Ontogenesis of GABA receptor sites in chick embryo cerebellum

Developmental Brain Research, 3 (1982) 263-275

263

Elsevier Biomedical Press

O N T O G E N E S I S OF GABA RECEPTOR SITES IN CHICK EMBRYO CEREBELLUM

SARA FISZER DE PLAZAS Instituto de Biologia Celular, Facultad de Medicina, Universidad de Buenos Aires - - Paraguay 2155, (I 121) Buenos Aires (Argentina)

(Accepted June 12th, 1981) Key words: GABA development of receptors -- ontogeny -- synaptogenesis -- Triton X-100 effects -- GABAmodulin -

-

SUMMARY The time-course of the development of GABA receptor sites in chick embryo cerebellum was correlated with the appearance of synaptic junctions in the cerebellar cortex. At 13 days of incubation, the earliest stage examined, specific [3H]GABA binding was only 19 °/o of that found in cerebella of adult chicks. Between 15 days of incubation and hatching, specific [ZH]GABA binding increased 3-fold, already reaching at birth adult values. During this period the number of synaptic junctions also increased. Scatchard analysis of the binding data obtained at birth revealed two binding sites of Kas 40 and 174 nM and a maximal number of binding sites (n) of about 1.6 and 4.0 pmol/mg protein, respectively. The high-affinity binding site for [ZH]GABA was inhibited by muscimol, GABA, imidazoleacetic acid and bicuculline (IC50: 0.007, 0.020, 0.1 and 10 /~M, respectively). These values correspond to the potencies shown by those compounds in the binding to the synaptic GABA receptor. Treatment of the synaptic membranes with Triton X-100 enhanced [ZH]GABA binding depending on the developmental stage studied, suggesting that GABAmodulin that inhibits the binding also appears during that period.

INTRODUCTION The amino acid neurotransmitter ~'-aminobutyric acid (GABA) plays an inhibitory role in the central nervous system of vertebratesS,%,zs, ',9. Such a transmitter role for GABA received support from the finding that bicuculline, a phtalideisoquinoline alkaloid from the species corydallis, blocked the effect of this amino acid on the crayfish stretch receptor 2". as well as on mammalian central neurons, probably through competitions with GABA receptors 4. 0165-3806/82/0000-0000/$02.75 (c) Elsevier Biomedical Press

264 in biochemical studies, it is important to differentiate the postsynaptic binding of GABA from the uptake or accumulation of this amino acid by neurot~ and nerve endings. While the latter is Na:-dependent :~°, GABA binding to postsynaptic sites is N a independent TM. Thus, the demonstration o f a GABA receptor site could be done by measuring the binding of [3H]GABA in a Na -flee medium in preparations of crude synaptic membranes from brainS,~.~.e2.:~L To analyze the developmental characteristics of GABA binding ~e studied the chick embryo cerebellum during ontogenesis, a system on which much embryological ~7, neuroanatomical~O.ls.za.:~:~ neurophysiologicaP; and neurochemical TM ~; information is available. Previous studies on the ontogenesis of chick embryo cerebellum have shown that GABA is present at early stages of development although the activity of lhe enzymes of the GABA system, glutamic acid decarboxylase (GAD) and GABA-transaminasc (GABA-T) increase much later in development 17. The appearance of [:~H]GABA binding in rat brain also precedes that of G A D :j, suggesting that the GABA receptor may develop independently from other components of the GABA system. The function of the GABA receptor seems to be influenced by a protein which acts as an endogenous inhibitor of GABA binding. This protein, knov, n as GABAmodulin, can be removed by freezing, thawing and treatment with Triton X-100v', :~e, procedures which also enhance the binding affinity for the GABA receptor sites:L The aim of the present study was to determine the time-course of the development of GABA receptor sites in chick embryo cerebellum as a parameter of neuronal differentiation in the central nervous system. We also analyzed the kinetic and pharmacological properties of the interaction of [3H]GABA with synaptic membranes isolated from chick cerebellum. To establish the developmental characteristics of GABAmodul[n and its possible coupling to the GABA receptor, the effects of Triton X-100 on the density of specific [JH]GABA binding sites and on the affinity of the receptors for [aH]GABA was studied during development. MATERIALS AND METHODS

Drugs used 4-Amino-n-[2,3-:~H]butyric acid (57 Ci;'mmol, Radiochemical Centre, Amersham) was diluted directly to 10 ..7 and 10 ~ M or with the additional unlabeled GABA to 10 r' M. Thc purity of the labeled GABA was determined by thin-layer chromatography on cellulose plates and found to be 99',,] or greater, using n-butanol:acetic acid :water (120:30:50). Drugs used were: GA BA, L-2-4-diaminobutyric acid ( DA BA). aminooxiacetic acid and imidazoleacetic acid (Sigma Chemical, St. I.ouis, MO), muscimol and bicuculline methochloride (gifts from Dr. E. Peck, Jr.) nipecotic acid and muscimol (gifts from Prof. P. Krosgaard-Larsen) and ( :. )-bicuculline methiodide and (--)-bicuculline methiodide (gifts from Dr. H. MOhler).

