Identification of GABA neurons in rat cortical cultures by GABA uptake autoradiography.

Autoradiographic studies of rat cortical cultures were conducted with tritiated transmitters and related drugs. Autoradiographs prepared from cultures incubated in [3H]GABA showed selective labeling: dense accumulations of silver grains over the somas and all processes of approximately 30-50% of the neuronal population, few grains over the non-neuronal cells. This labeling was blocked by diaminobutyric acid (DABA) and sodium-free media but not by beta-alanine and thus has the characteristics of GABA uptake in other neuronal systems. There were no obvious differences in the size, shape, number of processes or distribution in the culture between neurons which accumulated GABA and those which did not. Similar cultures incubated in either [3H]glycine or [3H]glutamate and processed by autoradiography resulted in a much different distribution of silver grains than that seen for [3H]GABA. Following incubation in [3H]glycine, silver grains were distributed uniformly over all cells in the culture, both neuronal and non-neuronal. This distribution suggests a metabolic and not a neurotransmitter role for glycine in the cultures, as would be expected of neuronal cells derived from cerebral cortex. Glutamate incubations resulted in the appearance of silver grains over only the non-neuronal cells with very few over the neuronal population. Autoradiograms were also prepared following incubation in the potent GABA receptor agonist [3H]muscimol. These autoradiograms were indistinguishable from those obtained following [3H]GABA incubation. Thus, a finite population of neurons was densely labeled, the labeling was blocked by the GABA uptake inhibitors DABA, nipecotic acid, guvacine and Na+-free media, while substances which interact with the GABA receptor, bicuculline methiodide, THIP, isoguvacine and the noncompetitive antagonist, picrotoxin, were without effect. These results demonstrate that the affinity of muscimol for the GABA uptake site far outweighs its affinity for the GABA receptor site in autoradiographic experiments where intact cells are employed, presumably because its binding to receptors is fleeting. Therefore, muscimol autoradiography may not be informative about GABA receptor localization. These autoradiographic studies suggest that nearly half the neurons in our culture system are GABA neurons but disclosed no morphological handle for GABA neurons.