Selective Stimulation of Neurotransmitter Release from Chick Retina by Kainic and Glutamic Acids

Journal of Neurochemistry Raven Press, New York @ 1982 International Society for Neurochemistry

0022-304218Z1001- 1169/$02.75/0

Selective Stimulation of Neurotransmitter Release from Chick Retina by Kainic and Glutamic Acids

Ricardo Tapia and Clorinda Arias

Departamento de Neurociencias, Centro de Investigaciones en Fisiologia Celular, Universidad Nacional Autonoma de MPxico, Mhxico D.F., Mhxico

Abstract: The excitatory action of kainic and glutamic acids in chick whole retina was demonstrated as an immediate stimulation of the release of labeled y-aminobutyric acid (GABA) and glycine in a superfusion system. This stimulatory effect was 3 ~ 10 times greater than that produced by a depolarizing K+ concentration; in addition, it was independent of Caz+ in the medium, but notably inhibited when Na+ was omitted from the medium. Under identical experimental conditions, neither kainic nor glutamic acid had any effect on the release of labeled dopamine or a-aminoisobutyric acid, thus indicating that their effect is not unspecific or due to cell damage. Similar although less marked stimulation of labeled GABA and glycine release by kainic acid was obtained in subcellular retinal fractions, particularly in fraction PI, which contained photoreceptor terminals and outer segments. This stimulation was also Ca2+ independent and greatly reduced when Na+ was omitted from the medium. It is suggested that the stimulation of GABA release by kainic and glutamic acids is probably due to a Na+-dependent, carrier-mediated mechanism that responds to the entry of Na+ produced by the interaction of glutamic and kainic acids with retinal membranes. In cortical or striatal slices from mouse brain, these acids had a negligible stimulatory effect on GABA and dopamine release. Key Words: Kainic acid-Glutamic acid-y Aminobutyric acid release-Glycine release-Retina. Tapia R. and Arias C. Selective stimulation of neurotransmitter release from chick retina by kainic and glutamic acids. J . Neurochern. 39, 1169- 1178 (1982).

It is known that glutamate (Glu) is a strong exci- tant of neuronal activity in several regions of the mammalian nervous system (Krjevic, 1974; Cur- tis, 1975). The systemic administration of this amino acid produces neurotoxic effects such as convul- sions and neuronal necrosis in some areas of the CNS in young animals (Olney, 1979). Among several compounds structurally related to Glu and also having neuroexcitatory and neurotoxic properties, kainic acid (KA) seems to be the most powerful (Shinozaki and Konishi, 1970; Johnston et al., 1974; Olney et al., 1974; Schwarcz et al., 1978). It has

Received December 1, 1981; accepted April 21, 1982. Address correspondence and reprint requests to Ricardo

Tapia, Departamento de Neurociencias, Centro d e Inves- tigaciones en Fisiologia Celular, Universidad Nacional Au- tbnoma de MCxico, Apartado Postal 70-600, 04510-Mexico, D.F.. Mexico.

been proposed that the mechanism of the neuroexcitatory and neurotoxic actions of KA involves a depolarizing effect through its interaction with a discrete population of receptors, although the majority of this population seems to be different from the glutamate receptors (Simon et al., 1976; Hall et al., 1978; Foster and Roberts, 1978; McLen- nan and Lodge, 1979; Love11 and Jones, 1980). Neurotoxic effects of KA have been observed in several areas of the CNS, including the striatum and the retina (McGeer et al., 1978).

In chick retina, KA produces a severe loss of

Abbreviations used: ACh, Acetylcholine; AIB, a-Aminoiso- butyric acid; GABA, y-Aminobutyric acid; GDEE, y-Glutamic acid diethylester; KA, Kainic acid.

1169

1170 R . TAPIA AND C . ARIAS

cells, mainly in the inner nuclear and plexiform layers, 2- 14 days after intraocular injection (Schwarcz and Coyle, 1977; Biziere and Coyle, 1979; Morgan and Ingham, 1981 ; Pasantes-Morales et al., 1981). Swelling of the inner and outer plexiform layers was evident 1 h after intraocular KA injection in rats (Goto et al., 198l), and cell necrosis in the inner nuclear layer was observed after 15-30 min incubation of rabbit retinas (Hamp- ton et al., 1981). In the present work we have studied the possibility of detecting the excitatory action of KA and Glu as an immediate effect on the release of several neurotransmitters in whole chick retina and in retinal subcellular fractions. For comparative purposes, the effect of Glu and KA on transmitter release was also studied in mouse cortical and striatal slices and in mouse brain synaptosomes.