Preparations of synaptic membranes Embryos of White l,eghorn chicks were purchased from a local hatchery. For

265 studies in chicks after hatching, eggs containing ! 8-day-old embryos were placed into a 37 °C high-humidity incubator and rotated twice until hatching, which took place between days 20 and 21 of incubation. After hatching, the chicks were kept at room temperature and were supplied ad libitum with water and chicken feed. At various stages of development, the cerebellum was rapidly removed and dissected on ice-cold petri dishes. For the preparation of synaptic membranes a sufficient number of cerebella was accumulated, weighed and homogenized in distilled water (pH 7.1 ) using a Potter-Elvehjem homogenizer with a Teflon plestle. After centrifugation at 100,000 g for 30 min, the supernatant was discarded and the resultant pellet (P~) was rehomogenized in distilled water and layered on a discontinuous sucrose density gradient. The sucrose gradient was made of 2 ml of 0.6 M sucrose, 1 ml of 1.2 M sucrose and 1 ml of 1.4 M sucrose. After centrifugation in a SW39 rotor at 103,600 g for 90 min, the fraction P1 was separated into various subfractions: P1A between the sample and 0.6 M sucrose layers, P1B between 0.6 M and 1.2 M sucrose, PIC between 1.2 M and 1.4 M sucrose and a pellet as P~D. These subfractions represented the myelin fragments (P~ A), a fraction enriched in synaptic membranes (P1B), a fraction with membranes of various origins but poor in synaptic membranes (P1C) and a crude mitochondrial fraction (P~ D) containing also nuclei and cellular debris. These 4 fractions were diluted with 3 vols. of distilled water and pelleted by centrifugation at 100,000 g for 60 min. The final pellet of synaptic membranes (PIB) was resuspended in 50 mM Tris-HCl buffer pH 7.1, and centrifuged at 1t30,000 g for 60 min. In some cases the membranes were immediately suspended in buffer for assay of GABA binding (fresh tissue), whereas in other cases the membranes were stored at --20 ':C for at least 18 h (frozen tissue). To study Na~-independent [aH]GABA binding the synaptic membranes were prepared according to a method previously described 7. The frozen pellets were resuspended in 50 mM Tris-HCl buffer pH 7.1 and Triton X-100 was added to the suspension to a final concentration of 0.05 ~o (v/v). This suspension was incubated at 37 cC for 30 min, centrifuged at 100,000 g for 30 min and the pellet was resuspended in the same buffer for the GABA binding assay.

Binding studies The binding of [3H]GABA to the synaptic membranes was measured by means of the centrifugation assay of Enna and Snyder a with minor modifications. Aliquots of fresh or frozen synaptic membranes (0.2--0.4 mg of protein) were incubated in triplicate in 0.6 ml 50 mM Tris-HCl buffer, pH 7.1 containing 50 nM [aH]GABA alone or in the presence of 1 mM GABA for 5 min at 0 °C in scintillation biovials. After incubation, the reaction was terminated by centrifugation of the vials in a SM24 rotor of a Sorvall RC-2 Centrifuge at 17,000 g for 10 min at 4 °C. The supernatant fluid was decanted, and the pellets were rinsed superficially with 1 mi of ice-cold buffer. Bound radioactivity was extracted with 0.2 ml Protosol (New England Nuclear), 3 ml of toluene phosphor were added, and radioactivity was assayed by liquid scintillation spectrometry at a counting efficiency of 25 ~o. Specific [3H]GABA binding was determined by subtracting from the total

266 binding the amount bound in the presence of I mM unlabeled GABA Inon-specitic binding).

