Comparison of neuroprotective efficacy of competitive and non-competitive AMPA antagonists in vitro

ELSEVIER

Environmental

Toxicology and Pharmacology 3 (1997) 69-72

Comparison of neuroprotective efficacy of competitive and non-competitive AMPA antagonists in vitro Attila D. KOV&ZS*, Gkza Szab6 Depariment of’Biochemistry,

EGIS Biological Laboratories, EGIS Pharmaceuticals Ltd., P.O. Box 100, H-1475 Budapest 10, Hungary

Received 3 June 1996; revised 13 September

1996; accepted

30 October

1996

Abstract

The neuroprotective efficacy of the most potent competitive a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) and three recently developed 2,3-benzodiazepine non-competitive AMPA antagonists (GYKI 52466, 53405 and 53655) was investigated in primary cultures of rat telencephalic neurons. NBQX protected cultured neurons against AMPA (20 PM for 21-23 h) induced toxicity with EC,, of 0.5 PM. In the same test GYKI 52466, 53405 and 53655 had EC,, values of 10.6, 9.3 and 5.1 PM, respectively. Thus we found the competitive antagonist NBQX to be ten times more effective neuroprotectant in vitro than the most potent non-competitive GYKI compound (GYKI 53655). 0 1997 Elsevier Science B.V. Keywords:

AMPA antagonists;

AMPA receptor; Excitotoxicity;

1. Introduction Glutamate and aspartate are the major excitatory neurotransmitters in the mammalian brain. Excessive or persistent activation of glutamate-gated ion channels may cause neuronal degeneration and this process is involved in the neuropathology of several age-related neurodegenerative disorders (e.g., Parkinson’s disease and Huntington’s disease), as well as in stroke and seizures (Bondy and Lebel, 1993; Coyle and Puttfarcken, 1993; Danysz et al., 1995; Dawson et al., 1995). Ionotropic glutamate receptors are categorised pharmacologically and genetically into three subclasses: Nmethyl-D-aspartate (NMDA) receptors, a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors and kainate receptors (Hollmann and Heinemann, 1994). Recently, interest has turned to

* Corresponding 2577932.

author.

Tel.:

1382-6689/97/$17.00 0 1997 Elsevier PI1 S1382-6689(96)00134-2

+ 36

1 2566387;

Science

fax:

B.V. All rights

+ 36

1

reserved.

Neuroprotection

the research of AMPA receptor antagonists as potential therapeutic agents for the treatment of the above-mentioned diseases and brain insults. The 2,3-dihydroxy-6nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) is the most potent competitive AMPA antagonist and it has been shown to be neuroprotective in several animal models of global and focal ischaemia (Gill, 1994). The 2,3-benzodiazepine GYKI 52466 [ I-(4-aminophenyl)-4methyl-7,8-methylenedioxy-5H-2,3_benzodiazepine hydrochloride] is a selective, non-competitive AMPA antagonist possessing neuroprotective effect in vitro and in vivo as well (Wilding and Huettner, 1995; Vizi et al., 1996). In the course of our research on new 2,3-benzodiazepines, we compared NBQX versus GYKI 52466 and its most effective analogues GYKI 53405 [1-(4aminophenyl)-3-acetyI-4-methyl-3,4-dihydro-7,8-methylenedioxy-5H-2,3_benzodiazepine] and GYKI 53655 [1-(4-aminophenyl)-3-methylcarbamoyl-4-methyl-3,4dihydro-7,8-methylenedioxy-5H-2,3-benzodiazepine] for neuroprotective efficacy in primary cultures of rat telencephalic neurons.

