Abstracts

Rationale: Temporal lobe epilepsy is associated with changes in the expression of GABA(A) receptor subunits that are presumed to be pathologic. Some changes could instead be homeostatic and compensatory, and still other changes could contribute paradoxically depending on cell-type expression and other factors. We sought to explore the role of individual GABA(A) receptor subunits in modulating activity in dissociated neuronal cultures by comparing spontaneously bursting and quiescent cultures. Methods: We used multi-electrode array (MEA) recordings to determine the firing characteristics of hippocampal neurons grown under different conditions. Neurons grown in the absence of a glial-feeder layer exhibited increased spontaneous bursting compared to cultures grown on a glial feeder layer. We hypothesized that differences in phasic and/or tonic GABAergic inhibition might contribute to this difference, so we compared expression of alpha1, alpha4 and alpha5 subunits using quantitative RT-PCR. We then used MEA recordings in bursting cultures to determine the effects of zolpidem and furosemide, agents that have relative selectivity for alpha1 (phasic) and alpha4-containing (tonic) GABA(A) receptors, respectively, on overall firing rate, number of bursts and firing rate during bursts. Results were scaled so that the baseline response was considered to be 100%. Results: The ratio of alpha1 subunit expression in quiescent cultures compared to bursting cultures was 1.3 ± 0.16 (n=11). The ratio of alpha4 subunit expression was 0.52 ± 0.066 (n=10), and the ratio of alpha5 subunit expression was 1.7 ± 0.34 (n=11). Increases in alpha1 and alpha5 subunit expression seem consistent with differences in neuronal activity under the two culture conditions. However, it is not immediately clear how decreased alpha4 subunit expression would contribute to diminished neuronal activity. One hypothesis is that this represents a compensatory response to neuronal quiescence. Alternatively alpha4 manipulations may have paradoxical effects on overall network activity. To test these hypotheses, we examined the effects of drugs with relative selectivity for GABA(A) receptor subtypes. We found that 500 nM zolpidem, which potentiates alpha1-containing receptors, decreased the overall firing rate to 60% ± 8.2% of baseline (n = 3) and decreased the number of bursts to 28% ± 9.8% of baseline. Interestingly, the firing rate during bursts was not significantly affected by zolpidem (100% ± 5.9% of baseline). We then examined the alpha4 subunit antagonist furosemide (500 µM). Surprisingly, furosemide decreased the overall firing rate to 61% ± 9.7% of baseline (n = 3) and decreased the number of bursts to 66% ± 20% of baseline. In contrast to zolpidem, which had no effect on the firing rate during bursts, furosemide increased the firing rate during bursts to 124% ± 8.5% of baseline. Conclusions: These findings suggest that increased expression of alpha4 subunits, which should result in increased tonic inhibition, may paradoxically contribute to spontaneous neuronal bursting, possibly by inhibiting inhibitory interneurons.