MATERIALS AND METHODS

[2-3HH]Glycine (specific activity, 15 Ciimmol), [2,3- 3H]GABA (specific activity, 36 CUmmol), ~ - [ u - ~ ~ C ] G l u (specific activity, 240 mCi/mmol), [methyl-3H]choline chloride (specific activity, 80 Ci/mmol), [ethyl-2- 3H]dopamine (specific activity, 21.5 Ci/mmol), and [methyl-3H]cr-aminoisobutyric acid (AIB; specific activity, 36 Ciimmol) were obtained from New England Nu- clear (Boston, MA). Kainic acid and L-G~u as well as a-glutamic acid diethylester (GDEE), EGTA, aminooxyacetic acid, and other metabolic inhibitors were from Sigma Chemical Co. (St. Louis, MO). Chicks (4-6 weeks old), dark adapted for 2 h, were used to obtain retinal tissue, and mice (local strain) were used for the experiments with brain tissue.

Subcellular fractionation of retina to obtain a nuclear pellet containing fragments of the outer photoreceptor segments and photoreceptor terminals (P,) and a crude synaptosomal fraction (Pz) which contains small synaptosomes derived from the plexiform layers and is only slightly contaminated with mitochondria was carried out as described by Lopez-Colome et al. (1978~). A purified synaptosomal fraction from mouse brain was obtained by the procedure of Hajos (1975). For the experiments with slices, the cortex or the striatum was dis- sected out and sliced in a McIlwain tissue chopper (0.4-mm thickness). In the experiments with whole retinas, each retina was cut in two halves and one half was used for each experimental condition. In this way it was possible to compare in parallel four experimental conditions using retinal tissue from one chick.

The release of neurotransmitters from chick whole retina and from mouse cortex and striatal slices was studied by the superfusion procedure previously described (Lopez-Colome et al., 1978b). The release from chick retinal fractions P, and P, and from mouse brain synaptosomes was studied also by a superfusion technique previously described in detail (Tapia and Meza-Ruiz, 1977). Briefly, the tissue was incubated for 10 min at 37°C in 2 or 2.5 ml of oxygenated Krebs-Ringer medium (118 mM NaCI, 1.2 mM KHZP04, 4.7 m M KCI, 1.17 mW MgS04, 2.5 m.W CaCI,, 25 mM NaHCO,, and 5.6 m M glucose, pH 7.4), and the labeled neurotransmitters or choline was

then added. The amount of radioactivity added for whole retina or slices and for subcellular fractions was, respectively (final concentrations shown in parentheses): ['HIGABA. 2.5 (0.22 p M ) and 0.83 pCi (0.5 pM); [ '4C]Gl~, 0.5 (3.35 p M ) and 0.4pCi(1.2 p M ) ; [3H]glycine, 1.0 (0.1 pM) and 1.0 pCi (0.1 p M ) ; [,H]AIB, 2.5 (0.03 pM) and 2.5 pCi (0.03 p M ) . In the case of slices and whole retina, [3H]dopamine (4 pCi , 2.8 p M ) and [3H]choline (4 pCi, 0.09 p M ) were also tested. After 10- min uptake incubation, the retina and slices were trans- ferred to four parallel glass superfusion chambers (0.25-ml volume), and the subcellular fractions to eight parallel chambers in which the bottom was a multiperfo- rated support holding 0.65 or 0.45 pm (for retinal P2 fraction) Millipore filters, on which the fractions were placed. Both for retina or slices and for the fractions, washing was carried out by superfusion with the Krebs-Ringer medium at high speed of the peristaltic pump (1.5 mumin); the speed was then adjusted to 0.5 mlimin and fractions were collected each min. After 4 or 5 min of spontaneous release, the superfusion medium was quickly substituted by medium containing 48 mM KCl (high K+), Glu, or KA. When necessary, the pH of the medium containing the acids was adjusted to 7.4 with NaOH or KOH. In some experiments Na+ or Ca2+ was omitted from the superfusion media. When Na+ was omitted, the osmolarity was maintained with 0.25 M sucrose and the buffer was 25 m M Tris-HC1, pH 7.4, instead of NaHCO.?. When Caz+ was absent, 1 mM EGTA was added. In the experiments with GABA, all media contained 0.1 mM aminooxyacetic acid, in those with choline, 0.1 mM eserine, and in those with dopamine, 0.1 mM pargyline and 0.1 mdml of ascorbic acid. The arrangement of four or eight parallel chambers (with four- or eight-channel peristaltic pumps) permitted the simultaneous study of different experimental conditions.

The radioactivity release in each fraction was counted in a Packard TriCarb scintillation spectrometer after the addition of 5 ml of Tritosol (Fricke, 1975). The results are expressed as percent of total radioactivity released per minute (total radioactivity = total released + that remain- ing in the tissue at the end of the superfusion). In most experiments with [?H]GABA and ['4C]Glu, their release was studied simultaneously by adding both amino acids to the uptake medium and adjusting the scintillation counter for double-label counting of the released fractions. In each experiment the overlapping of I4C and 'H was cal- culated from standard mixtures of the labeled amino acids, and this value (approximately 18%) was used to correct all experimental results.

