Strychnine Binding Associated with Glycine Receptors of the Central

Transcription

1 Proc. Nat. Acad. Sci. USA Vol. 70, No. 10, pp , October 1973 Strychnine Binding Associated with Glycine Receptors of the Central Nervous System ANNE B. YOUNG AND SOLOMON H. SNYDER* Departments of Pharmacology and Experimental Therapeutics, and Psychiatry and the Behavioral Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland Communicated by Cheves Walling, June 22, 1973 ABSTRACT ['HIStrychnine binds to synaptic-membrane fractions of the spinal cord in a selective fashion, indicating an interaction with postsynaptic glycine receptors. Displacement of strychnine-by glycine and other amino acids parallels their glycine-like neurophysiologic activity. The regional localiition of strychnine binding in the central nervous system correlates closely with endogenous glycine concentrations. In subcellular fractionation experiments, strychnine binding is most enhanced in synaptic-membrane fractions. Strychnine binding is saturable, with affinity constants for glycine and strychnine of 10 and 0.03 MM, respectively. Biochemical (1) and neurophysiological (2) evidence suggests that glycine is a major inhibitory neurotransmitter in the mammalian central nervous system. Strychnine antagonies the hyperpolariing actions of glycine at spinal synapses, where it also antagonies naturally occurring synaptic inhibition (3). Glycine best mimics a natural inhibitory transmitter in the spinal cord and brain stem, but not in the cerebral cortex (2). Similarly, endogenous glycine (4), highaffinity synaptosomal uptake of glycine (5), and unique glycine-accumulating synaptosomes (6) are most concentrated in the spinal cord and brain stem and not in the cerebral cortex. Here we report the binding of [3H]strychnine to membrane fractions of the spinal cord and brain stem, which appears to represent a specific interaction with the postsynaptic glycine receptor. METHODS Strychnine Was Labeled by catalytic tritium exchange at New England Nuclear Corp. 50 mg of strychnine sulfate was dissolved in 0.3 ml of trifluoroacetic acid, and to this was added 50 mg of 5% Rh/A1203 and 25 Ci of tritiated H20. The reaction mixture was stirred overnight at 800 and labile tritium was removed under reduced pressure with methanol as a solvent. After filtration from the solvent, the product was dissolved in 10 ml of methanol. In our laboratory, the product was purified by thin-layer chromatography on Silica Gel F-254 plates, 0.25 mm thickness (EM Laboratories, Inc., Elmsford, N.Y.), in three consecutive solvent systems (nbutanol-glacial acetic acid-h20 4:1:1, chloroform-methanol 1: 1, and methanol-h20 7:3). Purified [3H]strychnine moved as a single peak with authentic strychnine in all three systems. The specific activity of [3H]strychnine was 13 Ci/mmol, as determined by comparison with the ultraviolet absorption of standard solutions at 254 nm. * To whom reprint requests should be sent at the Department of Pharmacology. Tissue Preparation. Male Sprague-Dawley rats ( g) were decapitated and the brains and spinal cords were rapidly removed. For typical binding assays, the meninges were carefully removed and the medulla oblongata-pons and spinal cord were combined and homogenied in 20 volumes of ice-cold 0.32 M sucrose in a Potter-Elvehjem glass homogenier fitted with a Teflon pestle. The whole homogenate was centrifuged for 10 min at 1000 X g. The pellet (crude nuclear fraction) was discarded, and the resultant supernatant fluid was centrifuged for 20 min at 17,000 X g. The pellet (crude mitochondrial fraction), after resuspension in 20 volumes of ice-cold distilled H20, was homogenied with a ground-glass homogenier and centrifuged for 20 min at 9000 X g. The supernatant fluid was collected and the pellet, a bilayer with a soft buffy upper coat, wa's rinsed carefully with the supernatant fluid to collect the upper layer. The supernatant fluid was then centrifuged at 48,000 X g for 20 min. The final pellet was resuspended in 0.05 M sodium-potassium phosphate buffer (ph 7.4) ( mg of protein per ml) to obtain crude synaptic membranes. This suspension was divided into aliquots and stored at -30. Specific