Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Feb 1;85(3):189-201.
doi: 10.1016/j.biopsych.2018.06.019. Epub 2018 Jul 5.

Chronic Stress Remodels Synapses in an Amygdala Circuit-Specific Manner

Affiliations

Chronic Stress Remodels Synapses in an Amygdala Circuit-Specific Manner

Jun-Yu Zhang et al. Biol Psychiatry. .

Abstract

Background: Chronic stress exposure increases the risk of developing various neuropsychiatric illnesses. The behavioral sequelae of stress correlate with dendritic hypertrophy and glutamate-related synaptic remodeling at basolateral amygdala projection neurons (BLA PNs). Yet, though BLA PNs are functionally heterogeneous with diverse corticolimbic targets, it remains unclear whether stress differentially impacts specific output circuits.

Methods: Confocal imaging was used to reconstruct the morphology of mouse BLA PNs with the aid of retrograde tracing and biocytin staining. The synaptic activity in these neurons was measured with in vitro electrophysiology, and anxiety-like behavior of the mice was assessed with the elevated plus maze and open field test.

Results: Chronic restraint stress (CRS) produced dendritic hypertrophy across mouse BLA PNs, regardless of whether they did (BLA→dorsomedial prefrontal cortex [dmPFC]) or did not (BLA↛dmPFC) target dmPFC. However, CRS increased the size of dendritic spine heads and the number of mature, mushroom-shaped spines only in BLA↛dmPFC PNs, sparing neighboring BLA→dmPFC PNs. Moreover, the excitatory glutamatergic transmission was also selectively increased in BLA↛dmPFC PNs, and this effect correlated with CRS-induced increases in anxiety-like behavior. Segregating BLA↛dmPFC PNs based on their targeting of ventral hippocampus (BLA→ventral hippocampus) or nucleus accumbens (BLA→nucleus accumbens) revealed that CRS increased spine density and glutamatergic signaling in BLA→ventral hippocampus PNs in a manner that correlated with anxiety-like behavior.

Conclusions: Chronic stress caused BLA PN neuronal remodeling with a previously unrecognized degree of circuit specificity, offering new insight into the pathophysiological basis of depression, anxiety disorders, and other stress-related conditions.

