- Email: [email protected]
Chronic stress-induced dendritic reorganization and abundance of synaptosomal PKA-dependent CP-AMPA receptor in the basolateral amygdala in a mouse model of depression
Accepted Manuscript Chronic stress-induced dendritic reorganization and abundance of synaptosomal PKA-dependent CP-AMPA receptor in the basolateral amygdala in a mouse model of depression Eun-Surk Yi, Seikwan Oh, Jang-kyu Lee, Yea-Hyun Leem PII:
S0006-291X(17)30559-4
DOI:
10.1016/j.bbrc.2017.03.093
Reference:
YBBRC 37480
To appear in:
Biochemical and Biophysical Research Communications
Received Date: 7 March 2017 Accepted Date: 19 March 2017
Please cite this article as: E.-S. Yi, S. Oh, J.-k. Lee, Y.-H. Leem, Chronic stress-induced dendritic reorganization and abundance of synaptosomal PKA-dependent CP-AMPA receptor in the basolateral amygdala in a mouse model of depression, Biochemical and Biophysical Research Communications (2017), doi: 10.1016/j.bbrc.2017.03.093. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RESEARCH HIGHLIGHTS
RI PT
Chronic stress causes behavioral depression, which is reversed by fluoxetine. Chronic stress causes the change of dendritic morphology of BLA neurons.
SC
Chronic stress increases synaptosomal PKA-dependent CP-AMPARs levels of BLA.
M AN U
Chronic stress-induced structural and molecular changes are reversed by fluoxetine.
AC C
EP
TE D
Blockage of CP-AMPARs in BLA mitigates stress-elicited depressive phenotype.
ACCEPTED MANUSCRIPT Chronic stress-induced dendritic reorganization and abundance of synaptosomal PKAdependent CP-AMPA receptor in the basolateral amygdala in a mouse model of
Eun-Surk Yia,*, Seikwan Ohb,*, Jang-kyu Leec, Yea-Hyun Leemb
a
RI PT
depression
SC
Department of Exercise Rehabilitation & Welfare, Gachon University, Incheon, Republic of
Korea
Department of Molecular Medicine and TIDRC, School of Medicine, Ewha Womans
University, Seoul, Republic of Korea c
M AN U
b
Department of Physical Education, Dankook University, Cheonan-si, Chungnam-do 330-
951, Republic of Korea *
TE D
These authors contributed equally to the work
Corresponding author: Yea-Hyun Leem, PhD
EP
Department of Molecular Medicine and TIDRC, School of Medicine, Ewha Womans
AC C
University, Seoul 158-710, Republic of Korea Tel: +82-2-2650-2570, Fax: +82-2-2650-5850, E-mail: [email protected]
ACCEPTED MANUSCRIPT Abstract Chronic stress is a precipitating factor for disorders including depression. The basolateral amygdala (BLA) is a critical substrate that interconnects with stress-modulated neural
RI PT
networks to generate emotion- and mood-related behaviors. The current study shows that 3 hours per day of restraint stress for 14 days caused mice to exhibit long-term depressive behaviors, manifested by disrupted sociality and despair levels, which were rescued by
SC
fluoxetine. These behavioral changes corresponded with morphological and molecular
changes in BLA neurons, including chronic stress-elicited increases in arborization, dendritic
M AN U
length, and spine density of BLA principal neurons. At the molecular level, calciumpermeable α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (CP-AMPARs) within the synaptosome exhibited an increased GluR1:GluR2 subunit ratio. We also observed increased GluR1 phosphorylation at Ser 845 and enhanced cyclic AMP-dependent protein
TE D
kinase (PKA) activity in the BLA. These molecular changes reverted to the basal state posttreatment with fluoxetine. The expression of synaptophysin (SYP) and postsynaptic density protein 95 (PSD-95) at BLA neuronal synapses was also enhanced by chronic stress, which
EP
was reversed post-treatment. Finally, chronic stress-provoked depressive behavior was overcome by local blockage of CP-AMPARs in the BLA via stereotaxic injection (IEM-1460).
AC C
Chronic stress-elicited depressive behavior may be due to hypertrophy of BLA neuronal dendrites and increased of PKA-dependent CP-AMPAR levels in BLA neurons. Furthermore, fluoxetine can reverse chronic stress-triggered cytoarchitectural and functional changes of BLA neurons. These findings provide insights into depression-linked structural and functional changes in BLA neurons.
Keywords: chronic restraint stress, depression, BLA, CP-AMPAR, PKA, dendritic
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
remodeling
ACCEPTED MANUSCRIPT 1. Introduction
Chronic stress has high worldwide prevalence and contributes to the development of many
RI PT
affective disorders such as depression. The amygdalae are a pair of critical nuclei within the subcortical limbic system regulating emotion- and mood-related behaviors. They have nerve connections with a diversity of other structures involved in stress responses, including the
SC
mesolimbic reward pathway, the prefrontal cortex (PFC) , and the hippocampus. [1-2].
Several structural and functional neuroimaging approaches have revealed activity-dependent
M AN U
modification of the amygdalae such as hyperactivity and hypertrophy in patients with psychiatric illness [3-4]. In rodent studies, repeated or chronic stress caused reorganization of the dendritic architecture of BLA neurons, as indicated by greater dendritic growth, ramification, and spine density of BLA principal neurons [5-6]. Repeated stress-precipitated
TE D
morphological remodeling of BLA neurons occurred concomitantly with enhanced tonic excitatory synaptic input to the BLA [7-8], indicating that the morphological characteristics of neurites are coupled with the synaptic potential of excitatory afferents to BLA neurons.
EP
Furthermore, chronic stress-triggered changes to the structure of BLA neurons persisted for a
AC C
long time, as did enhanced anxiety-like behavior [9]. A study reported that chronic restraint stress-provoked hypertrophy of BLA neurons was reversed by lithium treatment [5], implying that stress-related abnormal remodeling of BLA neurites can be medically reversed in chronic stress-related disorders.
Stress is known to increase membrane excitability across BLA neurons via enhanced extracellular glutamate levels [7-8, 10]. Mechanistically, Ca2+ influx by local glutamatergic activity regulates activity-dependent modifications that modulate neuronal excitability, synaptogenesis, and synaptic plasticity [11-13]. N-methyl-D-aspartate receptor (NMDAR)-
ACCEPTED MANUSCRIPT dependent long-term potentiation (LTP) are associated with the synaptic insertion and calcium permeability of AMPARs [14-16]. Ca2+ influx or conductance of AMPARs is determined by their subunit composition, e.g., the GluR1:GluR2 ratio. In addition,
RI PT
phosphorylation of the GluR1 by diverse kinases such as cyclic AMP-dependent protein kinase A (PKA) is required for the incorporation of AMPARs in synapses, which induces NMDAR-dependent LTP [13, 17-18]. Synaptic recruitment of CP-AMPARs is thought to
SC
play a crucial role in activity-dependent synaptic plasticity. However, it remains unknown whether structural plasticity-correlated CP-AMPARs of BLA neurons regulate depressive
M AN U
phenotypes.
Therefore, we determined whether chronic stress-precipitated dendritic reorganization and changes to AMPAR phenotypes of BLA neurons were reversed in response to fluoxetine treatment. Furthermore, we explored the role of CP-AMPARs of BLA neurons in depressive
EP
model of depression.
TE D
behaviors by blocking them using stereotaxic microinjection with IEM-1460 in a mouse
2. Materials and Methods
AC C
2.1. Experimental mice
Male 7 week-old C57BL/6 mice were obtained from Daehan Biolink, Inc. (Eumsung, Chungbuk, Korea) and housed in clear plastic cages under specific pathogen-free conditions with a 12:12-h light-dark cycle (lights on at 08:00 and off at 20:00). Mice had free access to standard irradiated chow (Purina Mills, Seoul, Korea). All animal experimental procedures used in this study were approved by the Institutional Animal Care and Use Committee at Ewha Womans University based on the National Institutes of Health Guide for the care and
ACCEPTED MANUSCRIPT use of Laboratory Animals.
RI PT
2.2. Experimental design The mice were divided into three groups (control: CON, stress: RST, and stress with fluoxetine treatment: RST+Flu; 20 mice per group). To induce restraint, 8 week-old mice
SC
were individually placed into a well-ventilated 50-mL conical tube to prevent forward or backward movement at set times from 10 a.m. to 13 p.m. for the 14 consecutive days. Control
M AN U
mice remained undisturbed in their home cages. For the group treated with fluoxetine, daily intraperitoneal injections of fluoxetine (20 mg/kg) were administered. Control and restrained mice were treated with saline instead of fluoxetine. For the AMPA receptor blockage experiment, mice were anesthetized with 250 mg/kg body weight of tribromoethanol. A 0.5
TE D
µl of 27 pg IEM-1460 (30 µM; Tocris Bioscience) dissolved in 0.9% saline or a saline-only were stereotaxically injected into bilateral BLA (AP, -1.40 mm; ML, ±3.1 mm; DV, -4.9 mm) 7 days after the last exposure to restraint stress. Behavior tests was performed 2 hours after
AC C
EP
the end of drug injection (FST implementation about 10 min after ST).