A.D. Kot&s,

70

2. Materials

G. Szabd 1 Environmrntal Toxicology and Pharmacology 3 (1997) 69-72

and methods

2.1. Cell culture Primary culture of rat embryonic telencephalic cells was prepared from Sprague-Dawley rat embryos from the 17th-19th day of gestation as described (Kov&cs and Egyed, 1996), with minor modifications. Briefly, telencephalons were dissected and after removal of meninges, mechanically dissociated in culture medium by trituration with 1 ml automatic pipette above a nylon mesh with pore diameter of 42 pm. 7-8 x lo5 cells filtered through the mesh were plated onto 24-well plates coated with poly-L-lysine. For dissociation, plating and maintenance Eagle’s Minimal Essential Medium (EMEM) supplemented with 10% heat-inactivated fetal calf serum (FCS), 4 mM glutamine, 20 mM KCI, gentamicin sulphate (50 mg/l) and amphotericin B (2.5 mg/l) was used. Cultures were incubated at 37°C in a humidified atmosphere containing 5% CO,. Medium changes were carried out 24 h and 3-4 days after plating. On the 5th or 6th day in vitro (DIV) the proliferation of non-neuronal cells was arrested by addition of 10 PM cytosine arabinofuranoside (CAR) for 24 h after that the cells were shifted into EMEM, containing 10% heat-inactivated horse serum (HS), 4 mM glutamine, 12 mM glucose, gentamicin sulphate and amphotericin B (EMEM-10% HS). On the 1 lth day in vitro, to prevent the formation of a second layer of glia, cultures were treated with 10 PM CAR for 24 h. Following this CAR exposure medium change was carried out with EMEM-10% HS containing penicillin (lo5 U/l) as antibiotic instead of gentamicin and amphotericin B. No further medium change was made prior to the AMPA toxicity experiments. 2.2. Determination

of neuroprotection

Experiments were performed on 18-20-day-old cultures. Cultures were exposed for 21-23 h to AMPA (20 PM; Tocris Neuramin) and to various concentrations of its antagonists. (GYKI 52466xHCl was synthesised in the Institute for Drug Research, Budapest; NBQX, GYKI 53405 and GYKI 53655 were synthesised in EGIS Pharmaceuticals Ltd., Budapest). GYKI 52466xHCl was dissolved in distilled water, NBQX, GYKI 53405 and 53655 were dissolved in dimethyl sulfoxide (DMSO). At its final concentration (O.l%), DMSO failed to evoke detectable response in our cultures. The dilutions of antagonists and the treatments were made in treating medium (EMEM containing 5% HS, 4 mM glutamine, 12 mM glucose and lo5 U/l penicillin). Cultures were preincubated with NBQX, GYKI 52466, 53405 and 53655 for 30 min at 37°C before addition of AMPA (10 ~1 from a stock solution of 1 mM to a culture well), the antagonists were present during the whole AMPA exposure period.

The neuronal cell death was visually assessed by phase contrast microscopy and quantified by measuring the activity of LDH released into the culture medium. After 21-23 h of treatment with AMPA and the test compounds the culture supernatants (supernatants A) were removed and replaced with treating medium lacking HS and supplemented with 500 PM glutamate to destroy the surviving neurons. After 18-20 h of incubation the culture supernatants (supernatants B) were removed. LDH activities of supernatants A and B (LDH, and LDH,) were determined. Neurotoxicity was expressed as LDH activity in the culture supernatant after AMPA treatment as a percent of the total neuron specific LDH activity in the same culture well: [LDH,/(LDH, + LDH,)] x 100. One hundred per cent neuroprotection was defined as the percent LDH release in the absence of AMPA (spontaneous cell death), whereas the percent LDH release obtained in the presence of 20 ,uM AMPA alone was taken as zero protection. 2.3. Measurement

of LDH

actiuitJ

LDH activities were determined at 37”C, in 96-we11 microtiter plates in duplicate from 60 ~1 aliquots of culture supernatants. The decrease of NADH absorbance at 340 nm was measured by a kinetic ELISA reader. The LDH activity of 60 ~1 treating medium was subtracted as background (LDH activity of 5% HS) in the case of culture supernatants A.

