Adrenal factors regulate the effects of acute stress on synaptic potentiation
Previously, we and others have demonstrated effects of acute glucocorticoid exposure on synaptic potentiation12,13,15,16. Here, we expand on this finding to further examine the molecular pathway between the initial stress induction and the outcome on synaptic plasticity. Adrenal-intact and adrenalectomised (ADX) rats underwent a 30-minute inescapable restraint stress protocol26 while control animals were unstressed (figure 1A). Plasma samples were collected from each animal at times 0, 30, 60, 90 and 120 minutes, while whole hippocampi were collected at times 0 and 60 minutes from stress induction. The plasma corticosterone levels of all rats in this study were assessed by radioimmunoassay (RIA) and analysed by Two-way ANOVA, which revealed a significant effect of time (p<0.0001), a significant effect of ADX (p<0.0001), and a significant interaction of time*ADX (p<0.0001). ADX rats had extremely low levels of corticosterone at all time points during the restraint protocol, so the RIA minimal detectable concentration (1.7ng/ml) was used for analysis in cases where sample concentration was too low to detect. Adrenal-intact rats exhibited the expected rise in circulating corticosterone by the end of the 30-minute restraint stress protocol (1358.0±260.8 ng/ml, compared to 46.8±29.0 at baseline; p<0.001 Dunnett’s post hoc test, and 3.7±1.0 in time-matched ADX rats; p<0.0001 Bonferroni post hoc test). Corticosterone levels were still significantly elevated above baseline at 60 minutes (565.6±136.6ng/ml, compared to 46.8±29.0 at baseline; p<0.01 Dunnett’s post hoc test, and 3.8±0.8 in time-matched ADX rats; p < 0.001 Bonferroni post hoc test) before falling to baseline levels at 120 minutes.
GluA1 is known to be a key player in modulating plasticity at the hippocampal synapse, with PKA-dependent phosphorylation at its Ser-845 residue facilitating receptor insertion into the synapse widely accepted as an early step in the synaptic potentiation processes6,27,28. Here, we have measured pSer-845 GluA1 (pGluA1) as well as total GluA1 in synaptic plasma membrane (SPM) and postsynaptic density (PSD) cellular fractions prepared from whole hippocampi taken from adrenal-intact or adrenalectomised (ADX) rats at 60-minutes after the onset of a 30-minute restraint stress, compared to unstressed controls (figure 1 B, C). In the SPM, pGluA1 was enhanced in stressed adrenal-intact rats relative to unstressed controls, and the stress effect was ablated in ADX rats. Two-way ANOVA revealed a significant effect of stress (p=0.0483) and significant stress*ADX interaction (p=0.0381). Bonferroni post hoc multiple comparison tests showed that the stress-induced increase in pGluA1 in the SPM was significant (p=0.0104), and that this stress effect was significantly reduced by ADX (p=0.0274). In the PSD, no significant effect of stress, ADX, or interaction was found for pGluA1. No significant effect of stress, ADX, or interaction was found for total GluA1 in either the SPM or the PSD. These data show that the small but significant increase in SPM pGluA1 occurs in response to stress and is adrenal-dependent.
Corticosteroid receptor antagonism with RU-486 further modulates the glucocorticoid effect on pGluA1.