2.3. Behavioral assessment For the forced swim test (FST), mice were individually placed in an acrylic cylinder (100mm diameter × 250-mm height) containing water (24 ± 1°C) to a depth of 17 cm. All mice were exposed to a 15-min pre-test on day 1. The test was conducted 24 h later and the experiment recorded using a video camera. Mice were forced to swim for 6 min, and the immobile time was measured after the first min. The sociality test (ST) was slightly modified by Crawley’s sociability method [19]. In brief, the apparatus was partitioned into three equal
ACCEPTED MANUSCRIPT chambers (each 19 cm height × 19 cm length × 45 cm width) with dividing walls of clear Plexiglas that could be removed to allow free access to each chamber. The test mouse was first acclimatized by placing it in the closed-off center compartment for 5 min. An unfamiliar
RI PT
male (conspecific) mouse that had no prior contact with the test mouse was enclosed in a wire cup in either the a third of spot from the left or the right chamber for 1 min, and then the test mouse was allowed to explore for 5 min by removing both the dividing walls. Sociality was
SC
quantified using the time spent in interaction zone with the cup containing the novel conspecific (IZ) or non-interaction zone with the empty cup (Non-IZ). Social avoidance
M AN U
index (SAI) was calculated using the following equation: SAI (%) = (Non-IZ/IZ) x 100.
2.4. Golgi staining
TE D
Golgi-Cox staining of brain tissue was conducted using a NovaUltraTM Golgi-Cox stain Kit (IHC World, Woodstock, MD, USA) according to the procedure suggested by the manufacturer. Briefly, the fresh brain tissue was immediately immersed in a plastic jar filled
EP
with Golgi-Cox solution and stored in the dark at room temperature for 16 days. Brain tissue was washed two times with phosphate-buffered saline (PBS) for 2 days, and then 100-µm
AC C
thick sections prepared using a vibratome (Leica, Wetzlar, Germany) were washed with distilled water and stained with Post-Impregnation Solution for 10 min. Stained section was dehydrated through serial ethanol immersion. The slides were then cleared with xylene and coverslipped with Permount.
2.5. Neuronal reconstruction and morphometric analysis Golgi-stained neurons were assessed with Olympus BX51 microscope (Olympus, Tokyo,
ACCEPTED MANUSCRIPT Japan) equipped with a DP71 camera. We examined the principal neurons of the BLA (bregma: -1.07 ~ -2.03 mm). For each brain, 7–8 neurons that appeared to be completely filled were selected, and small somata with few dendrites and large somata with bipolar
RI PT
primary dendrite were excluded from this analysis. For neuronal reconstruction, each neuron was traced using the NeuronJ plugin for ImageJ software (National Institutes of Health, Image Engineering, Bethesda, MD, USA). The dendritic length, dendritic branching, and
SC
spine number were analyzed as the number of intersections at 20 µm interval distance points starting from the soma in Sholl analysis [20] using ImageJ software (NIH Image
M AN U
Engineering).
2.6. Synaptosome extraction and western blot analysis
TE D
To obtain synaptosome fraction, BLA from mice was isolated using tissue punches (1.25-mm diameter). Dissected tissues were homogenized in a homogenizing buffer (0.32 M sucrose; 20 mM HEPES, pH 7.4; 1 mM EDTA; 1X protease inhibitor cocktail; 5 mM NaF; 1 mM sodium
EP
vanadate). The homogenate was centrifuged at 3,000 rpm for 15 min. The pellet contains nuclei and large cell debris. The supernatant was centrifuged at 15,000 rpm for 15 min. After
AC C
centrifugation, the supernatant was removed and the pellet was resuspended and sonicated in protein lysis buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 2 mM EDTA, 1 mM NaVO3, 5 mM NaF and 1X protease inhibitor cocktail. Protein samples were electrophoretically separated on 10% polyacrylamide gels, transferred to nitrocellulose membranes (Amersham Bioscience, Buckinghamshire, UK), and incubated overnight with primary antibody in a blocking buffer at room temperature. The membranes were washed in the buffer and incubated with horseradish peroxidase-conjugated secondary antibody (anti-rabbit IgG 1:3,000) for 2 h at room temperature. The optical density of each
ACCEPTED MANUSCRIPT band was measured using Image J. Anti-PKA (1:2,000) and anti-phospho-PKA (1:1,000) were obtained from Cell Signaling Tech. Inc. (Danvers, MA, USA). Anti-GluR1 (1:1,000), and anti-phospho-GluR1 (Ser845, 1:500), and anti-β-actin (1:5,000) were purchased from
RI PT
Abcam (Cambridge, UK). Anti-GluR2 (1:1,000) and anti-PSD-95 (1:1,000), from Millipore (Billerica, MA, USA), Anti-Synaptophysin (1:1,000) from Santa Cruz Biotechnology (Santa
SC
Cruz, CA, USA).
M AN U
2.7. Statistical analysis
Significant differences between groups were determined using t-tests, one-way, and repeated two-way analysis of variance (SPSS for Windows, version 18.0, Chicago, IL, USA). Post-hoc comparisons were made using Student-Newman-Keuls tests. All values are reported as mean
EP
3. Results
TE D
± standard error of mean (SEM). Values of p < 0.05 were considered statistically significant.
3.1. Chronic stress induced long-lasting depressive behaviors
AC C
We found that 14 consecutive days of restraint stress (3 hours/day) produced behavioral depression that persisted for at least 4 weeks after the last exposure to the stress stimulus. In the ST, the time that control mice spent in the IZ was significantly longer than that in the Non-IZ at 7 (P7d: t7 = 3.04, p < 0.05), 14 (P14d: t7 = 3.36, p < 0.05), and 28 days (P28d: t7 = 4.04, p < 0.01) after the last exposure to the restraint (Fig. 1Ba). However, the time that restrained mice spent in the IZ was comparable to that in the Non-IZ at 7 (t7 = -0.55, p > 0.05), 14 (t14 = 0.06, p > 0.05), and 28 days (t7 = -0.22, p > 0.05) after the last exposure to the restraint (Fig. 1Ba). SAI corresponded well to sociality data (P7d: t14 = -2.46, p < 0.05; P14d:
ACCEPTED MANUSCRIPT t14 = -2.19, p < 0.05; P28d; t14 = -2.84, p < 0.05; Fig. 1Bb). In the FST, the immobile time of restrained mice was elevated as compared with that of control mice throughout the 4-week test period (P7d: t14 = -3.00, p < 0.01; P14d: t14 = -3.11, p < 0.01; P28d: t14 = -3.02, p < 0.01;
RI PT
Fig. 1Bd).
3.2. Chronic stress-induced behavioral depression was alleviated and structural
M AN U
was reversed by long-term post-treatment with fluoxetine
SC
reorganization of neuronal dendrites at proximal and/or intermediate segments in the BLA
Chronic stress-induced depressive behaviors were reversed 3 weeks after intraperitoneal treatment with fluoxetine (20 mg/kg). The decrease in the time that restrained mice spent in the IZ was reversed by fluoxetine (CON: t11 =3.16, p < 0.01; RST: t11 = -0.78, p > 0.05;
TE D
RST+Flu: t11 = 2.92, p < 0.05). The altered pattern of SAI was consistent with ST data (F2,33 = 5.09, p < 0.05). In the FST, the increased immobility of restrained mice was reduced by fluoxetine (F2,33 = 4.61, p < 0.05).
EP
We found that the dendritic structure of BLA neurons was altered by chronic stress and antidepressant treatment. The dendritic length of BLA neurons in restrained mice was
AC C
significantly longer than that in control mice (40–140 µm), which was reversed by fluoxetine (group × distance from soma, F32,336 = 5.89, p < 0.01; group, F1,21 = 25464.36, p < 0.01; distance from soma, F16,336 = 1367.56, p < 0.01; Fig. 2Da). The number of dendritic arborizations of BLA neurons was increased by stress (60–120 µm), but the number of such ramifications returned to baseline at the same segments following post-treatment with fluoxetine (group × distance from soma, F32,336 = 2.13, p < 0.01; group, F1,21 = 14175.56, p < 0.01; distance from soma, F16,336 = 901.38, p < 0.01; Fig. 2Db). Spine density was also
ACCEPTED MANUSCRIPT increased by stress, which in turn reduced after fluoxetine treatment (F2, 21 = 5.79, p < 0.01; Fig. 2Dc).
