102
:VL'ur~sci~'nct' Lett~'J',, i ~0 ~1992~ 1(12 it~: :i' 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 03~4-3~)41)t)2/S 05.00
NSL 08422
Sensitive procedures for measuring chloride fluxes mediated by the purified glycine receptor incorporated into phospholipid vesicles Margarita Garcia-Calvo, Fernando Valdivieso, Federico M a y o r Jr. and Jesfls Vgtzquez Departamento de Biologia Molecular UAM-CSIC, Facultad de Ciencias, Universidad Aut6noma de Madrid. Madrid ( Spain )
(Received 20 September 1991; Revised version received 21 November 1991; Accepted 27 November 199 l ) Key words." Glycine receptor; Reconstitution; Chloride; Channel; Rat; Spinal cord; Fluorescence quenching
Two novel methods have been developed to directly measure chloride influx into purified glycine receptor-containing phospholipid vesicles. Using a method based on the fluorescence quenching of a chloride-sensitive probe entrapped into the vesicles, a chloride influx was detected which could be enhanced by glycine and completely abolished by the antagonist strychnine. In addition, by tracing the ~6C1 influx into the proteoliposomes, a stimulatory effect of both glycine and fl-alanine could be seen, which can be inhibited by strychnine and other glycine antagonists. These data, together with a previous report demonstrating ligand-mediated iodide fluxes in the same preparation (Biochemistry, 28 (1989) 6405q6409), clearly demonstrate the utility of the reconstituted receptor preparation to investigate some ion channel and pharmacological properties of the glycine receptor.
The inhibitory neurotransmitter glycine appears to produce its effect through a hyperpolarization due to the opening of chloride channels associated with glycine receptors, which can be blocked by the convulsive alkaloid strychnine [1, 2]. This selective antagonist has been utilized as a tool for glycine receptor purification and characterization [13, 15, 17, 18, 21]. The glycine receptor has been reported to contain two subunits of M r 4 8 , 0 0 0 and 58,000 and a cytoplasmic associated peripheral protein of 93,000 Da [15, 19]. The 48 and 58 kDa polypeptides have been proposed to assemble the chloride channel in a pentameric core structure [3, 12, 16]. In a previous study, we attempted the reconstitution of the glycine receptor into phospholipid vesicles. This preparation exhibited optimal preservation of the pharmacological and [3H]strychnine binding properties [7]. Moreover, we described for the first time a technique for the determination of glycine receptor-induced iodide fluxes. The aim of the present work was to support and extend these observations, describing an agonist-activated, antagonist-sensitive chloride influx through the purified glycine receptor incorporated into liposomes. Glycine receptors were purified from rat spinal cord by affinity chromatography on 2-amino-strychnineCorrespondence: F. Mayor Jr., Departamento de Biologia Molecular UAM-CSIC, Facultad de Ciencias, Universidad Aut6noma de Madrid, Cantoblanco, 28049 Madrid, Spain.
agarose and reconstituted into liposomes by gel filtration through Sephadex G-50 as reported [7]. The column was pre-equilibrated and eluted with the reconstitution buffer (25 mM potassium phosphate, pH 7.4) containing either 120 mM potassium gluconate (for fluorescence quenching assays) or 120 mM potassium chloride (for 36C1 flUX experiments). The proteoliposomes, which eluted in the void volume, were fused with 1/10 volume of a preparation of sonicated liposomes through a freeze-thaw sonication cycle [7] and immediately used for functional studies. Two different experimental approaches were developed to demonstrate specific chloride influx through the purified, reconstituted glycine receptor. In the first method, a fluorescence quenching assay using the probe SPQ (6-methoxy-N-(3-sulfopropyl)quinolinium; Molecular Probes) [7] was employed to measure chloride permeation through the channel. For these studies, and prior to the freeze-thaw-sonication cycle, reconstituted vesicles were loaded with SPQ as described [7]. Fluorescence time courses were measured when 100/.tl (about 2 pmol of [3H]strychnine binding sites [13]) of large proteoliposomes were mixed with 2.9 ml of 12 mM NaC1, 108 mM potassium gluconate, 25 mM potassium phosphate, pH 7.4, in the absence (controls) or presence of 100 pM glycine. To determine the effect of strychnine, the proteoliposomes were incubated with 50/tM strychnine for 10 min at 4°C before the addition of NaCI and the agonist.
103 ~(;()
I¸
1 i~
I----
!
