Corticosterone after early adolescent stress prevents social avoidance, aversive behavior, and morphine-conditioned place preference in adulthood

Abstract

Rationale: Stress during childhood or adolescence increases vulnerability to psychiatric disorders in adults. In adult rodents, the delayed effects of stress can increase anxiety-like behavior. These effects, however, can be prevented with post-stress administration of corticosterone (CORT). The effectiveness of CORT in preventing adolescent stress-induced emotional behavior alterations in adulthood has yet to be investigated.

Objectives: Here, we investigated the interactions between early adolescent stress and exogenous corticosterone on adult social, aversive, and drug-seeking behavior in mice, which are translationally related to symptoms associated with psychiatric and substance abuse disorders.

Methods and results: A single administration of CORT in drinking water (400ug/mL) for 24 h after social defeat or context fear conditioning prevents defeat-induced social avoidance, alters fear processing, prevents adolescent stress-induced anhedonia, and prevents stress-potentiated morphine place preference in adulthood. Exogenous CORT did not immediately prevent stress-induced potentiation of morphine conditioned-place preference in adolescents but did so in adult mice. However, when administered to adolescent mice, CORT also prevented the incubation of morphine-conditioned place preference into adulthood. Lastly, exogenous CORT administration blunted endogenous corticosterone but was unrelated to freezing behavior during a fear test.

Conclusions: This is the first demonstration of adolescent post-stress CORT promoting socio-emotional resilience and preventing drug-seeking behavior. Our data suggest elevated corticosterone after a stress experience promotes resilience for at least 40 days across the developmental transition from adolescence to adulthood and is effective for socio-emotional and drug-seeking behavior. These results are critical for understanding how adolescent stress impacts emotional and drug-seeking behavior into adulthood.

Keywords: Drug-seeking; Fear conditioning; Foot-shock stress; Social defeat.

Fig. 1

Post-stress CORT prevents defeat-induced social avoidance and stress-potentiated morphine place preference. (A) Timeline for adolescent social defeat and corticosterone treatment on social interaction and morphine conditioned place preference. (B) Total number of attacks for mice that underwent social defeat stress, there was no difference in total number of attacks between vehicle (VEH, Red Bars) and corticosterone (CORT, Blue Bars)-treated mice. (C) Total fluid intake over the 24 h treatment period; defeated mice consumed more water than non-defeated controls. (D) Social interaction test in adolescence (P32) and adulthood (P62). There were no immediate effects of social defeat or CORT administration during the P32 test. When mice were tested as adults (P62) CORT or social defeat alone decreased social interaction compared to non-defeated vehicle treated controls. In defeated mice, CORT treatment normalized adult social interaction comparable to non-defeated vehicle controls. (E) Morphine conditioned place preference in adulthood; both corticosterone and social defeat alone increased preference for the morphine paired chamber but corticosterone treatment after social defeat prevented this potentiated preference, producing preference scores comparable to non-defeated vehicle treated controls. No defeat VEH (n = 13), No defeat CORT (n = 10), Defeat VEH (n = 13), Defeat CORT (n = 13). Bars with different letters indicate a significant difference between treatment groups, p < 0.05. Two-way ANOVA with Tukey’s or Sidaks multiple comparison tests. * = p < 0.05; ** = p < 0.01; **** = p < 0.0001

Fig. 2

Post-stress CORT maintains context memory specificity and prevents stress-potentiated morphine place preference. (A) Timeline for adolescent fear conditioning and CORT treatment on fear expression, morphine place preference and sucrose preference. (B) Freezing during context fear conditioning. There was no difference in fear acquisition (average of post-shock freezing across the 5 shocks) between vehicle-treated (VEH, Red Bars) and corticosterone (CORT, Blue Bars)-treated groups. Shocked mice displayed increased freezing compared to no shock control mice. (C) During a fear test in adolescence (Adol. P32) only mice exposed to fear conditioning (Circles) had high levels of freezing whereas no shock control groups (Triangles) did not exhibit freezing. VEH-treated mice had a significant reduction in freezing from the adolescent (P32) to adult (P62) test, but when mice were given CORT, this effect was eliminated. No other comparisons were significantly different. (D) Morphine conditioned place preference testing identified that CORT alone, and shock alone potentiated morphine place preference compared to no shock control mice administered VEH (though the increase in the VEH-treated group was no statistically significant). Corticosterone prevented the stress-potentiated morphine place preference; preference was no different than no shock VEH-treated controls. (E) Total fluid intake over the 24 h period, there was no effect of shock or treatment on fluid consumption. (F) Sucrose preference testing identified that shock exposure (in the absence of CORT) produced lower levels of sucrose preference compared to no shock controls and no shock CORT-treated mice. CORT prevented the reduced sucrose preference produced by adolescent stress. No shock VEH (n = 11), No shock CORT ( n = 12), Shock VEH (n = 11), Shock CORT (n = 11). Three-way ANOVA with Sidak’s post hoc tests across treatment groups and age (C) and two-way ANOVA with Sidak’s MC post hoc test in (D), * = p < 0.05; ** = p < 0.01; **** = P < 0.0001