RI PT
3.3. Chronic stress-induced changes to synaptic protein and AMPA receptor subunit expression were reverted to basal levels by post-treatment with fluoxetine
SC
Chronic stress-provoked increases in GluR1 and p-GluR1 (Ser 831) protein expression in BLA neurons were reversed after fluoxetine treatment (GluR1: F2,12 = 122.16, p < 0.01; p-
M AN U
GluR1: F2,12 = 78.85, p < 0.01; Fig. 4Aa-c). Chronic stress reduced GluR2 protein expression in BLA neurons, which was reversed by fluoxetine (F2,12 = 52.12, p < 0.01; Fig. 4Aa-b). The GluR1:GluR2 ratio was increased by chronic stress, but this was reversed after fluoxetine treatment (F2,12 = 107.83, p < 0.01; Fig. 4Ad). Furthermore, the p-PKA/PKA expression ratio
TE D
in the BLA was significantly increased by chronic stress, and this was reversed by fluoxetine (F2,12 = 116.47, p < 0.01; Fig. 4Ba-b). The expression of synaptic proteins PSD-95 and SYP in the BLA was upregulated after chronic stress, which was conversely downregulated after
AC C
d).
EP
fluoxetine treatment (PSD-95: F2,12 = 69.19, p < 0.01; SYP: F2,12 = 16.76, p < 0.01; Fig. 4Bc-
3.4. Depressive behavior in chronically restrained mice was mitigated by the blockage of AMPA receptors in BLA neurons Stereotaxic microinjection of IEM-1460 into the BLA 7 days after the last exposure to restraint significantly ameliorated chronic stress-induced depressive behaviors. In the ST, the reduction in the time that restrained mice spent in the IZ was reversed by IEM (CON: t9 =2.65, p < 0.05; RST+Saline: t9 = -0.72, p > 0.05; RST+IEM: t9 = 2.41, p < 0.05; Fig. 4Ba).
ACCEPTED MANUSCRIPT SAI corresponded well to SA data (F2,27 = 4.05, p < 0.05; Fig. 4Bb). Moreover, the immobile time of restrained mice was significantly more than that of control mice, and this was
RI PT
reversed by IEM (F2,27 = 5.66, p < 0.01).
SC
4. Discussion
The current study showed that chronic stress-induced structural abnormality and synaptic
M AN U
proliferation of PKA-dependent CP-AMPARs in BLA neurons were reversed by the antidepressant fluoxetine, as were changes in the expression of synaptic proteins. These changes corresponded well to depressive behavior. Moreover, chronic stress-induced depressive behaviors were alleviated by CP-AMPAR blockage in the BLA. First, we
TE D
implemented the mouse model of depression in which, mice were subjected to 3 hours of restraint stress per day for 14 days. At 7, 14, and 28 days after the last exposure to stress, the levels of social avoidance and despair in restrained mice were significantly higher than that of
EP
control mice, indicating that depressive signs induced by the stress regimen used in this study lasted minimum 4 weeks. This depressive behavior was overcome post-treatment with
AC C
fluoxetine for 3 weeks, suggesting that this experimental paradigm is relevant for exploring depression- and antidepressive response-related cellular and molecular mechanisms in BLA neurons.
The BLA is a critical structure modulating mood- and emotion-related behaviors, including anxiety and depression. The BLA anatomically projects excitatory glutamatergic afferents to the hypothalamus and the nucleus accumbens, receives dopaminergic and norepinephrinergic input from the ventral tegmental area and the dorsal raphe and also interconnects with the
ACCEPTED MANUSCRIPT PFA and the hippocampus [1-2], implying that this region can serve as the node of a stressrelated neural circuit involved in these disorders. To determine whether fluoxetine affects chronic stress-induced structural plasticity of BLA
RI PT
neurons, morphological assessments were performed using Sholl analysis in an independent experiment with a brain specimen from a mouse not subjected to behavior tests to control for the effect of the behavior tests themselves on neuronal morphology. In our experimental
SC
paradigm, chronic stress resulted in longer dendritic length and greater ramification at
proximal or/and intermediate segments (>140 µm from the somata), together with a greater
M AN U
spine density on BLA neurons. Amygdala hypertrophy is commonly observed in patients with psychiatric disorders [3-4]. The increase in dendritic length, arborization, and spine number occurred in repeated or chronic stressed BLA neurons [5-6, 8, 21]. The findings discussed above are concurrent with and support our data that chronic stress caused morphological
TE D
changes such as dendritic growth of BLA neurons. This structural modification of BLA neurons was reversed long-term post-treatment with the antidepressant fluoxetine. A study showed that tianeptine, an antidepressant with proven clinical efficacy, inhibited repeated
EP
stress-induced dendritic growth in the BLA and behavioral disturbance [22]. Their findings support ours and suggest that the antidepressant fluoxetine may reduce chronic stress-induced
AC C
enlargement of BLA neurons.
Dendritic morphology controls dendritic computation and integration, which locally modulates the intrinsic excitability, the input signal transduction and strength, and the synaptic connectivity of neurons [23-24]. Amygdala hypertrophy can arise from increased amygdala activity, which is due to repeated stress-elicited changes such as increased intrinsic membrane excitability or excitatory synaptic drive [7-8, 22]. We found that chronic stress led to increased GluR1 and reduced GluR2 protein expression in the synaptosomal fraction of
ACCEPTED MANUSCRIPT BLA neurons, representing an increased GluR1:GluR2 ratio. NMDAR-mediated LTP is achieved by long-term AMPAR phenotypes, such as synaptic recruitment and calcium permeability, primarily at excitatory glutamatergic synapses [14-16]. AMPARs are hetero-
RI PT
oligomeric assemblies of GluR1–4; those containing GluR1 or lacking GluR2 have a higher conductance or Ca2+ permeability [25]. The chronic stress-induced expression patterns of AMPAR subunits in our study imply that chronic stress causes BLA changes with
SC
hyperactivity due to increased presence of AMPAR assemblies with higher Ca2+ permeability. Chronic stress-induced changes to the expression of GluR1 and GluR2 were reverted to the
M AN U
baseline levels post-treatment with fluoxetine. A study showed that the antidepressant tianeptine reduced action potentials through NMDAR-mediated potentiation without affecting AMPA currents in lateral amygdala neurons in naïve rodents [22]. The differences between our data and the electrophysiological data in that study may be attributable to the
TE D
treatment duration and the conditions, including acute vs. chronic and normal vs. pathological conditions. Hence, we can speculate that long-term treatment with fluoxetine may affect AMPAR compositions with higher calcium permeability in BLA neurons under potentially
EP
pathophysiological conditions without affecting them in BLA neurons under normal conditions. Phosphorylation of GluR1 at Ser 845 was also increased by chronic stress, and
AC C
this was reversed after fluoxetine treatment. The trafficking of AMPA receptors into the plasma membrane by PKA-directed GluR1 phosphorylation at Ser 845 determines channel open time and synaptic strength [26-27]. PKA-directed GluR1 phosphorylation also primes CaMKII phosphorylation for synaptic plasticity such as LTP and protein synthesis [17-18]. Based on the previous information, our data suggest that chronic stress-induced synaptic incorporation of CP-AMPARs may trigger long-lasting synaptic plasticity of BLA neurons, thereby increasing BLA hyperactivity. This change can be reversed by prolonged treatment with the antidepressant fluoxetine. In terms of synaptic protein expression on BLA neurons,
ACCEPTED MANUSCRIPT synaptosomal PSD-95 and synaptophysin, post-synaptic and presynaptic markers, respectively, were increased by chronic stress, and this was reversed by fluoxetine, providing evidence that chronic stress-induced synaptogenesis in the BLA can be reversed by an
RI PT
antidepressant, although other antidepressants have not been tested in the current study. Finally, chronic stress led to a higher density of CP-AMPARs within BLA neuronal synapses, which led us to further explore the role of BLA neuronal CP-AMPARs in depressive
SC
phenotypes. Notably, chronic stress-provoked behavioral depression was mitigated by local blockage of CP-AMPARs in the BLA, suggesting that enhanced CP-AMPAR distribution in
M AN U
BLA neurons is strongly implicated in depressive symptoms, and long-term treatment with fluoxetine can suppress CP-AMPAR function in the BLA. Together, chronic stress can trigger PKA-dependent synaptic incorporation of CP-AMPARs and dendritic hypertrophy of BLA neurons, thereby precipitating a depressive phenotype. These cellular, molecular, and
such as fluoxetine.
