The Changes of Histone Methylation Induced by Adolescent Social Stress Regulate the Resting-State Activity in mPFC

Abstract

Early-life stress can lead to sustained alterations in regional resting-state brain functions, but the underlying molecular mechanism remains unclear. Stress can also induce sustained changes in epigenetic modifications across brain regions, which are not limited to a few genes; rather, they often tend to produce global levels of change. The functional implication of these changes also remains to be elucidated. We hypothesize that global epigenetic changes may partly modulate the resting-state functions of brain regions to influence behavior. To test this hypothesis, we used an adolescent social stress (ASS) model in mice and examined the relationship between epigenetic modifications and regional resting-state brain activity using resting-state functional magnetic resonance imaging (rs-fMRI). The results showed that, compared to the control mice, the stressed mice showed increased anxiety and social avoidance behaviors and greater levels of dimethylation of histone H3 at lysine 9 (H3K9me2) in the medial prefrontal cortex (mPFC). In addition, the resting-state activity represented by the amplitude of low-frequency fluctuation (ALFF) was significantly lower in the mPFC of stressed mice. To verify the relationship of H3K9me2 and ALFF, the specific inhibition of H3Kme2 was performed by using the drug UNC0642, which reversed the anxiety behavior induced by ASS and significantly increase the ALFF value of mPFC in both normal and ASS animals. Our study is the first to report an association between histone modifications and rs-fMRI findings, providing a new perspective for understanding of the significance of regional brain epigenetic changes and a possible molecular explanation for rs-fMRI findings.

Fig. 1.

The changes of behaviors in the ASS model. (A) Schematic timeline (PND, postnatal days; behavior tests are elevated plus maze test, social avoidance test, and 3-chamber test, performed in this order). (B) Social interaction ratio in the social avoidance test. (C) Open-arm ratio in the elevated plus maze test. (D) Social exploration ratio in the 3-chamber test. (E) Novel social preference in the 3-chamber test (*P < 0.05, **P < 0.01, ****P < 0.001; t test; control versus stress; n = 14 or 13).

Fig. 2.

The association of ALFF values and H3K9 methylations levels in mPFC in the ASS model. (A) Levels of H3K9me2 and H3K9me3 in the mPFC (**P < 0.01; t test; control versus stress; n = 14 or 13). (B) Typical immunoblot band of histone H3K9me2. (C) The ROI of the mPFC is shown on the template. (D) Extracted data for the ALFF in the mPFC (***P < 0.001; t test; control versus stress; n = 14 or 13). (E) Representative time series of BOLD signal in mPFC (0.01 to 0.1 Hz filtered).

Fig. 3.

Whole-brain voxel-based analysis of ALFF values. (A and B) ALFF values in the front of the brain (mainly in the mPFC) were significantly decreased in stressed mice compared to control mice (voxel P < 0.001, cluster P < 0.05, GRF correction). (C) Extracted data for the ALFF of the significant cluster (t test, ****P < 0.001, control versus stress; n = 14 or 13).

Fig. 4.

The inhibition of H3K9me2 increased ALFF value in mPFC. (A) Levels of H3K9me2 in the mPFC (*P < 0.05; t test; saline versus UNC0642; n = 8). (B) Typical immunoblot band of H3K9me2. (C) Extracted data for the ALFF in the mPFC (*P < 0.05; t test; saline versus UNC0642; n = 8). (D) Representative time series of BOLD signal in mPFC (0.01 to 0.1 Hz filtered).

Fig. 5.

The inhibition of H3K9me2 reversed the increase of anxiety and the reduction of ALFF value in mPFC. (A) Schematic timeline. (B) The open-arm ratio after the administration of UNC0642 (*P < 0.05; one-way ANOVA; n = 9 to 10). (C) Extracted data for the ALFF in the mPFC. (D) Representative time series of BOLD signal in mPFC (0.01 to 0.1 Hz filtered) (*P < 0.05; one-way ANOVA; n = 9 to 10). (E) Levels of H3K9me2 in the mPFC (*P < 0.05; one-way ANOVA; n = 6 to 8). (F) Typical immunoblot band of H3K9me2.