Chd1-Ume6 can disrupt Isw2-directed nucleosome positions and result in cryptic transcription. (A) Difference in nucleosome dyad signal between wild-type cells with and without addition of Chd1-Ume6, clustered (as in Figs. 2, 3) by direction of nucleosome repositioning (left) with average cluster signal from individual strains (right). (B) Average signals within cluster 3 in Figure 5A for indicated backgrounds showing cooperation of endogenous remodeling machinery with Chd1-Ume6 based on increased movement of distal nucleosomes to the +50-bp position. Red- and blue-shaded regions correspond to the range of nucleosomes mobilized by Chd1-Ume6 and endogenous Isw2/Ume6, respectively. The difference between the red and orange traces is due to sequential positioning by Isw2 and Chd1-Ume6 at Ume6 binding sites. (C) Example locus (upstream of RPC17) where initial positioning by Isw2/Ume6 is required for the final Chd1-Ume6 remodeled position observed in a wild-type background. In the absence of Isw2/Ume6 (red), nucleosomes are beyond the reach of Chd1-Ume6, while Isw2 action (blue) moves nucleosomes to a favorable position for further Chd1-Ume6 positioning in the wild-type background (orange). Black asterisk denotes the preferred Δume6 dyad position. (D) Changes in global (left; black dots) or motif-proximal (right; red dots) transcript abundance between the wild-type and wild-type +[Chd1-Ume6] strains. (E) Browser shot of the MEI5/VPS30 locus showing changes in nucleosome positions (top) and transcript abundance (bottom) upon addition of Chd1-Ume6 to a wild-type background. Blue and red signals reflect Watson (right-transcribed) and Crick (left-transcribed) strands, respectively, from normalized strand-specific RNA-seq profiles. (F) Same as E, showing differences in nucleosome dyad signal and strand-specific transcript abundance at the MEI4/ACA1 locus.
