Metabolism regulates the spontaneous firing of substantia nigra pars reticulata neurons via KATP and nonselective cation channels

Abstract

Neurons use glucose to fuel glycolysis and provide substrates for mitochondrial respiration, but neurons can also use alternative fuels that bypass glycolysis and feed directly into mitochondria. To determine whether neuronal pacemaking depends on active glucose metabolism, we switched the metabolic fuel from glucose to alternative fuels, lactate or β-hydroxybutyrate, while monitoring the spontaneous firing of GABAergic neurons in mouse substantia nigra pars reticulata (SNr) brain slices. We found that alternative fuels, in the absence of glucose, sustained SNr spontaneous firing at basal rates, but glycolysis may still be supported by glycogen in the absence of glucose. To prevent any glycogen-fueled glycolysis, we directly inhibited glycolysis using either 2-deoxyglucose or iodoacetic acid. Inhibiting glycolysis in the presence of alternative fuels lowered SNr firing to a slower sustained firing rate. Surprisingly, we found that the decrease in SNr firing was not mediated by ATP-sensitive potassium (KATP) channel activity, but if we lowered the perfusion flow rate or omitted the alternative fuel, KATP channels were activated and could silence SNr firing. The KATP-independent slowing of SNr firing that occurred with glycolytic inhibition in the presence of alternative fuels was consistent with a decrease in a nonselective cationic conductance. Although mitochondrial metabolism alone can prevent severe energy deprivation and KATP channel activation in SNr neurons, active glucose metabolism appears important for keeping open a class of ion channels that is crucial for the high spontaneous firing rate of SNr neurons.

Keywords: KATP; Trp channel; excitability; glycolysis.

Figure 1.

The ketone body βHB sustains spontaneous SNr firing in the absence of glucose, but glycolytic inhibition slows the firing rate. A, SNr spontaneous firing rate in the presence of glucose (G; 10 mm; 31.4 ± 3.2 spikes/s; black symbols, n = 10). Spontaneous firing rate (29.3 ± 4.5 spikes/s; blue symbols; n = 8) recorded from neurons in the absence of glucose but in the presence of βHB (2.5 mm) for at least 20 min was not significantly different from the firing rate in glucose (p > 0.05, one-way ANOVA with Bonferroni's test). In the absence of external fuel (0 mm glucose; 0 mm βHB), SNr spontaneous firing was almost completely silent (0.9 ± 0.4 spikes/s; red symbols; n = 10). B, When glycolysis was inhibited using 2-DG (5 mm) in the absence of glucose and βHB (black line), the spontaneous firing of an SNr neuron decreased and then sharply dropped until firing was silenced. Return of glucose in the external solution rapidly restored the spontaneous firing. In the presence of βHB (2.5 mm; blue line), glycolytic inhibition with 2-DG (5 mm) in the absence of glucose decreased the spontaneous firing of a SNr neuron but did not silence it. C, When glycolysis was inhibited with IAA (1 mm) in glucose solution (10 mm; black trace), the firing rate of a representative SNr neuron decreased and then sharply stoped firing action potentials. However, in the presence of the ketone body βHB (2.5 mm; blue trace) and absence of glucose, the firing of a representative SNr neuron was sustained after glycolytic inhibition with IAA at a lower firing rate. D, Average firing rate of SNr neurons in control (Ctrl) and test conditions. The initial firing rate of SNr neurons in glucose (10 mm) was 32.7 ± 2.9 spikes/s (black squares and line; n = 12), but after removal of glucose and addition of 2-DG (5 mm), the firing rate was almost completely silenced within 15 min (0.8 ± 0.3 spikes/s). The firing rate of SNr neurons in the absence of glucose but in the presence of βHB (2.5 mm) was 33.1 ± 3.7 spikes/s (blue square and line). After addition of 2-DG (5 mm or 10 mm), the firing rate of those SNr neurons was significantly reduced but not silenced (22.2 ± 2.6 spikes/s; n = 13; p = 0.0001, Student's paired t test). Addition of IAA (1 mm) to spontaneously firing SNr neurons (37.7 ± 9.7 spikes/s; n = 6; black circles and line) in glucose (10 mm) completely silenced SNr firing. When exogenous βHB (2.5 mm; blue circles and line) replaced glucose, the spontaneous firing of SNr neurons (37.2 ± 2.7 spikes/s) was significantly decreased after addition of IAA (1 mm) but was not silenced (17.8 ± 1.4 spikes/s; n = 20; p = 4.8 × 10⁻⁷, Student's paired t test). Similarly, when lactate (5 mm; n = 5; red circles and line) replaced glucose, SNr spontaneous firing (39.8 ± 4.5 spikes/s) was decreased after addition of IAA (21.6 ± 3.4 spikes/s; p = 0.002, Student's paired t test). The mitochondrial poisons rotenone (Rot; 1 μm) and oligomycin (Oligo; 1 μm) silenced SNr firing in glucose solution (33.7 ± 3.0 vs 0.8 ± 0.5 spikes/s; n = 6; black diamonds and line). In the presence of rotenone (1 μm) and oligomycin (1 μm), βHB (2.5 mm) did not sustain SNr firing after treatment with IAA (40.2 ± 7.4 vs 0.2 ± 0.2 spikes/s; n = 4; blue diamonds and line). E, Replacement of glucose with 2-DG and βHB decreased SNr firing by 32.3 ± 5.0% (n = 13), which was significantly less than the decrease observed when using IAA to inhibit glycolysis (p < 0.05, one-way ANOVA with Bonferroni's test). Inhibition of glycolysis with IAA was performed with βHB either in the absence of glucose (blue symbols) or with 10 mm glucose (red symbols). The percentage decrease in firing rate was not significantly different (p > 0.05, one-way ANOVA with Bonferroni's test) between experiments without glucose (49.5 ± 3.4%; n = 20) or while maintaining glucose (64.2 ± 5.8%; n = 12). In the presence of the antioxidant Tempol (2 mm), the percentage decrease in firing rate (64.2 ± 5.2%; n = 8) after inhibition of glycolysis with IAA in the presence of βHB and glucose was not significantly different from control experiments without Tempol. All error bars indicate SEM; *p < 0.05.

