pubmed.ncbi.nlm.nih.gov

Physiological role for GABAA receptor desensitization in the induction of long-term potentiation at inhibitory synapses - PubMed

  • ️Fri Jan 01 2021

Physiological role for GABAA receptor desensitization in the induction of long-term potentiation at inhibitory synapses

Martin Field et al. Nat Commun. 2021.

Abstract

GABAA receptors (GABAARs) are pentameric ligand-gated ion channels distributed throughout the brain where they mediate synaptic and tonic inhibition. Following activation, these receptors undergo desensitization which involves entry into long-lived agonist-bound closed states. Although the kinetic effects of this state are recognised and its structural basis has been uncovered, the physiological impact of desensitization on inhibitory neurotransmission remains unknown. Here we describe an enduring form of long-term potentiation at inhibitory synapses that elevates synaptic current amplitude for 24 h following desensitization of GABAARs in response to agonist exposure or allosteric modulation. Using receptor mutants and allosteric modulators we demonstrate that desensitization of GABAARs facilitates their phosphorylation by PKC, which increases the number of receptors at inhibitory synapses. These observations provide a physiological relevance to the desensitized state of GABAARs, acting as a signal to regulate the efficacy of inhibitory synapses during prolonged periods of inhibitory neurotransmission.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. GABA causes a long-term increase in inhibitory synapse efficacy.

a Representative spontaneous inhibitory postsynaptic currents (sIPSCs) recorded from cultured hippocampal neurons voltage-clamped at −60 mV and pretreated with either vehicle or 1 mM GABA for 20 min ~24 h prior to recording. b Bar graph of mean amplitudes (pA) of sIPSCs displaying clean, monophasic rise phases (two-tailed t-test: t(7) = 4.75, p = 0.0021), and c median interevent intervals (IEIs) following pre-exposure to vehicle (grey) or 1 mM GABA (pink); (two-tailed t-test: t(7) = −0.21, p = 0.83); n = 4 neurons for control, and 5 neurons for GABA). d Mean cumulative probability distribution of sIPSC amplitudes and e maximum rate of rise slopes for all sIPSCs recorded after vehicle or GABA treatment. n = 718 events from 4 neurons, and 2520 events from 5 neurons, respectively. f Bar graph of mean amplitude of mIPSCs recorded from neurons pretreated with 100 μM GABA for 20 min prior to recording (n = 11 neurons (Ctrl), 12 neurons (+GABA); two-tailed t-test: t(21) = 2.71, p = 0.013). g Cumulative probability distribution of mIPSC amplitudes. All data in the bar graphs in this figure and those proceeding it are presented as mean values +/− SEM. Source data are provided as a Source data file for Fig. 1.

Fig. 2
Fig. 2. α2V297L and γ2V262F alter the desensitization of GABAARs.

a Average GABA (10 mM)-activated currents showing macroscopic desensitization of wild-type (black) mutant α2 (red) and γ2L (blue) GABAARs in outside-out patches pulled from transiently transfected HEK293 cells expressing α2β2γ2L wild-type and mutated subunits, as indicated. b Extents of desensitization caused by 10 mM GABA measured as the % depression of the peak to steady-state GABA current, n = 5, 6 and 9 patches for receptors containing α2WT, α2V297L and γ2LV262F, respectively (one-way ANOVA: F(2, 17) = 92.45, p = 7.3 × 10−10; Tukey test (wt, α2V297L): p = 3.4 × 10−6; Tukey test (wt, γ2LV262F): p = 0.0004). c Weighted decay time constants for exponential fits to macroscopic desensitization curves for α2WT, α2V297L and γ2LV262F containing GABAARs (one-way ANOVA: F(2,15) = 0.18, p = 0.84). n = 5, 4 and 9 patches, respectively; note that these are the same patches as panel (b), but weighted time constants could not be calculated for two of the α2V297L patches due to insufficient decay of the currents. d Averaged GABA currents recorded from outside-out HEK cell patches expressing GABAARs (colour coded as in a) showing responses to two 1 ms pulses of 10 mM GABA separated by 500 ms. e Re-sensitization of the peak GABA current measured from the relative amplitude of the current evoked by the second GABA pulse over a series of interevent intervals, n = 4 patches for each. Source data are provided as a Source data file for Fig. 2.