We next tested whether acute corticosterone treatment would be sufficient to reinstate the effect of stress on pGluA1, and conversely whether pre-treatment with the corticosteroid receptor antagonist RU-486 would ablate the corticosterone treatment effect. Hippocampi taken from ADX rats killed exactly 60 minutes after receiving a stress dose of corticosterone (3mg/kg i.p.) with or without pre-treatment with RU-486 (20mg/kg i.p.) were again processed to generate SPM and PSD fractions to assess pGluA1 and total GluA1 (figure 2 A, B). Control ADX rats received subcutaneous injections of vehicle (ethanol diluted in saline). A significant effect of treatment was detected by One-way ANOVA in both SPM (p=0.0163) and PSD fractions (p=0.0005). Interestingly, Bonferroni post hoc tests revealed that the small but significant corticosterone-induced increase in pGluA1 in the SPM of corticosterone-treated ADX rats (p=0.0464), was significantly augmented by pre-treatment with RU-468 (p=0.0181). In striking contrast, in the PSD, the significant corticosterone-induced increase in pGluA1 (p=0.0021) was blocked by RU-486 (p=0.0007). Significant effects of treatment were also found for total GluA1 in the SPM (p=0.0054) and PSD (p<0.0001) by One-way ANOVA. Similar to effects seen with pGluA1, Bonferroni post hoc tests revealed that RU486 treatment caused a significant increase in corticosterone-induced total GluA1 levels in the synaptic plasma membrane (p=0.0055), and a significant decrease in the postsynaptic density (p=0.002). Taken together these results are consistent with a corticosterone-inducible RU-486-sensitive mechanism, potentially involving lateral movement of pGluA1 from the SPM to the PSD.
The glucocorticoid-induced increase in total cellular content of pGluA1 is translation-dependent and transcription-independent.
We further tested whether the corticosterone-dependent effects observed also affected total cellular GluA1 and pGluA1 levels, and whether these effects involved classical transcriptional mechanisms and de novo protein synthesis. Ex vivo application of the protein translation inhibitor Cycloheximide (CX) and the transcription inhibitor Actinomycin D (Act D) were applied to hippocampal slices, subsequently incubated for 60 minutes with corticosterone then homogenised to release total cellular content (figure 2 C, D). A significant effect of treatment was detected for pGluA1 in these whole cell extracts (p=0.0068; One way ANOVA). A significant decrease in corticosterone-induced pGluA1 was detected in the CX pre-treated tissue (p<0.001, Bonferroni post hoc test) without a significant reduction in total GluA1. Act D had no effect on the corticosterone-inducible pGluA1 or total GluA1 levels.
Together, these data indicate that a glucocorticoid-induced translation-dependent transcription-independent increase in phosphorylation of GluA1 is seen in total cellular pGluA1 content, and further that glucocorticoid-induced changes in pGluA1 enrichment in defined parts of the plasma membrane (namely the SPM and PSD) is disrupted by antagonism of corticosteroid receptors. While RU-486 is generally thought of as a Type2 corticosteroid receptor (GR) antagonist, we cannot rule out the possibility of off-target effects on the Type1 corticosteroid receptor (MR) which is also highly expressed in the hippocampus, reported to be membrane localised10,29,30, and involved in synaptic potentiation mechanisms9. Importantly, our data suggest that glucocorticoid regulation of GluA1 function may be acting via more than one mechanism. The first mechanism via rapid translation of a protein required for phosphorylation of GluA1, either a kinase or a kinase-regulating accessory protein, and the second potential mechanism via a corticosteroid receptor-dependent lateral movement of pGluA1 from the SPM to the PSD. Considering that pre-treatment with the corticosteroid receptor antagonist RU-486 resulted in far greater accumulation of pGluA1 in the SPM than seen in physiological conditions of stress or corticosterone treatment, a model involving lateral diffusion of pGluA1 (see Park et al, 20188 for review) is strongly supported by our data.
Calcium/Calmodulin-sensitive Adenylyl cyclases are required for glucocorticoid-mediated synaptic potentiation
Activation of GluA1 via phosphorylation by PKA is a well-described mechanism for functional neuroplasticity during stress31-34. Yet, the link between signal reception at the synapse, the increase in intracellular calcium and further downstream activation of cAMP35 to upregulate levels of the active catalytic subunit of protein kinase A, which mediates its phospho-transferase abilities36 is still unclear. One group of synaptic associated molecules with potential for both processes i.e. utilization of increased calcium following NMDAR activation, with consequent increase in cAMP-mediated activation of PKA, are the calcium stimulated enzymes, the adenylyl cyclases (AC).