RESULTS Chick whole retina

When the normal superfusing medium was substituted by a medium containing 10 mM Glu or 1 mM KA, there was an immediate and large stimulation of [3H]GABA release from whole chick retina. This stimulation was to a value approximately 12 times the prestimulation value, and was much larger than that produced by depolarization with high K+ concentration (Figs. 1 and 2). The stimulation by K+ depolarization was only approximately 80% and in the presence of Ca2+ it was much slower than in its

J . Neurochem., Vol. 39, No. 4, 1982

SELECTIVE STIMULATION OF NEUROTRANSMITTER RELEASE

W v) 1400- a w J W a: a m a 1000- 0

n I U

LL

Z

t- 6 J 3

600-

0

r 200-

8

1171

-

-

-

- -

FIG. 1. Effect of high K+ (48 mM; HK+) (A), 1 mM KA (B), and 10 mM Glu (C) on [3H]GABA release in chick whole retina (0). and effect of Caz+ (0) and Na+ (A) omission. After loading with labeled GABA, the retinas were superfused at a rate of 0.5 rnllmin as described in Materials and Methods, and the superfusing medium was quickly substituted by a medium containing high K+ concentration, KA, or Glu (arrow). Each point is the mean value of 8 experiments for HK+; 18 experiments for KA with Ca2+ and Na+, 6 for no Ca2+, and 10 for no Na+; and 10 for Glu with Ca2+ and Na+, 4 for no Ca2+, and 6 for no Na+. The differences of the peak stimulation by Glu and KA between control and Na+ ornis- sion were highly significant (p < 0.001).

MINUTES

absence (Fig. 1A). In contrast, the stimulation produced by Glu or KA was completely independent of Ca2+, but when Na+ was omitted from the medium, the stimulatory effect of both KA and Glu was diminished by approximately 70% (Fig. 1B and C). In an effort to test if this stimulation of GABA release was due to an interaction of Glu and KA with Glu receptors, in some experiments the putative Glu antagonist GDEE was added to the superfusing medium, and the concentration of KA and Glu was reduced 10 times. As can be seen in Fig. 2 , the

KA(mM): 1.0 0.1 0.1 GLU(mM GDEE(mM1: - - I00

3.0 1.0 1.0 - - 10.0

FIG. 2. Stimulation of [3H]GABA release by KA and Glu and effect of GDEE in whole retina in the presence of Ca2+ and Na'. The percent stimulation refers to the peak value vs the prestimulation value. Mean values ? SEM are shown (n = 10, 4, and 4 for KA; n = 10, 4, and 4 for Glu). The concentrations of KA, Glu, and GDEE used are indicated below each bar. The effect of GDEE was not significant (p > 1 .O for Glu effect and p < 0.1 for KA effect).

stimulation at 0.1 mA4 KA was approximately half that observed at 1 mM, and GDEE inhibited it approximately 50%, although this inhibition was not significant due to the large standard error. When the Glu concentration was reduced from 10 to 1 mM, the stimulation of GABA release was proportionally diminished, and, although some blocking effect of GDEE was observed, this was even less significant than in the case of KA.

The results obtained in the experiments with [3H]glycine are shown in Fig. 3 . Similarly to r3H]GABA, both KA and Glu produced a notable stimulation of labeled glycine release (approximately 600 and 250%, respectively) that was abolished when Na+ was omitted from the medium, although in the latter condition the basal release was considerably increased. The stimulation produced by Glu seemed to occur more slowly than that induced by KA, and it was also observed in the absence of Ca2+, whereas that by KA was decreased by approximately 50% when this cation was absent. K+ depolarization stimulated [3H]glycine release in a Ca'+-dependent manner, but this effect was 60% less than that produced by KA (Fig. 3).

In contrast to the results obtained with labeled GABA and glycine, neither high K+ concentration nor KA had any significant effect on [14C]Glu release from whole retina, with or without Ca2+ in the medium (Fig. 4). Glu had a clear stimulatory effect, although considerably less notable than that on GABA and glycine release (only 90% stimulation). Interestingly, this effect of Glu on [14C]Glu release was independent of both Ca2+ and Na+ (Fig. 4).

The lack of effect of KA on labeled Glu release, as compared to its potent action on GABA and glycine release, suggested some specificity of effects. In order to study this further, the release of

J . Neurochem., V d . 39, N o . 4, 1982


u 2 m' 6 8 10 12 - 2 a 6 8 1 0 1 2

MINUTES

It,,,, 2 &LA 6 8 10 I2

FIG. 3. Effect of high K+ (HK+) (A), KA (B), and Glu (C)on [3H]glycine release in whole retina. Details as for Fig. 1. Each point is the mean value of six experiments. The differences of the peak stimulation by KA and Glu between control and Ca2+ omission were not significant (p > 0.2).

labeled dopamine and of the nonmetabolite GABA analog AIB was studied. As shown in Figs. 5 and 6, neither dopamine nor AIB release was significantly affected by 10 mM Glu or 1 mM KA. High K+ produced a 100% stimulation of dopamine release, which was completely abolished in the absence of Ca2+, whereas the release of AIB was not affected by K+ depolarization (Figs. 5 and 6).

