On-Site Energy Supply at Synapses through Monocarboxylate Transporters Maintains Excitatory Synaptic Transmission

4-CIN exerted only slight effects on membrane potential and action potential amplitude. A, An original trace of the whole-cell membrane current. 4-CIN (1 mm) was applied during the horizontal bar. The current responses to pre-pulse (see Materials and Methods) and stimulation artifacts are nullified by digital sample-and-hold after the experiments. B, Effects of 4-CIN on the holding current. Open red circles, values from each neuron. Horizontal bar and vertical bar on the right, mean value and SEM (n = 9). NS, not significantly different (paired t test). C, Membrane potential traces of a neuron before (black) and during (red) 4-CIN (1 mm) application. A constant depolarizing current of 13.9 pA was injected to induce action potential firing throughout the recording. D, Traces showing action potential waveforms (n = 8) before (black) and during (red) 4-CIN application. E, Summarized results of the effects of 4-CIN on action potential amplitude. **p < 0.01 (Mann–Whitney U test); n = 8. Data are presented as the mean values ± SEM. F, Effects of 4-CIN on interspike membrane potential (open black circle, before 4-CIN application; open red circles, during 4-CIN application; horizontal and vertical bars outside, the mean value and SEM). NS, not significantly different (paired t test); n = 8. In each experiment, the membrane potential was slightly depolarized to a suprathreshold level by a constant current injection so that the neurons regularly fired at 2–7 Hz.

Effects of d-lactate and phloretin on excitatory transmission. A1, Averaged eEPSC waveforms (n = 8) evoked by paired-pulse stimulation before (black) and during (red) d-lactate (30 mm) application in the NTS. Right, Overlaid waveforms scaled to eEPSC1. A2, Time course of the effect of d-lactate (30 mm) on eEPSC amplitudes (open circles, the amplitude of each eEPSC1; line, moving average over nine consecutive responses). B1, Averaged eEPSC waveforms (n = 8) evoked by paired-pulse stimulation before (black) and during (red) phloretin (500 μm) application in the NTS. Right, Overlaid waveforms scaled to eEPSC1. B2, Time course of the effect of phloretin on eEPSC amplitudes. C, Summary of the effects of 4-CIN, d-lactate (20 mm, n = 10; 30 mm, n = 4), and phloretin (n = 5) on eEPSC amplitude. ***p < 0.001, **p < 0.01, *p < 0.05 (Mann–Whitney U test) versus predrug; n = 9. Data are presented as the mean values ± SEM. D, Summary of the effects of d-lactate and phloretin on PPR. **p < 0.01; NS, not significantly (paired t test). Open circles represent the values from each neuron. E, Effects of d-lactate (20 mm, n = 10; 30 mm, n = 4) and phloretin (n = 5) on the holding current. Vertical axis, the difference in the holding current between before and during inhibitor application. Open circles represent the values from each neuron. Horizontal bars and vertical bars on the right side of open circles indicate mean value and SEM, respectively. Differences in the holding currents before and during inhibitor application were examined with paired t test; NS, not significantly different.

Effect of 4-CIN and d-lactate on AMPA-evoked current (I_AMPA) in NTS neurons. A, An IR-DIC image showing recording and application pipettes around the neuron being recorded. P, recording pipette; puff, puffer pipette. B, Upper traces, Averaged I_AMPA waveforms (n = 5) in the absence (black) and presence (orange) of 4-CIN (1 mm; left) and d-lactate (30 mm; right). AMPA (100 μm) was applied at the filled black circles for 15 ms. Lower traces, averaged eEPSC waveforms (n = 8) evoked by paired-pulse stimulation in the absence (black) and presence (red) of 4-CIN (left) and d-lactate (right). C, Representative time courses of simultaneously recorded I_AMPA (orange-filled triangles) and eEPSC amplitudes (red open circles). 4-CIN (above) and d-lactate (below) were applied during the period with horizontal bars. D, Summary of the effects of 4-CIN (n = 4) and d-lactate (n = 5) on I_AMPA and eEPSC amplitude. **p < 0.01, *p < 0.05 versus predrug; NS, not significantly different between groups (Mann–Whitney U test). Data are presented as the mean values ± SEM.