3. Results and discussion Fig. l(a) shows the microscopic appearance of mature telencephalic cultures (20 DIV): the glia monolayer is not visible because of the intensive neurite network of neurons. Treatment with 20 PM AMPA for 21 h results in almost total neuronal cell death (Fig. l(b)). NBQX (Fig. l(c)) or GYKI 52466, GYKI 53405 and GYKI 53655 (Fig. l(d)) provided complete neuroprotection against this excitotoxic insult. We measured the concentration dependence of neuroprotection for NBQX and the three GYKI compounds in our cell culture system. The dose-response curves are shown in Fig. 2. The EC,, values were determined from these curves by GraphPad Prism program: 0.5 PM for NBQX and 10.6, 9.3 and 5.1 PM for GYKI 52466, 53405 and 53655, respectively. NBQX proved to be ten times more potent a neuroprotectant in our telencephalic cultures than the most effective 2,3_benzodiazepine GYKI 53655. It is interesting to confront the neuroprotective EC,, values determined by us and the published IC,, values for AMPA channel block, because the relation between the extent of ion channel block and neuroprotection is

A.D. Kovdrs, G. Srabd / Encironmencal Toxicology and Pharmacology 3 (1997) 69-72

unknown. In a comparative patch clamp study Parsons et al. (1994) investigated the effects of NBQX and GYKI 52466 on the kinetics of AMPA (50 PM) currents on cultured rat superior colliculus and hippocampal cells. They found that NBQX and GYKI 52466 antagonised the plateau components of AMPA responses with IC,, values of 0.7 PM and 4.4 FM, respectively. In another study Donevan et al. (1994), using whole cell voltage-clamp recording on cultured hippocampal cells, examined the inhibition of AMPA (100 ,uM) and kainate (100 PM) currents by GYKI 52466, 53405 and 53655. The ICSO values in the case of AMPA block were 8.8 ,uM for GYKI 52466, 3.8 ,uM for GYKI 53405 and 1 .l PM for GYKI 53655. The rank order of potency of the three compounds in neuroprotective efficacy in our telencephalic cultures is the same but the differences between them are much smaller. It should be noted, however, that in our telencephalic cultures the cortical neurons predominate. The subunit composition and functional properties of native

I

71

T

-

GYKI

52466

__A__

GYKI

53405

GYKI

53655

-e-d-

o.io

1

.bo

-NBQX

10‘00 rod.00100b.00

CONCENTRATION

(MM)

Fig. 2. Concentration-dependent neuroprotection by NBQX, GYKI 52466, 53405 and 53655 against AMPA excitotoxicity. Cultures of 18-20 days in vitro were treated for 21-23 h with 20 pM AMPA and various concentrations of its antagonists. The antagonists were added to the cultures 30 min prior to the AMPA exposure and were present during the exposure. 100% neuroprotection was defined as the percent LDH release in the absence of AMPA (spontaneous cell death), whereas the percent LDH release obtained in the presence of 20 PM AMPA alone was taken as zero protection. Points are mean + experiments, each in duplicate. S.E.M. of two or three independent The nonlinear curve fittings were performed by GraphPad Prism program.

AMPA receptors in hippocampal and in cortical neurons have been shown to differ markedly (Catania et al., 1995; Geiger et al., 1995). Thus AMPA receptor blocking activities determined on hippocampal cells by electrophysiology may not be compared directly to neuroprotective values measured in cultures others than hippocampal. Our results demonstrate in vitro neuroprotection by GYKI 53405 and 53655, and these two compounds were compared to GYKI 52466 and NBQX in this context. The competitive AMPA antagonist NBQX proved to be much more potent a neuroprotectant in our telencephalic cultures than the three non-competitive AMPA antagonist GYKI compounds, However, for therapeutic applications the development of noncompetitive antagonists is desirable, because they have effective blocking activity even in the presence of large excess of the agonist. The latter situation exists in the brain during hypoxic/ischaemic conditions when a great amount of glutamate is released into the extracellular space causing excitotoxic neuronal cell death (Coyle and Puttfarcken, 1993).