In some experiments the release of labeled acetylcholine (ACh) derived from [3H]choline was studied. Both the spontaneous release and that

stimulated by high K+ were highly variable in these experiments. Therefore, it was not possible to as- sess the effects of KA and Glu on ACh release in whole retina.

Chick retinal fractions In order to investigate whether some kind of

neural organization was required to observe the effects of KA and Glu on the release of GABA and glycine, a series of experiments was carried out with fractions PI and P2 from chick retina. As shown

FIG. 4. Effect of high K+ (HK+) (A), KA (B), and Glu (C) on [14C]GI~ release in whole retina. De- tails as for Fig. 1. Each point is the mean value of 6 experiments for HK+, 10 experiments for KA and Glu with Ca2+ and Na+, and 6 for no Ca2+ or no Na+.

h I \

'&-*

MINUTES

J . Neurochem., Vol. 39, No. 4, 1982

SELECTIVE STIMULATION OF NEUROTRANSMITTER RELEASE 1173

in Fig. 7, in fraction P, the release of [3H]GABA was stimulated by both KA and Glu, but the extent of this stimulation was much smaller (200-300%) and slower than in whole retina. Interestingly, the effect of KA and Glu in fraction PI was greatly decreased when Na+ was omitted from the medium (Fig. 8). In fraction P, only a small and delayed stimulatory effect of both acids was observed. K+ depolarization of P, fraction resulted in a marked stimulation (215%) of labeled GABA release, which was larger (440%) in a Ca'+-free medium. In contrast, the effect of high K+ was small in fraction P p , comparable with that observed in whole retina (Fig. 7).

[3H]Glycine, in contrast to ["HIGABA, was not retained well by fraction P, and particularly by fraction P2 (compare the spontaneous release values in Figs. 9 and 3). KA stimulated glycine release in fraction P, similarly to whole retina, whereas in fraction P, this effect was approximately 50% smaller, due probably to the high spontaneous release rate in this fraction. The stimulatory effect of Glu on glycine release in both fractions was very small (Fig. 9). In fraction P, high K+ concentration resulted in a stimulation of glycine release similar to that in whole retina, but, contrary to the latter, this stimulation was independent of Ca2+. A small effect of high K+, also Ca2+ independent, was observed in fraction P, (Fig. 9).

The spontaneous release of [I4C]Glu in fractions P, and Pz was much higher than in whole retina, and neither KA nor Glu had any effect on the release of labeled Glu. A small, Ca'+-independent stimulatory effect of high K+ was observed in fraction PI (Fig. 10).

["HIAIB was not retained by fractions P, and P',

5r

I t I I L - HK+ B

iR

2 M A 6 8 10 12 MINUTES

FIG. 5. Effect of high K' (HK') (top) and KA and Glu (bottom) on [3H]dopamine release in whole retina. Details as for Fig. 1. Each point is the mean value of four experiments.

I t -

X P I t I I I I 2 K+o 6 8 10 12

? $ U

- K A * MINUTES

FIG. 6. Effect of high K+ (HK+), KA, and Glu on the release of [3H]AIB in whole retina in the presence of Ca2+ and Na+. De- tails as for Fig. 1. Each point is the mean value of four experiments.

and no significant effect of high K+, KA, or Glu was observed on the release of this amino acid in either fraction (Fig. 11).

Mouse brain slices and synaptosomes For comparative purposes, the effect of KA and

Glu on the release of neurotransmitters was also studied in slices of mouse brain cortex and striatum and in synaptosomes (data not shown). KA at 1 mM concentration had no effect at all on the release of labeled GABA, Glu, and ACh (derived from labeled choline) in any of these brain preparations, and it had a slight stimulatory effect on dopamine release in striatal slices (25 k 6%; mean value k SEM offour to six experiments). Glu at concentrations ranging from 0.1 to 10 mM produced a 250-470% (with SEM = 14- 19% of the corresponding mean; n = 4- 6) stimulation of [14C]Glu release in cortical and striatal slices and in synaptosomes, which was either abolished or decreased by approximately 85% when Na+ was omitted from the superfusing media. Glu at 10 mM had no significant effects on GABA release in cortex slices or synaptosomes, and slightly stimulated (25-45% with SEM =

27-48% of the corresponding mean; n = 4-6) the release of GABA and dopamine in striatal slices.