Effects of 4-CIN on heterologously expressed AMPA-R-mediated currents. A1, Effect of 4-CIN on currents mediated by AMPA-Rs with GluR1 + GluR4 subunits in Xenopus oocytes. AMPA-R currents were activated by kainate (KA; 100 μm). Kainate and 4-CIN were applied at the time indicated by the horizontal bars (the concentrations are indicated above each bar). A2, Concentration–response relation for the effect of 4-CIN on the currents mediated by AMPA-R composed of GluR1 + GluR4 (solid line) and GluR1 + GluR2 (dash line) subunits. Estimated IC₅₀ values of 4-CIN against the AMPA-R currents mediated by GluR1 + GluR4 and GluR1 + GluR2 subunits are 1258 ± 49 and 1359 ± 46 μm, respectively. IC₅₀ values estimated from a full date and set of each oocyte, and the mean and SEM are shown. Numbers in parentheses indicate the number of oocytes. B1, B2, Effect of increasing KA concentration on currents mediated by AMPA-Rs with GluR1 + GluR4 subunits in Xenopus oocytes. B1, In the absence of 4-CIN; B2, in the presence of 1 mm 4-CIN. B3, Concentration–response relation of kainate-induced currents mediated by GluR1 + GluR4 subunits in the absence (black) and presence (red) of 4-CIN (1 mm). Numbers in parentheses indicate the number of oocytes. Curve-fitting of the Hill equation was made assuming that the inward current activated by 300 μm (for 0 mm 4-CIN) and 1000 μm (for 1 mm 4-CIN) kainite was 100%. The mean and SEM are shown.

Effects of intracellular energy supply on synaptic suppression by 4-CIN. A, Averaged eEPSC waveforms (n = 8) evoked by paired-pulse stimulation before (black) and during (red) 4-CIN (1 mm) application in the absence (left) and presence (middle) of intrapipette ATP. Right graph shows summary effects of 4-CIN on eEPSC amplitude in the absence (left bar) and presence (right bar) of intrapipette ATP (0 ATP, n = 8; 2 ATP, n = 9). There was a significant difference in the inhibition by 4-CIN of eEPSC amplitude between in the absence and presence of intracellular ATP (Mann–Whitney U test; M-W). B, Averaged eEPSC waveforms (n = 8) before (black) and during (red) 4-CIN (1 mm) application in the absence (left) and presence (middle and right) of intrapipette ATP. Lactate (5 mm) was added into the patch pipette (left and middle). C, Summary of the effects of 4-CIN in the absence and presence of intracellular ATP and lactate on eEPSC amplitude (0 ATP, n = 8; 0 ATP + 5 Lac, n = 6; 2 ATP, n = 9; 2 ATP + 5 Lac, n = 6; 4 ATP, n = 8). The data for 0 ATP and 2 ATP without lactate are the same as in A and duplicated here to show the results of multiple comparisons. K-W, a result of statistical comparisons with Kruskal–Wallis multiple-comparison test, which was followed by pairwise comparison between groups with Scheffé comparison. The values show p values and *p < 0.05. Data are presented as the mean values ± SEM. D, Summary of change in PPR by 4-CIN in the absence and presence of intracellular ATP and lactate. Horizontal and vertical bars indicate the mean values and SEM.

Extracellular site of action of 4-CIN. A, A schematic of the experimental protocol. B, Averaged eEPSC waveforms (n = 8) evoked by paired-pulse stimulation after membrane rupture at 0 (1), 10 (2), and 25 min (3) and during (4) bath application of 4-CIN (1 mm); 4-CIN was intracellularly applied by adding 4-CIN (1 mm) to the internal solution. C, Time course of eEPSC amplitudes (open circles, the amplitude of each eEPSC1; line, moving average over nine consecutive responses). D, Summary of the effects of pipette and bath application of 4-CIN on eEPSC amplitude. Values and “NS” indicate the results of Friedman test for repeated measurements. There was a significant difference (p = 0.004) among groups. p values above the graph show the results of post hoc pairwise comparison between pairs indicated with brackets. NS, not significantly different; n = 6. Data are presented as the mean values ± SEM.

Extracellular perfusion of lactate can partially rescue the synaptic suppression by glucose deprivation in a 4-CIN-sensitive manner. A, Averaged eEPSC waveforms (n = 8) before (left) and during (middle, at 5 min; right, at 15 min) each manipulation (from top to bottom: glucose deprivation; replacement of glucose with 20 mm lactate; replacement of glucose with 20 mm lactate in the presence of 1 mm 4-CIN) at 32°C. B, Summary of the time courses of eEPSC amplitudes (black circles, glucose deprivation; n = 6; blue triangles, replacement of glucose with 20 mm lactate; n = 10; red diamonds, replacement of glucose with 20 mm lactate in the presence of 1 mm 4-CIN; n = 9). The eEPSC amplitude was significantly different (p < 0.05; Mann–Whitney U test) at the times marked with “*” at the bottom of plots. Data are presented as the mean values ± SEM. C, Summary of the effects of extracellular milieu replacement on eEPSC amplitude at 5 (left) and 15 min (right) after replacement. K-W, results of statistical comparison with Kruskal–Wallis multiple-comparison test among three groups. The values below the top brackets indicate the p values obtained by pairwise Scheffé comparisons between each group (pairs are shown with brackets). *p < 0.05, ^###p < 0.001, ^##p < 0.01, ^#p < 0.05 versus pre-replacement (Mann–Whitney U test). Data are presented as the mean values ± SEM. D, Summary of the time required to achieve a 70% decrease in eEPSC amplitude (T₇₀) by glucose deprivation in the absence and presence of 4-CIN. A Student's t test between two groups resulted in p < 0.05.

4-CIN enhanced the inhibitory effects of γDGG on eEPSC amplitude. A, Averaged eEPSC waveforms (n = 8) before (dotted lines) and during (solid lines) γDGG (1 mm) application in the absence (black) and presence (red) of 4-CIN (1 mm). Right, An overlay of these four traces scaled to the eEPSC before γDGG application. B, Time course of eEPSC amplitudes (open circles, the amplitude of each eEPSC1; line, moving average over nine consecutive responses). C, Summary of the effects of γDGG on eEPSC amplitude in the absence of and presence of 4-CIN. *p < 0.05 (Wilcoxon sign test); n = 6. Open circles represent the values from each neuron.

Comparisons of the effects of 4-CIN on postsynaptic responses mediated by AMPA-Rs, NMDA-Rs, and GABA_A-Rs. A1, Averaged eEPSC_NMDA waveforms (n = 8) before (black) and during (red) 4-CIN (1 mm) application. The holding potential was at +40 mV. A2, Time courses of eEPSC_NMDA amplitudes (open circles, the amplitude of each eEPSC_NMDA; line, moving average over nine consecutive responses). A3, Summary of the effects of 4-CIN on the amplitudes of eEPSC_AMPA (at +40 mV; n = 5) and eEPSC_NMDA (at +40 mV; n = 7); eEPSC_NMDA was abolished by MK-801 (black bar; n = 6). ^##p < 0.01 versus pre-4-CIN; NS, not significantly different (Mann–Whitney U test). Data are presented as the mean values ± SEM. B1, Averaged eIPSC waveforms (n = 8) before (black) and during (red) 4-CIN (1 mm) application. The holding potential was −60 mV (estimated Cl⁻ reversal potential was ∼0 mV). B2, Time course of eIPSC amplitude (open circles, the amplitude of each eIPSC1; line, moving average over nine consecutive responses). B3, Summary of the effects of 4-CIN on the amplitudes of eEPSC (at −60 mV; n = 9) and eIPSC (at −60 mV; n = 6). *p < 0.05, ^###p < 0.001, ^##p < 0.01 versus pre-4-CIN (Mann–Whitney U test).

Effects of 4-CIN on eEPSC amplitude under various experimental conditions. A, Top, Averaged eEPSC waveforms (n = 8) before (black) and during (red) 4-CIN (1 mm) application under “control” conditions (i.e., 0.1 Hz stimulation, −60 mV holding potential, 2 mm Ca²⁺ and room temperature). Bottom, The original traces with slower sweeps. B, Top, Averaged eEPSC waveforms (n = 50) before (black) and during (red) 4-CIN (1 mm) at 1 Hz stimulation. Bottom, The original traces with slower sweeps. C, Averaged eEPSC waveforms (n = 8) before (black) and during 4-CIN (1 mm; red) and d-Lactate (30 mm; blue) under various conditions (see text for each condition). D, Summary of the effects of 4-CIN and d-Lactate on eEPSC amplitude under various conditions. Numbers in bars indicate the number of neurons. Values above brackets indicate p values given by Mann–Whitney U test (M-W; used for comparison between control and each condition) or Kruskal–Wallis multiple-comparison test (K-W; used for comparison of the inhibition at 3 different holding potentials). *p < 0.05, ^###p < 0.001, ^##p < 0.01, ^#p < 0.05 versus pre-inhibitor (Mann–Whitney U test). Data are presented as the mean values ± SEM.