Much of the energy used by the brain is expended on the Na⁺/K⁺ pump to restore ion gradients that are run down by neuronal activity (Attwell & Laughlin, 2001). When cells become highly loaded with Na⁺ as a result of synaptic or action potentials, they may need to express more Na⁺/K⁺ pump subunits, or subunits of a lower Na⁺ affinity, in order to maintain adequate transmembrane gradients of Na⁺ and K⁺. Exposure of cultured neurons and glia to glutamate has been shown to upregulate expression of Na⁺ pump subunits (Brines & Robbins, 1993; Inoue et al. 1999).

We examined modulation of Na⁺/K⁺ pump activity in papain-isolated neurons (Purkinje cells from rat cerebellar slices; Billups & Attwell, 1996) and glia (Mƒller cells from Ambystoma tigrinum retinae; Brew & Attwell, 1987). (Animals were killed humanely in accordance with Home Office regulations.) After soaking in normal external (low K⁺) solution containing either 200 µM or 0 µM glutamate, cells were whole-cell clamped with an internal solution containing 20 mM Na⁺ (glia) or 60 mM Na⁺ (neurons), in external solution containing 12 mM K⁺, to activate the Na⁺/K⁺ pump. Strophanthidin (50 µM) was used to block the pump, generating an inward current proportional to the rate of pumping. Data were compared between cells by normalizing to cell capacitance, to compensate for differences in cell membrane area.

In salamander glia strophanthidin evoked an inward current of ~20 pA at -20 mV, which was larger at more positive potentials. Soaking in glutamate for 45 min, which will load the cells with Na⁺ because of co-transport of 3 Na⁺ with each glutamate anion by the glutamate transporters these cells express (Brew & Attwell, 1987), did not influence the size of the current (177 ± 20 pA nF^-1 (mean ± S.E.M.) for 17 control cells, and 179 ± 15 pA nF^-1 for 23 glutamate-soaked cells, P = 0.95 by two-tailed t test).

In Purkinje cells strophanthidin evoked an inward current of ~10 pA at 0 mV. Soaking in glutamate for 30 min, to Na⁺-load the cells by activating AMPA receptors, increased the pump current by 53 %, from 253 ± 36 pA nF^-1 in 14 control cells, to 385 ± 35 pA nF^-1 in 15 glutamate-soaked cells (P = 0.013).

Further work is needed to ascertain the mechanism of the glutamate-evoked upregulation of neuronal pump rate, and to investigate why a similar upregulation is not seen in glia.

This work was supported by The Wellcome Trust. Jenny Pennack is an M Biochem undergraduate at Bath University.