Glucose deprivation activates diversity of potassium channels in cultured rat hippocampal neurons

Abstract

1. Glucose is one of the most important substrates for generating metabolic energy required for the maintenance of cellular functions. Glucose-mediated changes in neuronal firing pattern have been observed in the central nervous system of mammals. K(+) channels directly regulated by intracellular ATP have been postulated as a linkage between cellular energetic metabolism and excitability; the functional roles ascribed to these channels include glucose-sensing to regulate energy homeostasis and neuroprotection under energy depletion conditions. The hippocampus is highly sensitive to metabolic insults and is the brain region most sensitive to ischemic damage. Because the identity of metabolically regulated potassium channels present in hippocampal neurons is obscure, we decided to study the biophysical properties of glucose-sensitive potassium channels in hippocampal neurons. 2. The dependence of membrane potential and the sensitivity of potassium channels to glucose and ATP in rat hippocampal neurons were studied in cell-attached and excised inside-out membrane patches. 3. We found that under hypoglycemic conditions, at least three types of potassium channels were activated; their unitary conductance values were 37, 147, and 241 pS in symmetrical K(+), and they were sensitive to ATP. For K(+) channels with unitary conductance of 37 and 241, when the membrane potential was depolarized the longer closed time constant diminished and this produced an increase in the open-state probability; nevertheless, the 147-pS channels were not voltage-dependent. 4. We propose that neuronal glucose-sensitive K(+) channels in rat hippocampus include subtypes of ATP-sensitive channels with a potential role in neuroprotection during short-term or prolonged metabolic stress.

Fig. 1.

Effect of glucose on potassium channel activity. (A, B) Unitary activity records obtained in the cell-attached configuration from two different neurons at resting potential (V _P=0 mV), in Ringer solution, without glucose, and then perfused again with Ringer solution. Pipette solution high-K (see Table I). Closed (c) current level is indicated to the left of the traces. (C) Current–voltage relationships for the two types of K⁺ channel in the cell-attached configuration. Straight lines represent the experimental data fitting using the linear equation. Data are presented as means ± SEM (n=2,• 1, ▪)

Fig. 2.

Potassium channels in the inside-out patches. (A, B) Single-channel activity in the inside-out configuration obtained in the absence of glucose from one patch of a cultured neuron, displayed at different time scales (V _m=40 mV). The three (0–2) current levels are indicated to the left of the traces. Pipettes were filled with high K solution. Membrane patches were perfused with internal solution. (C) Current–voltage relationships for the three types of K⁺ channel found in the inside-out configuration. Straight lines represent the experimental data fitting using the linear equation. Data are presented as means ± SEM (n=3, ▴; 5, •; 2, ▪).

Fig. 3.

ATP blockade K⁺ channels. Single-channel activity for SK (A), IK (B), and LK (C) channels obtained under control conditions, in the presence of ATP (1 mM) in the intracellular face of the patch, and after washout (V _m=40 mV). Pipettes were filled with high K solution. Membrane patches were perfused with internal solution.

Fig. 4.

Voltage-dependence and kinetic properties of the SK and LK channels in symmetrical K⁺ concentration. Steady state open probability (P _o) of membrane potential (V _m) values for the SK (A) and LK (B) channels. The curve lines correspond to Boltzmann fits for experimental values of P _o. Open (τ_o, ▴) and closed (τ_c1, ▪; τ_c2, •) time constants for the SK (C) and LK (D) channels to different membrane potentials. Pipettes were filled with high K solution. Membrane patches were perfused with internal solution.