Strychnine antagonizes GABA-evoked 36~1-uptake in membrane vesicles from rat and rabbit cerebral cortex

Malgorzata Puka and Jerzy W. Lazarewicz

Department of Neurochemistry, Medical Research Centre, Polish Academy of Sciences, 3 Dworkowa St., 00-784 Warsaw, Poland

Abstract. The uptake of 36 C1- into the fraction of membrane vesicles from rat and rabbit cortex and rabbit hippocampus was measured to investigate the selectivity of agonists and antagonists of inhibitory amino acid receptors coupled to chloride channels in the forebrain ~Aminobutyricacid (GABA) produced a dose-dependent 36~1-uptake into membrane vesicles, whereas glycine and taurine were ineffective. The chloride uptake induced by GABA was inhibited in a dose-dependent manner by bicuculline and picrotoxin, established antagonists of GABAA receptors. GABA elicited chloride influx was also inhibited, with similar potency, by strychnine, thus indicating a limited selectivity of this glycine receptor antagonist.

Key words: GABAA receptors, GABA antagonists, glycine, strychnine, chloride uptake, cortical membranes, rat, rabbit 512 M. Puka and J.W. Lazarewicz INTRODUCTION ing: NaCl 124 rnM, KC1 5 mM, KH2P04 1.2 mM, MgS04 1.3 mM, CaC12 2 mM, NaHC03 2.6 mM, GABA (y-aminobutyric acid) and glycine induce HEPES 10 mM pH 7.4 and glucose 10 mM. The ho- inhibitory responses in neurons of the mammalian mogenate was centrifuged at 1,000xg for 15 min. central nervous system (CNS) by activating the in- The supernatant was decanted and the pellet resus- ward C1- current through chloride channels coupled pended in 10 volumes of buffer and centrifuged to their specific receptors (Aprison and Daly 1978, again at 1,000xg for 15 min. The final pellet was re- Matsumoto 1989, Langosch et al. 1990). It has been suspended in K-H buffer (final concentration 8-9 suggested that taurine interference with these recep- mg proteinlml). tors may partially explain its neuromodulatory, in- Suspensions of the membrane vesicles were pre- hibitory effects in the CNS (Medina and De incubated at 30'~for 10 min, then 200 pl aliquots Robertis 1984, Horikoshi et al. 1988, Hausser et al. were diluted in 200 p1 of the incubation buffer con- 1992). There are marked differences in antagonist taining varying concentrations of agonists (GABA, specificity and localization of GABA and glycine glycine,) and/or antagonists (bicuculline, picro- inhibitory receptors in the brain. According to a toxin, strychnine) and 0.2 pCi of 36C1 - . After 5-s in- generally accepted theory GABA is an inhibitory cubation, the reaction was stopped by addition of 5 neurotransmitter in the cortex and hippocampus. ml ice-cold buffer and rapid filtration through GABAA receptors may be selectively inhibited by Whatman GF/B filters. The filters were washed bicuculline and picrotoxin (Andrews and Johnston with 2 x 5 ml of ice-cold buffer and the 36~1- 1979, Guidotti 1989, Matsumoto 1989), while gly- radioactivity was measured by liquid scintillation cine receptors are localized in the spinal cord and counting. Agonist-stimulated chloride uptake was are inhibited specifically by strychnine (Aprison calculated as the difference between total chloride and Daly 1978, Langosch et al. 1990). Accordingly, uptake and basal uptake in the absence of agonists. glycine binding sites in the cortex and hippocampus Student's t-test was used to analyze the signific- were ascribed to strychnine-insensitivemodulatory ance of differences between mean values. Analysis sites at the N-methyl-D-aspartate (NMDA) recep- of variance (ANOVA), two - way, was applied to tors (Thomson 1989). However, new data suggest determine significance of differences between ef- the existence of glycine inhibitory receptors in the fects of inhibitors on different membrane prepara- hippocampus. Limited specificity of some inhibi- tions (Table 11). tory amino acid receptor antagonists has been also Radioisotope 36~1-(3 mCi/ g Cl) was purchased reported (Oja et al. 1990). from Amersham. GABA, glycine, bicuculline, pi- The aim of this study was to investigate the spe- crotoxin, strychnine and Hepes were from Sigma. cificity of agonists and antagonists for inhibitory All other reagents were of analytical grade. Water amino acid receptors coupled to chloride channels was purified using Milli-ROJMilli-Qplus water sys- in the cerebral cortex and hippocampus. tem (Millipore).