Keywords: Amygdala; Anxiety; Prefrontal cortex; Projection neuron; Spine; Stress.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Chronic restraint stress (CRS) causes dendritic hypertrophy in basolateral amygdala (BLA) projection neurons (PNs) targeting dorsomedial pre-frontal cortex (dmPFC) (BLA→dmPFC) and not targeting dmPFC (BLA↛dmPFC). (A) Schematic showing injection of red fluorescent RetroBeads into dmPFC to label BLA→dmPFC PNs in the BLA. (B) Examples of RetroBead-labeled BLA→dmPFC and BLA↛dmPFC PNs filled with biocytin using patch clamp recording under whole-cell configuration (scale bar = 10 μm). (C) Examples of reconstructed BLA→dmPFC PNs from unstressed control mice and CRS-exposed mice (scale bar = 30 μm). (D) Sholl analysis of dendritic length as a function of distance from the soma in BLA→dmPFC PNs. Averaged total dendritic length (inset). Distance × CRS two-way analysis of variance (ANOVA): interaction (F9,152 = 1.77, p = .079), main effect of distance (F9,152 = 45.14, p < .001), main effect of CRS (F1,152 = 13.60, p < .001). t test: *p < .05, **p < .01, n = 9 neurons/5 control mice, n = 12 neurons/6 CRS-exposed mice. (E) Average dendritic branch number in BLA→dmPFC PNs. t test: **p < .01. (F) Examples of reconstructed BLA↛dmPFC PNs from unstressed control mice and CRS-exposed mice (scale bar = 30 μm). (G) Sholl analysis of dendritic length as a function of its distance from the soma in BLA↛dmPFC PNs. Averaged total dendritic length (inset). Distance 3 CRS two-way ANOVA: interaction (F8,99 = 0.803, p = .602), main effect of distance (F8,99 = 36.91, p < .001), main effect of CRS (F1,99 = 9.65, p = .003). t test: *p < .05, **p < .01, n = 8 neurons/5 control mice, n = 8 neurons/5 CRS-exposed mice. (H) Average dendritic branch number in BLA↛dmPFC PNs. t test: **p < .01. (I, J) Average branch number in proximal and distal dendrites of BLA→dmPFC (I) and BLA↛dmPFC (J) PNs. For BLA→dmPFC PNs, distance 3 CRS two-way ANOVA: interaction (F1,36 = 0.426, p = .518), main effect of distance (F1,36 = 11.66, p = .002), main effect of CRS (F1,36 = 12.21, p = .001). t test: *p < .05. For BLA↛dmPFC PNs, distance 3 CRS two-way ANOVA: interaction (F1,24 = 1.52, p = .229), main effect of distance (F1,24 = 2.37, p = .137), main effect of CRS (F1,24 = 10.44, p = .004). t test: *p < .05. Data are means ± SEM.
Figure 2.
Figure 2.
Chronic restraint stress (CRS) remodels dendritic spines in basolateral amygdala (BLA) projection neurons (PNs) not targeting dorsomedial prefrontal cortex (dmPFC) (BLA↛dmPFC) but not BLA PNs targeting dmPFC (BLA→dmPFC). (A) Representative images of dendritic spines in BLA→dmPFC and BLA↛dmPFC PNs from unstressed control mice and CRS-exposed mice (scale bar = 2 μm). (B) Average spine density in BLA→dmPFC and BLA↛dmPFC PNs. PN group 3 CRS two-way analysis of variance: interaction (F1,31 = 1.21, p = .281), main effect of PN group (F1,31 = 0.54, p = .466), main effect of CRS (F1,31 = 2.53, p = .122). For BLA→dmPFC PNs, n = 9 neurons/8 control mice, n = 9 neurons/8 CRS-exposed mice. For BLA↛dmPFC PNs, n = 10 neurons/9 control mice, n = 7 neurons/7 CRS-exposed mice. (C) Cumulative probability of spine head diameter in BLA→dmPFC PNs. Control mice, n = 4108 spines/9 neurons; CRS-exposed mice, n = 4240 spines/9 neurons. Kolmogorov-Smirnov test: p = .956. Inset shows the average spine head diameter. t test: p = .123. (D) Cumulative probability of spine head diameter in BLA↛dmPFC PNs. Control mice, n = 5348 spines/10 neurons; CRS-exposed mice, n = 3459 spines/7 neurons. Kolmogorov-Smirnov test: p < .001. Inset shows the average spine head diameter. t test: **p < .01. (E, F) Average density in mushroom, thin, and stubby spines in BLA→dmPFC (E) and BLA↛dmPFC (F) PNs. t test: **p < .01. Data are means ± SEM.
Figure 3.
Figure 3.