High K+ concentration in the presence of Ca2+ resulted in a 600-750% (with SEM = 9-25% of the corresponding mean) stimulation of GABA release and 200-240% (with SEM = 23-26% of the corresponding mean) stimulation of Glu release in cortical or striatal slices, and a 710 % 129 and 250 k 62% stimulation of dopamine and ACh release, respectively, in striatal slices (results of four to five experiments).

DISCUSSION

Specificity of KA and Glu effects The results described in this work show that KA

and Glu at the relatively high concentrations used are able to promote a rapid release of labeled GABA and glycine from chick retina that is several times

J . Neurochem.. Vol. 39, No . 4, 1982


k- 0 a 0

2 _I

0

FIG. 7. Effect o f h igh K+ (HK+), KA, and Glu on [3H]GABA release in retinal subcellular fractions Pi (lower panels) and P, (upper panels). Details as for Fig. 1. Each point is the mean value of six experiments. The difference of the peak stimulation by HK+ between control and Ca2+ omission in fraction Pi was in the limit of significance (p = 0.02).

aR a m a

' 3 Ht 7 9 I1 13 3 .U: 7 9 11 13 I t I I 1 I I I I I I

- GLU MINUTES

larger than that produced by K+ depolarization. That this release is not due to an unspecific effect of the acids or to a general damage of the tissue is shown by the finding that under identical experimental conditions, neither KA nor Glu had any effect on the release of [3H]dopamine. This amine, however, was released in a Ca'+-dependent manner by K' depolarization, similar to previously reported results in goldfish retina (Sarthy and Lam, 1979) indicating that the lack of effect of KA and Glu was not due to a methodological problem. Since in all the experiments the stimulation by KA, Glu, and high K+ was carried out in pafallel superfusion chambers containing half retinas from the same animals, other interpretations related to the different state of preservation of the tissue cannot be considered. Furthermore, the release of AIB, a GABA analog normally absent in nervous tissue, was not stimulated by KA, Glu, or high K+ (Fig. 6). Also relevant to the specificity of effects was the finding that neither KA nor high K+ had any effect

on the release of ['TIGlu, although Glu itself pro- moted a 100% stimulation of release of labeled Glu. The latter observation suggested to us the possibility of Glu homoexchange process, as has been described in brain tissue (Levi et al., 1976). However, the homoexchange should be dependent on the presence of Na+, and in our experiments this was not the case (Fig. 4). Although we cannot offer an explanation for the latter result, it seems valid to conclude that the results with labeled Glu, particularly the lack of effect of KA, also indicate that the action of KA and Glu on GABA and glycine release is not the result of unspecific damage of the retinal tissue.

Mechanism of KA- and Glu-induced release Although KA and Glu might be acting on different

receptors (Hall et al., 1978; McLennan and Lodge, 1979), both seem to excite neurons by a receptor mechanism that probably involves depolarization. On these grounds, our expectation was that the re-

FIG. 8. Effect of Na+ omission on the stimulation of [3H]GABA release by KA and Glu in retinal subfraction P,. Details as for Fig. 1. Each point is the mean value of five experiments. The differences of the peak stimulation between control and Na+ omission were highly significant (p < 0,001).

MINUTES

.I. Neurochcm.. Vol. 39, No. 4, 1982


FIG. 9. Effect of h igh K ' (HK+), KA, and Glu on [3H]glycine release in retinal subcellular fractions P, (lower panels) and P, (upper panels). Details as for Fig. 1. Each point is the mean value of four experiments.

lease of GABA and glycine stimulated by KA and Glu should be dependent on external Ca2+. Con- trary to this expectation, the release of GABA and glycine evoked by both KA and Glu was very similar in a medium containing 2 mM Ca2+ and in a medium with no Ca2+ and with EGTA. In contrast, when Na+ was omitted from the medium, a striking decrease in the stimulation of GABA and glycine release by both KA and Glu was observed (Figs. 1 and 3), suggesting a relationship between Na+ and amino acid release. Since this is a Ca2+-independent phenomenon, the increased GABA and glycine release might be due to the Na+-dependent, carrier- mediated mechanism that has been described in brain tissue (Levi et al., 1976; 1978; Martin, 1976).

I I I I I I 1 I

3 E* 7 9 II 13 3 M A 7 9 II 13 G'

MINUTES

This interpretation is supported by the notable Na+ dependence of the GABA-releasing action of KA and Glu in fraction PI (Fig. 8). That high Glu concentrations may indeed rapidly increase the membrane permeability to Na+ has been shown in cortical slices (Harvey and McIlwain, 1968) and in crayfish muscle (Onodera and Takeuchi, 1976). Al- though to our knowledge there are no direct data on the effect of KA on Na+ transport, a role of this cation in KA-releasing action is strongly supported by the finding that the short-term damaging effect of 1 mM KA on horizontal cells and plexiform layers in rabbit retina did not occur in a Na+-free medium (Hampton et al., 1981). Furthermore, it has been re- cently reported, using autoradiographic techniques,

FIG. 10. Effect of h igh K- (HK+), KA, and Glu on [i4C]GI~ release in retinal subcellular fractions P, (lower panels) and P, (upper panels). Details as for Fig. 1. Each point is the mean value of four experiments.