TE D
behavioral events in the BLA can be reversed by long-term treatment with an antidepressant
EP
Acknowledgments
AC C
Funding: This work was supported by the Korea Health Technology R&D Project [grant number HI12C0003]; and the National Research Foundation of Korea [grant numbers NRF2013R1A1A2062984, NRF-2016R1A6A3A11932098].
Conflict of interest The authors have no conflicts of interest.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT References
[1] C. Osório, T. Probert, E. Jones, et al., Adapting to Stress: Understanding the Neurobiology
RI PT
of Resilience, Behav Med. 21 (2016) 1-16. [2] S.J. Russo, J.W. Murrough, M.H. Han, et al., Neurobiology of resilience, Nat Neurosci. 15 (2012) 1475-84.
SC
[3] R.J. Davison, W. Irwin, M.J. Anderle, et al., The neural substrates of affective processing in depressed patients treated with venlafaxine, Am J Psychiatry 160 (2003) 64-75.
M AN U
[4] J. Lagopoulos, G.S. Malhi, A functional magnetic resonance imaging study of emotional stroop in euthymic bipolar disorder, Neuroreport 18 (2007) 1583-1587. [5] S.A. Johnson, J.F. Wang, X. Sun, et al., Lithium treatment prevents stress-induced
TE D
dendritic remodeling in the rodent amygdala, Neuroscience 163 (2009) 34-39. [6] M.A. Padival, S.R., Blume, J.E. Vantrease, et al., Qualitatively different effect of repeated stress during adolescence on principal neuron morphology across lateral and basal nuclei of
EP
the rat amygdala, Neuroscience 291 (2015) 128-145. [7] G.W. Hubert, C. Li, D.G., Rainnie, et al., Effects of stress on AMPA receptor distribution
AC C
and function in the basolateral amygdala, Brain Struct Funct. 219 (2014) 1169-1179. [8] M.A. Padival, D. Quinette, J.A. Rosenkranz, Effects of repeated stress on excitatory drive of basal amygdala neurons in vivo, Neuropsychopharmacology 38 (2013) 1748-1762. [9] A. Vyas, A.G. Pillai, A. Chattarji, Recovery after chronic stress fails to reverse amygdaloid neuronal hypertrophy and enhanced anxiety-like behavior, Neuroscience 128 (2004) 667-673. [10] L.R. Reznikov, C.A. Grillo, G.G. Piroli, et al., Acute stress-mediated increase in
ACCEPTED MANUSCRIPT extracellular glutamate levels in the rat amygdala: differential effects of antidepressant treatment, Eur J Neurosci. 25 (2007) 3109-3114. [11] M. Graupner, N. Brune, Calcium-based plasticity medel explains sensitivity of synaptic
RI PT
change to spike pattern, rate, and dendritic location, Pro Natl Acad Sci USA. 109 (2012) 3991-3996.
[12] J.J. Lippman-Bell, C, Zhou, H. Sun, et al., Early-life seizures alter synaptic calcium-
SC
permeable AMPA receptor function and plasticity, Mol. Cell. Neurosci. 76 (2016) 11-20.
M AN U
[13] P. Park, T.M. Sanderson, M. Amici, et al., Calcium-permeable AMPA receptors mediate the induction of the protein kinase A-dependent component of long-term potentiation in the hippocampus, J Neurosci. 36 (2016) 622-631.
[14] R.L. Hugenir, R.A. Nicoll, AMPARs and synaptic plasticity: the last 25 years, Neuron 80
TE D
(2013) 704-717.
[15] K. Plant, K.A. Pelkey, Z.A. Bortolotto, et al., Transient incorporation of native GluR2-
602-604.
EP
lacking AMPA receptors during hippocampal long-term potentiation, Nat Neurosci 9 (2006)
AC C
[16] H. Makino, R. Malinow, AMPA receptor incorporation into synapses during LTP: the role of lateral movement and excitosis, Neuron 64 (2009) 381-390. [17] J.A. Esteban, S.H. Shi, C.Wilson, et al., PKA phosphorylation of AMPA receptor subunits controls synaptic trafficking underlying plasticity, Nat Neurosci 6 (2003) 136-143. [18] Y. Yang, X.B. Wang, M. Frerking, et al., Delivery of AMPA receptors to perisynaptic sites precedes the full expression of long-term potentiation, Proc Natl Acad Sci USA. 105 (2008) 11388-11393.
ACCEPTED MANUSCRIPT [19] O. Kaidanovich-Beilin, T. Lipina, I. Vukobradovic, et al., Assessment of social interactions behaviors, J Vis Exp. 48 (2011) 2473. [20] D.A. Sholl, Dendritic organization in the neurons of the visual and motor cortices of the
RI PT
cat, J Anat. 87 (1953) 387-406. [21] J.A. Rosenkranz, E.R. Venheim, M. Padival, Chronic stress causes amygdala
SC
hyperexcitability in rodents, Biol Psychiatry 67 (2010) 1128-1136.
[22] A.G. Pillai, S. Anilkumar, S. Chattarji, The same antidepressant elicits contrasting
M AN U
patterns of synaptic changes in the amygdala vs hippocampus, Neuropschopharm 37 (2012) 2702-2711.
[23] M. Ferrante, M. Migliore, G.A. Ascoli, Functional impact of dendritic branch-point morphology. J Neurosci. 33 (2013) 2156-2165.
TE D
[24] M. London, M. Hausser, Dendritic computation, Annu Rev Neurosci. 28 (2005) 503-532. [25] V.A. Derkach,, M.C. Oh, E.S. Guire, et al., Regulatory mechanisms of AMPA receptors
EP
in synaptic plasticity, Nat Rev Neurosci 8 (2007) 101-113. [26] T.G. Banke, D. Bowie, H. Lee, et al., Control of GluR1 AMPA receptor function by
AC C
cAMP-dependent protein kinase. J Neurosci. 20 (2000) 89-102. [27] H.K. Lee, M. Barbarosie, K. Kameyama, et al., Regulation of distinct AMPA receptor phosphorylation sites during bidirectional synaptic plasticity, Nature 405 (2000) 955-959.
Figure Legends
ACCEPTED MANUSCRIPT Fig. 1. Experimental design and depressive-like behavior tests. (A) Experimental design. (B) Quantitative analysis of sociality and FST: (a) Time spent in IZ and Non-IZ. (b) Social avoidance index. (c) Traveling pattern in sociality test arena. (d) Immobile time in FST. * and
RI PT
** denote differences at p < 0.05 and p < 0.01, respectively.
Fig. 2. Chronic restraint stress and fluoxetine affected dendritic morphology in the BLA.
SC
Photomicrographs showing experimental design (A), and quantitative analysis of depressive-
M AN U
like behaviors (B): (a) Time spent in IZ and Non-IZ of sociality test. (b) Social avoidance index. (c) Immobile time of FST, and Glogi-stained neuron and reconstruction (C): (a) Region measured. (b) Neural reconstruction. (c) neuronal dendrites in the BLA region. Photomicrographs showing quantitative analysis of dendritic morphology of BLA neurons (D): (a) Dendritic length. (b) Intersection number. (c) Spine density. † and †† denote
TE D
difference between CON and RST at p < 0.05 and p < 0.01, respectively. # and ## denote difference between RST and RST+Flu at p < 0.05 and p < 0.01, respectively. * and ** denote
EP
differences at p < 0.05 and p < 0.01, respectively.
AC C
Fig. 3. Chronic restraint stress and fluoxetine affected the expression and phosphorylation of GluR1 and GluR2 as well as PKA activity and synaptic proteins in the BLA. Photomicrographs showing synaptosomal GluR1 and GluR2 expression and phosphorylation (A): (a) GluR1, GluR2, and p-GluR1 expression. (b) Quantitative analysis of expression of GluR1 and GluR2. (c) Quantitative analysis of expression of p-GluR1. (d) The ratio of GluR1/GluR2. Photomicrographs showing synaptosomal p-PKA, PKA, PDS-95, and synaptophysin expression (B): (a) p-PKA and PKA expression. (b) Quantitative analysis of
ACCEPTED MANUSCRIPT expression of p-PKA/PKA. (c) PSD-95 and synaptophysin expression. (d) Quantitative analysis of expression of PSD-95 and synaptophysin. * and ** denote differences at p < 0.05
RI PT
and p < 0.01, respectively.