B (/3
150
"i'
' ~'"
7, 1 O0
=
.%
,i
c)
,i
-D
[±
U'
I~ ¸
.J L ,)
,
18-
A
,
!
._1
L.~ :!,()
7T'
16
/
14
!
0
4O
50 C I
10 ¸ -2 4 7ir/-/e ( s e c )
6
I
G ~ ,:i
Fig. 1. Chloride influx into glycine receptor-containing liposomes measured by the SPQ fluorescence quenching assay (A), and the ~"C1 tracing method (B). A: fluorescence quenching was monitored as described in the text. Data are representative of 3 independent experiments. I I, controls: g, 100/aM glycine; A, 100/aM glycine plus 50/aM strychnine. B: ~6('1 uptake is shown as percentage of controls (basal influx: 46.05~_3.35 nmol C1/s): each value represents the mean + S.E.M. of 10 separate experiments with different preparations and performed in triplicate. (a), control; (b), 100 /aM glycine; (c), 100 /aM glycine plus 50/aM strychnine.
The initial fluorescence (F~) in the absence of NaCI was set equal to 1 and the recordings were made as described [7, 20]. Fig. 1A shows a representative time-course of the fluorescence quenching observed when proteoliposomes entrapping SPQ are mixed with an equiosmotic solution containing chloride. Addition of 100 # M glycine clearly increases the rate of ion translocation into the vesicles, as compared to control conditions. This effect is completely blocked by the simultaneous presence of 50 # M strychnine. The zero-time slopes of fluorescence quenching kinetics can be related to the initial rate of anion influx [10], when the various slopes were compared, strychnine was seen to reduce the initial rate obtained in the presence of glycine by about 50% (34 _+ 5 s ~ vs. 70 + 4 s t when only glycine is present). We have also studied the biochemical and pharmacological properties of the reconstituted glycine receptor by tracing the ~"CI influx into proteoliposomes. This technique has been successfully applied to examine the function of the purified ?'-aminobutyric acid (GABA) receptor ~hanncl complex [5, 9]. To measure the 36C1 influx, 120 200 #1 of large proteoliposomes (2 3 pmol of [3H]strychnine binding sites [13]) were preincubated for 10 rain at 30°C in the absence (controls) or presence of glycine receptor ligands. Aliquots of a 3~'CINa solution (0.6 0.7 Ci/mmol: Amersham) were then added to a final concentration of 2 #Ci/ml, and after 3-5 s 30 #1 of 20 mg/ml ?'-globulin and 300 #1 of 50% (w/v) polyethyleneglycol 6000 were rapidly added and mixed, diluted with 3 ml of ice-cold 8% (w/v) polyethyleneglycol 6000 in 120 mM KC1, 20 mM potassium phosphate, pH 7.4 (8% PEG) and filtered through Whatman GF/C filters pre-
2O
T -7
-
log
t5
i
3
abcd Clycine
ef AlanM~
Fig. 2. Effect of glycine concentration (A), and different ligands IB) on ~CI influx into glycine receptor-containing liposomes. A: proteoliposomes (1.8 pmol [~H]strychnine binding sites) were incubated with various concentrations of glycine in the absence and presence of 50 #M strychnine. The strychnine-sensitive fluxes are plotted vs. glycine concentration. Data are the means _+ S.E.M. of quadruplicate determinations. B: proteoliposomes (2.5 pmol [~H]strychnine binding sites) were incubated in the presence of 100 # M glycine {a d), 100/aM fl-alanine (e,f), 5 0 # M lso-THAZ Ib), 5/aM RU-5135 (c) or 50/aM strychnine (d,f). Data are the means _+ S.E.M. of triplicate determinations.
