Adolescent social housing protects against adult emotional and cognitive deficits and alters the PFC and NAc transcriptome in male and female C57BL/6J mice

Abstract

Introduction: Adolescence is a critical period in cognitive and emotional development, characterized by high levels of social interaction and increases in risk-taking behavior including binge drinking. Adolescent exposure to social stress and binge ethanol have individually been associated with the development of social, emotional, and cognitive deficits, as well as increased risk for alcohol use disorder. Disruption of cortical development by early life social stress and/or binge drinking may partly underlie these enduring emotional, cognitive, and behavioral effects. The study goal is to implement a novel neighbor housing environment to identify the effects of adolescent neighbor housing and/or binge ethanol drinking on (1) a battery of emotional and cognitive tasks (2) adult ethanol drinking behavior, and (3) the nucleus accumbens and prefrontal cortex transcriptome.

Methods: Adolescent male and female C57BL/6J mice were single or neighbor housed with or without access to intermittent ethanol. One cohort underwent behavioral testing during adulthood to determine social preference, expression of anxiety-like behavior, cognitive performance, and patterns of ethanol intake. The second cohort was sacrificed in late adolescence and brain tissue was used for transcriptomics analysis.

Results: As adults, single housed mice displayed decreased social interaction, deficits in the novel object recognition task, and increased anxiety-like behavior, relative to neighbor-housed mice. There was no effect of housing condition on adolescent or adult ethanol consumption. Adolescent ethanol exposure did not alter adult ethanol intake. Transcriptomics analysis revealed that adolescent housing condition and ethanol exposure resulted in differential expression of genes related to synaptic plasticity in the nucleus accumbens and genes related to methylation, the extracellular matrix and inflammation in the prefrontal cortex.

Discussion: The behavioral results indicate that social interaction during adolescence via the neighbor housing model may protect against emotional, social, and cognitive deficits. In addition, the transcriptomics results suggest that these behavioral alterations may be mediated in part by dysregulation of transcription in the frontal cortex or the nucleus accumbens.

Keywords: adolescence; alcohol; sex differences; social interaction; stress; transcriptomics.

Figure 1

Experimental design male and female C57BL/6J mice were housed in same-sex groups in one of two conditions from post-natal day 29 (PND 29) until the end of the experiment: neighbor housed (4/complex) or single housed (1/cage). Single and neighbor-housed animals were given intermittent access to two bottles of water or a two-bottle choice between ethanol and water. Blue PNDs on the timeline indicate experiments run for the first cohort of animals, while green PNDs indicate experiments run for the second cohort of animals. In cohort one, behavioral testing for male (n = 36) and female (n = 36) animals began on PND 54. The assays tested were social interaction, novel object recognition (NOR), and anxiety-like behavior in the light–dark box. Adult ethanol intake and preference was also measured where all animals were given access to a two-bottle choice between ethanol and water from PND 76–106. In cohort two, male (n = 16) and female (n = 16) were single and neighbor housed and given access to intermittent ethanol or water concurrently with cohort one. On PND 57, PFC, and NAc tissue was collected for microarray and qPCR analysis. Male and female cohorts were run separately.

Figure 2

Adolescent social isolation decreases social interaction in males and females. Time spent in the interaction zone with a stimulus mouse was significantly increased in neighbor housed mice as compared with group housed mice in males (A) and females (B). Time spent in the corner zones far away from the stimulus mouse was significantly increased due to single housing in males (C) and females (D). Ethanol drinking did not alter the behavior of males or females in either the time in the interaction or corner zones. *p < 0.05 by two-way ANOVA.

Figure 3

Adolescent social isolation induces memory deficits in males and females. In the novel object recognition task, single housed mice had a lower discrimination index as compared to male (A) and female (B) neighbor housed mice. Ethanol drinking did not alter the behavior of males or females. *p < 0.05 by two-way ANOVA.

Figure 4

Neighbor housing reduces anxiety-like behavior in the light–dark box. In male animals, percent time in the light was significantly increased in neighbor housed males (A), but only showed a trend for an increase in females (B) as compared to single housed animals. Distance in the light (C,D), and total distance travelled (E,F) were significantly increased in neighbor housed mice as compared to single housed mice in both males and females. An adolescent history of ethanol drinking did not influence these behaviors in males or females. *p < 0.05 by two-way ANOVA.