GLU A MINUTES -

.I. Neurochem., Vol. 39, No. 4, 1982


> k 2

9 n a a

a

I- 2

J

I- 0 I-

FIG. l l . Effect of high K+ (HK+), KA, and Glu on the release of [3H]AIB in retinal subcellular fractions P, (left) and P, (right). Details as for Fig. 1. Each point is the mean value of four experiments.

& zR

a I- lx > I- 3 n m

that Glu induces the release of r3H]GABA from horizontal cells also in a Na+-dependent manner (Yazulla and Kleinschmidt, 1981). In another recent publication (Goto et al., 1981), it was observed that 1 h after intraocular KA injection in the rat, retinal GABA levels were diminished. Since this effect was observed prior to the time when a decrease of glutamate decarboxylase (EC 4.1.1.15) activity could be detected, it seems possible that a rapid release of GABA as a consequence of KA action might account for the reduction in GABA content (Goto et al., 1981).

The possibility that the above Na+-dependent, carrier-mediated mechanism of KA- and Glu- releasing action is mediated by specific receptors has not been confirmed. When GDEE was added at concentrations 100 and 10 times greater than KA and Glu, respectively, there was some inhibition of the stimulated release of GABA, but this inhibition was not significant (Fig. 2). This result, however, is difficult to evaluate, since, besides the still unclear specificity of GDEE action on Glu and KA receptors (McLennan and Lodge, 1979), the binding of Glu to chick retinal membranes was not completely antagonized by GDEE even at a 105-fold excess concentration with respect to Glu (Lopez-Colome, 1981).

The results obtained in retinal subcellular fractions, particularly those obtained in fraction PI, in which the effects of KA and Glu on GABA release were almost as notable as in whole retina (Fig. 7), suggest that if these effects are receptor mediated, such receptor must be present at a presynaptic level. This conclusion is supported by the Na+ dependence of the KA and Glu action on [3H]GABA release observed in fraction PI (Fig. 8). In the case of glycine release, there is a notable difference in the potency of releasing action between KA and Glu

8

'1 A I 1 1 I I I I l l 1 1 I I 1 3 m+* 7 9 II 13 3 %+* 7 9 I I 13

m A a* GLU A

MINUTES -

in fractions P, and P2 (Fig. 9). This suggests that if such presynaptic receptors are involved, they are more sensitive to KA than to Glu in the terminals from which glycine is released.

In agreement with the possibility that KA and Glu receptors might be involved in the mechanism of the GABA- and glycine-releasing effect, it has been reported that high-affinity binding sites for KA and Glu are present in chick retinal membranes (Biziere and Coyle, 1979; Lopez-Colome, 1981). In rat brain, the KA binding sites seem to be highly concentrated in synaptic plasma membranes, particularly in the synaptic junctions (Foster et al., 1981), a finding that could explain our results with retinal subcellular fractions. However, our data on the effect of KA and Glu on the release of GABA from mouse brain synaptosomes were negative (see Results).

Presynaptic Glu receptors have also been proposed to account for the GDEE-resistant and tetrodotoxin-resistant moderate stimulatory effect of Glu on dopamine release reported in rat striatal slices (Giorguieff et al., 1977). According to other authors (Roberts and Sharif, 1978), this moderate effect is also produced by KA. In our experiments with mouse striatal slices, we observed only a com- paratively very small effect of KA and Glu on dopamine release and also of Glu on GABA release (see Results). These results might indicate a much lower sensitivity of striatal receptors to KA and Glu as compared to retinal receptors, at least with regard to the rapid Na+-dependent effects of KA and Glu.

KA did not show any effect on the release of [14C]Glu, in either whole retina or retinal fractions, or in brain slices or synaptosomes. These results, therefore, do not support the hypothesis that the action of KA is mediated by the release of Glu from presynaptic endings (McGeer et al., 1978). The not-

J . Neurochem., Voi. 39, No. 4, 1982


able stimulation of [14C]Glu release by Glu observed in cortical and striatal slices and in brain synaptosomes can be accounted for by a homoexchange process (Levi et al., 1976), since it was abolished or greatly inhibited when Na+ was absent in the superfusion medium (see Results).