Fig. 3
Fig. 3. γ2V262F expression increases sIPSC amplitudes at room temperature.

a Representative sIPSC traces, recorded at 21 °C from voltage-clamped neurons expressing the indicated constructs (mock transfection, black; α2V297L, green; γ2V262F, orange). b Average waveforms of sIPSCs recorded for each receptor construct, aligned on the rising phase and not peak scaled. c Mean sIPSC amplitudes for neurons either mock-transfected, or expressing α2 or γ2L wild type, or α2V297L, γ2LV262F (one-way ANOVA: F(4, 57) = 5.68, p = 0.00064; Tukey test (mock, α2V297L): p = 0.99; Tukey test (mock, γ2lV262F): p = 0.0022; n = 14, 14, 11, 14 and 9 neurons, respectively). d Mean sIPSC amplitude plotted against mean sIPSC frequency for neurons transfected with the indicated constructs. e Mean 20–80% rise time of the synaptic events of the recorded cells (one-way ANOVA: F(4,57) = 3.24, p = 0.017; Tukey test (V297L, V262F): p = 0.013). f Median interevent intervals (IEI) for all cells (one-way ANOVA: F(4,57) = 1.72, p = 0.16). g Mean cumulative probability distributions of the amplitudes, and h maximum rate of rise for the slopes of all sIPSCs recorded; n = 4334 events from 14 cells (Mock), 3973 events from 11 cells (α2V297L) and 1596 events from 9 cells (γ2LV262F). Source data are provided as a Source data file for Fig. 3.

Fig. 4
Fig. 4. γ2V262F expression increases miniature IPSC amplitudes.

a Representative mIPSCs recorded in voltage-clamp mode from neurons in the presence of tetrodotoxin (500 nM) either mock-transfected (black) or expressing γ2V262F (orange). b Mean amplitudes of mIPSCs for the constructs shown in a (two-tailed t-test vs mock: t(9) = −2.87, p = 0.019). c Cumulative probability distribution of mIPSC amplitudes and d maximum rate of rise slopes of all mIPSCs recorded; n = 18,668 events from 6 cells (mock), and 3742 events from 5 cells (γ2LV262F). e Mean 20–80% rise times of the mIPSCs recorded from each cell (two-tailed t-test: t(9) = 2.84, p = 0.019). f Median IEI for the mIPSCs recorded from each cell (two-tailed t-test: t(9) = −2.52, p = 0.033). Source data are provided as a Source data file for Fig. 4.

Fig. 5
Fig. 5. Induction of iLTP by GABA requires both GABAAR desensitization and PKC activation.

a Representative sIPSC traces recorded in voltage-clamp mode from neurons ~20 min after treatment with either GABA (1 mM, pink) or vehicle (black). b Average, unscaled sIPSC waveforms aligned on the rising phase. c Cumulative probability distribution of the amplitudes of all sIPSCs recorded: n = 4178 events from 13 cells, 2799 events from 10 cells, 1387 events from 7 cells, and 1440 events from 5 cells, respectively, for each construct (vehicle, black; GABA treatment, pink; α2V297L + GABA, blue; GABA in the presence of bisindolylmaleimide-I (Bis), orange). d Mean amplitudes of sIPSCs recorded from neurons pretreated with either vehicle or GABA: while expressing α2V297L, or in the presence of bisindolylmaleimide-I (Bis) (one-way ANOVA: F(3, 31) = 13.85, p = 0.0000066; Tukey test (vehicle vs GABA): p = 0.00002; Tukey test (vehicle, α2V297L + GABA): p = 0.99; Tukey test (vehicle vs GABA + bisindolylmaleimide): p = 0.97; n = 13, 10, 7 and 5 cells, respectively). e Median IEI for all events recorded from neurons after the indicated treatments (one-way ANOVA: F(3, 31) =  1.33, p = 0.28). f Maximum rate of rise slopes for all sIPSCs recorded. g Weighted time constants of exponential fits of the decay phases of sIPSCS recorded after treatment with either vehicle or GABA (two-tailed t-test: t(21) = −1.72, p = 0.10). h Mean 20–80% rise times of all events recorded for each neuron in the indicated conditions (one-way ANOVA: F(3,31) = 0.97, p = 0.41). Source data are provided as a Source data file for Fig. 5.

Fig. 6
Fig. 6. Cluster analysis of sIPSCs.

a T-distributed stochastic neighbour embedding (t-SNE) plot of all mock or GABA-pretreated sIPSCs from Fig. 5. The t-SNE plot was produced using the same parameters used to analyse the clustering of events (see ‘Methods’). Events are colour coded according to the clustering determined by skewed mixture modelling of the data (10 total clusters). b Plot of 20–80% rise time against instantaneous frequency of all sIPSCs, coloured by cluster. c Proportion of vehicle (grey) and 1 mM GABA-pretreated (pink) IPSC events in each cluster identified by the skewed mixture modelling. d–h Comparison of vehicle against GABA-treated events for each of the 10 clusters, for d peak IPSC amplitude, e 10–90% rise time, f instantaneous frequency, g rise slope and h maximum rise slope. On all boxplots, the box represents the interquartile range, the line within the box represents the median, and the whiskers represent the 10–90% range. Source data are provided as a Source data file for Fig. 6.