RESULTS METHODS GABA produced a dose-dependent uptake of Fractions of membrane vesicles were prepared chloride into the membrane vesicles of rat cerebral according to the method described by Obata et al. cortex (Fig. I). Maximal uptake was induced by 300 (1988). The cerebral cortex of male Wistar rats or pM GABA. Glycine and taurine at concentrations albino rabbits was homogenized by hand (12 up to 1 mM failed to affect chloride uptake (Fig. I). strokes) using a glass-glass homogenizer in 10 vol- Higher, pharmacologically irrelevant, concentra- umes (wlv) of Krebs-Hepes (K-H) buffer contain- tions of agonists were not investigated. Similar re- Strychnine and GABA-evoked 36~1-uptake 513

d glycine

T taurine Fig. 1. Effect of GABA, gly- I cine and taurine on 36C1 - uptake into crude membrane vesicles isolated from rat cortex. Data represent** mean fSEM (n=3). , Statistical significance of differences between basal and agonist-stimulated uptake: P

AGONIST CONCENTRATION (mM) sults were obtained on membrane vesicles isolated were for bicuculline and strychnine 2.61 pM and from rabbit cortex and hippocampus (Table I). 5.49 pM, respectively (Fig. 2). Both, 10 pM bicu- GABA antagonists - picrotoxin (results not culline and 10 pM strychnine significantly in- shown) and bicucculline, as well as a glycine anta- hibited GABA-induced 36~1-uptake in rat and gonist str chnine, caused a dose-dependent inhibi- rabbit cortex and rabbit hippocampus. No signifi- tion of 32C1- uptake evoked by 100 pM GABA in cant differences were found in the effect of strych- the rat cortex membrane vesicles. Concentrations nine and bicuculline in these different preparations that produce 50% of maximal inhibition (IC5o) of membrane vesicles (Table 11).

TABLE I

Comparison of GABA- and glycine-stimulated 36C1 - uptake into membrane vesicles isolated from rat and rabbit cortex and rabbit hippocampus

36~~uptake

Rat cortex Rabbit cortex Rabbit hippocampus

pmoleslmg protein 100 pM GABA 27.9f1.9"" (n=4)

100 pM Glycine 3.8f2.0 2.Ok1.2 0s.1 (n=3) (n=4) (n=5)

The chloride uptake was calculated as the difference between total and basal uptake. Values of basal uptake for rat cortex, rabbit cortex and** rabbit hippocampus were 7.6a.7, 10.3f1.0,and 10.1B.8 pmoleslmg protein, respectively. Data represent mean f SEM. , Statistical significance of differences between basal and agonist-stimulated uptake: P<0.01. 514 M. Puka and J.W. Lazarewicz

bicuculline

strychnine

Fig. 2. Effect of bicuculline (n=4) and strychnine (n=4)on 100 pM GABA-induced 36~1-uptake into membrane vesicles of rat cortex. Data represent mean +SEM. 100 1 I I I I I 0.0 1 0.1 1 10 100

ANTAGONIST CONCENTRATION (pM)

TABLE I1

-- Effects of inhibitors on ~~~~-stimulated36~1-uptake into membrane vesicles isolated from rat and rabbit cortex and rabbit hippocampus

Rat cortex Rabbit cortex Rabbit hippocampus

% of inhibition 100 pM GABA + 73.2k2.8 100 pM bicuculline (n=4)

100 pM GABA + 10 pM strychnine

The effect of antagonist was calculated as percent of inhibition of the GABA-stimulatedportion of the uptake. Data represent mean +SEM. Differences between 36~1-uptake into vesicles from rat and rabbit cortex and rabbit hippocampus and between effects of bicuculline and strychnine were not significant (P<0.05)by ANOVA, whereas all inhibitory effects of bicuculline and strychnine were significant at P<0.01.

