Supplementary Figure 2: Disruption or conversion of Trr-dependent H3K4 methylation at enhancers in Drosophila.

(a) Western blots from lysates of adult fly brains confirm H3K4me1 bulk reduction in trr-C/A, and H3K4me2/3 bulk increase in trr-Y/F, similar to that shown in wing discs. (b) ChIP–seq data from adult fly brains plotted as average profiles, centered on non-TSS H3K27ac peaks; n = 1,440.

(c) Box plots of Z-score averages comparing H3K4 methylation and H3K27ac at 5,367 active promoters in adult brain tissue. Plots are centered on H3K27ac peaks overlapping annotated TSSs, ±2.5 kb. P values are shown below. In comparison to the effects at enhancers, H3K4 methylation levels often show the opposite effect at TSSs. This is most likely an artifact due to redistribution of sequencing reads in the trr mutants.

(d) Average meta-peak profiles for both Trr and Lpt at 6,807 genomic loci demonstrate that neither of the catalytic mutations substantially affects chromatin binding. (e) ChIP–seq track examples from wing imaginal discs demonstrate reduced H3K4me1 and H3K27ac in trr-C/A, as well as increased H3K4me3 and H3K27ac in trr-Y/F at putative enhancers, similar to adult brains. (f) Density bar plots comparing H3K4 methylation and H3K27ac Z scores at 787 active enhancers in wing imaginal discs. Plots are centered on wild-type H3K27ac (non-TSS) peaks, ±2.5 kb. Note the reductions in H3K4me1 in trr-C/A and conversion of H3K4me1 to H3K4me3 in hyperactive trr-Y/F. The Z-score scale for H3K4me1 and H3K27ac is 0 to 1.5 and for H3K4me3 is 0 to 1. Note the changes in H3K27ac that track with increased H3K4 methylation. (g) Box plots of ChIP Z-score averages, centered on wild-type H3K27ac peaks, were calculated the same as in b. Quantitative differences in H3K4me1/3 and H3K27ac are shown in wing imaginal disc enhancers (n = 787), similar to adult brain tissues. Changes in histone modification at promoters (TSSs) are negligible (N = 4077).