Histone methylation regulates memory formation

Abstract

It has been established that regulation of chromatin structure through post-translational modification of histone proteins, primarily histone H3 phosphorylation and acetylation, is an important early step in the induction of synaptic plasticity and formation of long-term memory. In this study, we investigated the contribution of another histone modification, histone methylation, to memory formation in the adult hippocampus. We found that trimethylation of histone H3 at lysine 4 (H3K4), an active mark for transcription, is upregulated in hippocampus 1 h following contextual fear conditioning. In addition, we found that dimethylation of histone H3 at lysine 9 (H3K9), a molecular mark associated with transcriptional silencing, is increased 1 h after fear conditioning and decreased 24 h after context exposure alone and contextual fear conditioning. Trimethylated H3K4 levels returned to baseline levels at 24 h. We also found that mice deficient in the H3K4-specific histone methyltransferase, Mll, displayed deficits in contextual fear conditioning relative to wild-type animals. This suggests that histone methylation is required for proper long-term consolidation of contextual fear memories. Interestingly, inhibition of histone deacetylases (HDACs) with sodium butyrate (NaB) resulted in increased H3K4 trimethylation and decreased H3K9 dimethylation in hippocampus following contextual fear conditioning. Correspondingly, we found that fear learning triggered increases in H3K4 trimethylation at specific gene promoter regions (Zif268 and bdnf) with altered DNA methylation and MeCP2 DNA binding. Zif268 DNA methylation levels returned to baseline at 24 h. Together, these data demonstrate that histone methylation is actively regulated in the hippocampus and facilitates long-term memory formation.

Figure 1.

Contextual fear conditioning alters histone H3 methylation in hippocampus. A, Outline of experimental design used with results displayed in B–D. B, Twenty-four hours after fear conditioning training, animals were exposed to the context and freezing behavior was measured. Animals exposed to the novel training chamber followed by three consecutive footshocks (context plus shock) displayed increased freezing when compared with animals exposure to the context alone (Student's two-tailed t test, t₍₁₁₎ = 6.794, p < 0.0001, n = 6–7/group). ***p value significant compared with context alone. C, Animals were fear conditioned and area CA1 of hippocampus was isolated at 1 h after training. A significant increase in histone H3 trimethylation at lysine 4 (H3K4me3) was observed after fear conditioning (C+S) when compared with animals that were handled without exposure to naive (N) or context-exposed (C) animal cohorts (one-way ANOVA with Tukey's multiple-comparison test, F_(2,28) = 5.560, p = 0.0098, n = 9–10/group). *p < 0.05 compared with naive-treated animals. D, There was a significant increase in regulation of histone H3 dimethylation at lysine 9 (H3K9me2) 1 h after either fear conditioning or context exposure alone when compared with naive-treated animal cohorts (one-way ANOVA with Tukey's multiple-comparison test, F_(2,35) = 5.978, p = 0.0061, n = 9–16/group). *p < 0.05 compared with naive-treated animals. Solid line represents normalized naive control levels. Error bars are SEM.

Figure 2.

Histone H3 dimethylation persists in hippocampus after contextual fear conditioning. A, Animals were fear conditioned and area CA1 of hippocampus was isolated at 24 h after training. Results are displayed in B and C. B, Twenty-four hours after fear conditioning histone H3 trimethylation at lysine 4 (H3K4me3) levels returned to baseline in area CA1 in hippocampus (one-way ANOVA with Tukey's multiple-comparison test, F_(2,12) = 0.06002, p = 0.9421, n = 4–5/group). C, Histone H3 dimethylation (H3K9me2) decreased 24 h following both fear conditioning (C+S) and context exposure alone (C) relative to naive control (N) (one-way ANOVA with Tukey's multiple-comparison test, F_(2,12) = 8.917, p = 0.006, n = 4–5/group). **p < 0.01 compared with naive-treated animals. Solid line represents normalized naive control levels. Error bars are SEM.