Triton X-IO0 treatment To examine the effects of Triton X-100 on specific [aH]GABA binding during development, the frozen pellets of synaptic membranes obtained from cerebella of chick embryos at different stages of development were treated as follow~: the frozen membranes were thawed and suspended in 50 mM Tris-HCl buffer pH 7. I containing 0.05,~; Triton X-100 to a final protein concentration of l mg/ml. This suspension was incubated at 37 C for 30 min and then centrifuged at 100,000 g for 30 mln. "[he pellet obtained was similarly washed 3 times by repeated suspension and centrifugation to remove endogenous GABA. The final pellet was rcsuspended in 50 mM Tris-HCI buffer pH 7. I for [aH]GABA binding assay. Protein was determined by lhc method of Lowry et al. using bovine serum albumin as standard "-'°. Preparation Jor electron microscopy Cerebellar cortex obtained from various developmental ages and aliquots of the subfractions were fixed overnight at 0 C with 5'~ glutaraldehydc in 0.3 M phosphate buffer pH 7.4 and then dehydrated in graded ethanol. Following dehydration, the blocks were stained overnight in 2",, phosphotungstic acid in absolute ethanol containing a trace of water (I.-2 drops of 95";i ethanol per 10 ml of absolute ethanoll essentially as described by Bloom and Aghajanian e. After this staining, procedure tissues were embedded in Epon 812 and thin sections were examined under thc electron microscope. R ES U t.TS

Ultrastructural identification of .wnaptic junctions in chick embryo cerebellum at d~ff'erent stages Of development. The density of synaptic junctions observed in the cerebellar cortex stained with phosphotungstic acid varied between I 1 and 20 days of incubation. Bet~een I I and 13 days of incubation few postsynaptic densities were idenfitied (Fig. lal but their frequency markedly increased between days 14 and 20. But at day 20 a great number of synaptic junctions could be observed standing out from a virtually unstained background (Fig. I b). At the earliest stage examined, 13 days of incubation, the neurons and their processes were separated by wide empty spaces while at daF, 21, the overall aspect of the tissue was much more compact. This made it impossible to express the results of the ultrastructural observations in a quantitative manner such as, for example, number of synaptic densities per unit surface. in order to determine which of the 2 membrane fractions, PtB and PtC, obtained by subcellular fractionation was richer in synaptic junctions, these subfi'actions were also stained with phosphotungstic acid. As shown in Fig. 2, subfraction P~B of chick embryo cerebellum at 20 days of incubation, had a great number of synaptic junctions while, at the same age, subfraction P~C had only few synaptic functions. Thus, sub-

267

-y

,(

la

lb

!

~..~ '~

..y i 5.4. ¢

4''.', b

~ti~.

'

"

•.

•

"~.. ! ....

<

h

•

•

..

.~

Fig. 1. Electron micrographs of the cerebellar cortex of the chicken at 13 days (a) and 21 days (b) of development. The tissue was stained with ethanolic phosphotungstic acid for the demonstration of synaptic densities. -: 44,000. Fig. 2. Electron micrograph of a subcellular fraction enriched in synaptic membranes (Pj B), isolated from the cerebellar cortex of chick embryos at 21 days of development and stained with ethanolic phosphotungstic acid for the demonstration of synaptic densities. ;~: 18,000.

fraction PxB constitutes a convenient material for the attempted correlation between the development of GABA receptors and synapse formations as shown by th¢ morphological criteria,

268

o,

&

.~ ..... ;; f

-.

,

+'

o2 t..

I i

;

/

i i

/i +

-

•

i

~.~--~?+-;7 m mRTH 24 (.,)'

;Ige

c !L+ttJ,d

'~'

A~"~t ?....

(da,,,f-

Fig. 3. Development of GA BA receptor sites in chick embryo cerebellum expressed as a lunction of the concentrations of protein. For binding experiments, fresh synaptic membranes (Pt B) were incubated at 0 C for 5 min with 50 nM [aH]GABA alone or in the presence of unlabeled GABA (I mMI, to displace specific [aH]GABA binding. Values given represent means of 5 experiments, each performed in triplicate.

GA BA receptor sites hi developing chick embryo cerebellum: changes o[specffic and nonspecific [3H]GA BA binding durhlg development Fig. 3 shows the p a t t e r n o f increase o f G A BA receptors in the chick e m b r y o cerebellum during development. At the earliest time examined, 13 d a y s o f incubation, the specific binding o f [ 3 H ] G A B A g r a d u a l l y increased between 13 and 15 days o f incubation reaching 30% o f a d u l t levels. Between 15 days o f i n c u b a t i o n a n d birth, the specific binding o f [ 3 H ] G A B A increased 3-fold. At birth, specific binding o f [ZH]GABA reached adult values o f a b o u t 0.4 p m o l / m g protein, estimated at a [ ~ H ] G A B A c o n c e n t r a t i o n o f 50 nM. The changes in specific a n d non-specific [ 3 H ] G A B A binding in fresh synaptic m e m b r a n e s o f chick e m b r y o cerebellum at different ages is shown in Fig. 4. At 13 days

SPECIFIC

.,

+j!-

. ~

1"+

....

s

L

7

J.