strychnine binding was retained intact for at least 35 days under these storage conditions. Binding Assay. For measurement of the specific binding of strychnine to spinal-cord membranes, aliquots of the crude synaptic membranes ( mg of protein) were incubated in triplicate at 40 or 250 for 10 min in 2 ml of 0.05 M Na-K phosphate buffer (ph 7.4) containing 0.01% Triton X-100 and 4 nm [3H]strychnine (38,000 cpm) alone or in the presence of 1 mm glycine or'0.1 mm strychnine. After incubation, the reaction was terminated by centrifugation for 10 min at 48,000 X g. The supernatant fluid was decanted, and the pellet was washed with 5 ml, then 10 ml of ice-cold phosphate buffer. Bound radioactivity was extracted into Triton X-100- toluene phosphor for 8-12 hr and assayed by liquid scintillation spectrometry (Packard' Tricarb model 3385 or 3375), at a counting efficiency of 33%. For studies requiring rapid binding determinations, samples were filtered under reduced pressure -through Whatman glass-fiber circles (GF-B) and washed once with 5 ml of cold phosphate buffer (the filtering and washing'required 8 see). The filters were shaken with 18 ml of Triton X-100- toluene phosphor for 20 min and radioactivity was determined by liquid scintillation spectrometry. "Specific [3H]strychnine binding" is obtained by subtracting from the total bound radioactivity the amount that is not displaced by high concentrations (1 mm) of glycine. 2832

2 Proc. Nat. Acad. Sci. USA 70 (1973) TABLE 1. Subcellular distribution of specific [8H]strychninebinding in rat central nervous system Specific [3H~strychnine binding Specific Total activity activity (cpm/mg of (cpm/g of Fraction protein) wet tissue) Whole homogenate Undialyed ,000 Dialyed ,300 Crude nuclear pellet (PI) Undialyed 240 1,670 Dialyed 260 1,870 Crude mitochondrial pellet (P2) Undialyed 525 8,700 Dialyed ,700 Crude microsomal pellet (PI) Undialyed ,700 Dialyed 1i00 4,700 Osmotically shocked P2 subfractions Mitochondria-myelin pellet Undialyed without Triton X ,870 with Triton X ,700 Dialyed without Triton X ,870 with Triton X ,700 Crude synaptic membrane pellet Undialyed without Triton X ,000 with Triton X ,000 Dialyed without Triton X ,000 with Triton X , 000 Tissues were prepared and subjected to differential centrifugation. The various pellets were suspended in 50 mm sodiumpotassium phosphate buffer (ph 7.4) and assayed as described with and without. 0.01% Triton X-100 before and after dialysis against 50 mm phosphate buffer for 18 hr at 4. Data represent the mean of three experiments whose results varied less than 20%. The crude microsomal pellet (P3) was obtained by centrifugation at 100,000 X g for 1 hr. Compounds were purchased as follows: Strychnine sulfate was obtained from Smith, Kline & French Labs., Philadelphia, Pa.; glycine, L-cystathioni-ne, and e-aminocaproic acid from Sigma Chemical Corp., St. Louis, Mo.; B-alanine, L-a-alanine, L-serine, y-aminobutyric acid from Calbiochem, Lia Jolla, Calif.; taurine from K&K Labs., Inc., Plainview, N.Y.; and amninomethane sulfonic acid from Aldrich Chemicals, Milwaukee, Wis. Protein Was Determined by the method of Lowry et al. (7), with bovine-serum albumin as a standard. RESULTS Subcellulkr Localiation of Strychnine Binding in Rat Spinal Cord. For determination of relative [3H]strychninebinding activity in various subcellular fractions, spinal-cord homogenates were subjected to differential centrifugation (Table 1). Both dialyed and undialyed samples were 0 0~ ~ ~ ~ PROTEIN, mg FIG. 1. Effect of concentration of synaptic-membrane protein on specific ['H]strychnine binding. Various amounts of synapticmembrane protein obtained from rat brain stem and spinal cord were incubated for 10 min at 25' with 4 nm [3H]strychnine (38,000 cpm) alone or in the presence of 1 mm glycine in a total volume of 2 ml. Each point is the mean of triplicate determinations. examined with and without Triton X-100 treatment. Dialysis enhanced strychnine binding to the whole homogenate and the crude mitochondrial (P2) fraction by 100%, but produced little change in binding to other subcellular fractions. The enhanced recovery in these fractions may reflect a dilution of endogenous