Htz1 is preferentially enriched at inactive promoters. (A) Histogram of Htz1 enrichment (YBL325). The frequency of Log2 ratio from Htz1 data set (same as Fig. 1) fallen into 0.02 intervals (bin size) were calculated and plotted. (B) Same as A, except that Htz1 enrichment was normalized to global nucleosomal occupancy measured by histone H3 enrichment. (C) For each ORF, the closest 5′ intergenic region was selected to represent promoter regions; thus, the transcription rate of such IGR was assigned to be the same as the downstream (3′) gene. Transcription rates used here originated from a previously published microarray analysis (34). The data set compiled with the transcription rate of each ORF and corresponding Htz1 enrichment at the ORF and IGR is first sorted by transcription rate. Next, moving averages (window size = 80, step = 1) of the Htz1 enrichment at both the ORF and IGR are calculated and plotted as a function of transcription rate. (D) K-mean cluster analysis of genome-wide distribution of Htz1 and pol II (8WG16). (E) K-mean cluster analysis of pol II and Htz1 occupancy at a group of genes whose transcription are different when grown in rich media (YPD) and synthetic minimal media (SMM) (absolute difference of pol II enrichment is >0.9 and is consistent with mRNA microarray analysis).

The influence of incorporation of Htz1 into nucleosomal arrays on chromatin structure. (A) Deletion of HTZ1 does not disrupt the nucleosome positioning at four examined genomic loci. (B) Htz1-containing nucleosomes were refractory to histone modifications. (Left) A schematic of the steps for histone modification assays using immobilized yeast recombinant nucleosomal arrays. Bacterially expressed rDot1, rSet2 and Set2-TAP and Dot1-TAP were used for histone methyltransferase assays (HMT). The number of nucleosomes used in each assay was normalized based on silver staining after reconstitution (data not shown). SAGA (Spt7-TAP) and NuA4 (Epl1-TAP) were used in histone acetyltransferase (HAT) assays. The radioactivity counts on canonical nucleosomes assembled without the spacing factors (Isw1-TAP) (which results in randomly spaced nucleosomes) after each reaction were arbitrarily denoted as 1. Error bars represent the average deviations. (C) Supercoiling assay. Recombinant Htz1 was assembled into nucleosomal arrays in a manner similar to that of canonical histones. (D) Htz1 was enriched in the nucleosomal population that did not included K79me2 and K36me2 modifications. Nuclei were prepared from YBL325. Extracted chromatin was digested by micrococcal nuclease, and the soluble portion was subsequently loaded on a gel-filtration column. Each fraction was examined by Western blotting with indicated antibodies. Purified DNA was separated on a 2% agarose gel to monitor the size of nucleosomes.

SWR1 controls global deposition of Htz1. The flow chart of quantitative chromatin binding (QCB) assays is illustrated in Upper. Yeast strains YBL556 (WT), YBL557(Δswr1), YBL560 (Δswc2), and YBL561 (Δswc6) were subjected to QCB assay by using rabbit IgG (Sigma). The relative IP efficiency (the ratio of Htz1 IP over input) in WT was assigned as 1, and IP efficiency in mutants was normalized accordingly. Mock IP is performed by using an untagged yeast strain (BY4741).

Htz1 enrichment and transcription dependency of Htz1. Moving average of Htz1 enrichment in WT was plotted as a function of mRNA changes in Δhtz1 (A) and Δhtz1Δsir2 (B) (6).

H2AZ interferes with chromatin remodeling in vitro. (A) Htz1 containing yeast recombinant nucleosomes reduce the sliding activity of chromatin remodeling complexes. A 216bp DNA fragment that was PCR amplified from pGUBdSH with ³²P end-labeled primer set was reconstituted into mononucleosomes with yeast recombinant histone octamers that included either H2A [yeast core histone (YCH)] or Htz1 (yHTZ). The resulting nucleosomes were directly applied to nucleosome sliding assay by using four TAP-purified chromatin-remodeling complexes Chd1, Isw1, Swi/Snf, and RSC in the presence of ATP or ATP-γ-S. Native PAGE (5% acrylamide, 37.5:1) was used to resolve the mixed population of nucleosomes after remodeling. Chd1-TAP and Isw1-TAP mobilize the nucleosomes toward the center position, which migrates slower, as indicated on the Left, whereas Swi/Snf and RSC move them to the end position which migrates faster, as shown on the Right. (B) Time course analysis of sliding assay (Chd1). The sliding assay was carried out as described in A. At each time point, the remodeling reaction was stopped by addition of excess competitor DNA and nucleosomes. (C) Quantification of sliding kinetics. Each gel band was quantified by using the software imagequant tl. Sliding efficiency was defined as the intensity of the substrate over total population. Error bars represented the SD from three independent experiments.