Calcineurin mediates synaptic scaling via synaptic trafficking of Ca2+-permeable AMPA receptors

Abstract

Homeostatic synaptic plasticity is a negative-feedback mechanism for compensating excessive excitation or inhibition of neuronal activity. When neuronal activity is chronically suppressed, neurons increase synaptic strength across all affected synapses via synaptic scaling. One mechanism for this change is alteration of synaptic AMPA receptor (AMPAR) accumulation. Although decreased intracellular Ca2+ levels caused by chronic inhibition of neuronal activity are believed to be an important trigger of synaptic scaling, the mechanism of Ca2+-mediated AMPAR-dependent synaptic scaling is not yet understood. Here, we use dissociated mouse cortical neurons and employ Ca2+ imaging, electrophysiological, cell biological, and biochemical approaches to describe a novel mechanism in which homeostasis of Ca2+ signaling modulates activity deprivation-induced synaptic scaling by three steps: (1) suppression of neuronal activity decreases somatic Ca2+ signals; (2) reduced activity of calcineurin, a Ca2+-dependent serine/threonine phosphatase, increases synaptic expression of Ca2+-permeable AMPARs (CPARs) by stabilizing GluA1 phosphorylation; and (3) Ca2+ influx via CPARs restores CREB phosphorylation as a homeostatic response by Ca2+-induced Ca2+ release from the ER. Therefore, we suggest that synaptic scaling not only maintains neuronal stability by increasing postsynaptic strength but also maintains nuclear Ca2+ signaling by synaptic expression of CPARs and ER Ca2+ propagation.

Figure 1. TTX-induced multiplicative synaptic scaling mediated by CPARs.

(a) Representative traces of mEPSC recordings in each condition (n = number of cells). (b) Average mEPSC amplitude (***p<.001 and ****p<.0001, one-way ANOVA, uncorrected Fisher's LSD). (c) Average mEPSC frequency. (d) Average decay time (peak to 10%) (*p<.05 and ***p<.001, one-way ANOVA, uncorrected Fisher's LSD). (e) Average cumulative probability of mEPSC amplitude. TTX distribution is significantly different from no TTX (p = .015, K-S test). Distribution of TTX scaled down by a factor of 1.42 fitted to no TTX distribution (p = .935, K-S test).

Figure 2. TTX increased GluA1 surface trafficking by phosphorylation but decreased synaptic calcineurin levels.

(a) Representative immunoblots and quantitative analysis of synaptosomes from cultured cortical neurons in the presence and absence of TTX treatment (n = 6 experiments, *p<.05, unpaired two-tailed Student's t test). (b) Representative immunoblots of surface biotinylation and a summary graph in the presence and absence of TTX treatment (n = 3 experiments, *p<.05, unpaired two-tailed Student's t test). (c) Representative immunoblots showing deficiency of phosphorylation in GluA1 S845A mutant neurons (n = 3 experiments). Representative traces of mEPSCs in the presence or absence of TTX (n = number of cells). Average mEPSC amplitude, frequency, and decay time (peak to 10%).

Figure 3. TTX reduced in vivo calcineurin activity in a time-dependent manner.

Representative images of CFP channel, FRET channel, and pseudocolored emission ratio (Y/C) in each condition [blue (L), low emission ratio; red (H), high emission ratio]. Scale bar is 10 µm. A summary graph showing average of emission ratio (Y/C) in each condition (n = number of cells) (****p<.0001, one-way ANOVA, uncorrected Fisher's LSD).

Figure 4. Inhibition of calcineurin activity induced a multiplicative pharmacologic form of synaptic scaling mediated by CPARs.