binding inhibitors. Glycine, which potently inhibits strychnine binding, is localied in cytoplasm occluded within subcellular particles (1, 6). Endogenous glycine probably suffices to explain the inhibition of strychnine binding in undialyed whole homogenate and P2 fraction. Dialysis would remove endogenous glycine and thereby enhance strychnine binding. Likewise, more total activity was recovered from the combined synaptic membrane and mitochondria-myelin fractions, which lack occluded cytoplasm, than from the P2 pellet from which they were prepared. For dialyed samples with Triton X-100, the total specific binding of the P2 subfractions exceeded that of the P2 fraction by 27%, while with undialyed samples, the total [3H]strychnine binding of the P2 subfractions was triple that of the P2 pellet. Treatment of synaptic membranes with 0.01% Triton X-100 enhanced their strychnine-binding capacity by about 25%, while similar treatment did not affect other subcellular fractions. If the Triton X-100 concentration is increased to 0.05% and 0.1% then specific [3H]strychnine binding is decreased by 20 and 45%, respectively. More than half the total strychnine binding of homogenates was recovered in the crude mitochondrial fraction, which is enriched in synaptosomes and mitochondria. When the crude mitochondrial pellet was subjected to hypotonic shock, more than 4/5 of its binding activity was recovered in the crude synaptic-membrane fraction ("synaptic membranes") whose specific activity of strychnine binding was about 4- times that of the fraction containing large mitochondria and myelin fragments ("mitochondria-myelin"). Specific strychnine binding of synaptic membranes was enriched 8- and 3-times over the undialyed and dialyed whole homogenate, respectively. In typical assays with synaptic membranes, [3H]strychnine binding is reduced 70% by 1 mm nonradioactive glycine, whereas in undialyed whole homogenates there is only a 15% reduction. In subsequent experiments undialyed synaptic membranes treated with Triton X-100 were used for strychnine-binding studies. Glycine Receptor 2833 ~~~2.0

3 2834 Biochemistry: Young and Snyder Proc. Nat. Acad. Sci. USA 70 (1973) Z C, TABLE 2. Aminoacid competition for 8pecific [3H]strychnine binding m -'o a x - ( 0. CLw 0. C,, 0O IO lo-? Io-$ - 10o-4 STRYCHNINE, M 51 0-I 0' _ C/) m 8 m m M I,): Io I Displacement of ['H]strychnine binding by nonradio- FIG. 2. active strychnine. Synaptic-membrane suspensions (1.0 mg of protein per tube) were incubated with 4 nm ['H]strychnine (38,000 cpm) and increasing amounts of nonradioactive strychnine at 250 for 10 min. Nonspecific binding obtained in the presence of 0.1 mm strychnine has been subtracted from all experimental points. Values are the means of triplicate determinations. The experiment has been replicated three times. Specific binding of [3H]strychnine by synaptic membranes was linear between 0.2 and 1.25 mg of spinal-cord protein (Fig. 1). Binding studies were routinely performed in this linear range. Saturability of (3H]Strychnine Binding to Synaptic Membranes of Spinal Cord. Specific [3H]strychnine binding was saturable with increasing concentration, with half-maximal binding occurring at 0.03 um. By contrast, nonspecific binding, reflected by binding of [3H]strychnine in the presence of 1 mm glycine or 0.1 mm nonradioactive strychnine, was not saturable and increased linearly with increasing [3H]strychnine.- [3H]Strychnine binding was displaced by nonradioactive strychnine with half-maximal displacement at MAM nonradioactive strychnine and maximal displacement at 100 /A'M (Fig. 2) at 250. At 40, half-maximal strychnine displacement was observed at 0.01 um. The fact that halfmaximal saturation occurs at the same concentrations of l-* 10-? lb s 10- GLYCINE, M FIG. 3. Displacement of [3H]strychnine binding by nonradioactive glycine. Synaptic-membrane suspensions (1.0 mg of protein per tube) were incubated with 4 nm [3H]strychnine (38,000 cpm) and increasing amounts of glycine at 250 for 10 min. Nonspecific binding obtained in the presence of 10 mm glycine has been subtracted from all experimental points. Values are the means of triplicate determinations. The experiment has been replicated twice. % Displace- Strychnine ment of antagonied [3H]strychnine depression bound of spinal dorsal horn interneurons Compound M M M (2)* ct-amino acids Glycine t Lcd-Alanine i-serine L-Cystathionine - 0 (1)+ Aminomethane sulfonic acid (3-Amino acids fl-alanine DL-g-Aminoisobutyric acid Taurine y-amino acids y-aminobutyric acid y-amino-fl-hydroxybutyric acid Higher w-amino acids 8-Aminovaleric acid c-aminocaproic acid 20 0 Others Proline Values are means of data from three separate experiments whose results varied less than 15%. * Data from Curtis et al. (2). t Depressant activity is expressed as 1 to 4+ relative to glycine inhibition. Enclosure of a number in brackets reduces its value. (-) indicates not tested. nonradioactive and radiolabeled strychnine indicates that [3H]strychnine is biologically equivalent to the nonradioactive drug in terms of receptor binding and confirms the validity of the determined specific activity of [3H]strychnine. About the same maximal displacement of [3H]strychnine binding was obtained, with nonradioactive glycine as with nonradioactive strychnine (Fig. 3). Half-maximal displacement was observed with 10 MAM glycine, while maximal displacement required 1 mm glycine. Aminoacid Competition for [3H]Strychnine Binding. Curtis et at. (2) compared the ability of various amino acids to mimic the action of glycine in eliciting strychnine-antagonied depression of spinal dorsal horn interneurons. The potency of these compounds in displacing [3H]strychnine binding closely paralleled their ability to mimic the neurophysiologic actions of glycine (Table 2). Glycine and fl-alanine, the most potent displacers of [8H]strychnine, are also the most potent amino acids in eliciting strychnine-antagonied depression of interneurons. Glycine and (3-alanine were additive in displacing [3H]strychnine binding. Displacement by a mixture of 0.05 mm glycine and 0.05 mm p-alanine equalled displacement with 0. i mm glycine. i,-a-alanine, DL-#-aminoisobutyric acid, and taurine, which have the second most potent glycine-like neurophysiologic activity, were also the second most active group of amino acids in displacing specific strychnine binding. All

4 Proc. Nat. Acad. Sci. USA 70 (1973) TABLE 3. Regional distribution of specific [3H]strychnine binding in the central nervous system Specific [3H]strychnine Region binding (cpm/mg of protein) Rat brain Spinal cord 2315 ± 110 Medulla oblongata-pons 1712 ± 104 Midbrain 863 i 67 Hypothalamus Thalamus 485 ± 65 Cerebellum < 80 Hippocampus < 80 Corpus striatum < 80 Cerebral cortex < 80 Monkey spinal cord Cervical gray matter 886 ±t 35 Thoracic gray matter 775 i 32 Lumbar gray matter 1042 i 74 Dorsal white columns 0 Lateral white columns 0 Anterolateral white columns 0 Brains and spinal cords from 20 rats were pooled and rapidly dissected by the method of Glowinski and Iversen (10). Spinal cords from two Rhesus monkeys were removed and put in ice-cold 0.32 M sucrose. The meninges were carefully removed and then the dorsal, lateral, and anterolateral white columns were dissected from the gray matter. The gray matter was divided into cervical, thoracic, and lumbar areas. The tissues were prepared as described for spinal cord in Methods; dilutions were such that samples were assayed by the standard technique at 250. The values represented are the means of triplicate determinations from two separate experiments ± the standard error of the mean. The ratio of specific to nonspecific binding was 3.0 in spinal cord, 1.4 in thalamus, and not significantly greater than 1.0 in cerebellum, hippocampus, corpus striatum, and cerebral cortex. spinal cord's binding. No specific strychnine binding was detected in the cerebellum, hippocampus, corpus striatum, or cerebral cortex. This regional localiation of specific strychnine binding closely parallels the distribution of endogenous glycine throughout the rat central nervous system (4). Within the spinal cord endogenous glycine varies, with highest amounts in the cervical and lumbar regions, correlated with the greater number of interneurons in these areas as compared with the thoracic cord, and lower amounts in the white matter