Fig. 2

Metabolomic changes during fear memory consolidation. a Fear conditioning protocols and timing of tissue sampling for each group. b Freezing in the context (left) and tone (right) tests (DS, n = 8 mice; IS, n = 9 mice; two-tailed unpaired t-test, *p < 0.05, **p < 0.01; error bars, SEM). c Latency between detection of the target sleep–wake state and sampling. One-way ANOVA, p > 0.05, error bars, 95% CIs. d Sampling timing for REM (top), NREM (middle), and Wake (bottom) groups. ZT, zeitgeber time. ZT = 0 means the start of the light-cycle. One-way ANOVA, p > 0.05, error bars, 95% CIs. e Durations of each sleep–wake state between fear conditioning and sampling. Kruskal–Wallis test, p > 0.05, error bars, 95% CIs. PLS-DA plots for REM (f), NREM (g), and Wake (h) groups. Each point reflects one mouse, and ellipses represent 95% CIs. DS-REM, n = 12 mice; IS-REM, n = 7; CO-REM, n = 9; DS-NREM, n = 10; IS-NREM, n = 8; CO-NREM, n = 9; DS-Wake, n = 10; IS-Wake, n = 9; CO-Wake, n = 10. i–k Volcano plots of p-values of each metabolite from permutation analysis: DS vs. IS (left) and DS vs. CO (right) during REM sleep; DS vs. IS (up) and DS vs. CO (bottom) during NREM sleep; DS vs. IS (up) and DS vs. CO (bottom) during wakefulness. Red dashed line, p = 0.05. l–n Venn diagrams indicating numbers of significantly changed metabolites for each comparison (e.g., DS/IS means significantly changed, either positively or negatively, in DS vs. IS groups) during REM (left), NREM (middle), and Wake (right) groups. o Fold change of each metabolite (*|fold change|> 1.5 and p < 0.05, **|fold change|> 1.5 and p < 0.01), sorted from the highest fold change for DS vs. IS. Error bars, 95% CIs. Enrichment analysis for changed metabolites satisfying the criterion of \(({\text{DS}}/\mathrm{IS})\bigcap (\overline{\mathrm{IS}\bigcap \mathrm{CO}})\)) during REM (p) and NREM (q) sleep. No enriched pathways were found during wakefulness