Article Figures & Data
Figures
- Figure 1.
Activation-dependent increases in the amplitude of postsynaptic glycine responses. A, Schematics of the culture model which allows postsynaptic GlyR activation specifically after the D–H shift of the glycinergic response in mature neurons. Gly, glycine; Depol, depolarization. B, Schematic timelines of the experimental protocol for patch-clamp recordings from Ctrl and strychnine-treated neurons (1 μm). Culture medium chronically containing STR from the 0 DIV was exchanged with a recording solution lacking strychnine (WASH). A Ctrl group acutely applied STR for 30 s, and the solution was exchanged with a recording solution lacking strychnine. Patch-clamp recordings were started within 10 min following the washout of strychnine and continued for >50 min after strychnine removal. Aa–c indicate the putative developmental stages shown in A. C, Glycine-induced currents recorded by the gramicidin-perforated patch clamp in Ctrl (left) and in cSTR (right) neurons at a holding potential of −53 mV. Voltage ramp pulses from −113 to +37 mV were applied before (arrowheads, a and c) and during (arrowheads b and d) 100 μm glycine application. D, Average I–V relationship for glycinergic responses in Ctrl group (n = 5 cells) and cSTR group (n = 6 cells), as used to derive Egly. Superimposed histogram shows the mean values of Egly ± SEM. E, Representative traces of spontaneous glycinergic mIPSC at three different time points in the Ctrl group (top) and WASH group (bottom). After each recording, strychnine was applied to confirm the strychnine-sensitive glycinergic responses. Representative traces in a and b show longer ranges and wide-stretched views from same neurons. F, The amplitude distribution histograms of glycinergic mIPSCs at three different time points (bin width, 2 pA). Each graph is constructed from pooled data of 5 min duration in the Ctrl condition (n = 8 cells) and the WASH condition (n = 7 cells). G, H, Time courses of mean amplitudes (G) and interevent intervals (H) of the glycinergic mIPSC in the Ctrl condition (open circles, n = 8 cells) and the WASH condition (filled circles, n = 7 cells). Each point and bar represents the mean and ±SEM, respectively. I, Mean values ± SEM of the decay time constant of the glycinergic mIPSC in the Ctrl group (n = 8 cells) and the WASH group (n = 7 cells) at 10 min after strychnine removal. J, Time courses of mean amplitudes of the bath-applied 100 µm glycine-induced IPSC in the Ctrl condition (open circle, n = 7 cells) and the WASH condition (filled circles, n = 6 cells). Glycine is rapidly applied and removed in the whole-cell body by Y-tube. Each point and bar represents the mean and ±SEM, respectively. The significance of the difference was tested by the Student’s unpaired two-tailed t test (D, I) and by the Kruskal–Wallis test followed by post hoc comparisons using the Mann–Whitney U test with Bonferroni’s correction between Ctrl and WASH conditions at each time point (G, H, and J). *p < 0.05 (D and I) or 0.05/3 after Bonferroni’s correction (G–I). n.s., Not significant.
- Figure 2.
Strychnine removal restores synaptic GlyR clusters. A, Schematic timelines of the experimental protocol. Three experimental groups consisted of neurons incubated with only the vehicle (Ctrl group), 1 μm strychnine chronically (cSTR group), or 1 μm strychnine until 1 h before the experiment (WASH group). Neurons in all groups were subsequently fixed by 4% PFA. B, Representative double-immunostaining images of VIAAT (green) and GlyRs (magenta) in fixed cultured neurons in Ctrl, cSTR, and WASH conditions. Scale bar, 2 μm. C, The number of total VIAAT and GlyR punctate signals per neurons in Ctrl, cSTR, and WASH conditions. D, The number of GlyR punctate signals apposed to VIAAT per neurons in Ctrl, cSTR, and WASH conditions. E, The relative areas of total VIAAT and GlyR punctate signals in Ctrl, cSTR, and WASH conditions. F, The relative areas of GlyR signals apposed to VIAAT in Ctrl, cSTR, and WASH conditions. G, Overlap coefficients of Manders et al. (1993) for VIAAT and GlyR signals. H, A representative immunoblot analysis of total and surface expression levels of glycine receptors in Ctrl, cSTR, and WASH conditions. Biotinylated glycine receptors (surface) were isolated from the detergent-soluble fraction (total). Expression levels of β-actin were used as an internal control for total GlyR. M, Molecular weight marker. I, Quantification of total and surface expression levels of glycine receptors. Expression levels of total GlyRs are normalized by β-actin, and those of surface GlyRs are normalized by total GlyRs. C–G, I, Statistical data are reported as the mean ± SEM derived from 14–17 cells in three independent experiments (C–G) and from three independent experiments (I). The relative numbers, areas, and intensities are normalized to the mean values of Ctrl conditions. C–G, I, The significance of the difference was tested by one-way ANOVA with Bonferroni’s post hoc test (C–G) and by Kruskal–Wallis test followed by post hoc comparisons using Mann–Whitney U test with Bonferroni’s correction (I). *p < 0.05, **p < 0.01, ***p < 0.001. n.s. Not significant.
- Figure 3.
Cell surface dynamics of glycine receptors in live neurons. A, FRAP in neuronal processes expressing SEP-GlyRα1. Fluorescence of SEP-GlyRα1 was bleached in the region of the neuronal process indicated by the yellow rectangles. Each row represents before (pre), immediately after (+0 s), and at a later time (+257 s) after photobleaching in each condition (Ctrl, cSTR, and WASH). Scale bar, 2 µm. B, Normalized fluorescence recovery curves after photobleaching in Ctrl (black, n = 15 cells), cSTR (orange, n = 16 cells), WASH (cyan, n = 12 cells), and cSTR+ BFA (green, n = 5 cells) conditions. Each plot and bar represents the mean and ±SEM. C, Averaged mobile fractions in Ctrl, cSTR, WASH, and cSTR with BFA conditions (mean ± SEM), as shown in B. The mobile fraction was defined as the extent of fluorescence recovery at the end of the imaging time. D, A representative image of SEP-GlyRα1 fluorescence and the regions of the neuronal processes used for the FRAP-FLIP experiments. Repetitive photobleaching (FLIP, green rectangles) occurred at regions bilateral to the central FRAP region (yellow rectangle). A magenta rectangle shows the buffer region used to minimize any effects of the leakage of light from the FLIP regions. Scale bar, 2 µm. E, A schematic of hypothetical SEP-GlyRα1 movements under the FRAP-FLIP configuration shown in D. F, Normalized fluorescence recovery curves (mean ± SEM) in Ctrl (n = 8 cells, black), cSTR (n = 15 cells, orange), and WASH (n = 12 cells, cyan) conditions in the FRAP-FLIP experiments. G, Averaged mobile fractions in the Ctrl, cSTR, and WASH conditions (mean ± SEM), as shown in F. The significance of difference was tested by Kruskal–Wallis test followed by post hoc comparisons using the Mann–Whitney U test with Bonferroni’s correction. ***p < 0.001/6 after Bonferroni’s correction for the six tests. n.s., Not significant.
- Figure 4.
Glycine receptor dynamics at gephyrin-identified synaptic zones. A, Representative double-immunostaining images of VIAAT (green) and GPHN (red) in fixed cultured neurons in Ctrl, cSTR, and WASH conditions. Scale bar, 2 μm. B, Overlap coefficients of Manders et al. (1993) of VIAAT and GPHN signals (n = 13–16 cells). C, Representative images of neuronal processes coexpressing SEP-GlyRα1 and mCherry-GPHN in Ctrl, cSTR, and WASH conditions. Top, Expression of mCherry-GPHN fluorescence. FRAP of SEP-GlyRα1 was measured at focal spots with (arrows) and without (arrow heads) mCherry-GPHN signals. Each row represents before (pre), immediately after (+0 s), and at a later time (+267 s) after photobleaching. Scale bar, 2 µm. D, Fluorescence recovery curves after photobleaching at focal spots expressing mCherry-GPHN in control (black), cSTR (orange), cGinkB (green), and WASH (cyan) conditions. Each point and bar represents the mean and ±SEM, respectively. E, Averaged mobile fractions at focal spots expressing mCherry-GPHN in Ctrl (n = 14 cells), cSTR (n = 8 cells), cGinkB (n = 8 cells), WASH (n = 15 cells), and cMEC (n = 8 cells) conditions (mean ± SEM) for the data shown in D. F, Fluorescence recovery curves after photobleaching at mCherry-GPHN− regions in control (black), cSTR (orange), cGinkB (green), and WASH (cyan) conditions. Each point and bar represents mean and ±SEM, respectively. G, Average mobile fraction at focal spots lacking mCherry-GPHN in Ctrl (n = 14 cells), cSTR (n = 8 cells), cGinkB (n = 8 cells), and WASH (n = 15 cells) conditions (mean ± SEM), for the data shown in F. H, The correlation between fractions of fluorescence recovery of SEP-GlyRα1 and mCherry-GPHN (n = 37 cells). Each plot represents individual samples in Ctrl, cSTR, and WASH groups. I, Representative fluorescence images of mCherry-GPHN and SEP-GlyRα1 at a single dendrite before (pre) and after (+60) local application of glycine for 60 min. Arrows and arrowheads shows GPHN+ and GPHN− zones. Scale bar, 1 µm. J, Average changes of SEP-GlyRα1 fluorescence intensity in the GPHN+ and GPHN− zones at single dendrite (n = 7 cells). The significance of the difference was tested by one-way ANOVA with Bonferroni’s post hoc test (B), by Kruskal–Wallis test followed by post hoc comparisons using the Mann–Whitney U test with Bonferroni’s correction (E, G) and Student’s paired (H) or unpaired (J) two-tailed t test. The Pearson’s correlation coefficients (r) and p values (p) are indicated in H. *p < 0.05/10 after Bonferroni’s correction for the 10 tests (E) and p < 0.05 (J). n.s., Not significant.
- Figure 5.
GlyR activation regulates the formation, but not the maintenance, of GlyR clustering. A, Schematic timelines of strychnine treatments on mature neurons. Neurons cultured over 14 DIV without previous application of strychnine were applied for the administration of 1 μm strychnine for 48 h (+STR 48h), 36 h (+STR 36h), 24 h (+STR 24h), 1 h (+STR 1 h), or immediate (+STR 0 h) before and during imaging. B, Normalized fluorescence curves at focal spots with mCherry-GPHN signals in Ctrl (n = 14 cells, black), Ctrl with acute (+STR 0h; n = 6 cells, orange), with 1 h (+STR 1h; n = 9 cells, green), with 24 h (+STR 24h; n = 15 cells, cyan), with 36 h (+STR 36h; n = 11 cells, magenta), and with 48 h (+STR 48h; n = 10 cells, purple) of STR treatment before each experiment. Each point and bar represents mean and ±SEM, respectively. C, Average mobile fractions (mean ± SEM) shown in B. D, Fluorescence recovery curves after photobleaching at focal spots expressing mCherry-GPHN in WASH (n = 15 cells, black, shown in Fig. 4D), WASH with BAPTA-AM (n = 5 cells, orange), WASH with KN62 (n = 6 cells, green), WASH with Rp-cAMP (n = 5 cells, cyan), and WASH with GFX (n = 11 cells, magenta) conditions. Each point and bar represents mean and ±SEM, respectively. E, Averaged mobile fractions at focal spots expressing mCherry-GPHN in WASH (n = 15 cells, shown in Fig. 4E), WASH with BAPTA-AM (n = 5 cells), WASH with KN62 (n = 6 cells), WASH with Rp-cAMP (n = 5 cells), and WASH with GFX (n = 11 cells) conditions (mean ± SEM), for the data shown in D. The significance of difference was tested by the Kruskal–Wallis test followed by post hoc comparisons using the Mann–Whitney U test with Bonferroni’s correction. **p < 0.01/15 and p < 0.01/10 after Bonferroni’s correction for the 15 and 10 tests, respectively. n.s., Not significant.
- Figure 6.
GlyR activation modulates its diffusion properties around active synapses. A, Representative images of QD-labeled endogenous GlyRs (magenta) and FM4-64-labeled active synapses (green) shown as maximum intensity projections of 30 s of image recordings in Ctrl, cSTR, and WASH conditions (top panels). The bottom panels show the trajectories of correspondent QD-labeled GlyRs (centroids) shown in the top panels inside (green) and outside (magenta) active synapses. Scale bars, 2 µm. B, Cumulative probabilities of the median diffusion coefficient of each QD-GlyR within the active synapse region in Ctrl (n = 101 particles, 7 cells), cSTR (n = 125 particles, 8 cells), and WASH (n = 54 particles, 7 cells) conditions. C, Median diffusion coefficients of each QD-GlyR at synapses comparing Ctrl, cSTR, and WASH conditions (median ± 25–75% interquartile range (IQR)). D, Average dwell times of QD-GlyRs within synapse regions (Ctrl: n = 150 particles, 7 cells; cSTR: n = 314 particles, 8 cells; WASH: n = 97 particles, 7 cells, mean ± SEM). E, Average confinement size of QD-GlyRs at synapses (mean ± SEM) in Ctrl (n = 49 particles, 7cells), cSTR (n = 38 particles, 8 cells), and WASH (n = 19 particles, 7 cells) conditions. Significance of difference was tested by Kruskal–Wallis test followed by post hoc comparisons using the Mann–Whitney U test with Bonferroni’s correction, ***p < 0.001/3 after Bonferroni’s correction for the three tests (C) and by one-way ANOVA with Bonferroni post hoc test (D, E). **p < 0.01, ***p < 0.001. n.s., Not significant.
- Figure 7.
Summary diagram of activation-dependent postsynaptic GlyR clustering. (Left) Antagonized GlyRs by chronic application of strychnine (STR) are highly diffusive even at synaptic site in the presence of gephyrin, a scaffolding protein. (Middle) Strychnine removal allows GlyRs to be activated and decreases lateral diffusion at synaptic sites. (Right) Density of GlyRs increases at synaptic sites due to relatively lower diffusive properties than the extrasynaptic area.
In this issue
