Block of activity with 48 hours TTX treatment increased mEPSC amplitudes in cortical cultures from wild type mice but not Rab3A-/- mice. A) 10 second current traces recorded at -60 mV in pyramidal cortical neurons from an untreated (“CON”) and TTX-treated (“TTX”) neuron in cultures prepared from Rab3A+/+ mice from the Rab3A+/- colony. B) Average traces for the recordings shown in (A). C) Box and whisker plots for average mEPSC amplitudes from 30 CON and 23 TTX cells (Rab3A+/+, Rab3A+/- colony; mean mEPSC amplitudes, CON, 13.9 ± 0.7 pA; TTX, 18.2 ± 0.9 pA, p = 4.58 * 10-4, Kruskal-Wallis test). D) Current traces recorded in pyramidal cortical neurons from a CON and TTX-treated neuron in cultures prepared from Rab3A-/- mice. E) Average traces for the recordings shown in (D). F) Box and whisker plots for average mEPSC amplitudes from 25 CON and 26 TTX cells, Rab3A-/- data (Rab3A-/-, mean mEPSC amplitudes, CON, 13.6 ± 0.1; TTX, 14.3 ± 0.6, p = 0.318, Kruskal-Wallis test). G) Analysis of pooled mEPSC amplitudes for cultures from Rab3A+/+ mice of the Rab3A+/- colony (i) Cumulative distribution functions (CDFs) of quantiles of mEPSC amplitudes pooled from CON (black curve) and TTX- treated cultures (red curve) (Kolmogorov Smirnov (KS) test, test statistic (D) = 0.149, p = 8.53 * 10-13. (ii) Quantiles pooled from CON and TTX-treated cells were ranked from smallest to largest, plotted against each other, and fit with linear regression functions. Gray line, identity; red dashed line, linear regression fit. Inset, full range of the data. (iii) The same ranked mEPSC quantiles were paired and the ratio of CON/TTX computed for each pair, then plotted as a function of the CON mEPSC for the pair. Dashed line at ratio = 1.0 (no plasticity). H) Analysis of pooled mEPSCs amplitudes for cultures from Rab3A-/- mice. (i) CDFs of quantiles pooled from CON (black curve) and TTX-treated cultures (red curve) (KS test, D = 0.070, p = 0.042). (ii) Quantiles pooled from CON and TTX-treated cells were ranked from smallest to largest, plotted against each other, and fit with a linear regression function. Gray lines, identity; red dashed lines, linear regression fits. Inset: full range of the data. (iii) The same ranked mEPSC quantiles were paired and the ratio of CON/TTX computed for each pair, then plotted as a function of the CON mEPSC for the pair. Dashed line at ratio = 1 (no plasticity).

Block of activity with 48 hours TTX treatment increased mEPSC amplitudes in cortical cultures from wild type mice but not Rab3AEbd/Ebd mice, which had an increased mEPSC amplitude at baseline. A) 10 second current traces recorded at -60 mV in pyramidal cortical neurons from an untreated (“CON”) and TTX-treated (“TTX”) neuron in cultures prepared from Rab3A+/+ mice from the Rab3A+/Ebd colony. B) Average traces for the recordings shown in (A). C) Box and whisker plots for average mEPSC amplitudes from 20 CON and 23 TTX cells (Rab3A+/+, Rab3A+/Ebd colony, mean mEPSCs amplitudes, CON, 11.0 ± 0.6 pA; TTX, 15.0 ± 1.3 pA, p = 0.02, Kruskal-Wallis test). D) Current traces recorded in pyramidal-shaped cortical neurons from a CON and TTX-treated neuron in cultures prepared from Rab3AEbd/Ebd mice. E) Average traces for the recordings shown in (D). F) Box and whisker plots of average mEPSC amplitudes from 21 CON and 22 TTX cells (Rab3AEbd/Ebd, mean mEPSC amplitudes, CON, 15.1 ± 1.0 pA; TTX, 14.6 ± 1.1 pA, p = 0.81, Kruskal-Wallis test). G) Analysis of pooled mEPSC amplitudes for cultures from Rab3A+/+ mice of the Rab3A+/Ebd colony. (i) CDFs of quantiles pooled from CON (black curve) and TTX-treated cultures (red curve) (KS test, D = 0.177, p value = 2.88 * 10-9). (ii) Quantiles pooled from CON and TTX-treated cells were ranked from smallest to largest, plotted against each other, and fit with linear regression function. Gray line, identity; red dashed line, linear regression fit. Inset: full range of the data. (iii) The same ranked mEPSC quantiles were paired and the ratio of CON/TTX computed for each pair, then plotted as a function of the CON mEPSC for the pair. Dashed line at ratio = 1 (no plasticity). H) Analysis of pooled mEPSC amplitudes for cultures from Rab3AEbd/Ebdmice. (i) CDFs of quantiles pooled from CON (black curve) and TTX- treated cultures (red curve). (KS test, D = 0.067, p value = 0.097 (TTX mEPSCs were slightly reduced)). (ii) Quantiles pooled from CON and TTX-treated cells were ranked from smallest to largest, plotted against each other, and fit with a linear regression function. Gray line, identity; red dashed line, linear regression fit. Inset: full range of the data. (iii) The same ranked mEPSC quantiles were paired and the ratio of CON/TTX computed for each pair, then plotted as a function of the CON mEPSC for the pair. Dashed line at ratio = 1 (no plasticity).