Figure 5

Adolescent social isolation does not alter ethanol drinking as compared to neighbor housed mice. In adolescence, ethanol consumption was measured using an intermittent access 2-BC paradigm where animals were given the choice between one bottle of ethanol (15% v/v) in tap water or tap water alone. Control groups were given two bottles of tap water (not shown). Mice were housed in single cages or in neighbor cages from PND 29–52 and we compared ethanol intake, preference, and total fluid between groups. There was no significant effect of housing on adolescent ethanol intake (A,B), preference (C,D), or total fluid consumption (E,F) in males or females.

Figure 6

A history of ethanol drinking does not alter adult intake. In adulthood, all groups were given 2-BC for ethanol to assess whether adolescent exposure increases adult intake and preference from PND 76–106. Two-way RMANOVA revealed no significant effect of housing on ethanol intake (A,B), preference (C,D), or total fluid consumption (E,F) in males or females. Significant interactions were found for ethanol intake in males (A) on PND 78, 99, 101, and 104 (single water > neighbor water) and in females (B) on PND 90 (single water < neighbor water). Total fluid consumed also differed in females (F) by housing and day on PND 85 (single ethanol > neighbor ethanol) and PND 90 (single water < neighbor water). *p < 0.05 by two-way RMANOVA.

Figure 7

GO analysis showing genes categories differentially impacted by housing condition in the PFC. (A) Number of DEGs due to housing condition in male and female PFC at p < 0.05. (B) GO analysis of differentially expressed genes unique to males. (C) GO analysis of differentially expressed genes unique to females.

Figure 8

GO analysis showing genes categories differentially impacted by ethanol treatment in the PFC. (A) Number of DEGs due to ethanol treatment in male and female PFC at p < 0.05. (B) GO analysis of differentially expressed genes unique to males. (C) GO analysis of differentially expressed genes unique to females.

Figure 9

GO analysis showing genes categories differentially impacted by the interaction of housing condition and ethanol treatment in the PFC. (A) Number of DEGs due to the interaction of housing condition and ethanol treatment in male and female PFC at p < 0.05. (B) GO analysis of differentially expressed genes unique to males. (C) GO analysis of differentially expressed genes unique to females. (D) GO analysis of differentially expressed genes in both males and females due to the interaction of housing condition and ethanol treatment.

Figure 10

GO analysis showing genes categories differentially impacted by housing condition in the NAc. (A) Number of DEGs due to housing condition in male and female NAc at p < 0.05. (B) GO analysis of differentially expressed genes unique to males. (C) GO analysis of differentially expressed genes unique to females.

Figure 11

GO analysis showing genes categories differentially impacted by ethanol treatment in the NAc. (A) Number of DEGs due to ethanol treatment in male and female NAc at p < 0.05. (B) GO analysis of differentially expressed genes unique to males. (C) GO analysis of differentially expressed genes unique to females.

Figure 12

GO analysis showing genes categories differentially impacted by the interaction of housing condition and ethanol treatment in the NAc. (A) Number of DEGs due to the interaction of housing condition and ethanol treatment in male and female NAc at p < 0.05. (B) GO analysis of differentially expressed genes unique to males. (C) GO analysis of differentially expressed genes unique to females. (D) GO analysis of differentially expressed genes in both males and females due to the interaction of housing condition and ethanol treatment.

Figure 13

Immediate-early genes are changed in the PFC due to housing or the interaction of housing and ethanol treatment in males and females. Male and female DBA/2J mice were housed in same-sex groups in one of two conditions from post-natal day 29 (PND 29) until the end of the experiment: neighbor housed (4/complex) or single housed (1/cage). Single and neighbor-housed animals were given intermittent access to two bottles of water or a two-bottle choice between ethanol and water. qPCR was carried out on PFC tissue collected on PND 57 (n = 4/group) for the following genes: Arc, (A, B); Egr1 (C, D); Npas4 (E, F); Egr3 (G, H);. *p < 0.05 by two-way ANOVA.