soaked on 0.05% polyethylenimine [6]. The filters were washed twice with 8% PEG and dried. The incorporated radioactivity was determined by liquid scintillation counting (LKB 1219 Rackbeta). As expected from the fluorescence quenching data, we detected a relatively high basal permeability to chloride under control conditions. In the presence of glycine, an enhancement of chloride influx of about 50% over the control value was observed (Fig. 1B). This activating effect of glycine was abolished in the presence of 50 # M strychnine: the 3~'C1 influx was even inferior to the basal levels, suggesting that some channels may be opened in the absence of agonist. Similar effects have been observed in different GABA receptor preparations, depending on ligand concentrations [11, 14]. The effect of increasing glycine concentration on '~C1 influx is shown in Fig. 2A. Glycine activated the ~"C1 influx in a dose-dependent manner, giving a half-maximal effect between 1 and 10 #M. The concentration response relationship for the glycine-stimulated chloride influx is consistent with electrophysiological studies [4, 8]. The modulation of agonist-stimulated ~'C1 influx into proteoliposomes is illustrated in Fig. 2B. In addition to strychnine, other antagonists such as Iso-THAZ (kindly provided by Dr. Krogsgaard-Larsen, Copenhagen, Denmark) and RU-5135 (donated by RousselUclaf Laboratories, Paris, France) also inhibited the glycine-induced 3~C1 influx at micromolar concentrations. When glycine was substituted by fl-alanine, the same stimulatory effect on ~6CI influx was observed, although to a slightly lower extent. This fi-alanine-stimu-
104
lated 36C1 influx was also antagonized by 50 ¢tM strychnine. These findings are in agreement with the data obtained by the fluorescence quenching method and with those presented in a previous study [7]. In conclusion, this paper describes how the chloride influx through the glycine receptor can be directly measured by at least two relatively simple biochemical methods. The results presented herein, produced using the physiologically relevant chloride anion, support and further extend the results of our previous work [7] in which a specific iodide influx was proposed to be indicative of the functional integrity of the reconstituted glycine receptor. In addition, since the basic pharmacological properties of the receptor seem to remain intact, this preparation could be very useful for the study of glycine receptor functionality. For instance, the proteoliposomes could be fused with a planar lipid bilayer, to study the properties of a single channel in a highly nonleaky environment, under controlled voltage and ionic conditions. We thank Dr. Ana Ruiz-Gdmez and Esperanza Morato for advice about glycine receptor preparations. This work was supported in part by CICYT-CSIC Grant PB870216, Boehringer Ingelheim and an institutional grant from Fundaci6n Ram6n Areces. 1 Aprison, M.H. and Daly, E.C., Biochemical aspects of transmission at inhibitory synapses: the role of glycine, Adv. Neurochem., 3 (1978) 203-294. 2 Betz, H., Biology and structure of the mammalian glycine receptor, Trends Neurosci., 10 (1987) 113-117. 3 Betz, H., Ligand-gated ion channels in the brain: the amino acid receptor superfamily, Neuron, 5 (1990) 383-392. 4 Borman, J., Hamill, O.P. and Sakmann, B., Mechanism of anion permeation through channels gated by glycine and GABA in mouse cultured spinal neurons, J. Physiol., 385 (1987) 243 286. 5 Bristow, D.R. and Martin, I.L., Biochemical characterization of an isolated and functionally reconstituted GABA/benzodiazepine receptor, J. Neurochem., 54 (1990) 751-761. 6 Bruns, R.F., Lawson-Wendling, K. and Pugsley, T.A., A rapid filtration assay for soluble receptors using polyethylenimine-treated filters, Anal. Biochem., 132 (1983) 74-81. 7 Garcia-Calvo, M., Ruiz-G6mez, A., Vfizquez, J., Morato, E., Valdivieso, F. and Mayor Jr., F., Functional reconstitution of the glycine receptor, Biochemistry, 28 (1989) 6405 6409.