Acknowledgements The authors thank Katalin Hadarics for the excellent technical assistance and Drs And& Egyed and Karoly Tihanyi for the helpful discussions. Fig 1. Nemoprotection against AMPA excitotoxicity by NBQX and GYKI 53655. Phase contrast photomicrographs of rat telencephalic cultures (20 days in vitro) after 21 h of treatments. (A) Control. (B) 20 /tM AMPA. (C) 20 PM AMPA+ 10 PM NBQX. (D) 20 ,IM AMPA + 30 /cM GYKI 53655. NBQX or GYKI 53655 were added to the cultures 30 min prior to the AMPA exposure and were present during the exposure. Scale bar, 100 /urn.

References Bondy, S.C. and C.P. Lebel, 1993, The relationship between excitotoxicity and oxidative stress in the central nervous system, Free Radic. Biol. Med. 14, 633.

72

A.D. KovLics, G. Szabd 1 Environmental Toxicology and Pharmacology 3 (1997) 69-72

Catania, M.V., T.R. Tiille and H. Monyer, 1995, Differential expression of AMPA receptor subunits in NOS-positive neurons of cortex, striatum, and hippocampus, J. Neurosci. 15, 7046. Coyle, J.T. and P. Puttfarcken, 1993, Oxidative stress, glutamate, and neurodegenerative disorders, Science 262, 689. Danysz, W., C.G. Parsons, I. Bresink and G. Quack, 1995, Glutamate in CNS disorders, Drug News Perspect. 8, 261. Dawson, R., M.F. Beal, S.C. Bondy, D.A. DiMonte and G.E. Isom, 1995, Excitotoxins, aging, and environmental neurotoxins: implications for understanding human neurodegenerative diseases, Toxicol. Appl. Pharmacol. 134, 1. Donevan, S.D., S.-I. Yamaguchi and M.A. Rogawski, 1994, Non-Nmethyl-n-asparatate receptor antagonism by 3-N-substituted 2,3benzodiazepines: relationship to anticonvulsant activity, J. Pharmacol. Exp. Ther. 271, 25. Geiger, J.R.P., T. Melcher, D.-S. Koh, B. Sakmann, P.H. Seeburg, P. Jonas and H. Monyer, 1995, Relative abundance of sub-unit mRNAs determines gating and Ca’+ permeability of AMPA receptors in principal neurons and interneurons in rat CNS, Neuron 15, 193.

Gill, R., 1994, The pharmacology of I-amino-3-hydroxy-5-methyl-4isoxazole propionate (AMPA)/kainate antagonists and their role in cerebral ischemia, Cerebrovasc. Brain Metab. Rev. 6, 225. Hollmann, M. and S Heinemann, 1994, Cloned glutamate receptors. Annu. Rev. Neurosci. 17, 31. Kovacs, A.D. and A. Egyed, 1996, Protection against non-NMDA receptor-mediated excitotoxicity by GYKI 52466 in mature telencephalic cultures of the rat, Neurobiology, 4, 59. Parsons, C.G., R. Gruner and J. Rozental, 1994, Comparative patch clamp studies on the kinetics and selectivity of glutamate receptor 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F) antagonism by 1-(4-amino-phenyl)-4-methyl-7,8quinoxaline (NBQX) and (GYKI 52466) Neumethylendioxyl-5H-2,3_benzodiazepine ropharmacology 33, 589. Vizi, ES., A. Mike and I. Tarnawa, 1996, 2,3_benzodiazepines (GYKl 52466 and analogs): negative allosteric modulators of AMPA receptors, CNS Drug Rev. 2, 91. Wilding, T.J. and J.E. Huettner, 1995, Differential antagonism of cc-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid-preferring and kainate-preferring receptors by 2,3_benzodiazepines, Mol. Pharmacol. 47, 582.