Metabolic autocrine regulation of neurons involves cooperation among pannexin hemichannels, adenosine receptors, and KATP channels

Abstract

Metabolic perturbations that decrease or limit blood glucose-such as fasting or adhering to a ketogenic diet-reduce epileptic seizures significantly. To date, the critical links between altered metabolism and decreased neuronal activity remain unknown. More generally, metabolic changes accompany numerous CNS disorders, and the purines ATP and its core molecule adenosine are poised to translate cell energy into altered neuronal activity. Here we show that nonpathological changes in metabolism induce a purinergic autoregulation of hippocampal CA3 pyramidal neuron excitability. During conditions of sufficient intracellular ATP, reducing extracellular glucose induces pannexin-1 hemichannel-mediated ATP release directly from CA3 neurons. This extracellular ATP is dephosphorylated to adenosine, activates neuronal adenosine A(1) receptors, and, unexpectedly, hyperpolarizes neuronal membrane potential via ATP-sensitive K(+) channels. Together, these data delineate an autocrine regulation of neuronal excitability via ATP and adenosine in a seizure-prone subregion of the hippocampus and offer new mechanistic insight into the relationship between decreased glucose and increased seizure threshold. By establishing neuronal ATP release via pannexin hemichannels, and hippocampal adenosine A(1) receptors coupled to ATP-sensitive K(+) channels, we reveal detailed information regarding the relationship between metabolism and neuronal activity and new strategies for adenosine-based therapies in the CNS.

Figure 1.

Reduced [glucose]_e causes hyperpolarization in rat CA3 pyramidal neurons. A, The membrane potential trace (top) and the time course of the changes in the input resistance (bottom) in the CA3 pyramidal neuron recorded with standard (2 mm) ATP in the intracellular solution. Negative currents of 100 pA were applied every 30 s to measure input resistance. The [glucose]_e was switched from 11 to 3 mm at the horizontal bar (reduced glucose). B, C, Summaries of the effect of low-glucose extracellular solution on the membrane potentials (B) and the input resistances (C). *p < 0.05; **p < 0.01; n = 5.

Figure 2.

The effect of different concentrations of intracellular ATP or extracellular glucose on the reduced [glucose]_e-induced outward current in CA3 pyramidal neurons. A, The membrane current traces of four different intracellular ATP concentrations ([ATP]_i = 5, 2, 1, and 0.5 mm) with reducing [glucose]_e from 11 to 3 mm (reduced glucose). The holding potential was −70 mV. B1, B2, Summaries of the significant effect of intracellular ATP concentration on the peak current amplitude (B1) and the amplitude 12 min after changing the glucose concentration (B2) in reduced glucose-induced outward currents ([ATP]_i = 0.5 mm, n = 4; [ATP]_i = 1 mm, n = 5; [ATP]_i = 2 mm, n = 9; [ATP]_i = 5 mm, n = 4). C, Summary of the significant dose-dependency of the effects of reduced [glucose]_e with [ATP]_i = 2 mm (3 mm [glucose]_e, n = 9; 7 mm [glucose]_e, n = 5). D, PSC frequency was decreased significantly by reduced [glucose]_e with a dose-dependent relationship to [ATP]_i. *p < 0.05; **p < 0.01.

Figure 3.

Reduced [glucose]_e-induced outward current is mediated by activation of A₁Rs. A, DPCPX (1 μm) reversed significantly (top) or prevented (bottom) completely the reduced [glucose]_e-induced outward current in CA3 pyramidal neurons recorded from Sprague Dawley rat hippocampal slices. Average data shown in right panel for reversal (top right; n = 5) and prevention (bottom right; n = 4). B, Reduced [glucose]_e caused a significant DPCPX-sensitive outward current in CA3 neurons recorded from wild-type (WT) mouse (top left) but had no effect on CA3 neurons recorded from the A₁R receptor knock-out (A₁R-KO) mouse (bottom left). Average data shown in right panel for WT mice (n = 6) and A₁R-KO mice (n = 8). NS, Not significantly different; **p < 0.01.

