Handling stress impairs learning through a mechanism involving caspase-1 activation and adenosine signaling

Abstract

Acute stressors can induce fear and physiologic responses that prepare the body to protect from danger. A key component of this response is immune system readiness. In particular, inflammasome activation appears critical to linking stress to the immune system. Here, we show that a novel combination of handling procedures used regularly in mouse research impairs novel object recognition (NOR) and activates caspase-1 in the amygdala. In male mice, this handling-stress paradigm combined weighing, scruffing and sham abdominal injection once per hr. While one round of weigh/scruff/needle-stick had no impact on NOR, two rounds compromised NOR without impacting location memory or anxiety-like behaviors. Caspase-1 knockout (KO), IL-1 receptor 1 (IL-1R1) KO and IL-1 receptor antagonist (IL-RA)-administered mice were resistant to handling stress-induced loss of NOR. In addition, examination of the brain showed that handling stress increased caspase-1 activity 85% in the amygdala without impacting hippocampal caspase-1 activity. To delineate danger signals relevant to handling stress, caffeine-administered and adenosine 2A receptor (A2AR) KO mice were tested and found resistant to impaired learning and caspase-1 activation. Finally, mice treated with the β-adrenergic receptor antagonist, propranolol, were resistant to handling stress-induced loss of NOR and caspase-1 activation. Taken together, these results indicate that handling stress-induced impairment of object learning is reliant on a pathway requiring A2AR-dependent activation of caspase-1 in the amygdala that appears contingent on β-adrenergic receptor functionality.

Keywords: Adenosine; Amygdala; Caffeine; Caspase-1; Learning; Stress; β-Adrenergic receptor.

Figure 1.. Handling stress paradigm.

At the start of the dark cycle (time 0 hr), mice were subjected to weighing, scruffing, and IP needle stick, as indicated. Six hour stress (6 hr stress) was defined as two rounds of weigh/scruff/needle stick and is illustrated. Three hour stress (3 hr stress) was defined as only one round of weigh/scruff/needle stick.

Figure 2.. Repeated stress handling impairs recognition memory but not spatial memory or anxiety-like behavior.

(A) Mice were exposed to control conditions (control) or 6 hr handling stress (stress) then tested for NOR. Results are expressed as mean ± SEM % novel object investigation; n=7 (*p<0.05). (B) Mice were treated as in A and tested for NLR was performed. Results are expressed as mean ± SEM % Novel location investigation; n=8. (C) Mice were treated as in A & B mice and tested for EZM performance. Results are expressed as mean ± SEM % time in open arms; n=12. (D) Mice were treated as in A & B mice and blood glucose was measured. Results are expressed as mean ± SEM Blood Glucose; n=9 (*p<0.05). (E) Mice were treated as in A & B mice and home-cage locomotor activity was measured. Results are expressed as mean ± SEM cm moved during each time period; n=6 (*p<0.05). A-E were analyzed using a One-way ANOVA.

Figure 2.. Repeated stress handling impairs recognition memory but not spatial memory or anxiety-like behavior.

(A) Mice were exposed to control conditions (control) or 6 hr handling stress (stress) then tested for NOR. Results are expressed as mean ± SEM % novel object investigation; n=7 (*p<0.05). (B) Mice were treated as in A and tested for NLR was performed. Results are expressed as mean ± SEM % Novel location investigation; n=8. (C) Mice were treated as in A & B mice and tested for EZM performance. Results are expressed as mean ± SEM % time in open arms; n=12. (D) Mice were treated as in A & B mice and blood glucose was measured. Results are expressed as mean ± SEM Blood Glucose; n=9 (*p<0.05). (E) Mice were treated as in A & B mice and home-cage locomotor activity was measured. Results are expressed as mean ± SEM cm moved during each time period; n=6 (*p<0.05). A-E were analyzed using a One-way ANOVA.

Figure 2.. Repeated stress handling impairs recognition memory but not spatial memory or anxiety-like behavior.