, , , ' +,;~tll,+l

,~q+.

.

.

:,,+

.

.

~ Pelt+

+,:~Dut"r

,

Fig. 4. Analysis of specific and non-specific [~H]GABA binding during development. l h e changes in specific and non-specific ['~H]GABA binding were determined in fresh synaptic membranes of chick cerebellum at various ages. Results are expressed in terms of percentage of the total binding. Each bar represents the average of 4-6 experiments, each performed in triplicate.

269

/

X

80k

~--

HOI~OGENATE~/eI

/

/

,J • -

•

i

.'~, "C- 40L. ~. ~-

2O-

2N ~

L

/

113

115

1'7

(~rv'~Coptual

'i'9

-

I

BIRTH

~]cle (O~y£)

Fig. 5. Protein content of whole homogenates of chick embryo cerebellum and of PIB fraction as function of age. Each point represents the average of 4-6 experiments, each performed in triplicate. of incubation, the specific binding of [3H]GABA represented only 6.6 % of the total binding, being 93.4 °/ non-specific. While specific [3H]GABA binding increased between 15 days of incubation and birth to reach a maximum of 42 %, non-specific binding of [aH]GABA during the same period decreased to 58%. N o specific ['~H]GABA binding was found in the myelin fraction isolated from chick embryo cerebellum at any time during development.

Developmental pattern of total protein and synaptic membrane protein in chick embryo cerebellum Fig. 5 shows that the protein content of the cerebellum, expressed per fresh

*)kk

['.] ~,o, B......... ~

CC © ~ CI3

H,g,

. . ~

3-

400 1140

Lo.

16 40

k -.

~o

~Y

.....

~o

4o

Bound / Free fmol/pmol

Fig. 6. Scatchard analysis of specific ['~HIGABA binding to synaptic membranes of chick cerebellum at birth. Synaptic membranes were incubated with varying concentrations of [aH]GABA up to 800 nM as described in Methods. Non-specific binding was determined in the presence of l mM G A B A and specific binding was obtained after substracting the non-specific binding from total binding. The

data of specifically bound [aH]GABA were plotted according to Scatchard. Values given represent means of 6 experiments, each performed in triplicate.

270 weight and as percentage of adult values, was approximately 52",, a~ 13 days ¢,f incubation and increased linearly reaching 90,~, of adult values at birth. On the other hand, the curve for synaptic membrane protein (PIB) only began to rise ~lt 17 days c~l" incubation, concomitant with the increase in both the number of synaptic ju~ction~ and GABA binding sites. Saturabilit)' o/" spe('(h'(" [3H]GABA b&ding sites ht .s'.)'naptic membranes Of ctt/cA cc'rebellum at birth Saturation studies with [:~HJGABA were performed on the frozen P~B lYaction isolated from chicks cerebella at birth using concentrations of thc ligand ranging flora 5 nM to 800 nM. Scatchard analysis of the specific binding data revealed a high- and low-affinity components of GABA receptor binding with. apparent dissociation constants (Ka) of 40 nM and 174 nM and density (n) of 1.6 and 4.0 pmol 'mg protein respectively (Fig. 6). Hill plot analysis of the same data revealed regression coefficients of 0.89 and 0.90 which shows a lack of cooperativity of the binding sites. Pharmacological characterization q/'[aH]GA BA binding to .s-vnaptic membram's q/~'hi~'k cerebellum at birth Pharmacological analysis of the specific [:~H]GABA binding to frozen membranes (Table I) showed that non-radioactive GABA, the GABA agonists muscimol and imidazoleacetic acid, and the GABA antagonist bicucullinc, potently inhibited [:~H]GABA binding with ICs~ values of 0.020, 0.007, 0.1 and 10 I~M, respectively. Muscimol was thc most potent among the compounds examined, being approximately 3 times more potent than GABA itself. ( : )-Bicuculline methiodide, a GABA antagonist which is a more potent convulsant than (--)-bicuculline methiodide ~:~was more effective in displacing [3H]GABA binding, showing the stereospeciticity of the

TABLE I Pharmacoh~gic speci[7cio' of .:~H,'GABA bhzding to .wnaptic membranes of chick cerehelhtm at bi~th Inhibition of specific [:~HIGABAbinding was measured as described in Methods using frozen synaptic membranes and [3H]GABA to a final concentration of 4 nM. lC~0 values correspond to the concentration of compound which inhibits 50% of specific [aH]GABA binding and calculated by linear regression on log-probit graphs. Results are the means of 2-4 determinations using 5-7 concentrations of each drug in triplicate.