of the spinal cord (4). In monkey spinal cord, we observed the most specific strychnine binding in the cervical and lumbar gray matter, with less in the thoracic gray matter and no detectable binding in the dorsal, lateral, and antero- other amino acids examined, including 7-amino acids and higher w-amino acids, had negligible glycine-like neurophysiologic activity and negligible potency in displacing specific strychnine binding. Proline, a glycine analog not tested by Curtis et al. (2), resembled L-a-alanine and L-serine in its potency for displacing [3H]strychnine. Besides the compounds listed in Table 2, we found that threonine, aspartic acid, and valine at 1 mm had no influence on specific strychnine binding. Regional Distribution of Specific Strychnine Binding in the Central Nervous System. Endogenous levels of glycine are highest in the spinal cord, somewhat lower in the brain stem, and decrease as one ascends the neuraxis (4). Similarly, the ability of glycine to mimic natural inhibitory synaptic activity is most pronounced in the spinal cord and brain stem and less in higher centers (2, 8). Accordingly, glycine receptors should be most abundant in the spinal cord, less concentrated in the brain stem, and still less pronounced in higher centers of the brain. We measured [3H]strychnine binding in synaptic membranes from different regions of the central nervous system of rats (Table 3). Specific strychnine binding was greatest in the spinal cord and second highest in the medulla oblongata-pons. In the mid-brain, specific strychnine binding was 35% of the activity in spinal cord, while the hypothalamus and thalamus displayed 29 and 21%, respectively, of the r~~~~~~~~~~~~~~~~~~~~~~~~ Glycine Receptor > 10- t -2.0 C C] 50- \ m Du-~~~ 0E MINUTES I \ E 10o- CO 0 U- w a. CO) 3.0 C1 1.0 Go SECONDS FIG. 4. Rates of association and dissociation of ['HIstrychnine binding to rat spinal-cord synaptic membranes. Upper: Rate of binding at 40 of 4 nm [3H]strychnine to synaptic membranes (1.0 mg of protein per tube). These experiments were terminated by filtration on Whatman glass-fiber circles. Radioactivity bound to the filter paper in the absence of tissue (800 cpm) has been subtracted from all experimental points. The nonspecific binding is the binding obtained in the presence of 1.0 mm glycine. Lower: Semilogarithmic plot of the dissociation of bound [3H]- strychnine as a function of time at 40. Synaptic membranes were incubated for 10 min at 250 with 4 nm ['HIstrychnine (38,000 cpm) in the standard binding assay. After samples were cooled in an ice bath, nonradioactive strychnine (0) or glycine (0) (0.1 mm and 1.0 mm final concentrations, respectively) was added rapidly. Samples were stirred and filtered immediately (time 0) or allowed to incubate for longer times before they were filtered and washed with 5 ml of ice-cold phosphate buffer. Points represent the means of triplicate determinations. lateral white columns (Table 3). '

5 2836 Biochemistry: Young and Snyder Association and Dissociation of Specific [3H]Strychnine Binding in Rat Spinal Cord. At 250, specific [3H]strychnine binding to spinal-cord synaptic membranes was maximal at 1 min. Accordingly, to exlmine the rate of association in detail we measured the time course of binding at 40 (Fig. 4). At this temperature, half-maximal binding occurred at about 36 sec and binding reached a plateau by 5 min. This information permits calculation (9) of the bimolecular rate constant of strychnine-receptor association, ki, which is 0.6 X 107 M-1 sec'. By contrast, nonspecific strychnine binding was not time dependent and was only about 1/5 of the total amount of specific strychnine binding. The rate of dissociation could not be measured at 250 because the strychnine-receptor complex was already fully dissociated at 15 sec. At 40 we examined the rate of dissociation using either 0.1 mm strychnine or 1.0 mm glycine to displace [3H]strychnine (Fig. 4). When plotted semilogarithmically, the half-life of the strychnine-receptorcomplexwas about 30 sec. The rate constant for dissociation at 40, k2, was 2.3 X 10-2 sec-. The dissociation constant (k2/k1) was 4nM. Ionic Effects and Thermal Stability. Specific [3H]strychnine binding was unaffected by several ionic manipulations. Incubations in which sodium was completely replaced by Tris failed to alter binding. Addition of calcium, magnesium, or EDTA at 1 and 5 mm concentrations did not affect binding. There was a broad ph optimum for strychnine binding between 6.5 and 7.5. We examined the thermal stability of specific strychnine binding by preincubating spinal-cord synaptic membranes for 10 min at various temperatures and then conducting the standard binding assay at 25. Binding was not affected by preincubation at temperatures from 20 to 550, but was completely abolished by such treatment at 70. DISCUSSION Binding of [3H]strychnine to synaptic membranes of the spinal cord appears to represent an interaction with the postsynaptic receptor for the neurotransmitter actions of glycine. Displacement of specific [3H]strychnine binding by amino acids closely parallels their glycine-like depressant actions on spinal neurons. The distribution of specific [3H]- strychnine binding within the central nervous system correlates well with endogenous glycine, high-affinity synaptosomal glycine uptake into unique glycine-accumulating synaptosomes (4-6), and the ability of glycine to mimic natural inhibitory transmission (2). Of the subcellular fractions examined, the highest specific strychnine-binding activity occurs in the synaptic-membrane fraction, which is consistent with localiation of binding sites on postsynaptic glycine receptors. It should, however, be Proc. Nat. Acad. Sci. USA 70 (1973) borne in mind that the synaptic-membrane fraction contains numerous tissue fragments besides synaptic membranes. It is unlikely that strychnine binding involves membranes of nerve terminals mediating high-affinity glycine uptake because strychnine has negligible affinity for this uptake system (5). [3H]Strychnine binding is a saturable process with considerable affinity for strychnine. The bimolecular nature of the association constant suggests that strychnine binds to a single species of receptor. The affinity of glycine for receptor binding is 3 orders of magnitude less than that of strychnine, which is consistent with neurophysiological evidence that strychnine's affinity for the glycine receptor exceeds that of glycine (2). The binding assay used in this investigation is specific, sensitive, and simple. It should provide a valuable tool for assessing the receptor affinity of potential glycine agonists and antagonists. A.B.Y. is a graduate student; fellowship from the Scottish Rite Foundation. 1. Aprison, M. H. & Werman, R. (1965) Life Sci. 4, ; Graham, L. T., Jr., Shank, R. P., Werman, R. & Aprison, M. H. (1967) J. Neurochem. 14, ; Davidoff, R. A., Graham, L. T., Jr., Shank, R. P., Werman, R. & Aprison, M. H. (1967) J. Neurochem. 14, ; Matus, A. I. & Dennison, M. E. (1971) Brain Res. 32, ; Hokfelt, T. & Ljungdahl, A. (1971) Brain Res. 32, Curtis, D. R., Hosli, L. & Johnston, G. A. R. (1968) Exp. Brain Res. 6, 1-18; Kelly, J. S. & Krnjevic, K. (1969) Exp. Brain Res. 9, ; Galindo, A., Krnjevic, K. & Schwart, S. (1967) J. Physiol. (London) 192, ; Werman, R., Davidoff, R. A. & Aprison, M. H. (1968) J. Neurophysiol. 31, Dusser de Barenne, J. G. (1933) Physiol. Rev. 13, ; Bradley, K., Easton, D. M. & Eccles, J. C. (1953) J. Physiol. (London) 122, ; Owen, A. G. W. & Sherrington, C. S. (1911) J. Physiol. (London) 43, ; Curtis, D. R., Duggan, A. W. & Johnston, G. A. R. (1971) Exp. Brain Res. 12, Aprison, M. H., Shank, R. P., Davidoff, R. A. & Werman, R. (1968) Life Sci. 7, ; Aprison, M. H., Shank, R. P. & Davidoff, R. A. (1969) Comp. Biochem. Physiol. 28, Logan, W. J. & Snyder, S. H. (1971) Nature (London) 234, ; Johnston, G. A. R. & Iversen, L. L. (1971) J. Neurochem. 18, ; Logan, W. J. & Snyder, S. H. (1972) Brain Res. 42, Arregui, A., Logan, W. J., Bennett, J. P. & Snyder, S. H. (1972) Proc. Nat. Acad. Sci. USA 69, Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, Krnjevic, K., Randic, M. & Straughan, D. W. (1966) J. Physiol. (London) 184, ; Biscoe, T. J. & Curtis, D. R. (1967) Nature 214, Cuatrecasas, P. (1971) Proc. Nat. Acad. Sci. USA 68, Glowinski, J. & Iversen, L. L. (1966) J. Neurochem. 13,