Skip to main content
. 2018 May 29;7:e35368. doi: 10.7554/eLife.35368

Figure 3. Long-lived PRC2 mutants diminish the epigenetic drift of H3K27me3 during aging.

(A) Circos plot of the H3K27me3 epigenome illustrates peak profiles that are highly preserved with age and in PRC2 mutants. Black boxes and lines (innermost circle) represented common peak regions, corresponding to their chromosomal locations. Chromosome ideogram was in grey (outermost ring). PRC2 target genes previously found in cells and during development were shown next to their epigenomic loci. ChIP-seq was from muscle tissues of 3d- and 30d-old male flies. Genotypes: WT: 5905 and Pclc421/+; Su(z)12 c253/+. (B) H3K27me3 modification decreases in PRC2 mutants. Scatter plot showed H3K27me3 levels for gene bodies of all protein-coding genes in Pclc421; Su(z)12c253 as compared to WT. (wilcoxon signed-rank test, p<2.2e-16). Each dot on the plot represents a single gene locus. X- and Y-axis represented log2 mean value of genes’ reference-adjusted reads. Contour lines indicated that H3K27me3 signals were generally higher in WT compared to mutants. ChIP-seq was from muscle tissues of 3d-old male flies. Genotypes as in (A). (C) Inter-peak genes receive relatively less H3K27me3 modification in PRC2 mutants. In PRC2 mutant, inter-peak genes received relatively less signal (25.1%) than those of peak genes (13.7%). Reduction of signals was calculated by a subtraction of the mean signal intensity between WT and PRC2 mutants. Bootstrapped 95% confidence intervals: 3d inter-peak genes: [1.317, 1.861]; 3d Pclc421; Su(z)12c253 inter-peak genes: [0.944, 1.361]; 3d peak genes: [3.946, 4.414]; 3d Pcl421; Su(z)12253 peak genes: [3.364, 3.861]. ChIP-seq and genotypes as in (A). (D) Modification of H3K27me3 during aging has reduced selectivity in PRC2 mutants. Bootstrapped 95% confidence intervals: Pcl421; Su(z)12253 late-onset inter-peak genes: [0.053, 0.147]; Pcl421; Su(z)12253 late-onset peak genes: [−0.778,–0.578]. ChIP-seq was from muscle tissues of 3d and 30d old male flies. Genotypes: Pclc421/+; Su(z)12 c253/+. (E) Age-associated drifting of H3K27me3 is dampened in PRC2 mutants. Scatter plot showed gained H3K27me3 signal for all protein-coding genes in aged Pclc421; Su(z)12c253 and WT flies compared to young flies. Each dot on the plot represents a single gene locus. X- and Y-axis represented signal intensity transformed by Log2. Contour lines indicated that the majority of the gene signals displayed higher levels—thus more rapid changes—in aged WT compared to mutants. ChIP-seq and genotypes as in (A).

Figure 3.

Figure 3—figure supplement 1. Long-lived PRC2 mutants diminish the epigenetic drift of H3K27me3 during aging.

Figure 3—figure supplement 1.

(A) Relative quantification of different histone marks between WT and PRC2 mutants. Only H3K27me2/3 were selectively reduced in PRC2 mutants. Western blot was from head tissues of 3d-old male flies. (for H3K27me3 quantification: mean ±SD of three biological repeats; student t-test, *p<0.05, **p<0.01, ***p<0.001). Genotypes: WT: 5905. Mut: Pclc421/+; Su(z)12 c253/+. (B) and (C) The modification of H3K27me3 is preserved in muscle tissue between WT and PRC2 mutant. Venn diagram shows the number of shared regions in H3K27me3 enriched regions between WT and PRC2 mutant. Peak region reproducibly identified by two replicates were used to generate the venn diagram. Method as in Figure 1—figure supplement 1D. ChIP-seq was from muscle tissues of 3d-old male flies. Genotypes: WT: 5905. Pclc421/+; Su(z)12 c253/+. escc289/+; E(z)c239/+. (D) H3K27me3 modification decreases in PRC2 mutants. Scatter plot showed H3K27me3 levels for all protein-coding genes in escc289; E(z)c239 as compared to WT. (wilcoxon signed-rank test, p<2.2e-16). Each dot on the plot represents a single gene locus. X- and Y-axis represented log2 mean value of genes’ reference-adjusted reads. Contour lines indicated that H3K27me3 signals were higher in WT compared to mutants. ChIP-seq was from muscle tissues of 3d-old male flies. Genotypes: WT: 5905. escc289/+; E(z)c239/+. (E) Inter-peak genes receive relatively less H3K27me3 modification in PRC2 mutants. In PRC2 mutant, inter-peak genes received relatively less signal (23.8%) than those of peak genes (19.4%). We used bootstrapping to generate the mean reference-adjusted ChIP intensity of peak genes or inter-peak genes (10,000 draws with replacement of n = 500). Bootstrapped 95% confidence intervals: WT inter-peak genes: [2.626, 3.601]; esc289; E(z)239 inter-peak genes: [2.029, 2.627]; WT peak genes: [9.161, 10.367]; esc289; E(z)239 inter-peak genes: [7.363, 8.355]. ChIP-seq and genotypes as in (D). (F) Modification of H3K27me3 during aging has reduced selectivity in PRC2 mutants. Bootstrapped 95% confidence intervals: esc289; E(z)239 late-onset inter-peak genes: [0.551, 0.804]; esc289; E(z)239 late-onset peak genes: [0.532, 0.845]. ChIP-seq was from muscle tissues of 3d and 30d old male flies. Genotypes as in (D). (G) Age-associated drifting of H3K27me3 is dampened in PRC2 mutants. Contour lines indicated that the majority of the gene signals displayed higher levels—thus more rapid changes—in aged WT compared to mutants. ChIP-seq and genotypes as in (C).