Chronic Hippocampal Abnormalities and Blunted HPA Axis in an Animal Model of Repeated Unpredictable Stress

Abstract

Incidence of post-traumatic stress disorder (PTSD) ranges from 3 to 30% in individuals exposed to traumatic events, with the highest prevalence in groups exposed to combat, torture, or rape. To date, only a few FDA approved drugs are available to treat PTSD, which only offer symptomatic relief and variable efficacy. There is, therefore, an urgent need to explore new concepts regarding the biological responses causing PTSD. Animal models are an appropriate platform for conducting such studies. Herein, we examined the chronic behavioral and neurobiological effects of repeated unpredictable stress (RUS) in a mouse model. 12 weeks-old C57BL/6J male mice were exposed to a 21-day RUS paradigm consisting of exposures to a predator odor (TMT) whilst under restraint, unstable social housing, inescapable footshocks and social isolation. Validity of the model was assessed by comprehensive examination of behavioral outcomes at an acute timepoint, 3 and 6 months post-RUS; and molecular profiling was also conducted on brain and plasma samples at the acute and 6 months timepoints. Stressed mice demonstrated recall of traumatic memories, passive stress coping behavior, acute anxiety, and weight gain deficits when compared to control mice. Immunoblotting of amygdala lysates showed a dysregulation in the p75NTR/ProBDNF, and glutamatergic signaling in stressed mice at the acute timepoint. At 6 months after RUS, stressed mice had lower plasma corticosterone, reduced hippocampal CA1 volume and reduced brain-derived neurotrophic factor levels. In addition, glucocorticoid regulatory protein FKBP5 was downregulated in the hypothalamus of stressed mice at the same timepoint, together implicating an impaired hypothalamus-pituitary-adrenal-axis. Our model demonstrates chronic behavioral and neurobiological outcomes consistent with those reported in human PTSD cases and thus presents a platform through which to understand the neurobiology of stress and explore new therapeutic interventions.

Keywords: HPA axis; PTSD; animal model; corticosterone; stress.

FIGURE 1

Study timeline and experimental procedures for the stress paradigm. Stress paradigm involved 21 days of daily unstable social housing with an alternate congener (i), unpredictable repetitive exposures to danger-related predator odor (fox urine, TMT), while under a decapicone restrainer for 30 min (ii), and physical trauma in the form of five repeated inescapable footshocks (iii). After 21 days of stress, animals in the stress group were singly housed until the end of the study (iv). A battery of behavioral testing was conducted at an acute timepoint (1–14 days), 3 and 6 months after the last footshock. Brain tissue and plasma were collected only at the acute timepoint (10 days) and 6 months after the last footshock.

FIGURE 2

Effect of RUS on bodyweight gain and anxiety-like behavior. During the 21-day stress paradigm, stressed mice showed a reduction in body weight gain when compared to control mice (A). Growth rate was significantly lower only during and after the first week of stress (day 7) (B). Exposure to stress increased anxiety-like behavior at the acute timepoint as evident by reduced center zone entries per a 3-min time bin (C) and in the total 15 min (D) in the open field test. No significant changes were observed in the EPM open arm time (E) and entries (F) at the acute timepoint. Data in (A,B) was analyzed using repeated measures Two-Way ANOVA followed by post hoc Tukey’s test (n = 14–17), while data in (C–F) were analyzed using a student t-test (n = 8–9). Asterisks denote statistical significance as follows: ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001.

FIGURE 3

Recall of contextual and cued fear memory at the acute timepoint, 3 and 6 months after RUS. Stressed animals recalled contextual fear memories only at the acute timepoint as evident by increased freezing time compared to controls at the acute timepoint (A), but not at 3 months (B) or 6 months (C) after RUS. Panels (E–G) depict% freezing scores in the first minute of the cued memory test in absence and presence of a cue (tone) at the acute, 3- and 6-month timepoints, respectively. There was no significant freezing for all treatment groups when tested in a new context without cues (no tone) at all timepoints (D–F). After tone introduction, RUS mice showed a significant increase in % freezing at the acute timepoint, 3 and 6 months after RUS (D–F). Stressed mice displayed increased immobility time in minutes 4 and 6 of the FST when compared to a control group (G). Average immobility time in the last 4 min was increased in stressed mice compared to the control group (H). RUS didn’t alter open arm entries in the elevated plus maze test at 6 months after stress. Data in (A–C,H,I) were analyzed using a student t-test (n = 6–12), while data in (D–F) were analyzed using repeated measures Two-Way ANOVA followed by post hoc Sidak test (n = 10–12). Asterisks denote statistical significance as follows: ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001.

FIGURE 4

Effect of stress on HPA axis regulation at 6 months after RUS. Mice exposed to stress showed lower plasma levels of corticosterone (A) and ACTH (B), but not CRH (C) when compared to a control group at 6 months after RUS (n = 8–12). Representative western blot images for markers of glucocorticoid signaling in the hypothalamus (D). Quantification of mineralocorticoid Receptor (E), glucocorticoid receptor (F), FKBP5 (G), and CRH (H) levels in the hypothalamus (n = 5–6). Only the levels of FKBP5 (G) were significantly reduced after RUS. Plasma corticosterone levels were inversely correlated with the% freezing scores in the cued fear memory test (I) and the immobility time in the forced swim test (J) (n = 21). There was a positive correlation between% freezing scores in the cued fear memory test and the immobility time in the forced swim test (K). Each solid black triangle represents an animal in the stress group, while white circles represent control animals. Asterisks denote statistical significance as follows: ^∗p < 0.05; ^∗∗p < 0.01.

FIGURE 5

Exposure to stress elevates (A) Orexin-A levels in the hypothalamus. Magnetic beads multiplex assay showed an increase in the hypothalamic levels of Orexin-A at 6 months after RUS. No significant changes were observed in the levels of Oxytocin (B), β-endorphin (C), Substance P (D), α-MSH (E), and Neurotensin (F). Data in Figure 5 were analyzed using a student t-test (n = 5–6). Asterisks denote statistical significance as follows: ^∗p < 0.05.

FIGURE 6

Effect of stress on the hippocampus at 6 months after RUS. Effect of stress on the volume of the dorsal hippocampus (A) and the volume other hippocampal regions (B). Stress resulted in a volume reduction only in the CA1 region (B). Hippocampal BDNF (C) and corticosterone (D) levels were also reduced at 6 months after stress. Representative western blot images for markers of glucocorticoid signaling in the Hippocampus (E). Quantification of mineralocorticoid receptor (F), glucocorticoid Receptor (G), FKBP5 (H), and CRH (I) levels in the hippocampus at 6 months after RUS. Only the levels of CRH (I) were significantly reduced after RUS. Data were analyzed using a student t-test (n = 5–6). Asterisks denote statistical significance as follows: ^∗p < 0.05; ^∗∗p < 0.01.

FIGURE 7

Effect of stress on the amygdala at the acute timepoint. Representative western blot images from amygdala lysates for neurotrophic markers, NMDA receptors, and other synaptic plasticity markers are shown in (A,F,K), respectively. Quantification of western blot images of TRKB (B), P75NTR (D), ProBDNF (E), and NMDAR2B (G), NMDAR2A (H), NMDAR1 (I), PSD95 (J), CAMKII (L), and pNF-κB/NF-κB (M) levels in the amygdala at the acute timepoint. No change in amygdala BDNF levels after RUS at the acute timepoint (C). All Data were analyzed using a student t-test (n = 5–6). Asterisks denote statistical significance as follows: ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001.

FIGURE 8

A summary of stress effects on the amygdala, hippocampus, and HPA axis at different timepoints after RUS.