection, Int. Rev. Neurobiol. 40 (1997) 233–258. [18] B. Meldrum, Protection against ischaemic neuronal damage by drugs acting on excitatory neurotransmission, Cerebrovasc. Brain Metab. Rev. 2 (1990) 27–57. [19] J. Noraberg, B.W. Kristensen, J. Zimmer, Markers for neuronal degeneration in organotypic slice cultures, Brain Res. Protoc. 3 (1999) 278–299. [20] S. Ohkuma, S.H. Chen, M. Katsura, D.Z. Chen, K. Kuriyama, Muscimol prevents neuronal injury induced by NMDA, Jpn. J. Pharmacol. 64 (1994) 125–128. [21] D.E. Pellegrini-Giampietro, F. Peruginelli, E. Meli, A. Cozzi, S. Albani-Torregrossa, R. Pellicciari, F. Moroni, Protection with metabotropic glutamate 1 receptor antagonists in models of ischemic neuronal death: time-course and mechanisms, Neuropharmacology 38 (1999) 1607–1619. [22] L.D. Pozzo Miller, N.K. Mahanty, J.A. Connor, D.M. Landis, Spontaneous pyramidal cell death in organotypic slice cultures from rat hippocampus is prevented by glutamate receptor antagonists, Neuroscience 63 (1994) 471–487. [23] M. Schlander, S. Hoyer, M. Frotscher, Glutamate decarboxylaseimmunoreactive neurons in the aging rat hippocampus are more resistant to ischemia than CA1 pyramidal cells, Neurosci. Lett. 91 (1988) 241–246. [24] A. Shuaib, R. Mazagri, S. Ijaz, GABA agonist ‘muscimol’ is neuroprotective in repetitive transient forebrain ischemia in gerbils, Exp. Neurol. 123 (1993) 284–288. [25] T.B. Stensbøl, L. Borre, T.N. Johansen, J. Egebjerg, U. Madsen, B. Ebert, P. Krogsgaard-Larsen, Resolution, absolute stereochemistry and molecular pharmacology of the enantiomers of ATPA, Eur. J. Pharmacol. 380 (1999) 153–162. [26] J.J. Vornov, R.C. Tasker, J.T. Coyle, Direct observation of the agonist-specific regional vulnerability to glutamate, NMDA, and kainate neurotoxicity in organotypic hippocampal cultures, Exp. Neurol. 114 (1991) 11–22. [27] M. Yoshikawa, S. Goto, A. Okamura, T. Hamasaki, Y. Ushio, Neurotoxicity evoked by N-methyl-D-aspartate in the organotypic static slice cultures of rat cerebral cortices: effect of GABA(A) receptor activation, Acta Neuropathol. (Berl) 95 (1998) 592–596.