Stress-induced enhancement of mouse amygdalar synaptic plasticity depends on glucocorticoid and ß-adrenergic activity

Abstract

Background: Glucocorticoid hormones, in interaction with noradrenaline, enable the consolidation of emotionally arousing and stressful experiences in rodents and humans. Such interaction is thought to occur at least partly in the basolateral nucleus of the amygdala (BLA) which is crucially involved in emotional memory formation. Extensive evidence points to long-term synaptic potentiation (LTP) as a mechanism contributing to memory formation. Here we determined in adolescent C57/Bl6 mice the effects of stress on LTP in the LA-BLA pathway and the specific roles of corticosteroid and β-adrenergic receptor activation in this process.

Principal findings: Exposure to 20 min of restraint stress (compared to control treatment) prior to slice preparation enhanced subsequent LTP induction in vitro, without affecting baseline fEPSP responses. The role of glucocorticoid receptors, mineralocorticoid receptors and β2-adrenoceptors in the effects of stress was studied by treating mice with the antagonists mifepristone, spironolactone or propranolol respectively (or the corresponding vehicles) prior to stress or control treatment. In undisturbed controls, mifepristone and propranolol administration in vivo did not influence LTP induced in vitro. By contrast, spironolactone caused a gradually attenuating form of LTP, both in unstressed and stressed mice. Mifepristone treatment prior to stress strongly reduced the ability to induce LTP in vitro. Propranolol normalized the stress-induced enhancement of LTP to control levels during the first 10 min after high frequency stimulation, after which synaptic responses further declined.

Conclusions: Acute stress changes BLA electrical properties such that subsequent LTP induction is facilitated. Both β-adrenergic and glucocorticoid receptors are involved in the development of these changes. Mineralocorticoid receptors are important for the maintenance of LTP in the BLA, irrespective of stress-induced changes in the circuit. The prolonged changes in BLA network function after stress may contribute to effective memory formation of emotional and stressful events.

Figure 1. Schematic overview of the experimental design and method.

A Schematic time line of the experimental design. Thirty minutes after injection with vehicle or antagonist, mice were subjected to either restraint stress or left undisturbed. After another 20 min, animals were rapidly decapitated and brains were collected for slice preparation. Approximately 1–4 hours after decapitation, in vitro electrophysiological recordings were carried out. B Positioning of the stimulation (S_LA) and the recording electrode (R_BLA) at their sites within the lateral (LA) and the BLA respectively in mouse coronal brain slices. C) Representative local fEPSP evoked by stimulation of the LA–BLA pathway. The amplitude of the signal was calculated according to the formula (a+b)/2 as indicated in the figure. * indicates position of the stimulus artefact.

Figure 2. The effect of restraint stress on baseline transmission and LTP.

A Input-output curves constructed from the fEPSP amplitude vs increasing stimulation intensities at the BLA from brain slices of control (open symbols, n = 8) and restrained stressed mice (filled symbols, n = 7). No significant differences in stimulus-response relationships or maximal fEPSP amplitude were found. B HFS (1×100 Hz, 1 s) of the LA afferents resulted in stable LTP at BLA synapses in slices of control mice which was more pronounced in stressed mice. Averaged mean values during C 0–10 min and D 50–60 min of the post-tetanus recording period indicate that compared to controls (white columns), late LTP is significantly enhanced in stressed mice (black columns). Dashed line indicate pre-tetanus baseline levels. Error bars indicate SEM. ** p<0.01.

Figure 3. BLA LTP and antagonist pretreatment in undisturbed mice.

A HFS evoked potent and stable LTP at BLA synapses in all experimental groups. Compared to vehicle injected mice (n = 15, white circles), BLA LTP was not affected by propranolol (n = 6, blue circles) or mifepristone pretreatment (n = 6; grey circles). Pretreatment with spironolactone (n = 7, red circles) however gradually attenuated LTP induced some hours later in vitro. C Bar chart illustrating the averages per treatment group for 0–10 min and D 50–60 min of the post-tetanus period, showing attenuated LTP in spironolactone treated mice at the later time-point. Dashed line indicates pre-tetanus baseline values. Error bars indicate SEM. Tukey's post hoc test * p<0.05, ** p<0.01.

Figure 4. BLA LTP in acutely stressed mice pretreated with antagonists.

A HFS resulted in attenuated and unstable LTP in mice injected with propranolol (n = 6; blue circles) or spironolactone (n = 8; red circles) before stress exposure. Mifepristone pretreatment (n = 8; grey circles) attenuated LTP compared to vehicle (n = 14, white circles). C Bar chart illustrating the averages per treatment group for 0–10 min and D 50–60 min of the post-tetanus period. Dashed line indicates pre-tetanus baseline values. Error bars indicate SEM. Tukey's post hoc test * p<0.05, ** p<0.01, *** p<0.001.