Figure 3.

Regulation of histone H3 trimethylation in hippocampus after contextual fear conditioning requires proper timing. A, Diagram of experimental design displaying the timeline for when animals were exposed to the contextual fear conditioning or the latent inhibition training paradigm with results displayed in B and C. B, Animals exposed to the latent inhibition (latent inhibition plus shock; LI+S) training paradigm displayed significantly less freezing behavior when compared with animals exposed to the fear conditioning learning paradigm (context plus shock; C+S) (one-way ANOVA with Tukey's multiple-comparison test, F_(2,26) = 8.377, p = 0.0017, n = 8–10/group). *p < 0.05 compared with naive-treated animals, ^##p < 0.01 compared with C+S. C, Histone H3 trimethylation levels in area CA1 of hippocampus was significantly greater 1 h after fear conditioning compared with latent inhibition training. Solid line represents normalized naive control levels. D, Deficits in behavioral memory formation with genetic knock-out of histone methylation modifying enzymes. The figure depicts the level of freezing for four genotypes 24 h following fear conditioning training. A 24 h test after fear conditioning revealed significantly decreased levels of freezing in eed^+/+;Mll^+/− and eed^+/−;Mll^+/− mice compared with wild-type littermates (F_(3,36) = 32.41, p < 0.0001, n = 8–10/group, all comparisons to wild-type littermates p < 0.015). Error bars are SEM.

Figure 4.

Inhibition of HDAC activity alters histone H3 trimethylation after contextual fear conditioning. A, Experimental design is outlined with results displayed in B to D. B, HDAC inhibition with NaB treatment 1 h before fear conditioning (NaB + CS) increased freezing levels compared with vehicle-treated cohorts (Student's two-tailed t test, t₍₁₂₎ = 2.524, p = 0.0267, n = 7/group). *p value significant compared with vehicle-treated context plus shock alone. C, Histone H3 trimethylation (H3K4me3) levels slightly increased with NaB treatment when compared with vehicle-treated naive animal cohorts (one-way ANOVA with Tukey's multiple-comparison test, F_(2,11) = 13.64, p = 0.0019, n = 4/group). **p < 0.01 compared with vehicle-treated naive controls, ns = not significant compared with (context plus shock; CS). D, Dimethylated histone H3 (H3K9me2) levels significantly decreased with NaB treatment with fear conditioning (NaB + CS) compared with vehicle-treated fear-conditioned animals (context plus shock; CS) or vehicle-treated naive animal cohorts (one-way ANOVA with Tukey's multiple-comparison test, F_(2,11) = 7.199, p = 0.0136, n = 4/group). *p < 0.05 compared with naive controls, ^#p < 0.05 compared with context plus shock. Solid line represents normalized naive control levels. Error bars are SEM.

Figure 5.

Contextual fear conditioning increases histone H3 trimethylation around the Zif268 and bdnf gene promoter. A, Real-time PCR graph of histone H3 trimethylation (H3K4me3) ChIP samples. B, Histone H3 trimethylation (H3K4me3) levels in area CA1 of hippocampus increase at the Zif268 promoter 30 min after fear conditioning (context plus shock; CS) relative to context exposure (C) or naive (N) animals. (Student's two-tailed t test, t₍₆₎ = 2.638, p = 0.0386, n = 4/group). *p < 0.05 compared with naive controls, ^#p < 0.05 compared with context exposure alone. C, Fear-conditioned animals demonstrated a significant increase in histone H3 trimethylation (H3K4me3) levels at bdnf promoter 1 (1-sample t test, t₍₄₎ = 5.714, p = 0.0293, n = 3–4/group). *p < 0.05 compared with naive controls, ^#p < 0.05 compared with context exposure alone. At the time point assessed, there were no significant changes in histone H3 trimethylation (H3K4me3) levels at bdnf promoter 4 after fear conditioning. Solid line represents normalized naive control levels. Error bars are SEM.