Effects of high K+ The observations described in the present paper

on the Ca2+-dependent stimulation of labeled GABA, glycine, and dopamine by high K+ in whole chick retina are in agreement with previous data suggesting that these compounds function as neurotransmitters in retinal tissue (Krarner, 1976; Voa- den, 1976; Lopez-Colome et al., 1976; 19786; Sar- thy and Lam, 1979; Chin and Lam, 1980). However, in retinal subfractions P, and P,, we observed that the release of GABA and glycine stimulated by K+ depolarization was not dependent on external Ca2+. This finding differs from the partial Caz+ dependence previously shown for these two amino acids in retinal P, synaptosomal fractions (Lopez- Colome et al., 19786). Although GABA was well retained by this fraction, the isolation procedure might have altered its synaptic membrane properties, a possibility supported by the low glycine retention observed in fraction P2 (Fig. 9). In the case of [l4C]G1u, our results on the failure of Kf to stim- ulate its release in whole retina and in fraction P2 agree with similar results obtained with stimulation by light in chick (Lopez-Colome et al., 19786) and rabbit (Neal and Massey, 1980) retinas, and could be considered as evidence against a transmitter role for the amino acid in this tissue (Fig. lo). Also against such a role was the finding that the small stimulation of labeled Glu release by high K+ in fraction P1 was completely Ca'+ independent.

Acknowledgments: The a u t h o r s t h a n k Dr. H . Pasantes-Morales for her help in the initial phases of this study, This work was supported in part by a grant from the Consejo Nacional de Ciencia y Tecnologia (Project PCCBNAL 790214).

REFERENCES Biziere K. and Coyle J. T. (1979) Localization of receptors for

kainic acid on neurons in the innernuclear layer of retina. Neuropharmucolugy 18, 409 -41 3.

Chin C.-A. and Lam D. M.-K. (1980) The uptake and release of ["Hlglycine in the goldfish retina. J . Physiol. (Lond.) 308,

Curtis D. R. (1975) Gamma-aminobutyric and glutamic acids as mammalian central transmitters, in Metabolic Com- partmenrution and Neurotrunsmission (Bed S . , Clarke D. D., and Schneider D., eds), pp. 11-36. Plenum Press, New York.

Foster A . C. and Roberts P. J . (1978) High affinity L - [3H]glutamate binding to postsynaptic receptor sites on rat cerebellar membranes. J . Neurochem. 31, 1467-1477.

Foster A. C., Mena E. E . , Monaghan D. T., and Cotman C . W.

185- 195.

(1981) Synaptic localization of kainic acid binding sites. Nature 289, 73-75.

Fricke U . (1975) Tritosol: a new scintillation cocktail based on Triton X-100. Anal. Biochem. 63, 555-558.

Giorguieff M. F., Kernel M. L., and Glowinski J. (1977) Pre- synaptic effect of r-glutamic acid on the release of doparnine in rat striatal slices. Neurosci. Lett. 6, 73-77.

Goto M., Inomata M., Ono H., Saito K.-I., and Fukuda H. (1981) Changes of electroretinogram and neurochemical as- pects of GABAergic neurons of retina after intraocular injection of kainic acid in rats. Brain Res. 211, 305-314.

Hajos F. (1975) An improved method for the preparation of synaptosomal fractions in high purity. Brain Res . 93,

Hall J . G., Hicks T. P., and McLennan H. (1978) Kainic acid and the glutamate receptor. Neurosci. Lef t . 8, 171- 175.

Hampton C. K., Garcia C., and Redburn D. A. (1981) Localiza- tion of kainic acid-sensitive cells in mammalian retina. J . Neurosci. Res. 6, 99- 11 1 .

Harvey J. A. and McIlwain H. (1968) Excitatory acidic amino acids and the cation content and sodium ion flux of isolated tissues from the brain. Biochem. J . 108, 269-274.

Johnston G. A. R., Curtis D. R . , Davis J., and McCulloch R. M. (1974) Spinal interneurone excitation by conformationally restricted analogues of L-glutamic acid. Nature 248, 804-805.

Kramer S . G. (1976) Dopamine in retinal transmission, in Tran.smitters in the Visual Process (Bonting S . L., ed), pp. 165- 198. Pergamon Press, Oxford.

Krnjevic K. (1974) Chemical nature of synaptic transmission in vertebrates. Physiol. Rev. 54, 418-540.

Levi G., Poce U., and Kaiteri M. (1976) Uptake and exchange of GABA and glutamate in isolated nerve endings, in Trunvport Phenomena in the Nervous System (Levi G., Battistin L., and Lajtha A., eds), pp. 273-289. Plenum Press, New York.

Levi G., Banay-Schwartz M., and Raiteri M. (1978) Uptake, exchange and release of GABA in isolated nerve endings, in Amino Acids as Chemical Transmitters (Fonnum F., ed), pp, 327-350, Plenum Press, New York.

LopezColorne A. M. (1981) High-affinity binding of L-glutamate to chick retinal membranes. Neurochem. R e s . 6, 1019-1033.

