Chd1 is required for the maintenance of high levels of H2B monoubiquitination but not H3K4 or H3K79 trimethylation. Cell extracts from the nonessential yeast gene deletion mutants were subjected to SDS-PAGE and analyzed by Western blotting using H2B monoubiquitination-specific antibodies. Lane H of plate #148 of the collection is shown in A. The red arrow indicates the position in the plate containing the chd1Δ strain. The green arrow points to the empty well in the collection, used as the plate marker. (B) Cell extracts from wild-type (WT) and chd1Δ and bre1Δ strains were prepared and used in a titration analysis. Each extract was analyzed by its application to SDS-PAGE and tested for the presence of modified H3K4 and H3K79 mono-, di-, and trimethylation (H3K4me1–3 and H3K79me1–3) and H2B monoubiquitination (ubH2B) or total H4 using the appropriate antibodies. (C) A direct role for Chd1 in the regulation of histone H2B monoubiquitination was demonstrated by complementation of the chd1Δ strains with a vector carrying the CHD1 coding sequence under the GAL1 promoter. In the presence of galactose, Chd1 is expressed (shown in lanes 7,8), as evident by Western signal with anti-Flag, and yeast cells are restored for monoubiquitinated H2B. (D) CHD1 was also deleted in a parental background that contains a sole copy of Flag-tagged histone H2B (to detect both wild-type and monoubiquitinated H2B). Extracts prepared from wild-type and chd1Δ strains were analyzed by SDS-PAGE and Western with the indicated antibodies.

Genome-wide occupancy of histone H3K4 and H3K79 methylation and gene expression pattern as a result of the loss of CHD1. Global occupancy for H3K4 and H3K79 trimethylation in the presence and absence of Chd1 were determined using ChIP-seq in yeast cells. ChIPs with antibodies specific to trimethylated H3K4 and H3K79 were performed and the resulting DNA sequenced. (A) A track file for chromosome I of yeast. (B,C) Scatter plots depict the wild-type versus chd1Δ occupancy enrichment for the immunoprecipitation indicated in the top left. Normalized read sums of the immunoprecipitation divided by the normalized read sums of the whole-cell extract control within a region ±500 bp of each gene start site is plotted for the wild-type and mutant strain. The correlation coefficient shown in the bottom right is the Pearson correlation computed about the log2 transformed occupancy enrichment. The dotted lines indicate positions of a twofold change in occupancy. Scatter plot point density is indicated by color. (Blue) Low; (yellow) high. (D) RNA-seq track diagram showing complete loss of CHD1 transcript in the deletion strain. (E) MA plot depicting genes differentially expressed at a false discovery rate (FDR) <0.05 as a result of loss of CHD1. (F,G) RNA-seq track diagrams of two genes, HOM3 (F) and ARN2 (G), that are down-regulated in the absence of CHD1.

Nucleosomal occupancy and gene expression as a result of the loss of Chd1. (A) Heat maps of nucleosomal occupancies in wild-type (WT) and chd1Δ strains. Each row represents a genic nucleosomal array; arrays are sorted by array length and aligned by array midpoint (illustrated above the plot). Blue areas are the 5′ and 3′ nucleosome-free regions (NFRs). Color bar represents the nucleosomal occupancy from zero (dark blue) to genomic average (green) to twice the genomic average (dark red). Occupancy was measured by extending the dyad position of each nucleosomal tag 73 bp in both directions, then summing all tag coverage at each base pair. A difference map is shown in the right panel. Details of plot generation can be found in Batta et al. (2011). (B,C) Composite distribution of nucleosomal tags in wild type (filled gray), H2BK123A, and the chd1Δ mutant, aligned by the wild-type consensus position of the first nucleosome dyad for the indicated subsets of genes. Individual tag positions were defined by a single coordinate representing the nucleosome dyad. The subset of genes that are either up-regulated or down-regulated in the K123A or chd1Δ mutant is shown.

A conserved role of Chd1 in the maintenance of H2B monoubiquitination levels from yeast to humans. Human 293 cells were transfected with control, hChd1, and hBre1A shRNA as indicated. Following 96 h of incubation, the RNAi efficiency for Chd1 and Bre1 and their effect on the patterns of H2B monoubiquitination and H3K4 and H3K79 methylation were tested by Western analyses using the appropriate antibodies. Histone H3 and Tubulin were used as load controls.