Fig. 7
Fig. 7. GABA is inhibitory under conditions that induce iLTP.

a Extracellular spike currents recorded from a cultured hippocampal neuron in cell-attached voltage-clamp mode showing the response to 1 mM GABA application. b Summary of spike current frequencies before and after GABA exposure, n = 4 neurons. c Representative current trace recorded from a cultured hippocampal neuron in cell-attached voltage-clamp mode showing the response to bicuculline (50 µM) application—two segments of the recording are also shown below with higher time resolution. d Summary of spike current frequencies before and after bicuculline application, n = 5 neurons (paired-sample t-test: t(4) = −7.87, p = 0.0014). Baseline measures signify the spike firing rate before any treatment. Source data are provided as a Source data file for Fig. 7.

Fig. 8
Fig. 8. GABAAR desensitization increases synaptic receptor numbers by phosphorylating γ2S327 and β2S410.

a Plots of synaptic current amplitude variance against current amplitude from peak-scaled non-stationary variance analysis of sIPSCs presented in Fig. 5 (n = 13 and 10 cells, respectively). Data are shown for sIPSCs recorded from all GABA pretreated neurons (pink) and from controls (grey). b Single-channel current amplitudes (Iu) are calculated for synaptic GABAARs for each treatment (two-tailed t-test: t(21) = −1.17, p = 0.25). c Synaptic receptor numbers (Nc) calculated for sIPSCs for each treatment (two-tailed t-test: t(21) = −4.48, p = 0.00021). d Averaged sIPSC waveforms recorded from untransfected cells pretreated with either vehicle (grey) or GABA (pink), and cells expressing γ2S327A pretreated with GABA (purple), aligned to their rise phases. e Mean cumulative probability distributions of the sIPSC amplitudes and f Maximum rate of rise slopes of all sIPSCs recorded; n = 3470 events from 14 cells, 3808 events from 9 cells, 2266 events from 8 cells, and 4125 events from 6 cells, respectively, for each construct (mock, grey; +GABA, pink; γ2S327 + GABA, purple; β2S410 + GABA, blue). g Mean sIPSC amplitudes for the same conditions as in panel (e) and (f) (one-way ANOVA: F(3,33) = 29.20, p = 2.1 × 10−9; Tukey test (vehicle, S327A + GABA): p = 0.09; Tukey test (vehicle vs S410A): p = 0.12). h Mean 20–80% rise times for sIPSCs recorded under the indicated conditions (one-way ANOVA: F(3,34) = 1.32, p = 0.28). i Median IEI of the sIPSCs recorded from cells under the indicated conditions (one-way ANOVA: F(3,34) = 0.66, p = 0.58). j Immunoblots of lysates taken from HEK293 cells transiently transfected with the indicated constructs treated with either vehicle or GABA (1 mM) for 20 min prior to lysis. Blots were probed for phosphorylated (p) γ2S327 (top row), and for total γ2 levels (bottom row). k Fold changes in the phosphorylation levels for γ2S327 in cells expressing the indicated constructs (control, black; α1V296L, red; γ2LV262F, blue), after GABA 20 min treatment (one-sample t-test with a onefold change as the null hypothesis: t(11) = 2.35, p = 0.038; one-way ANOVA: F(2, 18) = 15.52, p = 0.00020; Tukey test (wt vs α1V296L): p = 0.036; Tukey test (wt vs γ2LV262F): p = 0.042; n = 12, 9 and 11 lysates for WT, α1V296L and γ2V262F, respectively). Source data are provided as a Source data file for Fig. 8.