Figure 4.

Reduced [glucose]_e opens K⁺ channels in rat CA3 pyramidal neurons. A, I–V relationship of the outward current generated with reduced [glucose]_e. The ramp voltage command (−50 mV to −150 mV at 100 mV/s) was applied before (1) and during (2) reduced [glucose]_e application. The reversal potential (E _rev) of the outward current was −92.75 ± 1.67 mV (n = 4). The I–V relationship showed inward rectification and the average E _rev was similar to the equilibrium potential for K⁺: −92.89 mV. B, BaCl₂ (1 mm) reversed (left trace and right panel; n = 5) or prevented (pretreatment, right panel; n = 4) completely the outward current caused with reduced glucose.

Figure 5.

Postsynaptic K_ATP channels are opened by reduced [glucose]_e. A, Tolbutamide (500 μm) reversed (left trace and right panel; n = 5) or prevented (pretreatment, right panel; n = 4) significantly the outward current caused by reduced glucose. B, Tolbutamide did not prevent (right) or reverse (left traces and middle) the decrease in PSC frequency with reduced glucose. Left traces “1” through “3” are time-extended traces taken at the indicated points in A. NS, Not significantly different; *p < 0.05; **p < 0.01.

Figure 6.

Autocrine regulation due to ATP release through pannexin-1 hemichannels during reduced [glucose]_e. A, With 2 mm intracellular adenosine ([Ado]_i) and 0.5 mm [ATP]_i, reduced [glucose]_e caused a small, transient outward current. The peak amplitude with 2 mm [Ado]_i was significantly smaller than that with 2 mm [ATP]_i (left), and unlike with 2 mm [ATP]_i, no significant current was observed after 15 min (2 mm [ATP]_i, n = 9; 2 mm [Ado]_i and 0.5 mm [ATP]_i, n = 7; right). B, CBX (10 or 100 μm) reversed (top trace and lower left; n = 5 per group) or prevented (lower right; n = 4 per group) significantly the outward current caused with reduced glucose. C, Octanol (1 mm) reversed (top trace and lower left; n = 5) or prevented (lower right; n = 4) significantly the outward current caused with reduced glucose. D, ¹⁰panx (100 μm) reversed (top trace and lower left; n = 5) or prevented (lower right; n = 4) significantly the outward current caused with reduced glucose. NS, Not significantly different; **p < 0.01.

Figure 7.

I–V relationship of the reduced [glucose]_e-induced current. A, The I–V curve generated with reduced [glucose]_e with cesium-based, 2 mm [ATP]_i solution in the presence of tolbutamide (500 μm) in rat CA3 pyramidal neurons. The ramp voltage command (+60 to −120 mV, 180 mV/s) was applied before, during reduced [glucose]_e and after antagonist application. The E _rev was −35.40 ± 1.76 mV (n = 8), and this I–V curve was suppressed significantly by CBX (100 μm, n = 4) or ¹⁰panx (100 μm, n = 4). Ramp responses in baseline and after CBX are largely overlapping; **p < 0.01. B, In rat astrocytes recorded in the stratum radiatum, reduced [glucose]_e causes no current with cesium-based, 2 mm [ATP]_i solution in the presence of tolbutamide (500 μm). The ramp voltage command was as in A. Ramp responses in baseline and after 3 mm glucose are largely overlapping. Right graph shows the summary of the current at 60 mV (n = 4). Cell identification of the astrocytes is shown in supplemental Figure 4, available at www.jneurosci.org as supplemental material.

Figure 8.

Schematic of the purinergic autocrine regulation in CA3 pyramidal neurons. With sufficient [ATP]_i (1), reducing [glucose]_e (2) induces neuronal ATP release directly via pannexin hemichannels (3); released ATP is dephosphorylated to adenosine (4) to activate adenosine A₁ receptors (5) and, ultimately, decrease neuronal excitability by opening K_ATP channels (6).