Fig. 4. Chronic stress-induced behavioral depression was alleviated by stereotaxic injection with IEM-1460, CP-AMPAR blocker. Experimental design (A): (a) an overall experimental
SC
scheme. (b) a detail procedure for IEM treatment and behavior tests. Photomicrographs
M AN U
showing quantitative analysis of depressive-like behaviors (B): (a) Time spent in IZ and NonIZ of sociality test. (b) Social avoidance index. (c) Immobile time of FST. The data are presented as the mean ± SEM (n = 10 animals/group). * and ** denote differences at p < 0.05
AC C
EP
TE D
and p < 0.01, respectively.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0006-291X(17)30559-4
DOI:
10.1016/j.bbrc.2017.03.093
Reference:
YBBRC 37480
To appear in:
Biochemical and Biophysical Research Communications
Received Date: 7 March 2017 Accepted Date: 19 March 2017
Please cite this article as: E.-S. Yi, S. Oh, J.-k. Lee, Y.-H. Leem, Chronic stress-induced dendritic reorganization and abundance of synaptosomal PKA-dependent CP-AMPA receptor in the basolateral amygdala in a mouse model of depression, Biochemical and Biophysical Research Communications (2017), doi: 10.1016/j.bbrc.2017.03.093. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RESEARCH HIGHLIGHTS
RI PT
Chronic stress causes behavioral depression, which is reversed by fluoxetine. Chronic stress causes the change of dendritic morphology of BLA neurons.
SC
Chronic stress increases synaptosomal PKA-dependent CP-AMPARs levels of BLA.
M AN U
Chronic stress-induced structural and molecular changes are reversed by fluoxetine.
AC C
EP
TE D
Blockage of CP-AMPARs in BLA mitigates stress-elicited depressive phenotype.
ACCEPTED MANUSCRIPT Chronic stress-induced dendritic reorganization and abundance of synaptosomal PKAdependent CP-AMPA receptor in the basolateral amygdala in a mouse model of
Eun-Surk Yia,*, Seikwan Ohb,*, Jang-kyu Leec, Yea-Hyun Leemb
a
RI PT
depression
SC
Department of Exercise Rehabilitation & Welfare, Gachon University, Incheon, Republic of
Korea
Department of Molecular Medicine and TIDRC, School of Medicine, Ewha Womans
University, Seoul, Republic of Korea c
M AN U
b
Department of Physical Education, Dankook University, Cheonan-si, Chungnam-do 330-
951, Republic of Korea *
TE D
These authors contributed equally to the work
Corresponding author: Yea-Hyun Leem, PhD
EP
Department of Molecular Medicine and TIDRC, School of Medicine, Ewha Womans
AC C
University, Seoul 158-710, Republic of Korea Tel: +82-2-2650-2570, Fax: +82-2-2650-5850, E-mail: [email protected]
ACCEPTED MANUSCRIPT Abstract Chronic stress is a precipitating factor for disorders including depression. The basolateral amygdala (BLA) is a critical substrate that interconnects with stress-modulated neural
RI PT
networks to generate emotion- and mood-related behaviors. The current study shows that 3 hours per day of restraint stress for 14 days caused mice to exhibit long-term depressive behaviors, manifested by disrupted sociality and despair levels, which were rescued by
SC
fluoxetine. These behavioral changes corresponded with morphological and molecular
changes in BLA neurons, including chronic stress-elicited increases in arborization, dendritic
M AN U
length, and spine density of BLA principal neurons. At the molecular level, calciumpermeable α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (CP-AMPARs) within the synaptosome exhibited an increased GluR1:GluR2 subunit ratio. We also observed increased GluR1 phosphorylation at Ser 845 and enhanced cyclic AMP-dependent protein
TE D
kinase (PKA) activity in the BLA. These molecular changes reverted to the basal state posttreatment with fluoxetine. The expression of synaptophysin (SYP) and postsynaptic density protein 95 (PSD-95) at BLA neuronal synapses was also enhanced by chronic stress, which
EP
was reversed post-treatment. Finally, chronic stress-provoked depressive behavior was overcome by local blockage of CP-AMPARs in the BLA via stereotaxic injection (IEM-1460).
AC C
Chronic stress-elicited depressive behavior may be due to hypertrophy of BLA neuronal dendrites and increased of PKA-dependent CP-AMPAR levels in BLA neurons. Furthermore, fluoxetine can reverse chronic stress-triggered cytoarchitectural and functional changes of BLA neurons. These findings provide insights into depression-linked structural and functional changes in BLA neurons.
Keywords: chronic restraint stress, depression, BLA, CP-AMPAR, PKA, dendritic
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
remodeling
ACCEPTED MANUSCRIPT 1. Introduction
Chronic stress has high worldwide prevalence and contributes to the development of many
RI PT
affective disorders such as depression. The amygdalae are a pair of critical nuclei within the subcortical limbic system regulating emotion- and mood-related behaviors. They have nerve connections with a diversity of other structures involved in stress responses, including the
SC
mesolimbic reward pathway, the prefrontal cortex (PFC) , and the hippocampus. [1-2].
Several structural and functional neuroimaging approaches have revealed activity-dependent
M AN U
modification of the amygdalae such as hyperactivity and hypertrophy in patients with psychiatric illness [3-4]. In rodent studies, repeated or chronic stress caused reorganization of the dendritic architecture of BLA neurons, as indicated by greater dendritic growth, ramification, and spine density of BLA principal neurons [5-6]. Repeated stress-precipitated
TE D
morphological remodeling of BLA neurons occurred concomitantly with enhanced tonic excitatory synaptic input to the BLA [7-8], indicating that the morphological characteristics of neurites are coupled with the synaptic potential of excitatory afferents to BLA neurons.
EP
Furthermore, chronic stress-triggered changes to the structure of BLA neurons persisted for a
AC C
long time, as did enhanced anxiety-like behavior [9]. A study reported that chronic restraint stress-provoked hypertrophy of BLA neurons was reversed by lithium treatment [5], implying that stress-related abnormal remodeling of BLA neurites can be medically reversed in chronic stress-related disorders.
Stress is known to increase membrane excitability across BLA neurons via enhanced extracellular glutamate levels [7-8, 10]. Mechanistically, Ca2+ influx by local glutamatergic activity regulates activity-dependent modifications that modulate neuronal excitability, synaptogenesis, and synaptic plasticity [11-13]. N-methyl-D-aspartate receptor (NMDAR)-
ACCEPTED MANUSCRIPT dependent long-term potentiation (LTP) are associated with the synaptic insertion and calcium permeability of AMPARs [14-16]. Ca2+ influx or conductance of AMPARs is determined by their subunit composition, e.g., the GluR1:GluR2 ratio. In addition,
RI PT
phosphorylation of the GluR1 by diverse kinases such as cyclic AMP-dependent protein kinase A (PKA) is required for the incorporation of AMPARs in synapses, which induces NMDAR-dependent LTP [13, 17-18]. Synaptic recruitment of CP-AMPARs is thought to
SC
play a crucial role in activity-dependent synaptic plasticity. However, it remains unknown whether structural plasticity-correlated CP-AMPARs of BLA neurons regulate depressive
M AN U
phenotypes.
Therefore, we determined whether chronic stress-precipitated dendritic reorganization and changes to AMPAR phenotypes of BLA neurons were reversed in response to fluoxetine treatment. Furthermore, we explored the role of CP-AMPARs of BLA neurons in depressive
EP
model of depression.
TE D
behaviors by blocking them using stereotaxic microinjection with IEM-1460 in a mouse
2. Materials and Methods
AC C
2.1. Experimental mice
Male 7 week-old C57BL/6 mice were obtained from Daehan Biolink, Inc. (Eumsung, Chungbuk, Korea) and housed in clear plastic cages under specific pathogen-free conditions with a 12:12-h light-dark cycle (lights on at 08:00 and off at 20:00). Mice had free access to standard irradiated chow (Purina Mills, Seoul, Korea). All animal experimental procedures used in this study were approved by the Institutional Animal Care and Use Committee at Ewha Womans University based on the National Institutes of Health Guide for the care and
ACCEPTED MANUSCRIPT use of Laboratory Animals.
RI PT
2.2. Experimental design The mice were divided into three groups (control: CON, stress: RST, and stress with fluoxetine treatment: RST+Flu; 20 mice per group). To induce restraint, 8 week-old mice
SC
were individually placed into a well-ventilated 50-mL conical tube to prevent forward or backward movement at set times from 10 a.m. to 13 p.m. for the 14 consecutive days. Control
M AN U
mice remained undisturbed in their home cages. For the group treated with fluoxetine, daily intraperitoneal injections of fluoxetine (20 mg/kg) were administered. Control and restrained mice were treated with saline instead of fluoxetine. For the AMPA receptor blockage experiment, mice were anesthetized with 250 mg/kg body weight of tribromoethanol. A 0.5
TE D
µl of 27 pg IEM-1460 (30 µM; Tocris Bioscience) dissolved in 0.9% saline or a saline-only were stereotaxically injected into bilateral BLA (AP, -1.40 mm; ML, ±3.1 mm; DV, -4.9 mm) 7 days after the last exposure to restraint stress. Behavior tests was performed 2 hours after
AC C
EP
the end of drug injection (FST implementation about 10 min after ST).
