Noise and single channels activated by excitatory amino acids in rat cerebellar granule neurones.

Abstract
1. Glutamate-receptor ion channels in rat cerebellar granule cells maintained in explant cultures have been investigated with patch-clamp methods. Properties of these channels were determined from noise analysis of whole-cell currents and from noise and single-channel currents recorded in outside-out membrane patches. 2. Glutamate (10-20 .mu.M) evoked two types of response. Some granule cells gave small inward currents accompanied by clear increases in current noise (''large noise'' responses), whereas other cells gave larger inward currents and small noise increases (''small noise'' responses). 3. A mean single-channel conductance (.gamma.) of 46.6 pS was estimated for glutamate from four ''large noise'' cells. A mean .gamma. value of 8.4 pS was estimated for seven other ''large noise'' cells. The results suggest that in these latter cells glutamate activated both large (.simeq. 50 pS) and small conductance (.simeq. 140 fS) channels. 4. Applications of aspartate (10-30 .mu.M) or N-methyl-D-aspartate (NMDA, 10-30 .mu.M) produced small inward currents and large increases in noise; .gamma.noise = 48.5 pS (aspartate) and 46.7 pS (NMDA). 5. Large single-channel currents were evoked by glutamate, aspartate and NMDA in outside-out patches. The mean conductance values obtained for the largest amplitude openings were: .gamma.(glutamate) = 49.5 pS, .gamma.(aspartate) = 51.5 pS, and .gamma.(NMDA) = 53.0 pS. For each agonist, these 50 pS openings comprised 75-85% of the completely resolved currents in each patch. Openings to 40 and 30 pS conductance levels accounted for 10-15% and 3-7% of the total, and the presence of apparently direct transitions between these levels and the 50 pS level suggests they are sublevels of the same multi-conductance channels. 6. A mean channel conductance of 22.9 pS was estimated from noise evoked by quisqualate (10-30 .mu.M). Single-channel currents were examined in four patches. In two, quisqualate evoked predominantly small currents for two amplitudes, .gamma. = 8.4 pS and 16.5 pS; some 50 pS openings were also present. In the other two patches, most openings were 50 pS events. 7. Granule cells gave inward currents to kainate (10-30 .mu.M), and a mean conductance of 3.1 pS was estimated from kainate noise. In patches in which aspartate or NMDA produced mainly 50 pS openings, more than 74% of the single channel currents evoked by kainate were of smaller amplitude, with mean conductances of .gamma. = 8.1 and 15.1 pS. Kainate also activated 50 pS channels and produced some 30 and 40 pS openings. 8. A mean single-channel conductance of 140 fS was estimated from ''small noise'' whole-cell currents evoked by glutamate (10-20 .mu.M). Spectra of noise produced by glutamate in outside-out patches from ''small noise'' cells were similar to spectra of noise produced in patches by kainate. 9. Our results indicate that granule cells possess at least three types of glutamate-receptor channels. One type has a maximum conductance of approximately 50 pS; it displays multiple open levels and appears to be activated selectively by aspartate and NMDA. Openings with conductances of 8 and 15 pS are produced by kainate and quisqualate in large numbers and it appears unlikely they are sublevels of the 50 pS channels. Channels with an estimated conductance of 140 fS appear to be present at high density in most granule cells. These channels are activated by glutamate and kainate, but not by aspartate or NMDA to any significant extent.