Potassium accumulation around individual purkinje cells in cerebellar slices from the guinea-pig.

K+-selective micropipettes were used to measure external K+ concentration [( K+]o) in the immediate vicinity of Purkinje cells in slices from guinea-pig cerebellum. The cells were either spontaneously active or were polarized via a separate intracellular micro-electrode. The level of [K+]o rose by 1-3 mM around the soma and dendrites of Purkinje cells during spike activity. The increases in [K+]o were usually greater during Ca2+-mediated spikes than during Na+-mediated spikes. This was even true at the soma where the Ca2+ spike only invaded electrotonically from the dendrites, in contrast to the Na+ spikes which were generated at the soma. No [K+]o changes were seen in the vicinity of Purkinje cells when the cells were hyperpolarized with current passage nor was any [K+]o change seen during subthreshold depolarizations. In glial cells, however, a hyperpolarizing current reduced [K+]o while a depolarizing current increased [K+]o in a symmetrical manner. When Ba2+ was substituted for Ca2+ in the bathing Ringer solution, prolonged plateau-potential spikes could be evoked from Purkinje cells. These spikes were accompanied by large [K+]o elevations but the plateau potentials outlasted the [K+]o elevations. These experiments suggest that large [K+]o increases can occur in the absence of Ca2+-mediated K+ conductances. Substitution of Mn2+ for Ca2+ in the Ringer solution removed some of the [K+]o increases at the Purkinje cell soma. Addition of tetrodotoxin to normal Ringer solution also reduced, but did not abolish the [K+]o increases at the soma. These experiments confirmed that both Na+ and Ca2+ spikes were involved in the [K+]o change. The diffusion characteristics of the slices were determined by an ionophoretic method using tetramethylammonium and ion-selective micropipettes. The extracellular volume fraction of the slice averaged 0.28 while the tortuosity averaged 1.84. These values were close to those found previously in the intact rat cerebellum. These data were used to make quantitative estimates of the expected [K+]o accumulation in the vicinity of a single cell (see Appendix). Such estimates showed reasonable agreement with the measured values. Our data show that quite large increases in [K+]o may occur around single Purkinje cells. Such increases have previously only been evident during the activation of cell populations in mammalian preparations. The present results are probably due to the superior recording conditions of the slice. Implications for intercellular communication are discussed.