(A) Mice were exposed to control conditions (control) or 6 hr handling stress (stress) then tested for NOR. Results are expressed as mean ± SEM % novel object investigation; n=7 (*p<0.05). (B) Mice were treated as in A and tested for NLR was performed. Results are expressed as mean ± SEM % Novel location investigation; n=8. (C) Mice were treated as in A & B mice and tested for EZM performance. Results are expressed as mean ± SEM % time in open arms; n=12. (D) Mice were treated as in A & B mice and blood glucose was measured. Results are expressed as mean ± SEM Blood Glucose; n=9 (*p<0.05). (E) Mice were treated as in A & B mice and home-cage locomotor activity was measured. Results are expressed as mean ± SEM cm moved during each time period; n=6 (*p<0.05). A-E were analyzed using a One-way ANOVA.

Figure 3.. Repeated stress handling activates caspase-1 in the brain.

(A) Mice were exposed to control conditions (control) or 6 hr handling stress (stress). Activated caspase-1 was examined by IHC. Results are expressed as mean ± SEM vs. control group; n=10–12 (*p<0.05). Analyzed using a One-way ANOVA for each region. (B) is a representative image from A (40x).

Figure 3.. Repeated stress handling activates caspase-1 in the brain.

(A) Mice were exposed to control conditions (control) or 6 hr handling stress (stress). Activated caspase-1 was examined by IHC. Results are expressed as mean ± SEM vs. control group; n=10–12 (*p<0.05). Analyzed using a One-way ANOVA for each region. (B) is a representative image from A (40x).

Figure 4.. Mice deficient in caspase-1 or the IL-1R1 lack stress handling-induced memory impairment.

(A) C57 Wild-type (WT) and Caspase-1 KO (Casp-1 KO) mice were exposed to control conditions (control) or stress handling (stress) for 6hr. NOR was performed 3 hr following stress exposure. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (B) As in A, WT and IL1R1 KO (IL1R1) mice were subjected to either control conditions or stress handling for 6 hr and NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object recognition; n=7–9. (C) Mice were exposed stress or control conditions as in A & B immediately following administration of the IL1R1 inhibitor Kineret. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object recognition. For both A-C values without a common superscript are significantly different (p<0.05), and were analyzed using a Two-way ANOVA.

Figure 4.. Mice deficient in caspase-1 or the IL-1R1 lack stress handling-induced memory impairment.

(A) C57 Wild-type (WT) and Caspase-1 KO (Casp-1 KO) mice were exposed to control conditions (control) or stress handling (stress) for 6hr. NOR was performed 3 hr following stress exposure. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (B) As in A, WT and IL1R1 KO (IL1R1) mice were subjected to either control conditions or stress handling for 6 hr and NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object recognition; n=7–9. (C) Mice were exposed stress or control conditions as in A & B immediately following administration of the IL1R1 inhibitor Kineret. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object recognition. For both A-C values without a common superscript are significantly different (p<0.05), and were analyzed using a Two-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 5.. Adenosine A2AR antagonism prevents stress handling-induced memory impairment and the activation of caspase-1 in the amygdala.

