Modification of potassium movement through the retina of the drone (Apis mellifera male) by glial uptake.

Intracellular recordings were made in photoreceptors and glial cells (outer pigment cells) of the superfused cut head of the honey-bee drone (Apis mellifera male). When the [K+] in the superfusate was abruptly increased from 3.2 mM to 17.9 mM both photoreceptors and glial cells depolarized. The time course of the depolarization of the photoreceptors was slower with increasing depth from the surface. Half time of depolarization was plotted against depth: this graph was compatible with the arrival of K+ being exclusively by diffusion through the extracellular clefts. However, as we then showed, this interpretation is inadequate. The time course of depolarization of the glial cells was almost the same at all depths. This indicates that they are electrically coupled. Consequently, current-mediated K+ flux (spatial buffering) through glial cells will contribute to the transport of K+ through the tissue: K+ ions enter the glial syncytium in the region of high external potassium concentration, [K+]0, and an equivalent quantity of K+ ions leave in regions of low [K+]0. Intracellular K+ activity (aiK) was measured with double-barrelled K+-sensitive micro-electrodes in slices of retina superfused on both faces. When [K+] in the superfusate was increased from 7.5 mM to 17.9 mM an increase in aiK was observed in glial cells at all depths in the slice (initial rate 1.7 mM min-1, S.E. of the mean = 0.2 mM min-1), but there was little increase in the photoreceptors (0.3 +/- 0.2 mM min-1). The increase in aiK in glial cells near the centre of the slice could not have been caused by spatial buffering; it presumably resulted from net uptake. We conclude that when [K+] is increased at the surface of this tissue, the build up of K+ in the extracellular clefts depends on extracellular diffusion, spatial buffering and net uptake. The latter two processes, which have opposing effects, involve about 10 times as much K+ as the first. This is in rough agreement with less direct experiments on mammalian brain (Gardner-Medwin, 1977, 1983b).