Article Figures & Data
Figures
- Figure 1.
Transcriptome-wide profiling reveals the dynamic landscape of m6A in the mouse mPFC following behavioral training. A, Distribution analysis of mRNA-associated m6A peaks identified in the mPFC shown as a percentage (pie chart) or as enrichment index after normalization to the relative length of each segment in the transcriptome (column graphs). B, Overall consensus motif associated with m6A peaks. C, Clustering and heat-map diagrams show the modulation of m6A peaks in the CS-only or fear-conditioned (FC) group relative to the naive group as the baseline control. Red and blue bars represent an increase and decrease, respectively (p values determined by Student's t test). C′, Heat-map diagram based on normalized count for 188 peaks identified to be differentially methylated in the FC versus naive comparison (p < 0.05). D, Representative MeRIP-seq traces of Rab33b showing consistent increase of m6A in a region near the stop codon in the FC mice (arrow). Cyan and red traces represent immunoprecipitated coverages, gray traces for the input coverages. E, Top 15 gene ontology categories associated with genes with modulated m6A peaks in the FC group (p < 0.05 or fold-change ≥1.5).
- Figure 2.
Activity-dependent regulation of locus-specific m6A in cultured neurons is associated with mRNA stability. A, MeRIP-qPCR analyses of KCl-depolarized neurons (50 mm, 30 min) showing modulation of locus-specific m6A (n = 3–10; Mann–Whitney U test; *p < 0.05 vs −KCl control). B, qPCR analysis of Fto in neurons transduced with FG12-FTO shRNA [knockdown (KD)], FG12 control, and FUW-FTO [overexpression (OX); n = 8–14; Mann–Whitney U test; ****p < 0.0001]. C, Representative plots showing mRNA decay profiles of Rab33b, Arhgef17, and Crtc1 in actinomycin-D pulse-chase experiments. D, mRNA decay time constant, τ for each transcript (n = 4–14, one-way ANOVA; *p < 0.05, **p < 0.01, ***p < 0.001). E, Decay plots for Egr1 and the calculated τ for Egr1 and Arc (n.s., not significant). Data represent mean ± SEM.
- Figure 3.
Targeted knockdown of FTO in the mPFC leads to enhanced fear retention. A, qPCR analyses of Egr1, Mettl3, Mettl14, and Fto in the mPFC following context-exposure (CS-only) or a fear conditioning (FC) paradigm (n = 6–8, one-way ANOVA with Dunnett's test relative to the naive controls, *p < 0.05, **p < 0.01). B, Cortical neurons were transfected with FG12-FTO shRNA, fixed and stained with anti-FTO antibody and DAPI. GFP expression indicates transfected neurons. Scale bar, 25 μm. C, Western blot (left) and m6A immunoblot (right) analyses of cortical neurons transduced with lentivirus particles expressing GFP control (Ctrl) or FTO shRNA. D, Representative image of mouse brain following lentiviral injections into the cingulate (Cgl) and prelimbic (PrL) cortical regions (placement for the control lentivirus indicated in gray and FTO shRNA in red). Scale bar, 200 μm. E, Immunohistochemical analysis of the mPFC region using anti-FTO antibody showing levels of FTO immunoreactivity in shRNA-expressing (GFP-positive) and GFP-negative control cells (n = 80 cells per group from 8 independent slices; Mann–Whitney U test; ****p < 0.0001). Scale bar, 20 μm. F, Representative m6A immunoblot analysis of RNA extracted from the mPFC tissues of mice injected with FG12 control or FTO shRNA lentivirus (n = 4; t test, p < 0.05). G, Cued fear conditioning revealed similar acquisition of fear between groups (control, n = 7; FTO shRNA, n = 8; two-way ANOVA; p = 0.1711). H, The freezing level measured in the same context 24 h after training was significantly higher in the knockdown group (one-way ANOVA with Tukey's post hoc analysis; in CS: control vs FTO shRNA; *p < 0.05). I, An open-field test revealed a comparable duration spent in the center area (left) and total distance traveled (right) by both groups (control, n = 14; FTO shRNA, n = 15). J, An elevated plus maze analysis revealed no differences in the percentage of time spent in (left) or the number of entries (right) to the open arms. Data represent mean ± SEM.
