The inhibition of single N‐methyl‐D‐aspartate‐activated channels by zinc ions on cultured rat neurones.

Abstract

1. Single channels activated by N-methyl-D-aspartate (NMDA) were studied in outside-out patches of cultured hippocampal neurones in the presence of glycine and absence of magnesium. The effects of the transition metal ions zinc and cadmium on NMDA channels were tested by placing the membrane patch at the mouth of one of an array of large barrelled flow pipes. 2. Amplitude histograms revealed several conductance levels between 5 and 45 pS with the majority of NMDA-activated openings > 25 pS. Zinc (5-100 .mu.M) and cadmium (30-100 .mu.M) reduced the number of large conductance events in a voltage-independent manner. Zinc (30 .mu.M) reduced the large conductance openings by approximately 70-80%. The small number of events under 20 pS precluded quantitative assessment of the effects of zinc and cadmium on these conductance levels. Zinc inhibition of the calculated macroscopic due to NMDA-activated channels could be fitted with a single binding site isotherm with an IC50 of 12 .mu.M. 3. Zinc and cadmium also reduced the mean open time of the two largest conductance events of 38 and 43 pS; this reduction was voltage independent. Open-time histograms were fitted with the sum of two exponentials. In the presence of 5 .mu.M-NMDA at -60 mV, .tau.o2 = 10.49 ms and .tau.o1= 1.47 ms; in 30 .mu.M-zinc, .tau.o2 = 3.49 and .tau.o1 = 0.8 ms. The ''blocking'' rate constant calculated at a membrane potential of +40 mV from the slope of 1/.sigma.o2 vs. [zinc]o was 4 .times. 106 M-1 s-1. 4. Closed-time analysis revealed brief (.tau.c = 0.4-1.0 ms) zinc-insensitive gaps; longer closed-time intervals were not analysed since all patches contained more than one channel. Both burst duration and the number of bursts were reduced in the presence of zinc. 5. At holding potentials negative to -40 mV in magnesium-free solutions, zinc also induced high-frequency flickering of the open channel which included complete channel closures at 4 kHz filtering. No zinc-induced flickering was seen at positive membrane potentials. The flickering was dose dependent, becoming prominent at zinc concentrations above 30 .mu.M. Cadmium did not induce flickering at concentrations up to 100 .mu.M. 6. Increasing the extracellular concentration of either sodium (from 160 to 500 mM) or calcium (from 2 to 30 mM) had no effect on the mean lifetime of the 43 and 38 pS events. In 30 mM-external calcium or 500 mM-external sodium, zinc (30 .mu.M) still reduced the number of openings and the man open time; however, the zinc-induced flickering was much less apparent, suggesting that the flickering was due to zinc binding within the open channel. The difference between zinc and cadmium in evoking high-frequency flicker further supports the correlation between open-channel block of NMDA channels by divalent cations and their water substitution rates. 7. These results demonstrate that the major action of zinc on single NMDA channels is to reduce the mean open time and the frequency of large conductance openings in a voltage-independent manner. This accounts for the inhibition of the macroscopic current and is likely to result from a binding site on the extracellular domain of the receptor-channel complex. We suggest that zinc acts as an allosteric inhibitor at a co-ordination site containing both cysteine and histidine residues, perhaps analogous to the binding site of zinc-dependent DNA-binding proteins and enzymes.