The increase in mEPSC amplitudes in neurons from cultures prepared from Rab3AEbd/Ebd mice, relative to those from Rab3A+/+ mice in the same strain, showed less divergent scaling than that of the activity-block induced increase in mEPSC amplitudes typically observed in cultures from Rab3A+/+ mice. A) Cumulative distribution function of quantiles pooled from Rab3A+/+ (black curve) and Rab3AEbd/Ebd cultures (red curve), D = 0.245; p = 1.52 * 10-16, KS test. Inset: box and whisker plot of average mEPSC amplitudes from 20 Rab3A+/+ and 21 Rab3AEbd/Ebdcells, mean mEPSC amplitudes, Rab3A+/+, 11.0 ± 0.7 pA; Rab3AEbd/Ebd, 15.1 ± 1.0 pA, p = 0.0027, Kruskal-Wallis test. B) Quantiles pooled from Rab3A+/+ and Rab3AEbd/Ebd cells were ranked from smallest to largest, plotted against each other, and fit with linear regression functions. Gray line, identity; red dashed line, linear regression fit. C) The same ranked mEPSC quantiles were paired and the ratio of Rab3A+/+/Rab3AEbd/Ebd computed for each pair, then plotted as a function of the Rab3A+/+ mEPSC for the pair. Dashed line at ratio = 1.0 (no difference).

Ca2+-permeable, NASPM-sensitive glutamate receptors did not mediate the TTX- induced increase in mEPSC amplitude in cultures of dissociated mouse cortical neurons. A) mEPSC amplitudes for 11 mouse cortical neurons from untreated cultures (“CON”) and 11 neurons from TTX treated cultures (“TTX”) were recorded before and after application of 20 μM NASPM. mEPSC amplitudes were still significantly larger in TTX-treated cultures compared to untreated control cultures after NASPM treatment (TTX effect pre-NASPM, p = 0.009; TTX effect post-NASPM, p = 0.006, Kruskal-Wallis non-parametric test). B) Acute application of NASPM had a modest effect on mEPSC amplitude in cells from CON or TTX cultures that did not reach statistical significance (CON, pre-NASPM, 12.9 ± 1.1 pA, post-NASPM, 11.9 ± 0.8 pA, p = 0.08; TTX, pre-NASPM, 17.5 ± 0.9 pA, post-NASPM, 16.1 ± 1.0 pA, p = 0.08; student’s paired t-test). C) Acute application of NASPM consistently reduced mEPSC frequency in cells from untreated and cells from TTX-treated cultures (CON, pre-NASPM, 1.84 ± 1.82 mEPSCs sec-1, post-NASPM, 1.56 ± 1.74 mEPSCs sec-1; p = 0.003; TTX, pre-NASPM, 4.40 ± 3.51 mEPSCs sec-1, post-NASPM, 2.68 ± 2.25 mEPSCs sec-1, p = 0.02; student’s paired t-test).

Confocal images of GluA2 and VGLUT1 immunofluorescence on MAP2-positive primary dendrites of pyramidal-shaped neurons in high density dissociated cortical cultures prepared from Rab3A+/+ mice. Shown are 5X confocal sections of two primary dendrites, one from an untreated culture (CON) and the other from a culture treated with TTX (TTX). Postsynaptic GluA2-positive receptor clusters were identified by their proximity to VGLUT1 immunoreactive presynaptic terminals within the same confocal section. GluA2-VGLUT1 synaptic pairs, which were of maximal brightness in these confocal slices, are indicated by white trapezoids. It can be seen that substantial numbers of GluA2-positive clusters located on the dendrites are unopposed by VGLUT1-positive terminals, and that some GluA2-positive and VGLUT1-positive clusters are found away from any MAP2 positivity, suggesting localization to axons or astrocytes. Scale bar = 10 μm.