8 Hamill, O.R, Borman, J. and Sakmann, B., Actix.ation of multiple conductance state chloride channels in spinal neurons by glycine and GABA, Nature, 305 (1983) 805 -808. 9 Hirouchi, M., Taguchi, J., [leha, T. and Kuriyama, K.. GABAstimulated ~6C1 influx into reconstituted vesicles with purified GABAA/benzodiazepine receptor complex, Biochem. Biophys. Res Commun., 146 (1987) 1471 1477. 10 Illsley, N.E and Verkman, A.S., Membrane chloride transport measured using a chloride-sensitive fluorescent probe. Biochemistry, 26 (1987) 1215 1219. 11 Kardos, J., '6C1 measurements on GABA receptor-activated chloride exchange, Biochem. Pharmacol., 38 (1989)2587 2591. 12 Langosch, D., Thomas, L. and Betz, H., Conserved quaternary structure of ligand-gated ion channels: the postsynaptic glycine receptor is a pentamer, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 7394 7398. 13 Marviz6n, J.C.G., V/lzquez, J., Garcia-Calvo, M., Mayor ,It., F., Ruiz-G6mez, A., Valdivieso, F. and Benavides, J., The glycine receptor: pharmacological studies and mathematical modelling of the allosteric interaction between the glycine- and strychnine binding sites, Mol. Pharmacot., 30 (1986) 590-597. 14 Paul, S.M., Schwartz, R.D., Creveling, C.R., Hollingsworth, E.B., Daly, J.W. and Skolnick, R, GABA receptor-mediated chloride transport in a 'cell-free" membrane preparation from brain, Science, 233 (1986) 228 229. 15 Pfeiffer, F., Graham, D. and Betz, H., Purification by affinity chromatography of the glycine receptor of ral spinal cord, J. Biol. Chem., 257 (1982) 9389-9393. 16 Pfeiffer, F., Simler, R., Grenningloh, G. and Betz, H., Monoclona! antibodies and peptide mapping reveal structural similarities between the subunits of the glycine receptor of rat spinal cord, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 7224--7227. 17 Ruiz-G6mez, A., Garcia-Catvo, M., V~izquez, J. Marviz6n, J.C:G., Valdivieso, F. and Mayor Jr., F., Thermodynamics of agonist and antagonist interaction with the strychnine-sensitive glycine receptor, J. Neurochem., 52 (19891 1775-1780. 18 Ruiz-G6mez, A., Vaello, M.L., Valdivieso, F. and Mayor Jr., F., Phosphorylation of the 48-kDa subunit of the glycine receptor by protein kinase C, J. Biol. Chem., 266 (1991) 559 566. 19 Schmitt, B., Knaus, R., Becker, C.M. and Betz, H., The Mr 93.000 polypeptide of the postsynaptic glycine receptor complex is a peripheral membrane protein, Biochemistry, 26 (1987) 805 811. 20 V~izquez, J., Garcia-Calvo, M., Valdivieso, F., Mayor, F. and Mayor Jr., F., Interaction of bilirubin with the synaptosomal plasma membrane, J. Biol. Chem., 263 (1988) 1255-1265. 21 Young, A.B. and Snyder, S.H., Strychnine binding associated with glycine receptors of the central nervous system, Proc. Natl. Acad. Sci. U.S.A., 70 (1973) 2832 2836.
:VL'ur~sci~'nct' Lett~'J',, i ~0 ~1992~ 1(12 it~: :i' 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 03~4-3~)41)t)2/S 05.00
NSL 08422
Sensitive procedures for measuring chloride fluxes mediated by the purified glycine receptor incorporated into phospholipid vesicles Margarita Garcia-Calvo, Fernando Valdivieso, Federico M a y o r Jr. and Jesfls Vgtzquez Departamento de Biologia Molecular UAM-CSIC, Facultad de Ciencias, Universidad Aut6noma de Madrid. Madrid ( Spain )
(Received 20 September 1991; Revised version received 21 November 1991; Accepted 27 November 199 l ) Key words." Glycine receptor; Reconstitution; Chloride; Channel; Rat; Spinal cord; Fluorescence quenching
Two novel methods have been developed to directly measure chloride influx into purified glycine receptor-containing phospholipid vesicles. Using a method based on the fluorescence quenching of a chloride-sensitive probe entrapped into the vesicles, a chloride influx was detected which could be enhanced by glycine and completely abolished by the antagonist strychnine. In addition, by tracing the ~6C1 influx into the proteoliposomes, a stimulatory effect of both glycine and fl-alanine could be seen, which can be inhibited by strychnine and other glycine antagonists. These data, together with a previous report demonstrating ligand-mediated iodide fluxes in the same preparation (Biochemistry, 28 (1989) 6405q6409), clearly demonstrate the utility of the reconstituted receptor preparation to investigate some ion channel and pharmacological properties of the glycine receptor.
The inhibitory neurotransmitter glycine appears to produce its effect through a hyperpolarization due to the opening of chloride channels associated with glycine receptors, which can be blocked by the convulsive alkaloid strychnine [1, 2]. This selective antagonist has been utilized as a tool for glycine receptor purification and characterization [13, 15, 17, 18, 21]. The glycine receptor has been reported to contain two subunits of M r 4 8 , 0 0 0 and 58,000 and a cytoplasmic associated peripheral protein of 93,000 Da [15, 19]. The 48 and 58 kDa polypeptides have been proposed to assemble the chloride channel in a pentameric core structure [3, 12, 16]. In a previous study, we attempted the reconstitution of the glycine receptor into phospholipid vesicles. This preparation exhibited optimal preservation of the pharmacological and [3H]strychnine binding properties [7]. Moreover, we described for the first time a technique for the determination of glycine receptor-induced iodide fluxes. The aim of the present work was to support and extend these observations, describing an agonist-activated, antagonist-sensitive chloride influx through the purified glycine receptor incorporated into liposomes. Glycine receptors were purified from rat spinal cord by affinity chromatography on 2-amino-strychnineCorrespondence: F. Mayor Jr., Departamento de Biologia Molecular UAM-CSIC, Facultad de Ciencias, Universidad Aut6noma de Madrid, Cantoblanco, 28049 Madrid, Spain.