2.3. Behavioral assessment For the forced swim test (FST), mice were individually placed in an acrylic cylinder (100mm diameter × 250-mm height) containing water (24 ± 1°C) to a depth of 17 cm. All mice were exposed to a 15-min pre-test on day 1. The test was conducted 24 h later and the experiment recorded using a video camera. Mice were forced to swim for 6 min, and the immobile time was measured after the first min. The sociality test (ST) was slightly modified by Crawley’s sociability method [19]. In brief, the apparatus was partitioned into three equal
ACCEPTED MANUSCRIPT chambers (each 19 cm height × 19 cm length × 45 cm width) with dividing walls of clear Plexiglas that could be removed to allow free access to each chamber. The test mouse was first acclimatized by placing it in the closed-off center compartment for 5 min. An unfamiliar
RI PT
male (conspecific) mouse that had no prior contact with the test mouse was enclosed in a wire cup in either the a third of spot from the left or the right chamber for 1 min, and then the test mouse was allowed to explore for 5 min by removing both the dividing walls. Sociality was
SC
quantified using the time spent in interaction zone with the cup containing the novel conspecific (IZ) or non-interaction zone with the empty cup (Non-IZ). Social avoidance
M AN U
index (SAI) was calculated using the following equation: SAI (%) = (Non-IZ/IZ) x 100.
2.4. Golgi staining
TE D
Golgi-Cox staining of brain tissue was conducted using a NovaUltraTM Golgi-Cox stain Kit (IHC World, Woodstock, MD, USA) according to the procedure suggested by the manufacturer. Briefly, the fresh brain tissue was immediately immersed in a plastic jar filled
EP
with Golgi-Cox solution and stored in the dark at room temperature for 16 days. Brain tissue was washed two times with phosphate-buffered saline (PBS) for 2 days, and then 100-µm
AC C
thick sections prepared using a vibratome (Leica, Wetzlar, Germany) were washed with distilled water and stained with Post-Impregnation Solution for 10 min. Stained section was dehydrated through serial ethanol immersion. The slides were then cleared with xylene and coverslipped with Permount.
2.5. Neuronal reconstruction and morphometric analysis Golgi-stained neurons were assessed with Olympus BX51 microscope (Olympus, Tokyo,
ACCEPTED MANUSCRIPT Japan) equipped with a DP71 camera. We examined the principal neurons of the BLA (bregma: -1.07 ~ -2.03 mm). For each brain, 7–8 neurons that appeared to be completely filled were selected, and small somata with few dendrites and large somata with bipolar
RI PT
primary dendrite were excluded from this analysis. For neuronal reconstruction, each neuron was traced using the NeuronJ plugin for ImageJ software (National Institutes of Health, Image Engineering, Bethesda, MD, USA). The dendritic length, dendritic branching, and
SC
spine number were analyzed as the number of intersections at 20 µm interval distance points starting from the soma in Sholl analysis [20] using ImageJ software (NIH Image
M AN U
Engineering).
2.6. Synaptosome extraction and western blot analysis
TE D
To obtain synaptosome fraction, BLA from mice was isolated using tissue punches (1.25-mm diameter). Dissected tissues were homogenized in a homogenizing buffer (0.32 M sucrose; 20 mM HEPES, pH 7.4; 1 mM EDTA; 1X protease inhibitor cocktail; 5 mM NaF; 1 mM sodium
EP
vanadate). The homogenate was centrifuged at 3,000 rpm for 15 min. The pellet contains nuclei and large cell debris. The supernatant was centrifuged at 15,000 rpm for 15 min. After
AC C
centrifugation, the supernatant was removed and the pellet was resuspended and sonicated in protein lysis buffer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 2 mM EDTA, 1 mM NaVO3, 5 mM NaF and 1X protease inhibitor cocktail. Protein samples were electrophoretically separated on 10% polyacrylamide gels, transferred to nitrocellulose membranes (Amersham Bioscience, Buckinghamshire, UK), and incubated overnight with primary antibody in a blocking buffer at room temperature. The membranes were washed in the buffer and incubated with horseradish peroxidase-conjugated secondary antibody (anti-rabbit IgG 1:3,000) for 2 h at room temperature. The optical density of each
ACCEPTED MANUSCRIPT band was measured using Image J. Anti-PKA (1:2,000) and anti-phospho-PKA (1:1,000) were obtained from Cell Signaling Tech. Inc. (Danvers, MA, USA). Anti-GluR1 (1:1,000), and anti-phospho-GluR1 (Ser845, 1:500), and anti-β-actin (1:5,000) were purchased from
RI PT
Abcam (Cambridge, UK). Anti-GluR2 (1:1,000) and anti-PSD-95 (1:1,000), from Millipore (Billerica, MA, USA), Anti-Synaptophysin (1:1,000) from Santa Cruz Biotechnology (Santa
SC
Cruz, CA, USA).
M AN U
2.7. Statistical analysis
Significant differences between groups were determined using t-tests, one-way, and repeated two-way analysis of variance (SPSS for Windows, version 18.0, Chicago, IL, USA). Post-hoc comparisons were made using Student-Newman-Keuls tests. All values are reported as mean
EP
3. Results
TE D
± standard error of mean (SEM). Values of p < 0.05 were considered statistically significant.
3.1. Chronic stress induced long-lasting depressive behaviors
AC C
We found that 14 consecutive days of restraint stress (3 hours/day) produced behavioral depression that persisted for at least 4 weeks after the last exposure to the stress stimulus. In the ST, the time that control mice spent in the IZ was significantly longer than that in the Non-IZ at 7 (P7d: t7 = 3.04, p < 0.05), 14 (P14d: t7 = 3.36, p < 0.05), and 28 days (P28d: t7 = 4.04, p < 0.01) after the last exposure to the restraint (Fig. 1Ba). However, the time that restrained mice spent in the IZ was comparable to that in the Non-IZ at 7 (t7 = -0.55, p > 0.05), 14 (t14 = 0.06, p > 0.05), and 28 days (t7 = -0.22, p > 0.05) after the last exposure to the restraint (Fig. 1Ba). SAI corresponded well to sociality data (P7d: t14 = -2.46, p < 0.05; P14d:
ACCEPTED MANUSCRIPT t14 = -2.19, p < 0.05; P28d; t14 = -2.84, p < 0.05; Fig. 1Bb). In the FST, the immobile time of restrained mice was elevated as compared with that of control mice throughout the 4-week test period (P7d: t14 = -3.00, p < 0.01; P14d: t14 = -3.11, p < 0.01; P28d: t14 = -3.02, p < 0.01;
RI PT
Fig. 1Bd).
3.2. Chronic stress-induced behavioral depression was alleviated and structural
M AN U
was reversed by long-term post-treatment with fluoxetine
SC
reorganization of neuronal dendrites at proximal and/or intermediate segments in the BLA
Chronic stress-induced depressive behaviors were reversed 3 weeks after intraperitoneal treatment with fluoxetine (20 mg/kg). The decrease in the time that restrained mice spent in the IZ was reversed by fluoxetine (CON: t11 =3.16, p < 0.01; RST: t11 = -0.78, p > 0.05;
TE D
RST+Flu: t11 = 2.92, p < 0.05). The altered pattern of SAI was consistent with ST data (F2,33 = 5.09, p < 0.05). In the FST, the increased immobility of restrained mice was reduced by fluoxetine (F2,33 = 4.61, p < 0.05).
EP
We found that the dendritic structure of BLA neurons was altered by chronic stress and antidepressant treatment. The dendritic length of BLA neurons in restrained mice was
AC C
significantly longer than that in control mice (40–140 µm), which was reversed by fluoxetine (group × distance from soma, F32,336 = 5.89, p < 0.01; group, F1,21 = 25464.36, p < 0.01; distance from soma, F16,336 = 1367.56, p < 0.01; Fig. 2Da). The number of dendritic arborizations of BLA neurons was increased by stress (60–120 µm), but the number of such ramifications returned to baseline at the same segments following post-treatment with fluoxetine (group × distance from soma, F32,336 = 2.13, p < 0.01; group, F1,21 = 14175.56, p < 0.01; distance from soma, F16,336 = 901.38, p < 0.01; Fig. 2Db). Spine density was also
ACCEPTED MANUSCRIPT increased by stress, which in turn reduced after fluoxetine treatment (F2, 21 = 5.79, p < 0.01; Fig. 2Dc).