Figure 2.

K_ATP channel activation after glycolytic inhibition is conditional on the perfusion flow rate and the presence of mitochondrial fuels. A, The spontaneous firing rate of SNr neurons, recorded in the continuous presence of the K_ATP channel blocker Glib (200 nm), was significantly decreased (circle symbols; 28.8 ± 4.2 vs 19.3 ± 2.5 spikes/s; n = 8; p = 0.001, Student's paired t test) after inhibition of glycolysis with 2-DG (5 mm) in the absence of glucose but in the presence of βHB (2.5 mm). The decreased firing rate after inhibition of glycolysis with IAA (1 mm) in the presence of βHB (2.5 mm) was not reversed after addition of Glib (10 μm; square symbols; 18.2 ± 1.7 vs 17.0 ± 1.8 spikes/s; n = 6; p > 0.05, one-way ANOVA with Bonferroni's test). IAA (1 mm) significantly decreased the firing rate of Kir6.2 KO SNr neurons (triangle symbols; 33.3 ± 3.6 vs 15.7 ± 1.6 spikes/s; n = 8; p = 0.0003, Student's paired t test). B, With a flow rate of 5 ml/min, βHB (2.5 mm) sustained the spontaneous firing of an SNr neuron after inhibition of glycolysis with IAA (1 mm). Further addition of the K_ATP channel blocker Glib (10 μm) did not reverse the decrease in firing rate. a–c, Traces depict cell-attached recordings of spontaneous firing at the indicated times (calibration: 50 pA, 200 ms). C, With a lower flow rate of 1 ml/min, βHB (2.5 mm) was unable to sustain the firing rate of an SNr neuron. Addition of Glib (200 nm) could partially restore the firing rate. a–c, Traces depict cell-attached recordings of spontaneous firing at the indicated times (calibration: 20 pA, 200 ms). D, Cell-attached recordings of spontaneous firing rates with inhibition of glycolysis by IAA (1 mm) in the presence of glucose (10 mm). IAA completely silenced SNr firing of control neurons (n = 6; black trace). When K_ATP channels were inhibited using Glib (200 nm; 10 min preincubation; n = 4; blue trace) or eliminated in Kir6.2 KO mice (n = 6; red trace), SNr firing displayed a transient increase, followed by a complete silencing. E, Representative whole-cell recordings showing the time course of the effect of IAA (1 mm) application on the normalized firing rate of SNr neurons in the presence of glucose. In a control neuron, application of IAA promptly decreased the spontaneous firing rate without any increase in firing rate (black line). In a neuron preincubated in Glib (200 nm; blue line) or in a neuron from a Kir6.2 KO mouse (red line), the firing rate increased after application of IAA and then stopped firing. F, Summarized data from all whole-cell experiments with application of IAA in the presence of glucose. After application of IAA, control neurons had a hyperpolarized resting potential (−74.6 ± 3.5 mV; n = 9; black symbols). After IAA, neurons preincubated in Glib (200 nm; blue symbols) had more depolarized resting potentials (−60.5 ± 2.7 mV; n = 6; p < 0.05, one-way ANOVA with Bonferroni's test), and neurons from Kir6.2 KO animals (red symbols) also rested more depolarized (−53.2 ± 3.8 mV; n = 6; p < 0.05, one-way ANOVA with Bonferroni's test). All error bars indicate SEM; *p < 0.05. Ctrl, Control.

Figure 3.