LopezColome A. M., Erlij. D., and Pasantes-Morales H. (1976) Different effects of calcium flux-blocking agents on light and potassium stimulated release of taurine from retina. Brain

LopezColome A. M., Tapia R., Salceda R., and Pasantes- Morales H. ( 1 9 7 8 ~ ) K+-stimulated release of labeled y- aminobutyrate, glycine and taurine in slices of several regions of rat central nervous system. Neuroscience 3 , 1069- 1074.

Lopez-Colome A. M., Salceda R., and Pasantes-Morales H. (19786) Potassium-stimulated release of GABA, glycine and taurine from the chick retina. Neurochem. Res. 3,431 -441.

Love11 K. L. and Jones M. Z. (1980) Kainic acid neurotoxicity in the mouse cerebellum. Brain Res. 186, 245-249.

Martin D. L. (1976) Carrier-mediated transport and removal of GABA from synaptic regions, in GABA in Nervous System Function (Roberts E., Chase T. N., and Tower D. B., eds), pp. 347-386. Raven Press, New York.

McGeer P. L., McGeer E. G., and Hattori T. (1978) Kainic acid as a tool in neurobiology, in Kuinic Acid as a Tool in Neurobiology (McGeer E. G., Olney J. W., and McGeer P. L., eds), pp. 123-138. Raven Press, New York.

McLennan H. and Lodge D. (1979) The antagonism of amino acid-induced excitation of spinal neurons in the cat. Brain Res. 169, 83-90.

Morgan I. G. and Ingham C. A. (1981) Kainic acid affects both plexiform layers of chicken retina. Neurosci. Lett. 21,

Neal M. J. and Massey S. C. (1980) The release of acetylcholine

485-489.

Res. 113, 527-534.

275 -280.

J . Neurochern.. Vol. 39, N u . 4, IY82


and amino acids from the rabbit retina in vivo. Neurochemistry 1, 191-208.

Olney J. W. (1979) Excitotoxic amino acids: research applica- tions and safety implications, in Glutamic Acid: Advunces in Biochemistry and Physiology (Filer L. J., Jr., Garattini S., Kare M. R. , Reynolds W. A., and Wurtman R. J., eds), pp. 287-319. Raven Press, New York.

Olney J. W., Rhee V., and Ho 0. L. (1974) Kainic acid: a powerful neurotoxic analogue of glutamate. Bruin Res. 77, 507-512.

Onodera K. and Takeuchi A. (1976) Permeability changes produced by L-glutamate at the excitatory post-synaptic membrane of the crayfish. J . Physiol. (Lond.) 255, 669-685.

Pasantes-Morales H., Quesada O., and Carabez A. (1981) Light-stimulated release of taurine from retinas of kainic acid-treated chicks. J . Neurochem. 36, 1583- 1586.

Roberts P. J. and Sharif N . A. (1978) Effects of L-glutamate and related amino acids upon the release of [3H]dopamine from rat striatal slices. Bruin Res. 157, 391-395.

Sarthy P. V. and Lam M. K. (1979) The uptake and release of ['HI dopamine in the goldfish retina. J . Neurochem. 32, 1269- 1277.

Schwarcz R. and Coyle J. T. (1977) Kainic acid: neurotoxic effects after intraocular injections. Invest. Ophthalmol. Vis. Sci. 16, 141-148.

Schwarcz R., Scholz D., and Coyle J . T . (1978) Structure- activity relations for the neurotoxicity of kainic acid deriva- tives and glutamate analogues. Neurophurmucology 17, 145- 151.

Shinozaki H. and Konishi S. (1970) Actions of several antihelmintics and insecticides on rat cortical neurons. Brain Res. 24, 368-371.

Simon J . R., Contrera J . F., and Kuhar M. J . (1976) Binding of [3H]kainic acid, an analogue of L-glutamate, to brain membranes. J . Neurochem. 26, 141-147.

Tapia R. and Meza-Ruiz G. (1977) Inhibition by ruthenium red of the calcium-dependent release of [3H]GABA in synaptosomal fractions. Bruin Kes. 126, 160- 166.

Voaden M. J. (1976) Gamma-aminobutyric acid and glycine as rctinal transmitters, in Transmitters in the Visual Process (Bonting S . L., ed), pp 107-125. Pergamon Press, Oxford.

Yazulla S. and Kleinschmidt J. (1981) Carrier-mediated release of GABA from retinal horizontal cells. Soc. Neurosci. Abstr. 7, 110.

.I. Neurochem., Val. 39, No . 4, 1982

Selective Stimulation of Neurotransmitter Release from Chick Retina by Kainic and Glutamic Acids

Documents

Transcript of Selective Stimulation of Neurotransmitter Release from Chick Retina by Kainic and Glutamic Acids