Chronic restraint stress (CRS) increases excitatory transmission in basolateral amygdala (BLA) projection neurons (PNs) not targeting dorsomedial prefrontal cortex (BLA↛dmPFC). (A) Representative traces showing miniature excitatory postsynaptic currents (mEPSCs) in BLA PNs targeting dmPFC (BLA→dmPFC) (scale bar = 1 second, 10 pA). (B, C) Cumulative probability of the interevent interval (B) and amplitude (C) of mEPSCs in BLA→dmPFC PNs. Average mEPSC frequency and amplitude (inset). n = 10 neurons/5 control mice, n = 10 neurons/5 CRS-exposed mice. Kolmogorov-Smirnov test for EPSC frequency: p = .207. Kolmogorov-Smirnov test for EPSC amplitude: p = .759. (D) Representative traces showing mEPSCs in BLA↛dmPFC PNs (scale bar = 1 s, 10 pA). (E, F) Cumulative probability of the interevent interval (E) and amplitude (F) of mEPSCs in BLA→dmPFC PNs. Average mEPSC frequency and amplitude (inset). n = 10 neurons/5 control mice, n = 10 neurons/5 CRS-exposed mice. Kolmogorov-Smirnov test for EPSC frequency: p < .001. Kolmogorov-Smirnov test for EPSC amplitude: p = .362. t test: *p < .05. (G) Representative traces showing electrically evoked EPSCs in BLA→dmPFC and BLA↛dmPFC PNs (2 stimuli separated by 50 ms) (scale bar = 20 ms, 150 pA). (H) Average paired pulse ratio (PPR) in BLA→dmPFC and BLA↛dmPFC PNs. PN group 3 CRS two-way analysis of variance: interaction (F1,47 = 0.10, p = .755), main effect of PN group (F1,47 = 0.07, p = .796), main effect of CRS (F1,47 = 0.36, p = .550). For BLA→dmPFC PNs, n = 12 neurons/5 control mice, n = 13 neurons/5 CRS-exposed mice. For BLA↛dmPFC PNs, n = 12 neurons/5 control mice, n = 14 neurons/5 CRS-exposed mice. (I, J) Representative traces showing N-methyl-D-aspartate receptor–mediated currents in BLA→dmPFC (I) and BLA↛dmPFC (J) PNs during their progressive blockage by repetitive electric stimuli in the presence of MK-801. The first, fifth, 10th, 20th, 40th, and 80th current traces shown sequentially from top to bottom (scale bar = 200 ms, 150 pA). (K, L) N-methyl-D-aspartate receptor–mediated current amplitude in the presence of MK-801 as a function of stimulus number (first current amplitude set to 100%). Average tau values (inset). For BLA→dmPFC PNs, n = 8 neurons/4 control mice, n = 7 neurons/4 CRS-exposed mice. t test: p = .955. For BLA↛dmPFC PNs, n = 7 neurons/4 control mice, n = 7 neurons/5 CRS-exposed mice. t test: p = .899. Data are means ± SEM.
Figure 4.
Figure 4.
Excitatory transmission in basolateral amygdala (BLA) projection neurons (PNs) not targeting dorsomedial prefrontal cortex (BLA↛dmPFC) correlates with chronic restraint stress (CRS)–induced anxiety-like behavior. (A) Schematic of experimental designs. (B) Average time mice spent and entries into the open arms in the elevated plus maze (EPM). n = 14 control mice, n = 17 CRS-exposed mice. t test: **p < .01. (C) Average time mice spent in the center square in the open field test (OFT). n = 9 control mice, n = 11 CRS-exposed mice. t test: *p < .05. (D) Correlations between miniature excitatory postsynaptic current (mEPSC) frequency in BLA PNs targeting dmPFC (BLA→dmPFC) and BLA↛dmPFC and open arm exploration in unstressed control mice. (E) Correlations between mEPSC frequency in BLA→dmPFC and BLA↛dmPFC PNs and open arm exploration in CRS-exposed mice. (F) Correlations between mEPSC frequency in BLA→dmPFC and BLA↛dmPFC PNs and time in center square in unstressed control mice. (G) Correlations between mEPSC frequency in BLA→dmPFC and BLA↛dmPFC PNs and time in center square in CRS-exposed mice. Data are means ± SEM. PND, postnatal day.
Figure 5.
Figure 5.