Mice were exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) prior to stress exposure mice were administered the pan-adenosine receptor antagonist caffeine. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=7–8. (B) Mice were treated as in A and caspase-1 activity was measured by IHC. Results are expressed as mean ± SEM % control vs. saline treated controls; n=8–10. (C) Representative images (40x) from the amygdala from B. (D)Mice were stressed as in A and administered the A2AR specific antagonist DMPX prior to stress exposure or a 2% DMSO vehicle control. NOR was performed 3 hr after stress exposure. Results are expressed as mean ± SEM % novel object investigation; n= 7–8. (E) Mice were treated as in D and caspase-1 activity was measured via IHC. Results are expressed as % control vs. DMSO treated control mice; n=7–8. (F) Representative amygdala images (40x) from E. (G) BALB/c control (WT) and A2AR KO mice were stressed as in A. NOR was performed 3 hr following stress exposure. Results are expressed as mean ± SEM % novel object investigation; n=8. (H) WT and A2AR KO mice were treated as in G and caspase-1 activity was measured via IHC. Results are expressed as mean ± SEM % control vs. BALB/c controls; n=10–11. (I) Representative amygdala images (40x) from H. (J) Mice were administered IP saline or IP adenosine. NOR was performed 30 min following adenosine administration. Results are expressed as mean ± SEM % novel object investigation; n= 8 (*p<0.05). (K) Mice were treated as in J and caspase-1 activity was measured by IHC. Results are expressed as % control vs. saline; n= 6–8 (*p<0.05). (L) Representative amygdala images (40x) from K. (M) Mice were treated as in J but were also ICV injected with biotin-YVAD-cmk immediately following adenosine injection. NOR was performed 30 min following YVAD injection. Results are expressed as mean ± SEM % novel object investigation; n= 4–5). For A-M values without a common superscript are significantly different (p<0.05). A-I & M were analyzed using a Two-way ANOVA procedure, J-K were analyzed using a One-way ANOVA.

Figure 6.. Catecholamine but not glucocorticoid antagonism blocks stress handling-induced memory impairment and caspase-1 activation.

Mice exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) Prior to stress exposure mice were administered the β-adrenergic receptor antagonist propranolol or vehicle control. NOR was performed 3 hr following stress exposure. Results are expressed as means ± SEM % novel object investigation; n=8. (B) Mice were treated identically to A and active caspase-1 was measured via IHC. Results are expressed as mean ± SEM % Control vs. saline treated control mice; n=7–9. (C) Representative images (40x) from the amygdala of B. (D) Mice were treated identically to A, but prior to stress exposure mice were administered the glucocorticoid receptor antagonist mifepristone or a 5% DMSO vehicle control. Results are expressed as means ± SEM % novel object investigation; n=8. For A-D values without a common superscript are significantly different (p<0.05), and all were analyzed using a Two-way ANOVA procedure.

Figure 6.. Catecholamine but not glucocorticoid antagonism blocks stress handling-induced memory impairment and caspase-1 activation.

Mice exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) Prior to stress exposure mice were administered the β-adrenergic receptor antagonist propranolol or vehicle control. NOR was performed 3 hr following stress exposure. Results are expressed as means ± SEM % novel object investigation; n=8. (B) Mice were treated identically to A and active caspase-1 was measured via IHC. Results are expressed as mean ± SEM % Control vs. saline treated control mice; n=7–9. (C) Representative images (40x) from the amygdala of B. (D) Mice were treated identically to A, but prior to stress exposure mice were administered the glucocorticoid receptor antagonist mifepristone or a 5% DMSO vehicle control. Results are expressed as means ± SEM % novel object investigation; n=8. For A-D values without a common superscript are significantly different (p<0.05), and all were analyzed using a Two-way ANOVA procedure.

Figure 6.. Catecholamine but not glucocorticoid antagonism blocks stress handling-induced memory impairment and caspase-1 activation.

Mice exposed to control conditions (control) or stress stimuli (stress) for 6 hr. (A) Prior to stress exposure mice were administered the β-adrenergic receptor antagonist propranolol or vehicle control. NOR was performed 3 hr following stress exposure. Results are expressed as means ± SEM % novel object investigation; n=8. (B) Mice were treated identically to A and active caspase-1 was measured via IHC. Results are expressed as mean ± SEM % Control vs. saline treated control mice; n=7–9. (C) Representative images (40x) from the amygdala of B. (D) Mice were treated identically to A, but prior to stress exposure mice were administered the glucocorticoid receptor antagonist mifepristone or a 5% DMSO vehicle control. Results are expressed as means ± SEM % novel object investigation; n=8. For A-D values without a common superscript are significantly different (p<0.05), and all were analyzed using a Two-way ANOVA procedure.