DISCUSSION that the recently described glycine-evoked chloride currents in the hippocampal pyramidal cells were In the present study we investigated the activity observed on neurones dissected from 1- to 2-week- of chloride channels coupled to amino acid recep- old rats (Shirazaki et al. 1991), whereas we used tors. The results indicate that these channels in the material prepared from adult animals. It is well cortical and hippocampal membranes are sensitive known that some receptors for amino acid neuro- to GABA and practically insensitive to glycine and transmitters, including GABAA, undergo drastic taurine, which is in agreement with previously de- changes in the first weeks of development (Kossut scribed data (Oja et al. 1990). It should be stressed et al. 1993). Our finding is in agreement with the Strychnine and GABA-evoked 36 C1 - uptake 515 known preferential localization of strychnine-sensi- GABA receptor-mediated C1- uptake in the dose- tive inhibitory glycine receptors in the spinal cord dependent manner in hippocampal and cortical (Aprison and Daly 1978). In the light of our data a membranes of adult rats and rabbits. Some degree possible role of taurine as an inhibitory neuromodu- of structural homology of GABA and glycine re- lator in the brain acting on the C1- channels may be ceptors coupled to chloride channels as well as of also questioned. the nicotinic acetylcholine receptor, has been sug- Thus, our data indicate that the agonist-evoked gested (Barnard et al. 1987, Smith et al. 1989). 36~1-translocation observed in these experiments Therefore, the lack af antagonists specificity may be may be attributed exclusively to GABAA receptor- caused by similarities between these two types of operated C1- channels, but not to the glycine chan- inhibitory amino acid receptors. nel. This is also supported by the sensitivity of 36 C1 - influx to bicuculline and picrotoxin, which are well ACKNOWLEDGEMENTS known inhibitors of GABAA receptor complex (Andrews and Johnston 1979, Matsumoto 1989). We wish to thank Mrs Anna Sobczuk for her ex- Strychnine is regarded as a specific antagonist of in- cellent technical assistance. This study was sup- hibitory glycine receptors (Langosch et al. 1990). In ported by the Medical Research Centre, Polish s ite of this, strychnine inhibits GABA-evoked Academy of Sciences. 31C1- uptake into the membrane vesicles fraction isolated from the cerebral cortex, resembling in that respect bicuculline and picrotoxin. These results REFERENCES show that in the cerebral cortex strychnine acts as an antagonist of GABA receptors coupled to Andrews P.R., Johnston G.A.R. (1979) Commentary: GABA agonists and antagonists. Biochem. Pharmacol. 28: 2697- chloride channels (GABAA receptors), which are 2702. insensitive to glycine. It indicates a limited speci- AprisonM.H., Daly E.C. (1978) Biochemical aspects of trans- ficity of this glycine receptor inhibitor. mission at inhibitory synapses: The role of glycine. In: Ad- There have been some previous reports sugges- vances in neurochemistry (Eds. B.W. Agranoff and M.H. ting unspecific effects of strychnine (Oja et al. Aprison), Vol 3. Plenum Press, New York, p. 203-294. BarnardE.A., DarlisonM.G., Seeburg P. (1987) Molecularbi- 1990). It has been shown that strychnine binds to ology of the GABAAreceptor: the receptorlchannel super- forebrain GABAA receptors (Squires and Seaderup family. Trends Neurosci. 10: 502-509. 1987), and that in isolated CAI neurones strychnine Carter C., Fage D., Nicolas C. (1 993) Ifenprodil may selec- blocks the GABA-evoked C1- current (Shirasaki et tively antagonise a subtype of NMDA receptor. J. Neu- al. 1991). However, in the latter study both GABA rochem. 61, suppl: S281D. Guidotti A. (1 989) Allosteric regulation of the GABAA iono- and glycine were shown to activate C1- current in the tropic receptor domain. In: Neurochemical pharmacology hippocampal neurones from immature rats, there- - a tribute to B.B.Brodie (Ed. E.Costa). Fidia Research fore, a sharp distinction between glycine and Foundation Symposium Series. Vol. 2. Raven Press, New GABA receptors in these experiments may be ques- York, p. 243-260. tioned. Regional differences in distribution of dif- Hausser M.A., Yung W.H., 1,acey M.G. (1992) Taurine and glycine activate the same C1- conductance in substantia ferent amino acid receptor subtypes has been nigra dopamine neurones. Brain Res. 57 1 : 103-108. demonstrated. For instance important differences in Horikoshi T., Asanuma A., Yanagisawa K., Anzai K., Goto S. regional expression pattern of NMDA receptor sub- (1988) Taurine and IJ-alanine act on both GABA and gly- types have been shown (Monyer et al. 1992). These cine receptors in Xenopus oocyte injected with mouse subtypes may differ in sensitivity to selective recep- brain messenger RNA. Molec. Brain Res. 4: 97-105. Kossut M., Glazewski S., SiucinskaE., Skangiel - Kramska J. tor antagonists (Carter et al. 1993). Some differen- (1993). Functional plasticity and neurotransmitter recep- ces between species may be also expected. Here we tor binding in the vibrissal barrel cortex. Acta Neurobiol. clearly demonstrated that strychnine inhibits Exp. 53: 161-173. 516 M. Puka and J.W. tazarewicz