(a) Representative traces of mEPSCs in each condition (n = number of cells). (b) Average mEPSC amplitude (****p<.0001, one-way ANOVA, uncorrected Fisher's LSD). (c) Average mEPSC frequency (****p<.0001, one-way ANOVA, uncorrected Fisher's LSD). (d) Average decay time (peak to 10%) (*p<.05, **p<.01, and ***p<.001, one-way ANOVA, uncorrected Fisher's LSD). (e) Average cumulative probability of mEPSC amplitude. FK506 distribution is significantly different from DMSO (p = .031, K-S test). Distribution of FK506 scaled down by a factor of 1.44 fitted to DMSO distribution (p = .984, K-S test).

Figure 5. Inhibition of calcineurin activity increased GluA1 surface trafficking by phosphorylation but decreases calcineurin.

(a) Representative immunoblots and quantitative analysis of synaptosomes from cultured cortical neurons in the presence and absence of FK506 treatment (n = 5 experiments, *p<.05, unpaired two-tailed Student's t test). (b) Representative immunoblots of surface biotinylation (D, DMSO; F, FK506) and a summary graph in the presence or absence of FK506 treatment (n = 6 experiments, *p<.05, unpaired two-tailed Student's t test).

Figure 6. Constitutively active calcineurin mutant inhibited TTX-induced synaptic scaling.

(a) Representative immunoblots showing overexpressing GluA1 with or without truncated mutant calcineurin (CaN-ΔAI) in HEK293 cells and a summary graph of pGluA1(S845) levels in each condition (n = 3 experiments, *p<.05, unpaired two-tailed Student's t test). (b) Representative traces of mEPSCs in each condition (n = number of cells). (c) Average mEPSC amplitude (****p<.0001, one-way ANOVA, uncorrected Fisher's LSD). (d) Average mEPSC frequency. (e) Average decay time (peak to 10%) (**p<.01, one-way ANOVA, uncorrected Fisher's LSD).

Figure 7. Synaptic scaling restored Ca²⁺ signals via TTX-induced CPARs.

(a) Schematic of Ca²⁺ imaging protocol. A red line indicates Ca²⁺ imaging, a black line represents the presence of TTX, and a blue line shows the addition of naspm. (b) Example bar graphs of Ca²⁺ activity in each condition. Each bar represents the GCaMP5 fluorescence intensity detected in a single exposure frame. Scale bars are 50 s. (c) Normalized average of total Ca²⁺ activity in each condition (n = number of neurons, **p<.01, ***p<.001, and ****p<.0001, one-way ANOVA, uncorrected Fisher's LSD).

Figure 8. Synaptic scaling restored Ca²⁺ signals via FK506-induced CPARs.

(a) Schematic of Ca²⁺ imaging protocol. A red line indicates Ca²⁺ imaging, a purple line shows the presence of DMSO, a black line represents the addition of TTX, a blue arrow shows the treatment of napsm, and a green line reveals the incubation of FK506. (b) Example bar graphs of Ca²⁺ activity in each condition. Each bar represents the GCaMP5 fluorescence intensity detected in a single exposure frame. Scale bars are 50 s. (c) Normalized average of total Ca²⁺ activity in each condition (n = number of neurons, **p<.01 and ****p<.0001, one-way ANOVA, uncorrected Fisher's LSD).

Figure 9. Synaptic scaling maintained CREB activity via ER Ca²⁺ release.

(a) Example bar graphs of Ca²⁺ activity in each condition. Each bar represents the GCaMP5 fluorescence intensity detected in a single exposure frame. Scale bars are 50 s. (b) Normalized average of total Ca²⁺ activity in each condition (n = number of neurons, ****p<.0001, one-way ANOVA, uncorrected Fisher's LSD). (c) Representative immunoblots of nuclear fraction from each condition. A bar graph showing normalized ratio of pCREB/CREB in each condition (n = 3 experiments, *p<.05 and **p<.01, one-way ANOVA, uncorrected Fisher's LSD).

Figure 10. A model of CPAR/calcineurin-dependent synaptic scaling for homeostasis of Ca²⁺ signaling.

A model showing regulation of synaptic insertion of CPAR by calcineurin leading to activity deprivation-induced synaptic scaling, followed by homeostasis of CREB activation in cultured cortical neurons.