(i A BA Muscimol lmidazole acetic acid ( • )-Bicucullinemethiodide (---)-Bicucullinemethiodide L-2,4-diaminobutyric acid Nipecoticacid Aminooxyacetic acid fl-alanine .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

0.020 0.007 0.1 10 500 . 1000 -1000 100() 1000 .

.

.

.

.

.

.

.

.

.

.

.

271 3C

o

~-

Q

: - - - 0

©

5

-AIte, TRITON X 100

I

,o

I /

O~

13

15

17

19

BIRTH

24

//

ADULTA . .

Conceptual age (days)

Fig. 7. Triton x-100 effects on specific [3H]GABA binding to synaptic membranes of chick cerebellum during development. Frozen synaptic membranes were treated with 0.05% Triton ×-100 as described in Methods and specific [aH]GABA binding was determined at 50 nM [3H]GABAusing the standard technique. O, Freshly prepared membranes; (,, frozen Triton ×-100 treated membranes. Each point represents the average of 4-6 experiments, each performed in triplicate. binding. On the other hand, L-2-4-diaminobutyric acid and nipecotic acid, inhibitors of neuronal GABA uptake, and fl-alanine, which inhibits GABA uptake by glial cells 14 had little affinity for the [3H]GABA binding site.

Effects of 0.05",(, Triton X-IO0 on specific [3H]GA BA binding during development Fig. 7 shows the action of 0.05 ?~oTriton X-100 on specific [:~H]GABA binding to frozen synaptic membranes of chick embryo cerebellum at different stages of development. While the effect of Triton X-100 was very low at the earliest stages, there was a marked increase in the amount of specific [3H]GABA binding to cerebella of embryos between 17 days of incubation and birth. During this period the specific binding of [3H]GABA to Triton-treated membranes increased 6-fold, and this value remained constant up to the adult stage. This increase in specific [3H]GABA binding is in agreement with previous findings made using adult rat brain membranes 9. The analysis of the changes in specific and non-specific [3H]GABA binding in Triton X-100 treated membranes showed that at 13 days of incubation the specific [3H]GABA binding represented 250,~ of the total binding, an increase due to the reduction in non-specific [3H]GABA binding since total binding had the same value as that found in fresh membranes. On the contrary, the increase of 6-fold in specific [:~H]GABA binding observed between 17 days of incubation and birth was due to an increase in the total binding capacity and represented 70 0~ of the total binding. At the earliest stages, when the action of Triton X-100 was not marked, the saturation studies revealed only the site of low-affinity with a Ka of about 190 nM (data not shown) but in this case the density (n) of GABA binding sites was lower (n :-: 1.3 pmol/mg protein). This results suggest that at early stages of development Triton X-100 treatment dissolves structural proteins not related to GABA receptor binding sites while

272 between 17 days of incubation and birth produces the liberation of an endogenous protein inhibitor involved in the high-affinity binding sites and which probabl3 appears during this period of development. These concepts are supported by the fact that at early stages of development, only the site of low affinity (K,~ 190 n M) was present. DISCUSSION

