The physiology of excitatory amino acids in the vertebrate central nervous system

Increasing the hydrogen ion concentration of the bathing medium reversibly depresses the sodium permeability of voltage-clamped frog nerves. The depression depends on membrane voltage: changing from pH 7 to pH 5 causes a 60% reduction in sodium permeability at +20 mV, but only a 20% reduction at +180 mV. This voltage-dependent block of sodium channels by hydrogen ions is explained by assuming that hydrogen ions enter the open sodium channel and bind there, preventing sodium ion passage. The voltage dependence arises because the binding site is assumed to lie far enough across the membrane for bound ions to be affected by part of the potential difference across the membrane. Equations are derived for the general case where the blocking ion enters the channel from either side of the membrane. For H+ ion blockage, a simpler model, in which H+ enters the channel only from the bathing medium, is found to be sufficient. The dissociation constant of H+ ions from the channel site, 3.9 x 10-6 M (pK a 5.4), is like that of a carboxylic acid. From the voltage dependence of the block, this acid site is about one-quarter of the way across the membrane potential from the outside. In addition to blocking as described by the model, hydrogen ions also shift the responses of sodium channel "gates" to voltage, probably by altering the surface potential of the nerve. Evidence for voltage-dependent blockage by calcium ions is also presented.

The interaction of two dissociative anaesthetics, ketamine and phencyclidine, with the responses of spinal neurones to the electrophoretic administration of amino acids and acetylcholine was studied in decerebrate or pentobarbitone-anaesthetized cats and rats. Both ketamine and phencyclidine selectively blocked excitation by N-methyl-aspartate (NMA) with little effect on excitation by quisqualate and kainate. Ketamine reduced responses to L-aspartate somewhat more than those of L-glutamate; the sensitivity of responses to these two putative transmitters was between that to NMA on one hand and that to quisqualate or kainate on the other. On Renshaw cells, ketamine and phencyclidine reduced responses to acetylcholine less than those to NMA but more than those to quisqualate or kainate. Dorsal root-evoked synaptic excitation of Renshaw cells was reduced to a greater extent than that following ventral root excitation. Intravenous ketamine, 2.5-20 mg/kg, and phencyclidine, 0.2-0.5 mg/kg, also selectively blocked excitation of neurones by NMA. Ketamine showed no consistent or selective effect on inhibition of spinal neurones by electrophoretically administered glycine or gamma-aminobutyricacid (GABA). The results suggest that reduction of synaptic excitation mediated via NMA receptors contributes to the anaesthetic/analgesic properties of these two dissociative anaesthetics.