Abstract
1. The effect of gamma-aminobutyric acid (GABA) on extracellular pH (pHo) was investigated in the turtle cerebellum, in vitro, using double-barrelled, H(+)-selective microelectrodes. Responses evoked by GABA were compared with pHo shifts evoked by repetitive stimulation of the parallel fibres. 2. In media buffered with 35 mM-HCO3- and 5% CO2, superfusion of GABA (1 mM) elicited an abrupt alkaline shift in the molecular layer, which averaged 0.05 +/- 0.02 pH units (+/- S.D., range 0.02-0.12 pH units). pHo often recovered in the continued presence of GABA, and displayed a rebound acidification upon wash-out. 3. The GABA-evoked alkaline shift was blocked by picrotoxin and was mimicked by the GABAA agonists isoguvacine and muscimol. The GABAB agonist baclofen did not elicit an alkaline shift. Alkaline shifts evoked by stimulation of the parallel fibres were unaffected by picrotoxin. 4. In nominally HCO3(-)-free solutions, buffered with 35 mM-HEPES, superfusion of GABA caused either no pHo change or a slow acid shift. In contrast, the alkaline shift evoked by stimulation of the parallel fibres became enhanced in HEPES-buffered media. 5. The alkaline shift evoked by GABA was accompanied by an increase in extracellular K+ ([K+]o) which averaged 1.7 mM above baseline. Experimental elevation of [K+]o to a comparable level always caused a pure acid shift in the extracellular space. 6. The GABA-evoked alkaline shift persisted when synaptic transmission was blocked using 4 mM-kynurenic acid or saline prepared with nominally zero Ca2+ and 10 mM-Mg2+. The alkaline shift evoked by repetitive stimulation of the parallel fibres was completely abolished in these media. 7. Although the GABA-evoked alkaline shift was blocked in nominally HCO3(-)-free media, substitution of 35 mM-formate for HCO3- restored the GABA response. Superfusion of 1 mM-GABA in formate saline produced an alkaline shift of 0.040 +/- 0.034 pH units. 8. These results indicate that gating of GABAA channels in the vertebrate CNS gives rise to an HCO3- efflux which can significantly increase the pH of the brain microenvironment. However, this mechanism cannot account for the extracellular alkalinization caused by parallel fibre stimulation. Extracellular alkaline shifts capable of modulating local synaptic operations may therefore be a consequence of either excitatory or inhibitory synaptic transmission.