Langosch D., Becker C.-M., Betz H. (1990) The inhibitory Oja S.S., Korpi E.R. and Saransaari P. (1990) Modification of glycine receptor: A ligand-gated chloride channel of the chloride flux across brain membranes by inhibitory amino central nervous system. Eur. J. Biochem. 194: 1-8. acids in developing and adult mice. Neurochem. Res. 15: Matsumoto R.R. (1989) GABA receptors: are cellular dif- 797-804. ferences reflected in function? Brain Res. Rev. 14: 203- Shirasaki T., Klee M.R., Nakaye T. and Akaike N. (1991) Dif- 225. ferential blockade of bicuculline and strychnine on Medina J.H., De Robertis E. (1984) Taurine modulation of the GABA- and glycine-induced responses in dissociated rat benzodiazepine-y-arninobutyric acid receptor complex in hippocampal pyramidal cells. Brain Res. 561: 77-83. brain membranes. J. Neurochem. 42: 1212- 12 17. Smith M.R., Zorec R. and McBurney R.N. (1989) Conduct- Monyer H., Sprengel R., Schoepfer R., Herb A., Higuchi M., ance states activated by glycine and GABA in rat cultured Lomeli H., Burnashev N., Sakmann B., Seeburg P. H. spinal neurones. J. Membr. Biol. 108: 45-52. (1992) Heteromeric NMDA receptors: molecular and func- Squires R.F. and Seaderup E. (1987) GABAA receptor bloc- tional distinction of subtypes. Science 256: 1217-1221. kers reverse the inhibitory effect of GABA on brain-spe- Obata T., Yamamura H.I., Malatynska E., Ikeda M., Laird H., cific [35~]~~~~binding. Brain Res. 414: 357-364. Palmer C.J., Casida J.E. (1988) Modulation of y-aminobu- Thomson A.M. (1989) Glycine modulation of the NMDA re- tyric acid-stimulated chloride influx by bicycloorthocarbox- ceptorlchannel complex. Trends Neurosci. 12: 349-353. ylates, bicyclophosphorus esters, polychlorocycloalkanes and other cage convulsants. J. Pharmacol. Exp. Ther. 244: 802-806. Received 14 May 1993, accepted 30 August 1993