Acute restraint stress reverses impaired LTP in the hippocampal CA1 region in mouse models of Alzheimer’s disease

Acute stress facilitates long-term potentiation (LTP) in the mouse hippocampus by modulating glucocorticoid receptors and ion channels. Here, we analysed whether this occurs in mouse models of Alzheimer’s disease (AD) with impaired LTP induction. We found that a brief 30 min restraint stress protocol reversed the impaired LTP assessed with field excitatory postsynaptic potential recordings at cornu ammonis 3-1 (CA3-CA1) synapses in both Tg2576 and 5XFAD mice. This effect was accompanied by increased phosphorylation and surface expression of glutamate A1 (GluA1) -containing α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs). Moreover, enhanced LTP induction and GluA1 phosphorylation were sustained up to 4 h after the stress. Treatment with 200 nM dexamethasone produced similar effects in the hippocampi of these mice, which supports the glucocorticoid receptor-mediated mechanism in these models. Collectively, our results demonstrated an alleviation of impaired LTP and synaptic plasticity in the hippocampal CA1 region following acute stress in the AD mouse models.

levels 22 and a selective decrease in AMPAR-mediated synaptic excitation 23 . The suppressed AMPAR and NMDAR expression and associated synaptic transmission lead to cognitive impairment 24 . By contrast, short or acute stress, such as the forced swim test and elevated plus maze task for 20 min, facilitates NMDAR and AMPAR expression and glutamatergic transmission in rat prefrontal cortex, a brain region mainly involved in working memory 25 . Furthermore, acute stress exposure increases AMPAR surface expression and enhances LTP induction in the CA1 region of rat hippocampus 26 .
The present study was conducted to determine whether acute stress can reverse impaired hippocampal LTP in mouse models of AD. We used restraint as a paradigm to induce acute stress 27 in well-characterised transgenic mouse models, Tg2576 and 5XFAD, which express genes associated with familial AD and exhibit impaired LTP 28,29 . To characterise LTP and synaptic plasticity in the hippocampi of these mice, we recorded field excitatory postsynaptic potentials (fEPSPs) in the CA1, and measured the phosphorylation and surface expression of GluA1 subunits of AMPARs. The effects of stress were validated in vivo and ex vivo in experiments using the glucocorticoid receptor agonist dexamethasone.

Results
Acute stress rescues impaired LTP in the hippocampal CA1 region of AD mice. In the first set of experiments, we determined whether acute stress alters LTP in the CA1 region by recording fEPSPs in acutely prepared hippocampal slices from either 5XFAD or Tg2576 mice (Fig. 1a). The baseline recordings were stable for 30 min, with no significant difference between stressed and control unstressed mice. However, further recordings for 60 min following HFS revealed that LTP was not induced in slices from control 5XFAD (n = 6 [six slices from six animals per group]) or control Tg2576 (n = 6) mice (Fig. 1b,c, respectively). By contrast, LTP was induced by HFS in slices from stressed 5XFAD [t = −6.855, df = 9; p < 0.01, n = 6, unpaired t test; Fig. 1b] and stressed Tg2576 [t = −11.135, df = 6; p < 0.01, n = 6, unpaired t test; Fig. 1c] mice. Additional experiments in slices from wild-type mice confirmed that LTP is induced by HFS in slices from unstressed control mice (n = 5) and that acute stress significantly enhances the magnitude of the LTP [t = −8.8, df = 9; p < 0.01, n = 5-6, unpaired t test; Supplementary Fig. 1] (Table 1). Altogether, these data confirm that acute stress potentiates LTP and suggest that acute stress can rescue impaired LTP in mouse models of AD.

Sustained hippocampal LTP and GluA1 phosphorylation.
To determine the time course of enhanced LTP and hippocampal GluA1 phosphorylation, we analysed fEPSPs 1, 3 and 5 h after the restraint stress (Fig. 3a).   Glucocorticoid receptor agonist mimics the acute stress effects in AD mouse models. We next evaluated whether a glucocorticoid receptor agonist would induce effects similar to acute restraint stress on hippocampal LTP in 5XFAD and Tg2576 mice. Acutely prepared hippocampal slices were perfused with 200 nM dexamethasone (a dose that enhances LTP in mice 26 ) while fEPSP recordings were acquired. Baseline levels were similar between treated and untreated slices. fEPSP slopes following HFS were significantly higher in slices treated with 200 nM dexamethasone than in untreated slices from 5XFAD mice [t = −7.1, df = 9; p < 0.01, n = 6, unpaired t test; Fig. 4a] and Tg2576 mice [t = −7.2, df = 8; p < 0.01, n = 6, unpaired t test; Fig. 4b], demonstrating enhanced LTP in the presence of the glucocorticoid agonist.
Results from the surface biotinylation assays revealed that 30 min dexamethasone treatment significantly increased the surface levels of GluA1 in hippocampal tissues from 5XFAD [t = −4.4, df = 6; p < 0.01,   www.nature.com/scientificreports www.nature.com/scientificreports/ results support the previous observations on the effects of acute stress and glucocorticoids on hippocampal LTP enhancement in mice.

Discussion
The results of the present study show that exposure to 30 min of restraint stress reverses hippocampal LTP impairments and increases AMPAR GluA1 phosphorylation and surface expression. These effects persisted for more than 3 h. Moreover, we confirmed that the effects of the stress were due to glucocorticoid activation, as treatment with dexamethasone produced similar results. These findings suggest a beneficial role of acute stress on hippocampal synaptic plasticity.
Previous studies have shown that acute stress facilitates LTP by modulating glutamatergic receptors. For example, it has been shown that acute stress induced enhancement of glutamatergic transmission depends on the modifications of NMDA and AMPA receptors 25 . In mice, exposure to 30 min restraint stress shown to elevate hippocampal CA1 LTP levels through increased surface expression and GluA1 phosphorylation 26 . Furthermore, acute stress induced by immobilization and tail-shock, reported to increase the excitability of CA1 pyramidal neurons 30 , which plays an essential role in the hippocampal LTP induction. Acute stress was also shown to facilitate long-term potentiation of population spikes (PS-LTP) in the CA1 region of mouse hippocampus 31 . Similarly, acute restraint stress elevates acetylcholinesterase levels and enhances LTP in the CA1 region of the hippocampus 32 . Consistent with these observations, we also found increased surface expression of GluA1 containing AMPAR and elevated LTP in transgenic mice hippocampus which furtherly supports the LTP enhancing effect of acute stress.
A stressful brief period of swimming or immobilization elevates CaMKII levels, which are essential for AMPAR trafficking and LTP maintenance, in rat hippocampus 33,34 . We show here that acute restraint stress enhanced surface expression of GluA1 subunits of AMPARs as well as their phosphorylation. These support the phosphorylation of GluA1 subunit at S845 and their homomeric assembly in AMPARs on the membrane surface as a mechanism by which restraint stress induces LTP 26 . Notably, GluA1 phosphorylation and AMPAR www.nature.com/scientificreports www.nature.com/scientificreports/ surface expression are reported to be lower in mice overexpressing APP or exposed to Aβ. For instance, exposure to soluble oligomers of Aβ results in dephosphorylation and reduced surface expression of GluA1 containing AMPARs 17 . Overexpression of APP has been shown to decrease synaptic AMPARs and depress synaptic transmission 35 . Moreover, mice overexpressing APP/PS1 gene, showed to express lower levels of AMPARs 36 . Previous reports on the effect of acute stress in AD mice showed increased levels of amyloid production which results in decreased glutamatergic transmission and LTP impairment 37,38 . However, the stress paradigms used in these studies are longer, such as several hours to several days. In the present study, we show that acute stress for 30 min enhanced the phosphorylation and surface expression of AMPARs in the hippocampi of 5XFAD and Tg2576 mice for more than 3 h. The fact that total levels of these receptor subunits were unchanged in stressed mice rules out the possibility of new receptor synthesis in response to acute stress.
Corticosterone is the major glucocorticoid responsible for the effects of stress 39 . Whereas increased or sustained glucocorticoid levels impair long-term memory, acute exposure enhances memory and cognitive performance. In mice, acute stress for 15 min showed to elevate intra-hippocampal corticosterone levels up to 240 nM 40 . In addition, 30 min exposure to corticosterone or dexamethasone at 200 nM concentrations shown to enhance LTP induction, through facilitated AMPAR trafficking in rat hippocampus 26 . Similarly, a short exposure to corticosterone increases the surface mobility and synaptic content of the AMPAR GluA2 subunit in rat hippocampal neurons 41 . Accordingly, we showed that dexamethasone elevated the phosphorylation and surface expression of GluA1-containing AMPARs and enhanced LTP in the CA1 regions of 5XFAD and Tg2576 mice.
In summary, the present study demonstrated an alleviation of impaired LTP and synaptic plasticity in the hippocampal CA1 following acute stress in the AD mouse models. We show that acute restraint stress, via glucocorticoid activation, enhances LTP by facilitating AMPAR trafficking and excitatory synaptic transmission in this region. Thus, we suggest that these molecular alterations modulated by acute stress may benefit memory function and cognition in these AD mice. Restraint stress and hippocampal slice preparation. Mice were physically restrained in well-ventilated 50 ml Falcon tubes for 30 min. Control mice were housed in their usual cages under normal conditions. Animals were sacrificed immediately following restraint stress, (between 9:00 a.m. and 10:00 a.m.), and the brains were quickly removed and transferred to ice-cold artificial cerebrospinal fluid (aCSF; 124 mM NaCl, 3 mM KCl, 26 mM NaHCO 3 , 1.25 mM NaH 2 PO 4 , 2 mM CaCl 2 , 1 mM MgSO 4 and 10 mM glucose) (Fig. 1a). For each mouse, a mid-sagittal cut was made in the brain, and one hemisphere was returned to ice-cold aCSF until required. Transverse hippocampal slices (400 μm thick) were cut using a McIlwain tissue chopper (Mickle Laboratory Engineering Co. Ltd., Guildford, UK) and allowed to stabilise in aCSF for 1 h with perfusion of 95% O 2 and 5% CO 2 at room temperature.

Animals
Electrophysiology. Hippocampal slices recovered for approximately 60 min after the slice procedure to allow stable responses to be obtained. Two stimulating bipolar electrodes comprising 66 μM twisted nichrome wire were set on the Schaffer collateral pathway (for LTP input) and subiculum region (for control input). Extracellular field potentials were recorded in the CA1 region using microcapillary electrodes containing NaCl (3 M). Stimuli were delivered alternatively to the two electrodes (0.016 Hz each). After establishing a stable baseline for 30 min, LTP was evoked by two trains of tetanic stimuli (100 Hz for 1 s with a 30 s interval) and fEPSPs were recorded for at least 60 min. The slope of the evoked field potential response was measured and expressed relative to the normalised preconditioning baseline. Data were collected by a NI USB-6251 data acquisition module (National Instruments, Austin, TX), amplified by an Axopatch 200B amplifier (Axon Instruments, Foster City, CA) and captured and analysed using WinLTP software (www.winltp.com).
Biotinylation and streptavidin pull-down assays. Surface biotinylation in acute slices was performed as described previously with some modifications 42 . Briefly, slices were washed twice in aCSF and then incubated in aCSF containing 1 mg/ml sulpho-NHS-SS-biotin for 45 min at 4 °C to label surface membrane proteins. Biotinylated tissue was then homogenised in lysis buffer containing 25 mM Tris (pH 7.6), 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS, 1 mM NaF and a cocktail of protease inhibitors (Sigma-Aldrich, St. Louis, MO). The lysate was centrifuged at 11,000 × g for 15 min at 4 °C to remove nuclei and cellular debris. Total protein concentration was determined with a bicinchoninic acid assay (Pierce of Thermo Fisher Scientific, Waltham, MA). A small amount of the lysate was removed for later whole-cell analysis. Subsequently, 100 μl of streptavidin beads (Thermo Fisher Scientific) was added to 600 µg of protein lysate, and the mixture was placed on a rotator at 4 °C for 2 h. Samples were then washed five times in wash buffer (25 mM Tris [pH 7.6], 150 nM NaCl, 0.5% Triton X-100), and the beads were pulled down after each wash by gentle centrifugation. Bound proteins were eluted in 2× SDS reducing buffer and gently heated at 60 °C for 30 min. The resulting supernatant was transferred to new tubes and heated at 90 °C for 5 min before gel loading.
Immunoblotting. Hippocampal tissue lysates were prepared using radio immune precipitation buffer with protease inhibitors (Cell Biolabs, Inc., San Diego, CA). Protein concentrations were determined with a bicinchoninic acid assay. Proteins were resolved in 10-12% gels and transferred to polyvinylidene difluoride fEPSP slope sGluA1 pS845-GluA1   Table 2. Time dependent effect of acute stress on LTP, and phosphorylation of AMPA -GluA1 levels.