The variations in specific [aH]GABA binding to synaptic membranes isolated from the cerebellum during development, should be interpreted on the basis of the maturational changes occurring in this structure. It has been previously shown that in developing chick cerebellum there are morphological signs of synapse formation already at 11 days of incubation. Although at this early stage of development thc number of synapses is very small, it increases between 14 and 19 days of incubation ~7. In the present study, the phosphotungstic acid method revealed few synaptic junctions in the chick cerebellum at 13 days of incubation but between days 15 and 20, there was a rapid increase in the number of junctional structures. Although these observations could not be quantitated, marked changes in the frequency of synaptic densities were evident during development. This agrees with the observations of K uriyama et al.~7 who similarly studied the development of synapses in chick embryo cerebellum but with conventional electron microscopic techniques. It should be noted that this pattern of appearance of synaptic junctions correlated with the development of bioetectrical activity in chick embryo cerebellum. Approximately at day 15 of incubation, spontaneous discharges of small amplitude were detected in the chick cerebellum for the first time and these increased with age '7. To study the development of GABA receptor sites, a subcellular fraction ~ich in synaptic membranes (PxB) was isolated from chick embryos' cerebella. At the earliest developmental stage examined, 13 days of incubation, specific [ZH]GABA binding only represented 19'!~i of the adult value. At this time, only few synaptic junctions could be observed. Between 15 and 20 days of incubation, there was a fast rise in the GABA binding capacity which increased 3-fold. This coincides with the proliferation of the granulle cells 17, on which approximately 70 !~,',iof cerebellar GABA receptors arc located 3~. The timing and rate of development of the GABA binding sites found in these experiments is similar to that previously demonstrated in chick whole brain I~. Thc study of the developmental pattern of total protein and of synaptic membrane protein showed that while total protein increased linearly from 13 days of incubation to birth, the protein content of the synaptic membrane fraction started to rise at 17 days of incubation, a time when there was an increase of the number of synaptic junctions and of GABA binding sites. These data indicated that there is a temporal correlation between the increase in specific [3H]GABA binding, in synaptic membrane protein and in the appearance of synaptic junctions, suggesting that these events may be related to one another. Specific [aH]GABA binding to synaptic membranes obtained from chick cerebellum at birth was saturable and Scatchard analysis of the data indicated the

273 presence of two binding sites: one with high-affinity (Ka :-~ 40 nM, n --- 1.6 pmol/mg protein) and other with low-affinity (Ka =- 174 riM, n = 4.0 pmol/mg protein). These are nearly identical to those shown in CIOa- treated membranes of adult rat cerebellum"~. Although the present is the first report on [ZH]GABA binding to a purified fraction containing synaptic membranes isolated from chick embryo cerebellum, the data agree well with that previously obtained with crude membrane fractions isolated from adult rat brain, suggesting that the specific [ZH]GABA binding observed in chick cerebellum represents the true binding to GABA receptor sites. Thus, Enna and Snyder9 reported that after treatment with 0.05 °/jo Triton X-100 specific binding of [3H]GABA was resolved into 2 components having Ka values of 16 and 130 nM and n values of 0.6 and 5 pmol/mg protein for the high- and low-affinity sites, respectively. Similarly, treatment with 0.03 °',i Triton X-100 revealed 2 binding sites with K,l values of 21.7 and 170 nM and n values of 1.9 and 5.8 pmol/mg protein, respectively 1. The specificity shown by the high-affinity binding site with respect to the ligand used, agrees with the relative potencies of the GABA-agonists and GABA-antagonists studied in neurophysiological experiments, indicating that specific [ZH]GABA binding reflects an interaction with the synaptic GABA rcceptor. The IC~0s of 7 nM for muscimol, 20 nM for GABA and 10 uM for bicuculline to inhibit the specific [3H]GABA binding coincides with data previously published 9,~'.. Likewise, the specific [:~H]GABA binding was not affected by inhibitors of GABA uptake such as nipecotic acid, t-2-4-diaminobutyric acid, indicating that the specific [3H]GABA binding determined behaves similarly to the postsynaptic GABA receptors previously characterized in adult rat braing,'-'k It has been recently reported that an endogenous protein inhibitor, GABAmodulin, isolated from flesh synaptic membrane preparations by freezing, thawing and treatment with Triton X-100, reduces non-competitively the aflSnity of GABA for its Na'-independent recognition site ~').. Therefore, the analysis of GABA binding in membranes isolated from developing cerebella and treated with Triton X-100 could give some clue as to the ontogenesis of GABAmodulin and its association with the GA BA receptor. While the effect of Triton X-100 was not important at early stages, i.e. before 17 days of incubation, there was a marked increase in the amount of specific ['~H]GABA binding between 17 days of incubation and birth, when it reached adult values. During this period the specific binding of [ZH]GABA in Triton treated synaptic membranes increased 6-fold. This increase in specific ['~HJGABA binding seems to be associated with the washing out, by Triton X-100, of the inhibitor of GABA binding, suggesting that GABAmodulin also appears during this period of development. A similar pattern of development for the Triton X-100 effect was found in rat cerebellum at different stages of development using the GABA agonist [3H]muscimol for measuring GABA binding sites '~. Moreover, the lack of effect of Triton X-100 at early stages of development of chick cerebellum indicates not only the lack of GABAmodulin but also the absence of the high-affinity binding sites. This is supported by the presence of only one type of binding sites, those of low-affinity with a Ka of about 190 nM when saturation studies are done in frozen cerebellar membranes obtained at 13 days of incubation (data not shown). This evidence taken together suggest that

