Abstracts

RATIONALE: Benzodiazepines are among the first medications used to treat neonatal seizures. While benzodiazepines effectively potentiate [gamma]-aminobutyric acid[sub]A[/sub] receptors (GABA[sub]A[/sub]R) in adult neurons, they are much less effective in potentiating GABA[sub]A[/sub]R in neonatal neurons. Thus, benzodiazepines may have additional sites of action in the neonatal brain. The inhibitory glycine receptor (GlyR) is a candidate site based on the ability of benzodiazepines to inhibit strychnine binding. In this study, we examined the effect of benzodiazepines on GlyR mediated currents in cultured embryonic mouse hippocampal neurons.
METHODS: Hippocampal neurons were cultured from embryonic day 15 Swiss Webster mice using standard techniques, which were approved by the Washington University Animal Studies Committee. GlyR and GABA[sub]A[/sub]R mediated currents were studied using the whole-cell patch clamp technique in neurons cultured for 4-14 days.
RESULTS: Glycine evoked a dose-dependent, strychnine sensitive, chloride current when applied to neurons voltage-clamped at -65mV. At 20-100[mu]M, chlordiazepoxide, nitrazepam, lorazepam, alprazolam, and triazolam inhibited 50[mu]M glycine ([italic]EC[sub]50[/sub][/italic]) currents by 20% or more. In contrast, 100[mu]M diazepam and 100[mu]M clobazam inhibited 50[mu]M glycine currents by 15% or less. Chlordiazepoxide and nitrazpam were studied in further detail. Chlordiazepoxide completely blocked 50[mu]M glycine currents with an [italic]IC[sub]50[/sub][/italic] of 230 [plusminus] 52[mu]M. Nitrazepam completely blocked 20[mu]M ([italic]EC[sub]20[/sub][/italic]) and 50[mu]M glycine currents with an [italic]IC[sub]50[/sub][/italic] of 100 [plusminus] 5[mu]M and 100 [plusminus] 29[mu]M, respectively. 80[mu]M nitrazepam increased the [italic]EC[sub]50[/sub][/italic] for the glycine dose-response curve from 52 [plusminus] 3[mu]M to 120 [plusminus] 10[mu]M, increased the Hill coefficient from 1.8 [plusminus] 0.2 to 3.7 [plusminus] 1.0, and decreased the maximum response by 22%. Nitrazepam block was not voltage dependent and could not be prevented by 10[mu]M flumazenil. Based on noise analysis, 50[mu]M nitrazepam decreased the mean channel burst duration for 10[mu]M glycine from 67 [plusminus] 4ms to 32 [plusminus] 4ms. In addition, nitrazepam increased the GlyR single channel conductance from 3.9 [plusminus] 0.9pS to 8.3 [plusminus] 1.6pS. Although 100[mu]M nitrazepam significantly prolonged the fast time constant of desensitization for 300[mu]M glycine ([italic]EC[sub]100[/sub][/italic]) currents from 0.9 [plusminus] 0.2s to 1.6 [plusminus] 0.2s, it did not alter other parameters of desensitization. Specifically, it did not alter the slow time constant of onset, the weighted mean time constant of onset, the relative contributions of the fast and slow components, the degree of desensitization, or the weighted mean time constant for recovery from desensitization for 300[mu]M glycine currents. Similarly, 50[mu]M nitrazepam did not alter desensitizing currents elicited by 10[mu]M glycine. We confirmed that benzodiazepines do not consistently potentiate GABA[sub]A[/sub]R currents in this preparation.
CONCLUSIONS: We conclude that benzodiazepines are noncompetitive inhibitors of GlyR in embryonic hippocampal neurons. The findings exclude mechanisms involving simple open channel block and enhanced desensitization. The inhibition results in part from impaired channel gating, though an effect on binding cannot be excluded. Since GlyR mediate a depolarizing response in neonatal neurons, their inhibition by benzodiazepines may be important in treating neonatal seizures.
[Supported by: NIH]