RI PT
3.3. Chronic stress-induced changes to synaptic protein and AMPA receptor subunit expression were reverted to basal levels by post-treatment with fluoxetine
SC
Chronic stress-provoked increases in GluR1 and p-GluR1 (Ser 831) protein expression in BLA neurons were reversed after fluoxetine treatment (GluR1: F2,12 = 122.16, p < 0.01; p-
M AN U
GluR1: F2,12 = 78.85, p < 0.01; Fig. 4Aa-c). Chronic stress reduced GluR2 protein expression in BLA neurons, which was reversed by fluoxetine (F2,12 = 52.12, p < 0.01; Fig. 4Aa-b). The GluR1:GluR2 ratio was increased by chronic stress, but this was reversed after fluoxetine treatment (F2,12 = 107.83, p < 0.01; Fig. 4Ad). Furthermore, the p-PKA/PKA expression ratio
TE D
in the BLA was significantly increased by chronic stress, and this was reversed by fluoxetine (F2,12 = 116.47, p < 0.01; Fig. 4Ba-b). The expression of synaptic proteins PSD-95 and SYP in the BLA was upregulated after chronic stress, which was conversely downregulated after
AC C
d).
EP
fluoxetine treatment (PSD-95: F2,12 = 69.19, p < 0.01; SYP: F2,12 = 16.76, p < 0.01; Fig. 4Bc-
3.4. Depressive behavior in chronically restrained mice was mitigated by the blockage of AMPA receptors in BLA neurons Stereotaxic microinjection of IEM-1460 into the BLA 7 days after the last exposure to restraint significantly ameliorated chronic stress-induced depressive behaviors. In the ST, the reduction in the time that restrained mice spent in the IZ was reversed by IEM (CON: t9 =2.65, p < 0.05; RST+Saline: t9 = -0.72, p > 0.05; RST+IEM: t9 = 2.41, p < 0.05; Fig. 4Ba).
ACCEPTED MANUSCRIPT SAI corresponded well to SA data (F2,27 = 4.05, p < 0.05; Fig. 4Bb). Moreover, the immobile time of restrained mice was significantly more than that of control mice, and this was
RI PT
reversed by IEM (F2,27 = 5.66, p < 0.01).
SC
4. Discussion
The current study showed that chronic stress-induced structural abnormality and synaptic
M AN U
proliferation of PKA-dependent CP-AMPARs in BLA neurons were reversed by the antidepressant fluoxetine, as were changes in the expression of synaptic proteins. These changes corresponded well to depressive behavior. Moreover, chronic stress-induced depressive behaviors were alleviated by CP-AMPAR blockage in the BLA. First, we
TE D
implemented the mouse model of depression in which, mice were subjected to 3 hours of restraint stress per day for 14 days. At 7, 14, and 28 days after the last exposure to stress, the levels of social avoidance and despair in restrained mice were significantly higher than that of
EP
control mice, indicating that depressive signs induced by the stress regimen used in this study lasted minimum 4 weeks. This depressive behavior was overcome post-treatment with
AC C
fluoxetine for 3 weeks, suggesting that this experimental paradigm is relevant for exploring depression- and antidepressive response-related cellular and molecular mechanisms in BLA neurons.
The BLA is a critical structure modulating mood- and emotion-related behaviors, including anxiety and depression. The BLA anatomically projects excitatory glutamatergic afferents to the hypothalamus and the nucleus accumbens, receives dopaminergic and norepinephrinergic input from the ventral tegmental area and the dorsal raphe and also interconnects with the
ACCEPTED MANUSCRIPT PFA and the hippocampus [1-2], implying that this region can serve as the node of a stressrelated neural circuit involved in these disorders. To determine whether fluoxetine affects chronic stress-induced structural plasticity of BLA
RI PT
neurons, morphological assessments were performed using Sholl analysis in an independent experiment with a brain specimen from a mouse not subjected to behavior tests to control for the effect of the behavior tests themselves on neuronal morphology. In our experimental
SC
paradigm, chronic stress resulted in longer dendritic length and greater ramification at
proximal or/and intermediate segments (>140 µm from the somata), together with a greater
M AN U
spine density on BLA neurons. Amygdala hypertrophy is commonly observed in patients with psychiatric disorders [3-4]. The increase in dendritic length, arborization, and spine number occurred in repeated or chronic stressed BLA neurons [5-6, 8, 21]. The findings discussed above are concurrent with and support our data that chronic stress caused morphological
TE D
changes such as dendritic growth of BLA neurons. This structural modification of BLA neurons was reversed long-term post-treatment with the antidepressant fluoxetine. A study showed that tianeptine, an antidepressant with proven clinical efficacy, inhibited repeated
EP
stress-induced dendritic growth in the BLA and behavioral disturbance [22]. Their findings support ours and suggest that the antidepressant fluoxetine may reduce chronic stress-induced
AC C
enlargement of BLA neurons.
Dendritic morphology controls dendritic computation and integration, which locally modulates the intrinsic excitability, the input signal transduction and strength, and the synaptic connectivity of neurons [23-24]. Amygdala hypertrophy can arise from increased amygdala activity, which is due to repeated stress-elicited changes such as increased intrinsic membrane excitability or excitatory synaptic drive [7-8, 22]. We found that chronic stress led to increased GluR1 and reduced GluR2 protein expression in the synaptosomal fraction of
ACCEPTED MANUSCRIPT BLA neurons, representing an increased GluR1:GluR2 ratio. NMDAR-mediated LTP is achieved by long-term AMPAR phenotypes, such as synaptic recruitment and calcium permeability, primarily at excitatory glutamatergic synapses [14-16]. AMPARs are hetero-
RI PT
oligomeric assemblies of GluR1–4; those containing GluR1 or lacking GluR2 have a higher conductance or Ca2+ permeability [25]. The chronic stress-induced expression patterns of AMPAR subunits in our study imply that chronic stress causes BLA changes with
SC
hyperactivity due to increased presence of AMPAR assemblies with higher Ca2+ permeability. Chronic stress-induced changes to the expression of GluR1 and GluR2 were reverted to the
M AN U
baseline levels post-treatment with fluoxetine. A study showed that the antidepressant tianeptine reduced action potentials through NMDAR-mediated potentiation without affecting AMPA currents in lateral amygdala neurons in naïve rodents [22]. The differences between our data and the electrophysiological data in that study may be attributable to the
TE D
treatment duration and the conditions, including acute vs. chronic and normal vs. pathological conditions. Hence, we can speculate that long-term treatment with fluoxetine may affect AMPAR compositions with higher calcium permeability in BLA neurons under potentially
EP
pathophysiological conditions without affecting them in BLA neurons under normal conditions. Phosphorylation of GluR1 at Ser 845 was also increased by chronic stress, and
AC C
this was reversed after fluoxetine treatment. The trafficking of AMPA receptors into the plasma membrane by PKA-directed GluR1 phosphorylation at Ser 845 determines channel open time and synaptic strength [26-27]. PKA-directed GluR1 phosphorylation also primes CaMKII phosphorylation for synaptic plasticity such as LTP and protein synthesis [17-18]. Based on the previous information, our data suggest that chronic stress-induced synaptic incorporation of CP-AMPARs may trigger long-lasting synaptic plasticity of BLA neurons, thereby increasing BLA hyperactivity. This change can be reversed by prolonged treatment with the antidepressant fluoxetine. In terms of synaptic protein expression on BLA neurons,
ACCEPTED MANUSCRIPT synaptosomal PSD-95 and synaptophysin, post-synaptic and presynaptic markers, respectively, were increased by chronic stress, and this was reversed by fluoxetine, providing evidence that chronic stress-induced synaptogenesis in the BLA can be reversed by an
RI PT
antidepressant, although other antidepressants have not been tested in the current study. Finally, chronic stress led to a higher density of CP-AMPARs within BLA neuronal synapses, which led us to further explore the role of BLA neuronal CP-AMPARs in depressive
SC
phenotypes. Notably, chronic stress-provoked behavioral depression was mitigated by local blockage of CP-AMPARs in the BLA, suggesting that enhanced CP-AMPAR distribution in
M AN U
BLA neurons is strongly implicated in depressive symptoms, and long-term treatment with fluoxetine can suppress CP-AMPAR function in the BLA. Together, chronic stress can trigger PKA-dependent synaptic incorporation of CP-AMPARs and dendritic hypertrophy of BLA neurons, thereby precipitating a depressive phenotype. These cellular, molecular, and
such as fluoxetine.
TE D
behavioral events in the BLA can be reversed by long-term treatment with an antidepressant
EP
Acknowledgments
AC C
Funding: This work was supported by the Korea Health Technology R&D Project [grant number HI12C0003]; and the National Research Foundation of Korea [grant numbers NRF2013R1A1A2062984, NRF-2016R1A6A3A11932098].