We therefore next investigated the requirement of these calcium sensitive enzymes in the glucocorticoid-induced increase in total cellular pGluA1 levels in ex vivo hippocampal slices. Pre-treatment with the AC inhibitor (SQ22536) prior to corticosterone treatment resulted in inhibition of the corticosterone-induced increase in pGluA1 without affecting total cellular levels of GluA1 (figure 3A, B). A significant effect of treatment on pGluA1 in the whole cell extracts (p=0.0083; One-way ANOVA) was found, with Bonferroni post hoc tests revealing a significant corticosterone-induced increase in pGluA1 (p=0.021) and significant inhibition of the corticosterone-induction by SQ22536 (p=0.020). Together, these findings support a necessary role for adenylyl cyclases in corticosterone-induced phosphorylation of GluA1.
To assess whether there was a functional sequela to the reduction in corticosterone-dependent pGluA1 after AC inhibition, we next evaluated the effect of SQ22536 on corticosterone-enhanced LTP (figure 3C). A high frequency stimulation (HFS) protocol was applied to acute hippocampal slices. This triggered an increase in field excitatory postsynaptic potentials (150.8 ±2.3% of baseline). As we have shown previously15, corticosterone treatment markedly augmented the LTP response (184.9 ± 7.9 % of baseline, p < 0.001). The augmenting effect of corticosterone was completely abolished by pre-treatment with the AC inhibitor SQ22536 (152.6 ± 10.6 % of baseline, One-way ANOVA p = 0.011, with Bonferroni post hoc tests; control vs corticosterone, p < 0.05, corticosterone vs corticosterone + SQ, p < 0.05; figure 3D). These results demonstrate for the first time that glucocorticoid-mediated amplification of EPSPs following LTP stimulation is dependent on adenylyl cyclases.
NMDAR-dependence for rapid corticosterone-induced increases in both Adenylate cyclase protein expression and phosphorylation of GluA1.
Pertinent to the mechanisms we have explored in this paper, we next interrogated the upstream role of NMDARs on adenylyl cyclases and PKA activation-dependent phosphorylation of GluA1. While it is already known that AC activity is regulated by calcium influx after NMDAR-dependent membrane depolarisation35,37, the role that glucocorticoids play in this part of the signalling cascade in relation to GluA-1 dependent synaptic potentiation is not yet understood. We found that corticosterone treatment of ex vivo hippocampal slices was able to rapidly induce an increase in total protein levels for both AC1 and AC8 as early as 60 minutes of treatment time (Figure 4A). However, only the AC8 corticosterone-dependent increase was sensitive to pre-treatment with the selective NMDAR antagonist, AP5. As AP5 competitively binds the receptor’s ligand binding site, blocking ligand binding, and consequently preventing calcium influx via the receptor’s cation channel38, our results indicate that rapid glucocorticoid-induced AC8 translation may be NMDAR-dependent in a similar manner to that already described for AC1 activity, albeit in cortical synapses39. In contrast, the corticosterone-induced increase in AC1 was not sensitive to AP5 pre-treatment, supporting a lack of NMDAR and calcium-dependence for its rapid translation in the hippocampus. Finally, and as predicted based upon our previous work15, NMDAR inhibition with AP5 also ablated the corticosterone-induced increase in total pGluA1 levels (figure 4D).
As indicated, significant effects of treatment were found by One-way ANOVA for AC1 (p=0.0191), AC8 (p=0.0007) and pGluA1 (p=0.0001) (figures 4B, C, E). Bonferroni post hoc test results are indicated on each graph, showing that the corticosterone-induced increase in AC1 (p<0.05) was not blocked by AP5 (figure 4B), the corticosterone-induced increase in AC8 (p<0.001) was ablated by pre-treatment with AP5 (p<0.01) (figure 4C) and finally that the corticosterone-induced GluA1 phosphorylation (p<0.05) was also completely abolished by AP5 (p<0.001; figure 4E).
Taken together, our findings indicate that - in addition to known calcium-dependent activation required for consequent PKA-dependent phosphorylation events40 - glucocorticoids induce increased phosphorylation of GluA1 via a mechanism involving both NMDAR-dependent and AC-dependent activity.