Fig. 9
Fig. 9. Induction of iLTP by PKC requires GABAAR activation and desensitization.

a Representative sIPSC traces recorded from cells pretreated with either vehicle (black) or phorbol 12-myristate 13-acetate (PMA, green). b Average waveforms of sIPSCs for each condition. c Mean cumulative probability distributions of the amplitudes of all sIPSCs recorded for each receptor construct (WT or α2V297L) and condition (vehicle, grey; +PMA, turquoise; α2V297L + PMA, light green; +PMA & picrotoxin (Ptx), purple); n = 1482 events from 7 neurons, 12,532 events from 10 neurons, 11,170 events from 5 neurons and 4745 events from 7 neurons, respectively. d Cumulative probability distributions of the maximum rate of rise slopes of all sIPSCs recorded as in panel (c). e Summary data for mean sIPSC amplitudes recorded from cells pretreated with either vehicle or PMA, or cells expressing α2V297L and pretreated with PMA, or pretreated with PMA in the presence of picrotoxin (100 µM; one-way ANOVA: F(3,25) = 5.84, p = 0.0036; Tukey test (vehicle vs PMA): p = 0.022; Tukey test (vehicle vs PMA + Ptx): p = 0.98; Tukey test (vehicle, α2V297L + PMA): p = 0.99). f Median sIPSC IEIs under the same four conditions as in (e) (one-way ANOVA: F(3,25) = 8.03, p = 0.0006; Tukey test (vehicle vs PMA): p = 0.0013; Tukey test (vehicle vs α2V297L + PMA): p = 0.0016; Tukey test (vehicle vs PMA + Ptx: p = 0.030). g Mean 20–80% rise time of sIPSCs for neurons recorded under the indicated conditions (one-way ANOVA: F(3,25) = 1.15, p = 0.35). Source data are provided as a Source data file for Fig. 9.

Fig. 10
Fig. 10. Allosteric modulators of GABAARs that increase the occupancy of the desensitized state also induce iLTP.

a Averaged GABA currents of GABAARs in outside-out patches pulled from transiently transfected HEK293 cells, to sequential pulses of 10 mM GABA in the presence of vehicle, 10 μM pregnenolone sulfate (PS, light blue), or 5 μM etomidate (purple). b Re-sensitization of the second GABA current with increasing interevent intervals in the presence of the indicated modulators; n = 4, 3, and 3, respectively, for each condition. c Weighted decay time constants for GABA current responses in control and exposed to the modulators (Eto = etomidate, PS = pregnenolone sulfate; one-way ANOVA: F(2,7) = 55.9, p = 0.00005; Tukey test (vehicle vs etomidate): p = 0.00007; Tukey test (etomidate vs pregnenolone sulfate): p = 0.0001). d Representative sIPSCs recorded from neurons pretreated with the indicated modulators (vehicle, 1 µM pregnenolone sulfate or 5 µM etomidate). e Average sIPSC waveforms for each modulator aligned on the current rising phases. f Mean sIPSC amplitudes (one-way ANOVA: F(2, 29) = 7.19, p = 0.0029; Tukey test (mock vs PS): p = 0.0099; Tukey test (mock vs etomidate): p = 0.0058) and g weighted decay time constants for sIPSCs recorded after treatment with the indicated modulators (one-way ANOVA: F(2,29) = 0.11, p = 0.89). h Median IEIs for the sIPSCs recorded from neurons after treatment under the indicated conditions (one-way ANOVA (IEI) F(2,29) = 4.25, p = 0.27). i 20–80% rise times of sIPSCs recorded after treatment (one-way ANOVA (rise times) F(2,29) = 4.25, p = 0.024; Tukey test (Mock vs Etomidate): p = 0.029; Tukey test (Mock vs PS): p = 0.078). j Cumulative probability distribution of the sIPSC amplitudes. k Maximum rate of rise slopes of all sIPSCs recorded; n = 3930 events from 11 cells, 8646 events from 10 cells and 4467 events from 11 cells, respectively. Source data are provided as a Source data file for Fig. 10.

Similar articles

Cited by

References

    1. Bliss TVP, Collingridge G, Morris RGM, Reymann KG. Long-term potentiation in the hippocampus: discovery, mechanisms and function. Neuroforum. 2018;24:A103–A120. doi: 10.1515/nf-2017-A059. - DOI
    1. Nicoll RA. A brief history of long-term potentiation. Neuron. 2017;93:281–290. doi: 10.1016/j.neuron.2016.12.015. - DOI - PubMed
    1. Lisman, J. Glutamatergic synapses are structurally and biochemically complex because of multiple plasticity processes: long-term potentiation, long-term depression, short-term potentiation and scaling. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 372, 10.1098/rstb.2016.0260 (2017). - PMC - PubMed
    1. Malenka RC, Bear MF. LTP and LTD: an embarrassment of riches. Neuron. 2004;44:5–21. doi: 10.1016/j.neuron.2004.09.012. - DOI - PubMed
    1. Barberis, A. Postsynaptic plasticity of GABAergic synapses. Neuropharmacology169, 10.1016/j.neuropharm.2019.05.020 (2019). - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources