1. N-Methyl-D-aspartate (NMDA), alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA), and kainate receptor channels have been examined in rat cerebellar granule neurones with whole-cell and single-channel patch-clamp methods. The whole-cell peak and steady-state aspartate and NMDA currents were reversibly inhibited by extracellular protons; the IC50 (concentration producing half-maximal inhibition) for the full H+ inhibition curve for NMDA receptors corresponded to pH 7.3, near to physiological pH. (S)-AMPA and kainate whole-cell currents were inhibited by protons with IC50 values that corresponded to pH 6.3 and 5.7, respectively; these receptors were, however, insensitive to H+ concentrations that inhibited NMDA receptor responses. 2. Proton inhibition of the NMDA, AMPA and kainate receptor-mediated responses was voltage insensitive, and did not involve a shift in reversal potential. 3. The EC50 (concentration producing half-maximal effect) for aspartate calculated from the whole-cell dose-response curve was similar at pH 6.8 and 7.6 (mean 11.2-mu-M). Although the EC50 for glycine potentiation of the aspartate response was marginally increased from 273 nM at pH 7.6 to 373 nM at pH 6.8, H+ inhibition was not overcome by up to 1 mM-external glycine. Inhibiting concentrations of H+ appropriate for AMPA and kainate receptors did not markedly alter the EC50 values determined for (S)-AMPA (3.4-mu-M) and kainate (114-mu-M) at pH 7.2. 4. Treatment of neurones with N-ethylmaleimide, iodoacetic acid, dithiothretiol or diethyl pyrocarbonate did not influence proton inhibition of NMDA receptor responses. However, treatment with diethyl pyrocarbonate, which potentiated aspartate responses, appeared to reduce the effectiveness of Zn2+ inhibition of NMDA receptors. 5. Desensitization of whole-cell NMDA and (S)-AMPA currents was studied with ionophoretic application of agonist to the cell soma. Whole-cell aspartate currents desensitized rapidly, irrespective of the glycine concentration. Increased H+ concentration did not detectably alter the ratio of peak/steady-state current, or the time constants describing the onset of, or recovery from, desensitization. The time constant describing desensitization of (S)-AMPA-induced whole-cell currents also appeared unchanged by inhibiting pH (6.2). 6. The amplitudes of aspartate- or NMDA-activated single-channel multiple conductance levels were unchanged by decreasing the pH to 6.8. Similarly, concentrations of H+ that inhibit (S)-AMPA (pH 6.2) and kainate currents (pH 5.6) did not change the single-channel conductance for either respective channel, as estimated from fluctuation analysis of whole-cell currents. 7. Recording of NMDA receptor channels from outside-out, inside-out, and cell-attached patches suggested that protons act at a site on or near the extracellular face of the receptor. 8. The opening frequency both of aspartate-activated individual openings and bursts of openings was decreased by 58.2 and 49.5 % with reductions in pH from 7.6 to 6.8. Inhibiting concentrations of H+ (pH 6.8) did not significantly alter the time constants of the three briefest exponential components describing shut-time distributions fitted to five components, but substantially increased the time constants of the longer shut-time components. 9. Two or three exponential components described the burst-length distributions constructed either exclusively from 50 pS events, or from 30-50 pS events, respectively. The overall mean burst length for NMDA receptor channels was decreased by 35.2 % in pH 6.8. The time constants describing either distribution were unchanged by inhibiting concentrations of H+, although the relative area of the longest fitted component was reduced. Inhibiting concentrations of H+ had analogous effect on the fitted exponential components describing distributions of individual open-time durations, as well as distributions of cluster lengths. 10. We conclude that protons inhibit NMDA, AMPA and kainate receptors with different IC50 values; only NMDA receptors were sensitive to changes in H+ within the physiological range. The properties of H+ inhibition appeared similar for all three receptors. Single-channel recordings suggested that protons might inhibit NMDA receptors by influencing the events preceding the opening of an agonist-bound receptor. The implications of proton sensitivity of NMDA receptors in both normal and pathological situations are discussed.