GluA2 receptor cluster size and average intensity increased following activity blockade in parallel with mEPSC amplitudes recorded in the same cortical cultures prepared from Rab3A+/+ mice, but the effects on receptor levels were less robust and differed from mEPSC amplitudes in characteristics of scaling. A) Cumulative distribution functions (CDFs) of mEPSC amplitudes recorded from untreated (“CON”) and TTX-treated cells (“TTX”) pooled from 3 cultures were significantly different (KS test, D = 0.162, p = 1.42 * 10-8, CON, n = 690, TTX, n = 720). Inset, mean mEPSC amplitudes for 23 CON cells and 24 TTX cells were significantly different (CON, 13.7 ± 4.5 pA; TTX, 16.4 ± 4.3 pA, p = 0.016, Kruskal-Wallis test). B) 24 mEPSC amplitude quantiles from each of 23 untreated cells were sorted from smallest to largest and plotted vs. 23 mEPSC amplitude quantiles from each of 24 TTX-treated cells, and the resulting relationship fit with a linear regression equation, Y = 1.106x + 1.22 (R2 = 0.984). Inset, full range of data used for fit. C) Sorted data from (B) were used to calculate the ratio of TTX mEPSC amplitude/CON mEPSC amplitude and the ratios plotted vs. CON mEPSC amplitude. D) CDFs of size of GluA2 receptor clusters imaged from primary dendrites in the same untreated cultures and TTX-treated cultures as in (A) – (C) were significantly different (KS test, D = 0.089, p = 0.002, CON, n = 870, TTX, n = 870). Inset, mean GluA2 cluster size for 29 CON dendrites and 29 TTX dendrites were not significantly different (CON, 0.97 ± 0.38 μm2; TTX, 1.15 ± 0.59 μm2, p = 0.44, Kruskal-Wallis test). E) 29 GluA2 cluster size quantiles from each of 29 untreated dendrites were sorted from smallest to largest and plotted vs. 29 cluster size quantiles from each of 29 TTX-treated dendrites, and the resulting relationship fit with a linear regression equation, Y = 1.328x -0.136 (R2 = 0.98). Inset, full range of data used for fit. F) Sorted data from (E) were used to calculate the ratio of TTX GluA2 cluster size/CON GluA2 cluster size and the ratios plotted vs. CON GluA2 cluster size. G) CDFs of intensity of GluA2 receptor clusters imaged from primary dendrites in untreated cultures and TTX-treated cultures were significantly different (KS test, D = 0.120, p = 6.73 * 10-6, CON, n = 870, TTX, n = 870). Inset, mean GluA2 cluster intensity for 29 CON dendrites and 29 TTX dendrites were not significantly different (CON, 673 ± 90 a.u.; TTX, 687 ± 72 a.u., p = 0.25, Kruskal-Wallis test). H) 29 GluA2 cluster intensity quantiles from 29 untreated dendrites were sorted from smallest to largest and plotted vs. 29 cluster intensity quantiles from 29 TTX-treated dendrites, and the resulting relationship fit with a linear regression equation, Y = 0.899x + 82 (R2 = 0.987). Inset, full range of data used for fit. I) Sorted data from (H) were used to calculate the ratio of TTX cluster intensity/CON cluster intensity and plotted vs. CON cluster intensity.