Glycolytic inhibition hyperpolarizes the membrane potential by decreasing a constitutively active conductance. A, Action potentials were recorded immediately after establishing a whole-cell recording from control (Ctrl) neurons or neurons preincubated in IAA (1 mm) and βHB (2.5 mm). Representative traces from two separate neurons, one in the control condition (black line) and one preincubated in IAA and βHB (blue line), show that IAA increases interspike intervals. B, The action potential waveform of neurons incubated in IAA in the presence of βHB (n = 7; blue trace) was similar to action potentials from control neurons (n = 5; black trace) but did have a more prominent afterhyperpolarization (calibration: 20 mV, 5 ms). C, With action potentials blocked using lidocaine (1 mm) and with the K_ATP channel blocker Glib (200 nm) present, application of IAA (1 mm) in the continued presence of βHB (2.5 mm) decreased the membrane potential of SNr neurons recorded in perforated-patch configuration (−53.8 ± 2.0 to −67.5 ± 1.3 mV; n = 8; p = 3.0 × 10⁻⁵, Student's paired t test). D, Perforated-patch voltage-clamp recordings (holding potential of −70 mV) of SNr neurons in the presence of lidocaine (1 mm) exhibited a decrease in inward current after application of IAA (1 mm) with βHB (2.5 mm; −41.9 ± 7.4 to −6.1 ± 3.7 pA; n = 4; p = 0.0009, Student's paired t test). All error bars indicate SEM; *p < 0.05.

Figure 4.

Glycolytic inhibition decreases a nonselective cationic current. A, Steady-state I–V relationship of control (Ctrl) neurons (n = 10; black filled squares) versus neurons preincubated (>10 min) in IAA (1 mm) and βHB (2.5 mm; n = 10; blue squares) over the voltage range from −100 to −10 mV. B, Steady-state currents were measured in the presence of barium chloride (1 mm) in the control condition (10 mm glucose; black diamonds; n = 10) and in neurons preincubated with IAA and βHB (blue diamonds; n = 10). Neurons in IAA and βHB had decreased steady-state inward current. C, With lowered external sodium (L.S.), the steady-state I–V relationship of control neurons (n = 10; black circles) was similar to that of neurons preincubated in IAA and βHB in lowered sodium (n = 9; blue circles). For comparison, the I–V plot of control neurons in standard sodium condition is shown (gray squares). D, Steady-state I–V relationship was similar for neurons in IAA and βHB (blue squares) and neurons preincubated (>10 min) in the nonspecific TRP channel blocker 2-APB (200 μm; n = 6; red squares). Both I–V plots from neurons in IAA and βHB and neurons in 2-APB were different than the control neurons (gray squares). E, The steady-state current (pA) at −50 mV of control neurons (−66.0 ± 16.7; n = 10) was significantly different from neurons preincubated in IAA and βHB (54.9 ± 10.7; n = 10; p < 0.05). The steady-state current at −50 mV from neurons preincubated in 2-APB (200 μm) was also significantly different from control neurons (57.9 ± 22.3; n = 6; p < 0.05) but not significantly different from neurons in IAA and βHB. Neurons in low external sodium (27 mm NaCl) had an average steady-state current at −50 mV that was also significantly different from control neurons (95.9 ± 10.3; n = 10; p < 0.05). In the low sodium condition, IAA in the presence of βHB did not significantly alter steady-state current at −50 mV compared with control neurons in low sodium (112.3 ± 20.4; n = 9; p > 0.05). Significance of pairwise comparisons at the p < 0.05 level was determined by one-way ANOVA with Bonferroni's test. F, The steady-state current (pA) at −50 mV in the presence of barium from control neurons (−205.7 ± 14.8; n = 10) was significantly different from neurons preincubated in IAA and βHB (−73.7 ± 17.4; n = 10; p = 1.7 × 10⁻⁵, Student's unpaired t test). The decrease in inward current at −50 mV generated by IAA and βHB in the presence of barium (132.0 ± 22.8 pA) was similar in magnitude as the decreased observed without barium (120.5 ± 19.8 pA). *p < 0.05. Comparisons that are not significantly different (p > 0.05) are also indicated (n.s.). All error bars indicate SEM. All experiments were performed in the presence of lidocaine (1 mm) and TEA (1 mm).

Figure 5.

The reduction in firing rate produced by glycolytic inhibition does not require TRPC channels. A, Loose-patch cell-attached recording of a TRPC3 KO SNr neuron showed a reduction in firing rate after application of IAA (1 mm) in the presence of βHB (3 mm). a, b, Traces depict cell-attached recordings of spontaneous firing before and after application of IAA in the presence of βHB (calibration: 50 pA, 200 ms). B, Glycolytic inhibition in the presence of βHB (2.5 or 3 mm) decreased the spontaneous firing rate of TRPC3 KO SNr neurons (28.6 ± 4.3 to 11.2 ± 2.8 spikes/s; n = 10; p = 6 × 10⁻⁵, Student's paired t test). C, SNr neurons lacking all seven TRPC channels are spontaneously active. Inhibition of glycolysis with IAA in the presence of βHB (2.5 mm) significantly reduces the firing rate of these neurons (28.7 ± 3.4 to 12.6 ± 1.8 spikes/s; n = 7; p = 0.001, Student's paired t test). Ctrl, Control.