Figure 6.

Increases in histone H3 trimethylation correlate with altered DNA methylation (Zif268 promoter) after fear conditioning. A, Top, Methylation sites within CpG island 1 region of the Zif268 promoter. Sequencing primer pair positions are indicated by the left and right arrows. Methylation levels at the Zif268 promoter increased in area CA1 of hippocampus 30 min following fear conditioning (context plus shock) relative to context exposure (context) or naive animals (2-way ANOVA with Bonferroni's post hoc test; significant effect of behavior condition (F = 18.51, p < 0.0001, n = 4–8/group). B, Average percentage of DNA methylation across the Zif268 promoter (CpG island 1) in area CA1 of hippocampus after fear conditioning (context plus shock) compared with context exposure (context) or naive animals. ***p < 0.001 compared with naive controls. C, Although surprising, after training Zif268 gene expression remained significantly elevated with increased DNA methylation (context plus shock) (1-sample t test, t₍₅₎ = 3.238, p = 0.023, n = 3–6/group). *p < 0.05 compared with naive controls. Error bars are SEM.

Figure 7.

Increases in DNA methylation at the Zif268 promoter after fear conditioning is transient 24 h after fear conditioning. A, Top, Zif268 sequencing primer pair positions indicated by the left and right arrows. Methylation levels at the Zif268 promoter increased in area CA1 of hippocampus 24 h following context exposure (context) relative to fear conditioning (context plus shock) or naive animals (2-way ANOVA with Bonferroni's post hoc test; significant effect of behavior condition (F = 3.84, p < 0.05, n = 4–5/group). B, Average percentage of total DNA methylation across the Zif268 promoter (Fig. 6, CpG island 1) in area CA1 of hippocampus after fear conditioning (context plus shock) or context exposure (context) compared with naive animals. *p < 0.05 compared with naive controls. C, Interestingly, after context exposure Zif268 gene expression significantly decreased with increased DNA methylation (context) (1-sample t test, t₍₃₎ =, p = 0.006, n = 4–5/group). *p < 0.05 compared with naive controls. Error bars are SEM.

Figure 8.

Contextual fear conditioning triggers altered MeCP2 levels around the Zif268 gene promoter. A, MeCP2 levels in area CA1 of hippocampus decrease at the Zif268 promoter region that corresponds with increased histone H3 trimethylation (ChIP 1) 30 min after fear conditioning (context plus shock) relative to context exposure (context) or naive animals (1-sample t test, t₍₃₎ = 7.402, p = 0.0051, n = 3–4/group). *p < 0.05 compared with naive controls. B, Fear-conditioned animals demonstrated a significant increase in MeCP2 levels at the Zif268 promoter region that corresponds with increased DNA methylation (ChIP 2) 30 min after fear conditioning (context plus shock) relative to naive animals (1-sample t test, t₍₃₎ = 5.412, p = 0.0124, n = 3–4/group). *p < 0.05 compared with naive controls. At the time point assessed, there were no significant changes in MeCP2 levels at Zif268 promoter after context exposure (context) relative to naive animals. Error bars are SEM. C, Schematic: Proposed mechanism on how the chromatin microenvironment at the Zif268 promoter regulates its expression; following contextual fear conditioning, an enhancement in histone H3 lysine 4 trimethylation is observed upstream of the transcription start site (TSS) coinciding with a reduction in MeCP2 DNA binding (ChIP 1). At CpG island 1 (Fig. 6), increased DNA methylation is associated with increased MeCP2 binding (ChIP 2). We postulate that MeCP2 associated with CpG island 1 interacts with CREB bound to a cAMP response element (CRE) site situated near the TSS. This interaction allows increased transcription of the Zif268 gene during long-term memory formation. MeCP2, Methyl CpG binding protein; me, methyl groups; CREB, cAMP response element binding protein; ds-DNA, double-stranded DNA.