Selective early-acquired fear memories undergo temporary suppression during adolescence

Abstract

Highly conserved neural circuitry between rodents and humans has allowed for in-depth characterization of behavioral and molecular processes associated with emotional learning and memory. Despite increased prevalence of affective disorders in adolescent humans, few studies have characterized how associative-emotional learning changes during the transition through adolescence or identified mechanisms underlying such changes. By examining fear conditioning in mice, as they transitioned into and out of adolescence, we found that a suppression of contextual fear occurs during adolescence. Although contextual fear memories were not expressed during early adolescence, they could be retrieved and expressed as the mice transitioned out of adolescence. This temporary suppression of contextual fear was associated with blunted synaptic activity in the basal amygdala and decreased PI3K and MAPK signaling in the hippocampus. These findings reveal a unique form of brain plasticity in fear learning during early adolescence and may prove informative for understanding endogenous mechanisms to suppress unwanted fear memories.

Fig. 1.

Hippocampal-dependent memory across adolescent development. (A) Mice of all ages were fear-conditioned with three tone-shock pairings. Twenty-four hours later, they were returned to the conditioning context (context A) and freezing behavior was scored. Twenty-four hours after the contextual fear test, mice were placed in a novel context (context B) and scored for freezing behavior during five tone presentations. (B) P29 mice froze significantly less than older mice when tested for hippocampal-dependent, contextual fear. (C) In a hippocampal-dependent, spatial, novel object placement task, all age groups spent significantly more time than chance exploring objects in a novel location. (D) Adolescent mice (P29–P39) froze significantly less than both younger (P23–P27) and older (P49–P70) mice. All results are presented as a mean ± SEM determined from analysis of 7–10 mice per group (fear conditioning) and 28 mice per group (novel object placement) (*P < 0.05, ***P < 0.001). See also Figs. S1 and S2.

Fig. 2.

Early-adolescent mice display impaired contextual fear at early but not late time points. (A) Mice were fear-conditioned (Fear Cond) with three tone-shock pairings (P29; n = 7 control, 10 fear-conditioned; adult, n = 10 control, 10 fear-conditioned). Twenty-four hours later, freezing behavior was scored in the conditioning context, context A. Fourteen days later, mice were retested for contextual fear in context A. Twenty-four hours after the second contextual fear test, mice were tested for cued fear in a novel context, context B. (B) Mice fear-conditioned at P29 did not display significant freezing to context A when tested 24 h postconditioning, but showed significantly enhanced freezing levels when tested 14 d postconditioning. (C) Mice conditioned as adults displayed significant freezing to context A when tested both 24 h and 14 d postconditioning. All results are presented as a mean ± SEM determined from analysis of 7–10 mice per group (**P < 0.01, ***P < 0.001). See also Figs. S2 and S3.

Fig. 3.

Contextual fear conditioning–induced synaptic potentiation is present in adult mice but not in early-adolescent mice. Input–output curve of slopes of fEPSPs recorded from BA in adult mice (A) (n = 17 slices in each group, n = 5 mice in each group), early-adolescent mice (B) (n = 14 slices in each group, n = 5 mice in each group), and P43 mice fear-conditioned at P29 (C) (n = 15 slices in each group, n = 5 mice in each group). Examples of fEPSPs are shown on the left. All results are presented as a mean ± SEM. See also Figs. S4 and S5.

Fig. 4.

PI3K and ERK/MAPK activation occurs in the hippocampi of adult but not early-adolescent mice during contextual fear memory retrieval. (A) Mice were fear-conditioned with three tone-shock pairings in context A (P29, n = 14; adult, n = 19). Twenty-four hours postconditioning, mice in the retrieval group were returned to context A for a contextual fear retrieval session and killed 15 min postretrieval, whereas control mice were killed without being given a retrieval session. (B and C) Representative Western blots showing activation of PI3K (B) and ERK/MAPK (C) signaling upon contextual fear memory retrieval in adult but not P29 mice. (D and E) Quantification of pAkt (D) and pERK2 (E) normalized to total Akt and ERK protein levels, respectively (n = 6–10 mice per group; **P < 0.01).