274 between 17 days o f i n c u b a t i o n a n d birth, the synthesis o f the protein in~ oivcd in the high-affinity binding sites takes place, a process which seems to be associated with the a p p e a r a n c e o f the e n d o g e n o u s protein inhibitor acting on those sites. in conclusion, the present results indicate that the rate o f a p p e a r a n c e o f G A B A binding sites is t e m p o r a l l y correlated with the f o r m a t i o n o f synaptic jttnction~, with the a p p e a r a n c e o f the protein involved in the high-affinity binding sites a , ~ell as with that o f G A B A m o d u l i n . Studies on the ontogencsis o f the G A B A receptor sites, as well as ot the tactors controlling its functions, m a y help to clarify the nature o f the synaptic tncchanisms in which G A B A is involved. ACKNOWLEDGEMENTS The a u t h o r wish to t h a n k the following tor their generous gifts o f d r u g s : Prof. I'. K r o s g a a r d - L a r s e n , Royal D a n i s h School o f P h a r m a c y , (nipecotic acid and muscimol); Dr. E. J. Peck, Jr., Baylor College o f Medicine, (muscimol and bicuculline methochloride) and Dr. H. Mi~hler, H o f f m a n - k a Roche. (( ~ ) bicucullinc m e t h i o d i d e and (---)-bicuculline methiodide). i am also grateful to Mrs. A l b a M i t r i d a t e de N o v a r a tbr her excellent technical assistance and Mrs. Alicia F. de C a n d a m e for typing the manuscript. This research was s u p p o r t e d by grants t¥om the Consejo N a c i o n a l de Investigaciones Cientificas y T~cnicas, A r g e n t i n a ( k e g a j o No. 8819;;79).

REFERENCES 1 Aldino, C., Balzano, M. A. and Toffano, G., Ontogenic development of GABA recognition sites in different brain areas, Pharmacol. Res. Commun., 12 (1980) 495-500. 2 Bloom, F. E. and Aghajanian, G. K., Cytochemistry of synapses: a selective staining method l'c~r electron microscopy, Science, 154 (1966) 1575-1577. 3 Coyle, J. T. and Enna, S. H., Neurochemical aspects on the ontogenesis of gabanergic neurons in the rat brain, Brain Res., 111 (1976) 119-113. 4 Curtis, D. R., Duggan, A. W., Felix, D. and Johnston, G. A. R., Gaba, bicucullmc and central inhibition, Nature (Lond.), 226 (1970) 1222-1224. 5 Curtis, D. R. and Watkins, J. C., The pharmacology of amino acids related to ;,-aminobutyric acid. Pharmacol. Rev., 17 (1965) 347--39I. 6 Eccles, J. C., Functional meaning of the patterns of synaptic connections in the cerelx~llum. Perspect. biol. Med., 8 (1965) 289 310. 7 Enna, S. J., Collins, J. F. and Snyder, S. H., Stereospecificity and structure-activity requirements of GABA receptor binding in rat brain, Brabl Res., 124 (1977) 185-.190. 8 Enna, S. J. and Snyder, S. H., Properties of ;.,-aminobutyric acid (GABA) receptor binding in rat brain synaptic membrane fractions, Brain Res., 100 (1975) 81.-97. 9 Enna, S. J. and Snyder, S. H., Influences of ions, enzymes, and detergents on ~,-aminobutyric acid receptor binding in synaptic membranes of rat brain, Molec. Pharmacol., 13 (1977j 442-453. 10 Enna, S. J., Yamamura, H. I. and Snyder, S. H., Development of muscarinic cholinergic and GABA receptor binding in chick embryo brain, Brain Res., 101 (1976) 177-183. I 1 Gray, E. G., The granule cells, mossy fibers and Purkinje spine synapses of the cerebellum: light and microscope observations, J. Anat. (Lond.), 95 (1961) 345-356. 12 Guidotti, A., Toffano, G. and Costa, E., An endogenous protein modulates the attinity of GABA and bcnzodiazepin¢ rcceptor:i in rat ~ain, Nature (Lond.), 275 (1978) 553-555,

275 13 Hirsch, H. E. and Robbins, E., Distribution of 7-aminobutyric acid in the layers of the cerebral and cerebellar cortex. Implications for its physiological role, J. Neurochem., 9 (,1962) 63-70. 14 Johnston, G. A. R., Physiologic pharmacology of GABA and its antagonists in the vertebrate nervous system. In E. Roberts, T. N. Chase and D. B. Tower (Eds.), GABA in Nervous System Fanction, Raven Press, New York, 1976, pp. 395-41 I. 15 Johnston, G. A. R., Neuropharmacology of amino acid inhibitory transmitters, Ann. Rev. Pharmacol. Toxicol., 18 (1978) 269-289. 16 Kuriyama, K., Haber, B., Sisken, B. and Roberts, E., The ~,-aminobutyric acid system in rabbit cerebellum, Proc. nat. Acad. Sci. (U.S.A.), 55 ('1966) 846-852. 17 Kuriyama, K., Sisken, B., ito, J., Simonsen, D. G., Haber, B. and Roberts, E., The ),-aminobutyric acid system in the developing chick embryo cerebellum, Brain Res., 11 (1968) 412 430. 18 Larramendi, L. M. H., Fichkenscher, L. and Lcmkey-Johnston, N., Synaptic vesicles of inhibitory and excitatory terminals in the cerebellum, Science, 156 (1967) 968-969. 19 Lester, B. R. and Peck, E. J., Jr., Kinetic and pharmacological characterization of gammaaminobutyric acid receptive sites from mammalian brain, Brain Res., 161 (1979) 79-97. 20 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with Folin phenol reagent, J. biol. Chem., 193 (,1951) 265-275. 21 M~.~hler, H., GABA receptor binding with [ZH]GABA and [ZH]- ( ! ) -bicuculline-methiodide: an improved method. In P. Mandel and F. V. DeFeudis (Eds.), GABA-Biochemistry and C N S Functh~ns, Advances in L-werimental Medicine and Biolog),, Vol. 123, Plenum Press, New York, 1979, pp. 355-362. 22 McLennan, H., Bicuculline and inhibition of crayfish stretch receptor neurones, Nature (Lond.), 228 (1970) 674-675. 23 Mugnaini, E. and Forstronen, P. F., Ultrastructural studies on the cerebellar histogenesis. 1. Differentiation of granule cells and development of glumeruli in the chick embryo, Z. Zellfi~rsch., 77(1967) 115 143. 24 Olsen, R. W., Ticku, M. K., VanNess, P. C. and Greenle, D., Effects of drugs on gammaaminobutyric acid receptors, uptake, release and synthesis in vitro, Brain Res., 139 (1978) 277-294. 25 Oja, S. S., Kontro, P. and L/ihdesm/iki, P., Amino acids as inhibitory neurotransmitters, Progr. Pharmacol., 1 (1977) 1-19. 26 Palacios, J. M., Niehoff, D. L. and Kuhar, M. J., Ontogeny of GABA and benzodiazepine receptors: effects of Triton X-100, bromide and muscimol, Brahz Res., 179 (1979) 390 395. 27 Peters, J. J., Vonderahe, A. R. and Huesman, A. A., Chronological development of electrical activity in the optic lobes, cerebellum, and cerebrum of the chick embryo, Physiol. Zool., 33 (1960) 225-231. 28 Roberts, E., GABA and nervous system function - - a perspective, Bh~chem. Pharmacol., 23 (1974) 2637-2649. 29 Roberts, I_. and Eidclbcrg, E., Metabolic and neurophysiological roles of ;,-aminobutyric acid, Int. Rev. Neurobiol., 2 (1960) 279-332. 30 Sano, K. and Roberts, E., Binding of ;'-aminobutyriuc acid to mouse brain preparations, Bh~chem. Pharmacol., 12 (1963) 489 502. 31 Simantov, R., Oster-Granite, M. L., Herndon, R. M. and Snydcr, S. H., '7-aminobutyric acid (GABA) receptor binding selectively depleted by viral induced granule cell loss in hamster cerebellum, Brain Res., 105 (1976) 365 371. 32 Toffano, G., Guidotti, A. and Costa, E., Purification of an endogenous protein inhibitor of the high affinity binding of 7-aminobutyric acid to synaptic membranes of rat brain, Proc. nat. Acad. Sci., ( U.S.A.j, 75 (1978) 4024-4028. 33 Uchizon,,), K., Characteristics of excitatory and inhibitory synapses in central nervous system of the cat, Nature (LontL), 207 (1965) 642-643. 34 Zukin, S. R., Young, A. B. and Snyder, S. H., Gamma-aminobutyric acid binding to receptor sites in the rat central nervous system, Proc. nat. Acad. Sci. (U.S.A.), 71 (1974) 4802-4807.