Figure 2: EAAT2 activity gates STDP polarity and time window.
From: Astrocytes gate Hebbian synaptic plasticity in the striatum
(a) Current–clamp recording of MSN in the absence of cortical stimulation showing that brief DHK application (300 μM for 5 min) induced significant depolarization, indicating the presence of ambient glutamate in the slice. This depolarization was fully reversed after 15 min of DHK washout and was dependent on AMPAR and type-I/II mGluR, but not NMDAR. (b,c) DHK application had no effect on long-term synaptic efficacy changes estimated from 15 min after DHK washout (example in b and averaged time-course of experiments in c). The brief application of DHK without the STDP protocol induced a transient decrease in EPSC amplitude and an inward shift in Iholding (light gray area). Both EPSC amplitude and Iholding had fully recovered 15 min after DHK washout. Ri remained unchanged during and after DHK application. The effects of DHK were fully reversible and, thus, compatible with the estimation of long-term synaptic efficacy changes. (d) Example of LTD induced by 100 pre–post pairings (ΔtSTDP=+38 ms) with a transient blockade of EAAT2 by DHK (300 μM for 5 min, dark gray area; the light gray area indicates DHK washout). Top, EPSC strength before and after pairings. Bottom, time course of Ri (baseline, 47±0.2 MΩ; 50–60 min after pairings, 51±0.1 MΩ; change of 10%). (e) Averaged time-course of experiments with the transient blockade of EAAT2 with DHK, showing the induction of LTD for both post–pre (−70<ΔtSTDP<0 ms) and pre–post (0<ΔtSTDP<+70 ms) pairings. (f) LTD expression for −70<ΔtSTDP<+70 ms with DHK. Synaptic strength was assessed 45–60 min after pairings (light blue circles: individual neurons; dark blue circle: average). (g) Example of LTP induced by 100 post–pre pairings (ΔtSTDP=−175 ms) during the transient blockade of EAAT2 with DHK (Ri, baseline: 136±0.5 MΩ; 50–60 min after pairings: 145±1 MΩ; change of 6%). (h) Averaged time-course of experiments with transient EAAT2 blockade with DHK during pairings, inducing LTP for both post–pre (−250<ΔtSTDP<−100 ms) and pre–post (+100<ΔtSTDP<+250 ms) pairings. (i) Averaged time-course of experiments with transient EAAT2 blockade with DHK during pairings, inducing LTP for ΔtSTDP∼±500 ms. (j) Time window for long-term synaptic strength for post–pre and pre–post pairings (−500<ΔtSTDP<+500 ms) in control conditions and in the presence of DHK. In controls, bidirectional plasticity was induced over a narrow time window (−30<ΔtSTDP<+30 ms) and no plasticity was observed with uncorrelated pairings (−500<ΔtSTDP<−30 ms and +30<ΔtSTDP<+500 ms). During transient EAAT2 blockade, plasticity was observed regardless of the ΔtSTDP value: LTD for narrow ΔtSTDP (−70<ΔtSTDP<+70) and LTP for a larger ΔtSTDP (−500<ΔtSTDP<−70 ms and +70<ΔtSTDP<+500 ms). Insets correspond to the mean of 60 EPSCs during baseline and at 1 h after STDP pairings. Error bars represent the s.d. (except in panel a,f,j: s.e.m.) *P<0.05; **P<0.01; ***P<0.001; NS: not significant by unpaired t-test, two-tailed, inside groups after one-way ANOVA; post hoc Bonferroni comparisons test (a), unpaired t-test, two-tailed (b,d,g) or one sample t-test (c,e,f,h,i,j).