Fig. 3

Post stress CORT prevents fear generalization into adulthood. (A) Timeline for context fear training and CORT treatment on fear expression in the training context [A] or a novel context [B]. Mice that were tested in the training context in adolescence (P32) were tested in the training context as adults (P62). The same procedure was applied to those mice tested in the novel context. (B) Context fear acquisition of vehicle (VEH, Red Bars) and corticosterone (CORT, Blue Bars)-treated mice in their “to-be” testing contexts. There was no difference in fear acquisition (average of post-shock freezing across the 5 shocks) (C) Context fear expression as percent time spent freezing over three minutes in the training context or in a novel context in adolescence (P32) and adulthood (P62). No differences were observed between VEH- and CORT-treated mice in the training context. However, CORT-treated mice displayed lower freezing compared to VEH-treated mice when tested in the novel context. VEH-treated mice demonstrate a time-dependent increase in freezing when tested in the novel context, whereas CORT treatment blocks this increase. (D) Total fluid intake over the 24 h period, there was no effect of grouping on fluid consumption. VEH Training Context (n = 10), CORT Training Context (n = 9), VEH Novel Context (n = 10), CORT Novel Context (n = 10). **** = P < 0.0001; Two-way ANOVA with Sidak’s post hoc test

Fig. 4

Post-stress CORT blocks morphine place preference in adults. (A) Timeline for adult context fear conditioning and corticosterone (CORT) treatment on fear expression and morphine place preference. (B) Freezing during context fear conditioning. There was no difference in fear acquisition (average of post-shock freezing across the 5 shocks) between vehicle-treated (VEH, Red Bars) and corticosterone (CORT, Blue Bars)-treated groups. Shocked mice displayed increased freezing compared to no shock control mice. (C) Context fear expression as percent time spent freezing over three minutes. Context fear conditioned mice that were administered CORT displayed less freezing compared to VEH-treated mice. Shocked mice displayed greater freezing compared to no shock control mice regardless of treatment. (D) Total fluid intake over the 24 h period, there was a main effect of treatment on fluid intake; CORT-treated mice drank more water than VEH-treated mice, but no individual post hoc comparison was significant. (E) CORT administration or shock increased morphine conditioned place preference. CORT-administration in shocked mice prevented potentiated morphine place preference. No shock VEH (n = 7), No shock CORT ( n = 7), Shock VEH (n = 7), Shock CORT (n = 6). * = p < 0.05; ** p < 0.01; **** p < 0.0001; Two-way ANOVA with Sidak’s post hoc test

Fig. 5

Post stress CORT in adolescence blocks incubation of morphine place preference. (A) Timeline for adolescent context fear conditioning and incubation of morphine place preference. (B) Freezing during context fear conditioning. There was no difference in fear acquisition (average of post-shock freezing across the 5 shocks) between vehicle-treated (VEH, Red Bars) and corticosterone (CORT, Blue Bars)-treated groups. Shocked mice displayed increased freezing compared to no shock control mice. (C) Context fear expression in adolescence as percent time spent freezing over three minutes. Shocked mice diplayed more freezing than no shock controls but there were no differences across treatment groups. (D) Total fluid intake over the 24 h period. (E) Morphine conditioned place preference in adolescence showed that CORT does not reduce preference for the morphine paired side in adolescence. However, CORT treatment blocked incubation of preference – all groups increased their preference scores when tested in adulthood except for the CORT-treated mice that received shock. No shock VEH (n = 12), No shock CORT (n = 9), Shock VEH (n = 10), Shock CORT (n = 11). **** P < 0.0001; *** P < 0.001 ** = P < 0.01; * = p < 0.05; Two-way ANOVA with Sidak’s post hoc test and Three- way ANOVA with Sidak’s in (E)

Fig. 6

Post stress CORT blunts serum corticosterone. (A) Timeline for adolescent fear conditioning and serum corticosterone analysis. (B) Freezing during context fear conditioning. There was no difference in fear acquisition (average of post-shock freezing across the 5 shocks) between vehicle-treated (VEH, Red Bars) and corticosterone (CORT, Blue Bars)-treated groups. Shocked mice displayed increased freezing compared to no shock control mice. (C) Context fear expression in early adolescence (P32; Left) and adulthood (P62; Right). At both tests, shocked mice displayed more freezing than no shock controls but there were no differences across treatement groups. (D) Serum corticosterone during adolescence (P32; Left) and adulthood (P62; Right). In adolescence, serum corticosterone was suppressed in shocked mice that were administered CORT compared to VEH-treated mice. In adulthood, serum corticosterone was elevated in shocked mice compared to no shock controls regardless of treatment. No shock VEH (n = 10 P32; n = 6 P62), No shock CORT ( n = 9 P32; n = 6 P62), Shock VEH (n = 6 P32; n = 6 P62), Shock CORT (n = 6 P32; n = 6 P62). Two-way ANOVA with Sidak’s post hoc test; * p < 0.05; ** p < 0.01; **** p < 0.0001