Chronic restraint stress (CRS) causes neuronal remodeling and increased excitatory transmission in basolateral amygdala (BLA) projection neurons (PNs) targeting ventral hippocampus (BLA→vHPC). (A) Schematic showing injection of red fluorescent RetroBeads into vHPC and green fluorescent RetroBeads in nucleus accumbens (NAc) and examples of labeled PNs in BLA. Very few PNs were colabeled (scale bar = 100 μm). (B) Examples of dendritic spines in BLA→vHPC and BLA→NAc PNs (scale bar = 2 μm). (C) Average spine density in BLA→vHPC and BLA→NAc PNs. PN group 3 CRS two-way analysis of variance (ANOVA): interaction (F1,26 = 2.75, p = .109), main effect of PN group (F1,26 = 0.37, p = .54), main effect of CRS (F1,26 = 3.05, p = .093). For BLA→vHPC PNs, n = 7 neurons/4 control mice, n = 8 neurons/4 CRS-exposed mice. For BLA→NAc PNs, n = 7 neurons/6 control mice, n = 9 neurons/7 CRS-exposed mice. (D) Average spine head diameter in BLA→vHPC and BLA→NAc PNs. PN group 3 CRS two-way ANOVA: interaction (F1,16198 = 4.64, p = .03), main effect of PN group (F1,16198 = 95.13, p < .001), main effect of CRS (F1,16198 = 1.51, p = .218). For BLA→vHPC PNs, n = 2165 spines/7 control neurons, n = 4914 spines/8 CRS neurons. For BLA→NAc PNs, n = 3966 spines/7 control neurons, n = 5157 spines/9 CRS neurons. (E, F) Average density in mushroom, thin, and stubby spines in BLA→vHPC (E) and BLA→NAc (F) PNs. For BLA→vHPC PNs, spine type 3 CRS two-way ANOVA: interaction (F2,36 = 2.55, p = .092), main effect of spine type (F2,36 = 183.4, p < .001), main effect of CRS (F1,36 = 18.55, p < .001). t test: *p < .05, **p < .01. For BLA→NAc PNs, spine type 3 CRS two-way ANOVA: interaction (F2,36 = 0.01, p = .988), main effect of spine type (F2,42 = 99.61, p < .001), main effect of CRS (F1,42 = 0.003, p = .953). (G, H) Example traces and average miniature excitatory postsynaptic current (mEPSC) frequency and amplitude in BLA→vHPC (G) and BLA→NAc (H) PNs. For BLA→vHPC PNs, n = 10 neurons/5 control mice, n = 13 neurons/5 CRS-exposed mice. For BLA→NAc PNs, n = 10 neurons/4 control mice, n = 14 neurons/5 CRS-exposed mice. t test: *p < .05 (scale bar = 1 second, 10 picoamperes). (I, J) Correlations between mEPSC frequency in BLA→vHPC PNs and open arm exploration in elevated plus maze in CRS-exposed mice. (K, L) Correlations between mEPSC frequency in BLA→NAc PNs and open arm exploration in elevated plus maze in CRS-exposed mice. (M, N) Correlations between mEPSC frequency in BLA→vHPC (M) and BLA→vHPC (N) PNs and time in center square in open field test in CRS-exposed mice. Data are means ± SEM.
Figure 6.
Figure 6.
Summary schematic showing projection-specific remodeling of basolateral amygdala (BLA) projection neurons (PNs) after chronic stress. BLA PNs targeting dorsomedial prefrontal cortex (BLA→dmPFC) and a subset of those not targeting dmPFC (BLA↛dmPFC) projecting to the nucleus accumbens (NAc) show dendritic hypertrophy with no change in spine density or number of mushroom-type mature spines and no alteration in excitatory transmission. BLA→ventral hippocampus (vHPC) PNs exhibit dendritic hypertrophy, increased spine density, and a preponderance of mature spines. BLA→vHPC PNs also show augmented excitatory transmission, which correlates with stress-induced anxiety-like behavior. mEPSC, miniature excitatory postsynaptic current.

Comment in

Similar articles

Cited by

References

    1. McEwen BS, Morrison JH (2013): The brain on stress: Vulnerability and plasticity of the prefrontal cortex over the life course. Neuron 79:16–29. - PMC - PubMed
    1. McEwen BS, Bowles NP, Gray JD, Hill MN, Hunter RG, Karatsoreos IN, et al. (2015): Mechanisms of stress in the brain. Nat Neurosci 18:1353–1363. - PMC - PubMed
    1. Dias-Ferreira E, Sousa JC, Melo I, Morgado P, Mesquita AR, Cerqueira JJ, et al. (2009): Chronic stress causes frontostriatal reorganization and affects decision-making. Science 325(5940):621–625. - PubMed
    1. de Kloet ER, Joels M, Holsboer F (2005): Stress and the brain: From adaptation to disease. Nat Rev Neurosci 6:463–475. - PubMed
    1. Luthi A, Luscher C (2014): Pathological circuit function underlying addiction and anxiety disorders. Nat Neurosci 17:1635–1643. - PubMed

Publication types