Conflict of interest The authors have no conflicts of interest.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT References
[1] C. Osório, T. Probert, E. Jones, et al., Adapting to Stress: Understanding the Neurobiology
RI PT
of Resilience, Behav Med. 21 (2016) 1-16. [2] S.J. Russo, J.W. Murrough, M.H. Han, et al., Neurobiology of resilience, Nat Neurosci. 15 (2012) 1475-84.
SC
[3] R.J. Davison, W. Irwin, M.J. Anderle, et al., The neural substrates of affective processing in depressed patients treated with venlafaxine, Am J Psychiatry 160 (2003) 64-75.
M AN U
[4] J. Lagopoulos, G.S. Malhi, A functional magnetic resonance imaging study of emotional stroop in euthymic bipolar disorder, Neuroreport 18 (2007) 1583-1587. [5] S.A. Johnson, J.F. Wang, X. Sun, et al., Lithium treatment prevents stress-induced
TE D
dendritic remodeling in the rodent amygdala, Neuroscience 163 (2009) 34-39. [6] M.A. Padival, S.R., Blume, J.E. Vantrease, et al., Qualitatively different effect of repeated stress during adolescence on principal neuron morphology across lateral and basal nuclei of
EP
the rat amygdala, Neuroscience 291 (2015) 128-145. [7] G.W. Hubert, C. Li, D.G., Rainnie, et al., Effects of stress on AMPA receptor distribution
AC C
and function in the basolateral amygdala, Brain Struct Funct. 219 (2014) 1169-1179. [8] M.A. Padival, D. Quinette, J.A. Rosenkranz, Effects of repeated stress on excitatory drive of basal amygdala neurons in vivo, Neuropsychopharmacology 38 (2013) 1748-1762. [9] A. Vyas, A.G. Pillai, A. Chattarji, Recovery after chronic stress fails to reverse amygdaloid neuronal hypertrophy and enhanced anxiety-like behavior, Neuroscience 128 (2004) 667-673. [10] L.R. Reznikov, C.A. Grillo, G.G. Piroli, et al., Acute stress-mediated increase in
ACCEPTED MANUSCRIPT extracellular glutamate levels in the rat amygdala: differential effects of antidepressant treatment, Eur J Neurosci. 25 (2007) 3109-3114. [11] M. Graupner, N. Brune, Calcium-based plasticity medel explains sensitivity of synaptic
RI PT
change to spike pattern, rate, and dendritic location, Pro Natl Acad Sci USA. 109 (2012) 3991-3996.
[12] J.J. Lippman-Bell, C, Zhou, H. Sun, et al., Early-life seizures alter synaptic calcium-
SC
permeable AMPA receptor function and plasticity, Mol. Cell. Neurosci. 76 (2016) 11-20.
M AN U
[13] P. Park, T.M. Sanderson, M. Amici, et al., Calcium-permeable AMPA receptors mediate the induction of the protein kinase A-dependent component of long-term potentiation in the hippocampus, J Neurosci. 36 (2016) 622-631.
[14] R.L. Hugenir, R.A. Nicoll, AMPARs and synaptic plasticity: the last 25 years, Neuron 80
TE D
(2013) 704-717.
[15] K. Plant, K.A. Pelkey, Z.A. Bortolotto, et al., Transient incorporation of native GluR2-
602-604.
EP
lacking AMPA receptors during hippocampal long-term potentiation, Nat Neurosci 9 (2006)
AC C
[16] H. Makino, R. Malinow, AMPA receptor incorporation into synapses during LTP: the role of lateral movement and excitosis, Neuron 64 (2009) 381-390. [17] J.A. Esteban, S.H. Shi, C.Wilson, et al., PKA phosphorylation of AMPA receptor subunits controls synaptic trafficking underlying plasticity, Nat Neurosci 6 (2003) 136-143. [18] Y. Yang, X.B. Wang, M. Frerking, et al., Delivery of AMPA receptors to perisynaptic sites precedes the full expression of long-term potentiation, Proc Natl Acad Sci USA. 105 (2008) 11388-11393.
ACCEPTED MANUSCRIPT [19] O. Kaidanovich-Beilin, T. Lipina, I. Vukobradovic, et al., Assessment of social interactions behaviors, J Vis Exp. 48 (2011) 2473. [20] D.A. Sholl, Dendritic organization in the neurons of the visual and motor cortices of the
RI PT
cat, J Anat. 87 (1953) 387-406. [21] J.A. Rosenkranz, E.R. Venheim, M. Padival, Chronic stress causes amygdala
SC
hyperexcitability in rodents, Biol Psychiatry 67 (2010) 1128-1136.
[22] A.G. Pillai, S. Anilkumar, S. Chattarji, The same antidepressant elicits contrasting
M AN U
patterns of synaptic changes in the amygdala vs hippocampus, Neuropschopharm 37 (2012) 2702-2711.
[23] M. Ferrante, M. Migliore, G.A. Ascoli, Functional impact of dendritic branch-point morphology. J Neurosci. 33 (2013) 2156-2165.
TE D
[24] M. London, M. Hausser, Dendritic computation, Annu Rev Neurosci. 28 (2005) 503-532. [25] V.A. Derkach,, M.C. Oh, E.S. Guire, et al., Regulatory mechanisms of AMPA receptors
EP
in synaptic plasticity, Nat Rev Neurosci 8 (2007) 101-113. [26] T.G. Banke, D. Bowie, H. Lee, et al., Control of GluR1 AMPA receptor function by
AC C
cAMP-dependent protein kinase. J Neurosci. 20 (2000) 89-102. [27] H.K. Lee, M. Barbarosie, K. Kameyama, et al., Regulation of distinct AMPA receptor phosphorylation sites during bidirectional synaptic plasticity, Nature 405 (2000) 955-959.
Figure Legends
ACCEPTED MANUSCRIPT Fig. 1. Experimental design and depressive-like behavior tests. (A) Experimental design. (B) Quantitative analysis of sociality and FST: (a) Time spent in IZ and Non-IZ. (b) Social avoidance index. (c) Traveling pattern in sociality test arena. (d) Immobile time in FST. * and
RI PT
** denote differences at p < 0.05 and p < 0.01, respectively.
Fig. 2. Chronic restraint stress and fluoxetine affected dendritic morphology in the BLA.
SC
Photomicrographs showing experimental design (A), and quantitative analysis of depressive-
M AN U
like behaviors (B): (a) Time spent in IZ and Non-IZ of sociality test. (b) Social avoidance index. (c) Immobile time of FST, and Glogi-stained neuron and reconstruction (C): (a) Region measured. (b) Neural reconstruction. (c) neuronal dendrites in the BLA region. Photomicrographs showing quantitative analysis of dendritic morphology of BLA neurons (D): (a) Dendritic length. (b) Intersection number. (c) Spine density. † and †† denote
TE D
difference between CON and RST at p < 0.05 and p < 0.01, respectively. # and ## denote difference between RST and RST+Flu at p < 0.05 and p < 0.01, respectively. * and ** denote
EP
differences at p < 0.05 and p < 0.01, respectively.
AC C
Fig. 3. Chronic restraint stress and fluoxetine affected the expression and phosphorylation of GluR1 and GluR2 as well as PKA activity and synaptic proteins in the BLA. Photomicrographs showing synaptosomal GluR1 and GluR2 expression and phosphorylation (A): (a) GluR1, GluR2, and p-GluR1 expression. (b) Quantitative analysis of expression of GluR1 and GluR2. (c) Quantitative analysis of expression of p-GluR1. (d) The ratio of GluR1/GluR2. Photomicrographs showing synaptosomal p-PKA, PKA, PDS-95, and synaptophysin expression (B): (a) p-PKA and PKA expression. (b) Quantitative analysis of
ACCEPTED MANUSCRIPT expression of p-PKA/PKA. (c) PSD-95 and synaptophysin expression. (d) Quantitative analysis of expression of PSD-95 and synaptophysin. * and ** denote differences at p < 0.05
RI PT
and p < 0.01, respectively.
Fig. 4. Chronic stress-induced behavioral depression was alleviated by stereotaxic injection with IEM-1460, CP-AMPAR blocker. Experimental design (A): (a) an overall experimental
SC
scheme. (b) a detail procedure for IEM treatment and behavior tests. Photomicrographs
M AN U
showing quantitative analysis of depressive-like behaviors (B): (a) Time spent in IZ and NonIZ of sociality test. (b) Social avoidance index. (c) Immobile time of FST. The data are presented as the mean ± SEM (n = 10 animals/group). * and ** denote differences at p < 0.05
AC C
EP
TE D
and p < 0.01, respectively.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT