EFFECT OF EXCITATORY AMINO-ACIDS AND ANALOGS ON [H-3]ACETYLCHOLINE RELEASE FROM AMACRINE CELLS OF THE RABBIT RETINA

The pharmacology of cholinergic amacrine cells has been further studied by examining the effects of excitatory amino acids and antagonists on [3H]acetylcholine (ACh) release from the retinas of anaesthetized rabbits. Exposure of the retina to glutamate (5 mM), aspartate (5 mM), kainate (8 .mu.M) and quisqualate (8 .mu.M) abolished the light-evoked release of [3H]ACh but increased the spontaneous resting release four- to fivefold. N-methyl-D-aspartate (NMDA) (5 mM) in normal Krebs bicarbonate medium abolished the light-evoked release of [3H]ACh but did not affect the resting release. However, in Mg-free medium, NMDA (0.5 mM) abolished the light-evoked release of [3H]ACh and increased the resting release fivefold. The effects of other agonists were not altered in Mg-free medium. The amplitude of the electroretinogram (e.r.g.) b-wave was not significantly reduced by glutamate, aspartate or NMDA (in normal or Mg-free medium). Kainate and quisqualate reduced the b-wave amplitude to approximately 50 and 30% of controls respectively. The general excitatory amino acid antagonist, cis-2,3-piperidine dicarboxylic acid (PDA) (2 mM) blocked the light-evoked of [3H]ACh, but had no significant effect on the e.r.g. b-wave amplitude or on the resting release of [3H]ACh. L(+)-2-amino-4-phosphonobutyrate (L(+)-APB) decreased the light-evoked release of [3H]ACh and the amplitude of the e.r.g. b-wave in parallel (correlation coefficient 0.995). D(-)-APB had similar effects but was fifteen times less potent. Since the L(+)-compound is known to mimic the photoreceptor transmitter on the depolarizing, but not hyperpolarizing, bipolar cells, these results strongly suggest that the [3H]ACh released in response to light originates mainly from the ''on'' (displaced) cholinergic amacrine cells. Our experiments give no information on the origin of the spontaneously released [3H]ACh. PDA (2-5 mM) blocked the effects of glutamate, aspartate, kainate, quisqualate and NMDA on the resting release [3H]ACh. In contrast, D(-)-APB (5 mM), which is a relatively non-specific excitatory amino acid antagoist in the spinal cord, blocked only the actions of kainate and had no blocking effect on the actions of glutamate or aspartate (the putative bipolar cell transmitters) or NMDA. D(-)-2-amino-5-phosphovalerate (APV) which is a relatively selective NMDA antagonist in the spinal cord failed to discriminate between the effects of kainate and NMDA on the resting release [3H]ACh, D-.alpha.-aminoadipate at concentrations up to 5 mM had no effect on any of the agonists. NMDA (5 mM) in normal (Mg-containing) medium selectively abolished the effect of kainate on the resting release of [3H]ACh. It is not possible in these experiment to determine the exact site of action of drugs affecting ACh release. However, the effects of glutamate, aspartate and PDA on [3H]ACh release, together with their lack of effect on the e.r.g. b-wave amplitude, are most easily explained by the presence on cholinergic amacrine cells of excitatory amino acid receptors and an input from bipolar cells using glutamate or perhaps aspartate as their transmitter. The pharmacology of excitatory amino acid receptors associated with amacrine cell [3H]ACh release is substantially different from that described in the spinal cord.