GluA2 receptor cluster size, average intensity, and mEPSC amplitudes recorded in the same cortical cultures prepared from Rab3A-/- mice were not increased following activity blockade. A) Cumulative distribution functions (CDFs) of mEPSC amplitudes recorded from untreated (“CON”) and TTX-treated cells (“TTX”) pooled from 3 Rab3A-/- cultures were not significantly different (KS test, D = 0.072, p = 0.085, CON, n = 630, TTX, n = 570). Inset, mean mEPSC amplitudes for 21 CON cells and 19 TTX cells were not significantly different (CON, 14.9 ± 3.8 pA; TTX, 14.0 ± 4.0 pA, p = 0.34, Kruskal-Wallis test). B) 19 mEPSC quantiles from each of 21 untreated cells were sorted from smallest to largest and plotted vs. 21 mEPSC quantiles from each of 19 TTX-treated cells, and the resulting relationship fit with a linear regression equation, Y = 0.93x + 0.13 (R2 = 0.997). Inset, full range of data used for fit. C) Sorted data from (B) were used to calculate the ratio of TTX mEPSC amplitude/CON mEPSC amplitude and plotted vs. CON mEPSC amplitude. D) CDFs of GluA2 receptor cluster size imaged from primary dendrites in the same untreated cultures and TTX-treated cultures as in (A) – (C) were not significantly different (KS test, D = 0.045, p = 0.31, CON, n = 900, TTX, n = 870). Inset, mean GluA2 cluster size for 30 CON dendrites and 29 TTX dendrites were not significantly different (CON, 0.93 ± 0.27 μm2; TTX, 0.91 ± 0.28 μm2, p = 0.74, Kruskal-Wallis test). E) 29 GluA2 cluster size quantiles from each of 30 untreated dendrites were sorted from smallest to largest and plotted vs. 30 cluster size quantiles from each of 29 TTX-treated dendrites, and the resulting relationship fit with a linear regression equation, Y = 1.01x -0.03 (R2 = 0.993). Inset, full range of data used for fit. F) Sorted data from (E) were used to calculate the ratio of TTX GluA2 cluster size/CON GluA2 cluster size and the ratios plotted vs. CON GluA2 cluster size. G) CDFs of GluA2 receptor cluster average intensity imaged from primary dendrites in untreated cultures and TTX-treated cultures were not significantly different (KS test, D = 0.0.060, p = 0.08, CON, n = 900, TTX, n = 870). Inset, mean GluA2 cluster intensity for 30 CON dendrites and 29 TTX dendrites were not significantly different (CON, 766 ± 68 a.u.; TTX, 776 ± 79 a.u., p = 0.47, Kruskal-Wallis test). H) 29 GluA2 cluster intensity quantiles from 30 untreated dendrites were sorted from smallest to largest and plotted vs. 30 cluster intensity quantiles from 29 TTX-treated dendrites, and the resulting relationship fit with a linear regression equation, Y = 1.09x - 60 (R2 = 0.993). Inset, full range of data used for fit. I) Sorted data from (H) were used to calculate the ratio of TTX cluster intensity/CON cluster intensity and the ratios plotted vs. CON cluster intensity.

Comparison of effect of activity blockade with TTX on mEPSC amplitude and GluA2 receptor cluster characteristics. Voltage clamp recordings of mEPSCs and confocal imaging of GluA2 receptor immunofluorescence at VGLU1-positive synapses were performed on coverslips from the same 3 Rab3A+/+ cultures (left) and 3 Rab3A-/- cultures (right). p values are not corrected for multiple comparisons.

mEPSC amplitudes and GluA2 receptor cluster sizes did not always change in parallel after treatment with TTX in cultures from Rab3A+/+ and Rab3A-/- mice. A) Rab3A+/+ Culture #1: mEPSC amplitudes and GluA2 receptor cluster sizes increased in parallel after treatment in TTX. i) CDFs of mEPSC amplitudes of cells from untreated and TTX treated dishes were significantly different (KS test, D = 0.16, p = 0.010; CON, n = 180, TTX, n = 210). ii) CDFs of GluA2 receptor cluster sizes of dendrites from untreated and TTX treated dishes were significantly different (KS test, D = 0.17, p = 3.79 * 10-4; CON, n = 300, TTX, n = 300). iii) 35 mEPSC amplitude quantiles from each of 6 untreated cells and 30 quantiles from each of 7 TTX-treated cells were sorted from smallest to largest and used to calculate the ratio of TTX mEPSC amplitude/CON mEPSC amplitude, which was plotted vs. CON mEPSC amplitude. iv) 30 GluA2 receptor cluster size quantiles from each of 10 untreated dendrites and 30 GluA2 receptor cluster size quantiles from each of 10 TTX-treated dendrites were used to calculate the ratio of TTX GluA2 cluster size/CON GluA2 cluster size, and plotted vs. CON GluA2 cluster size. Inset, full range of data. B) Rab3A+/+ Culture #2: mEPSC amplitudes and GluA2 receptor cluster sizes increased in parallel after treatment in TTX. i) CDFs of mEPSC amplitudes of cells from untreated and TTX treated dishes were significantly different (KS test, D = 0.21, p = 1.06 * 10-4; CON, n = 210, TTX, n = 240). ii) CDFs of GluA2 receptor cluster sizes of dendrites from untreated and TTX treated dishes were significantly different (KS test, D = 0.21, p = 5.39 * 10-6; CON, n = 270, TTX, n = 300). iii) 32 mEPSC amplitude quantiles from each of 7 untreated cells, and 28 mEPSC amplitude quantiles from each of 8 TTX treated cells were sorted from smallest to largest and were used to calculate the ratio of TTX mEPSC amplitude/CON mEPSC amplitude, and plotted vs. CON mEPSC amplitude. iv) 30 GluA2 receptor cluster size quantiles from each of 9 untreated dendrites and 27 GluA2 receptor cluster size quantiles from each of 10 TTX-treated dendrites were sorted from smallest to largest and used to calculate the ratio of TTX GluA2 cluster/CON GluA2 cluster size, and plotted vs. CON GluA2 cluster size. Inset, full range of data. C) Rab3A+/+ Culture #3: mEPSC amplitudes increased, but GluA2 receptor cluster sizes decreased, after treatment in TTX. i) CDFs of mEPSC amplitudes of cells from untreated and TTX treated dishes were significantly different (KS test, D = 0.15, p = 0.002; CON, n = 300, TTX, n = 270). ii) CDFs of GluA2 receptor cluster sizes of dendrites from untreated and TTX treated dishes were significantly different, but the TTX CDF is shifted to the left (KS test, D = 0.16, p = 0.001; CON, n = 300, TTX, n = 270) iii) 27 mEPSC amplitude quantiles from each of 10 untreated cells, and 30 mEPSC amplitude quantiles from each of 9 TTX-treated cells, were sorted from smallest to largest, used to calculate the ratio of TTX mEPSC amplitude/ CON mEPSC amplitude, and plotted vs. CON mEPSC amplitude. iv) 27 GluA2 receptor cluster size quantiles from each of 10 untreated dendrites and 30 GluA2 receptor cluster sizes from each of 9 TTX-treated dendrites were sorted from smallest to largest, used to calculate the ratio of TTX GluA2 receptor cluster size/CON GluA2 receptor cluster size, and plotted vs. CON GluA2 receptor cluster size. D) Rab3A-/- Culture #1 (note, this culture is distinct from that of Rab3A+/+ Culture #1). Neither mEPSC amplitudes, nor GluA2 receptor cluster size, increased following TTX treatment. i) CDFs of mEPSC amplitudes of cells from untreated and TTX-treated cultures were not significantly different (KS test, D = 0.05, p = 0.90; CON, n = 210, TTX, n = 180). ii) CDFs of GluA2 receptor cluster sizes of dendrites from untreated and TTX (Figure 8 continued) treated dishes were not significantly different, although the TTX CDF is shifted slightly to the left (KS test, D = 0.10, p = 0.08; CON, n = 300, TTX, n = 300). iii) 30 mEPSC amplitude quantiles from each of 7 untreated cells and 35 mEPSC quantiles from each of 6 TTX-treated cells were sorted from smallest to largest, used to calculate the ratio TTX mEPSC amplitude/CON mEPSC amplitude, and plotted vs. CON mEPSC amplitude. iv) 30 GluA2 receptor cluster size quantiles from each of 10 untreated dendrites and 10 TTX-treated dendrites were sorted from smallest to largest, used to calculate the ratio TTX cluster size/CON cluster size, and plotted vs. CON cluster size. E) Rab3A-/- Culture #2 (distinct from Rab3A+/+ Culture #2) mEPSC amplitudes and GluA2 receptor cluster sizes decreased after TTX treatment. i) TTX mEPSC amplitudes CDF is shifted slightly to the left compared to the CON mEPSC amplitudes CDF but the difference is not significant (KS test, D = 0.08, p = 0.28, CON, n = 240, TTX, n = 240). ii) TTX GluA2 receptor cluster sizes CDF is significantly shifted to the left compared to CON GluA2 receptor cluster sizes CDF (KS test, D = 0.13, p = 0.01; CON, n = 300, TTX, n = 300). iii) 24 mEPSC amplitude quantiles from each of 8 untreated cells and 8 TTX- treated cells were sorted from smallest to largest, used to calculate the ratio TTX mEPSC amplitude/CON mEPSC amplitude, and plotted vs. CON mEPSC amplitude. iv) 27 GluA2 receptor cluster size quantiles from each of 10 untreated dendrites, and 30 GluA2 receptor cluster size quantiles from each of 9 TTX treated dendrites, were sorted from smallest to largest, used to calculate the ratio TTX GluA2 cluster size/CON cluster size, and plotted vs. CON cluster size. Both mEPSC amplitude and GluA2 receptor cluster size ratios were below 1 across the majority of data. F) Rab3A-/- Culture #3 (distinct from Rab3A+/+ Culture #3). mEPSC amplitudes were decreased following TTX treatment, but GluA2 receptor cluster sizes were increased. i) TTX CDF of mEPSC amplitudes was significantly shifted to the left of CON CDF (KS test, D = 0.21, p = 4.85 * 10-4, CON, n = 180, TTX, n = 180). ii) TTX CDF of GluA2 receptor cluster sizes was significantly shifted to the right of CON CDF (KS test, D = 0.15, p = 0.003, CON, n = 300, TTX, n = 300). iii) 30 mEPSC amplitude quantiles from each of 6 untreated and 6 TTX-treated cells were sorted from smallest to largest, used to calculate the ratio TTX mEPSC amplitude/CON mEPSC amplitude, and plotted vs. CON mEPSC amplitude. iv) 30 GluA2 receptor cluster size quantiles from each of 10 untreated dendrites and 10 TTX-treated dendrites were sorted from smallest to largest, used to calculate the ratio TTX GluA2 cluster size/CON GluA2 cluster size, and plotted vs. CON CluA2 cluster size.

Rab3A in neurons, not astrocytes, was required for full homeostatic plasticity after activity blockade. A-C, Data from dissociated cortical neurons plated on an astrocyte feeder layer, each prepared separately from the type of mice depicted in the schema at left: A) Neurons from Rab3A+/+ mice plated on astrocytes from Rab3A+/+ mice; i) CDFs for mEPSC amplitudes from untreated (CON, black curve) and TTX-treated (TTX, red curve) cultures (KS test, test statistic, D = 0.141, p = 2.74 * 10-5, n = 510 CON, 600 TTX samples; inset, means, CON, 13.3 ± 0.5 pA, n = 17 cells, TTX, 16.7 ± 1.2 pA, n = 20 cells, p = 0.03, Kruskal Wallis test; ii) rank order plot and linear regression fit, slope = 1.89, intercept = -8.51, R2 = 0.97; iii) ratio plot of TTX mEPSC amplitude/CON amplitude vs. CON amplitude; B) Neurons from Rab3A+/+ mice plated on astrocytes from Rab3A-/- mice; i) CDFs of mEPSC amplitudes, CON, black curve; TTX, red curve (KS test, test statistic 0.273, p = 2.98 * 10-11), n = 330 CON, 330 TTX samples; inset, means, CON, 13.3 ± 1.0 pA, n = 11 cells, TTX, 18.8 ± 1.4 pA, n = 11 cells, p = 0.005, Kruskal-Wallis test; ii) rank order plot and linear regression fit, slope = 1.76, intercept = -4.64, R2 = 0.98; iii) ratio plot of TTX mEPSC amplitude/TTX mEPSC amplitude vs. CON mEPSC amplitude; C) Neurons from Rab3A-/- neurons plated on astrocytes from Rab3A+/+ mice; i) CDFs of mEPSC amplitudes, CON, black curve, TTX, red curve (KS test, test statistic 0.126, p = 0.005), n = 420 CON, 330 TTX samples; inset, means, CON, 15.2 ± 1.1 pA, n = 14 cells, TTX, 16.9 ± 0.7 pA, n = 11 cells, p = 0.125, Kruskal-Wallis test;; ii) rank order plot and linear regression fit; slope = 1.01, intercept = 1.53, R2 = 0.97; ratio plot, TTX mEPSC amplitude/CON mEPSC amplitude vs. CON mEPSC amplitude. Gray line, identity; red dashed line, linear regression fit. Insets in rank order plots depict full range of data.

Rab3A is required for increased mEPSC amplitude after activity blockade via 2 mechanisms: 1. increased amount of transmitter released from a single vesicle and 2. increased postsynaptic receptors. Left: in the presence of normal activity, Rab3A is in its control state (“State 1,” red), maintaining mEPSC amplitude within a limited range by regulating either vesicle size or transmitter content, or both, and by regulating the number of postsynaptic receptors, likely through release of an anterograde signal. Right: after activity is blocked, Rab3A accumulates in an alternate functional state (“State 2,” gray) resulting in quantal size expanding to larger ranges because of increased vesicle size or transmitter content and because of increased numbers of receptors no longer being under the normal control of the anterograde signal.