agarose and reconstituted into liposomes by gel filtration through Sephadex G-50 as reported [7]. The column was pre-equilibrated and eluted with the reconstitution buffer (25 mM potassium phosphate, pH 7.4) containing either 120 mM potassium gluconate (for fluorescence quenching assays) or 120 mM potassium chloride (for 36C1 flUX experiments). The proteoliposomes, which eluted in the void volume, were fused with 1/10 volume of a preparation of sonicated liposomes through a freeze-thaw sonication cycle [7] and immediately used for functional studies. Two different experimental approaches were developed to demonstrate specific chloride influx through the purified, reconstituted glycine receptor. In the first method, a fluorescence quenching assay using the probe SPQ (6-methoxy-N-(3-sulfopropyl)quinolinium; Molecular Probes) [7] was employed to measure chloride permeation through the channel. For these studies, and prior to the freeze-thaw-sonication cycle, reconstituted vesicles were loaded with SPQ as described [7]. Fluorescence time courses were measured when 100/.tl (about 2 pmol of [3H]strychnine binding sites [13]) of large proteoliposomes were mixed with 2.9 ml of 12 mM NaC1, 108 mM potassium gluconate, 25 mM potassium phosphate, pH 7.4, in the absence (controls) or presence of 100 pM glycine. To determine the effect of strychnine, the proteoliposomes were incubated with 50/tM strychnine for 10 min at 4°C before the addition of NaCI and the agonist.
103 ~(;()
I¸
1 i~
I----
!
B (/3
150
"i'
' ~'"
7, 1 O0
=
.%
,i
c)
,i
-D
[±
U'
I~ ¸
.J L ,)
,
18-
A
,
!
._1
L.~ :!,()
7T'
16
/
14
!
0
4O
50 C I
10 ¸ -2 4 7ir/-/e ( s e c )
6
I
G ~ ,:i
Fig. 1. Chloride influx into glycine receptor-containing liposomes measured by the SPQ fluorescence quenching assay (A), and the ~"C1 tracing method (B). A: fluorescence quenching was monitored as described in the text. Data are representative of 3 independent experiments. I I, controls: g, 100/aM glycine; A, 100/aM glycine plus 50/aM strychnine. B: ~6('1 uptake is shown as percentage of controls (basal influx: 46.05~_3.35 nmol C1/s): each value represents the mean + S.E.M. of 10 separate experiments with different preparations and performed in triplicate. (a), control; (b), 100 /aM glycine; (c), 100 /aM glycine plus 50/aM strychnine.
The initial fluorescence (F~) in the absence of NaCI was set equal to 1 and the recordings were made as described [7, 20]. Fig. 1A shows a representative time-course of the fluorescence quenching observed when proteoliposomes entrapping SPQ are mixed with an equiosmotic solution containing chloride. Addition of 100 # M glycine clearly increases the rate of ion translocation into the vesicles, as compared to control conditions. This effect is completely blocked by the simultaneous presence of 50 # M strychnine. The zero-time slopes of fluorescence quenching kinetics can be related to the initial rate of anion influx [10], when the various slopes were compared, strychnine was seen to reduce the initial rate obtained in the presence of glycine by about 50% (34 _+ 5 s ~ vs. 70 + 4 s t when only glycine is present). We have also studied the biochemical and pharmacological properties of the reconstituted glycine receptor by tracing the ~"CI influx into proteoliposomes. This technique has been successfully applied to examine the function of the purified ?'-aminobutyric acid (GABA) receptor ~hanncl complex [5, 9]. To measure the 36C1 influx, 120 200 #1 of large proteoliposomes (2 3 pmol of [3H]strychnine binding sites [13]) were preincubated for 10 rain at 30°C in the absence (controls) or presence of glycine receptor ligands. Aliquots of a 3~'CINa solution (0.6 0.7 Ci/mmol: Amersham) were then added to a final concentration of 2 #Ci/ml, and after 3-5 s 30 #1 of 20 mg/ml ?'-globulin and 300 #1 of 50% (w/v) polyethyleneglycol 6000 were rapidly added and mixed, diluted with 3 ml of ice-cold 8% (w/v) polyethyleneglycol 6000 in 120 mM KC1, 20 mM potassium phosphate, pH 7.4 (8% PEG) and filtered through Whatman GF/C filters pre-
2O
T -7
-
log
t5
i
3
abcd Clycine
ef AlanM~
Fig. 2. Effect of glycine concentration (A), and different ligands IB) on ~CI influx into glycine receptor-containing liposomes. A: proteoliposomes (1.8 pmol [~H]strychnine binding sites) were incubated with various concentrations of glycine in the absence and presence of 50 #M strychnine. The strychnine-sensitive fluxes are plotted vs. glycine concentration. Data are the means _+ S.E.M. of quadruplicate determinations. B: proteoliposomes (2.5 pmol [~H]strychnine binding sites) were incubated in the presence of 100 # M glycine {a d), 100/aM fl-alanine (e,f), 5 0 # M lso-THAZ Ib), 5/aM RU-5135 (c) or 50/aM strychnine (d,f). Data are the means _+ S.E.M. of triplicate determinations.
soaked on 0.05% polyethylenimine [6]. The filters were washed twice with 8% PEG and dried. The incorporated radioactivity was determined by liquid scintillation counting (LKB 1219 Rackbeta). As expected from the fluorescence quenching data, we detected a relatively high basal permeability to chloride under control conditions. In the presence of glycine, an enhancement of chloride influx of about 50% over the control value was observed (Fig. 1B). This activating effect of glycine was abolished in the presence of 50 # M strychnine: the 3~'C1 influx was even inferior to the basal levels, suggesting that some channels may be opened in the absence of agonist. Similar effects have been observed in different GABA receptor preparations, depending on ligand concentrations [11, 14]. The effect of increasing glycine concentration on '~C1 influx is shown in Fig. 2A. Glycine activated the ~"C1 influx in a dose-dependent manner, giving a half-maximal effect between 1 and 10 #M. The concentration response relationship for the glycine-stimulated chloride influx is consistent with electrophysiological studies [4, 8]. The modulation of agonist-stimulated ~'C1 influx into proteoliposomes is illustrated in Fig. 2B. In addition to strychnine, other antagonists such as Iso-THAZ (kindly provided by Dr. Krogsgaard-Larsen, Copenhagen, Denmark) and RU-5135 (donated by RousselUclaf Laboratories, Paris, France) also inhibited the glycine-induced 3~C1 influx at micromolar concentrations. When glycine was substituted by fl-alanine, the same stimulatory effect on ~6CI influx was observed, although to a slightly lower extent. This fi-alanine-stimu-
104
lated 36C1 influx was also antagonized by 50 ¢tM strychnine. These findings are in agreement with the data obtained by the fluorescence quenching method and with those presented in a previous study [7]. In conclusion, this paper describes how the chloride influx through the glycine receptor can be directly measured by at least two relatively simple biochemical methods. The results presented herein, produced using the physiologically relevant chloride anion, support and further extend the results of our previous work [7] in which a specific iodide influx was proposed to be indicative of the functional integrity of the reconstituted glycine receptor. In addition, since the basic pharmacological properties of the receptor seem to remain intact, this preparation could be very useful for the study of glycine receptor functionality. For instance, the proteoliposomes could be fused with a planar lipid bilayer, to study the properties of a single channel in a highly nonleaky environment, under controlled voltage and ionic conditions. We thank Dr. Ana Ruiz-Gdmez and Esperanza Morato for advice about glycine receptor preparations. This work was supported in part by CICYT-CSIC Grant PB870216, Boehringer Ingelheim and an institutional grant from Fundaci6n Ram6n Areces. 1 Aprison, M.H. and Daly, E.C., Biochemical aspects of transmission at inhibitory synapses: the role of glycine, Adv. Neurochem., 3 (1978) 203-294. 2 Betz, H., Biology and structure of the mammalian glycine receptor, Trends Neurosci., 10 (1987) 113-117. 3 Betz, H., Ligand-gated ion channels in the brain: the amino acid receptor superfamily, Neuron, 5 (1990) 383-392. 4 Borman, J., Hamill, O.P. and Sakmann, B., Mechanism of anion permeation through channels gated by glycine and GABA in mouse cultured spinal neurons, J. Physiol., 385 (1987) 243 286. 5 Bristow, D.R. and Martin, I.L., Biochemical characterization of an isolated and functionally reconstituted GABA/benzodiazepine receptor, J. Neurochem., 54 (1990) 751-761. 6 Bruns, R.F., Lawson-Wendling, K. and Pugsley, T.A., A rapid filtration assay for soluble receptors using polyethylenimine-treated filters, Anal. Biochem., 132 (1983) 74-81. 7 Garcia-Calvo, M., Ruiz-G6mez, A., Vfizquez, J., Morato, E., Valdivieso, F. and Mayor Jr., F., Functional reconstitution of the glycine receptor, Biochemistry, 28 (1989) 6405 6409.
8 Hamill, O.R, Borman, J. and Sakmann, B., Actix.ation of multiple conductance state chloride channels in spinal neurons by glycine and GABA, Nature, 305 (1983) 805 -808. 9 Hirouchi, M., Taguchi, J., [leha, T. and Kuriyama, K.. GABAstimulated ~6C1 influx into reconstituted vesicles with purified GABAA/benzodiazepine receptor complex, Biochem. Biophys. Res Commun., 146 (1987) 1471 1477. 10 Illsley, N.E and Verkman, A.S., Membrane chloride transport measured using a chloride-sensitive fluorescent probe. Biochemistry, 26 (1987) 1215 1219. 11 Kardos, J., '6C1 measurements on GABA receptor-activated chloride exchange, Biochem. Pharmacol., 38 (1989)2587 2591. 12 Langosch, D., Thomas, L. and Betz, H., Conserved quaternary structure of ligand-gated ion channels: the postsynaptic glycine receptor is a pentamer, Proc. Natl. Acad. Sci. U.S.A., 85 (1988) 7394 7398. 13 Marviz6n, J.C.G., V/lzquez, J., Garcia-Calvo, M., Mayor ,It., F., Ruiz-G6mez, A., Valdivieso, F. and Benavides, J., The glycine receptor: pharmacological studies and mathematical modelling of the allosteric interaction between the glycine- and strychnine binding sites, Mol. Pharmacot., 30 (1986) 590-597. 14 Paul, S.M., Schwartz, R.D., Creveling, C.R., Hollingsworth, E.B., Daly, J.W. and Skolnick, R, GABA receptor-mediated chloride transport in a 'cell-free" membrane preparation from brain, Science, 233 (1986) 228 229. 15 Pfeiffer, F., Graham, D. and Betz, H., Purification by affinity chromatography of the glycine receptor of ral spinal cord, J. Biol. Chem., 257 (1982) 9389-9393. 16 Pfeiffer, F., Simler, R., Grenningloh, G. and Betz, H., Monoclona! antibodies and peptide mapping reveal structural similarities between the subunits of the glycine receptor of rat spinal cord, Proc. Natl. Acad. Sci. U.S.A., 81 (1984) 7224--7227. 17 Ruiz-G6mez, A., Garcia-Catvo, M., V~izquez, J. Marviz6n, J.C:G., Valdivieso, F. and Mayor Jr., F., Thermodynamics of agonist and antagonist interaction with the strychnine-sensitive glycine receptor, J. Neurochem., 52 (19891 1775-1780. 18 Ruiz-G6mez, A., Vaello, M.L., Valdivieso, F. and Mayor Jr., F., Phosphorylation of the 48-kDa subunit of the glycine receptor by protein kinase C, J. Biol. Chem., 266 (1991) 559 566. 19 Schmitt, B., Knaus, R., Becker, C.M. and Betz, H., The Mr 93.000 polypeptide of the postsynaptic glycine receptor complex is a peripheral membrane protein, Biochemistry, 26 (1987) 805 811. 20 V~izquez, J., Garcia-Calvo, M., Valdivieso, F., Mayor, F. and Mayor Jr., F., Interaction of bilirubin with the synaptosomal plasma membrane, J. Biol. Chem., 263 (1988) 1255-1265. 21 Young, A.B. and Snyder, S.H., Strychnine binding associated with glycine receptors of the central nervous system, Proc. Natl. Acad. Sci. U.S.A., 70 (1973) 2832 2836.
