Genome-wide dynamics of Htz1, a histone H2A variant that poises repressed/basal promoters for activation through histone loss

Abstract

Histone variants help specialize chromatin regions; however, their impact on transcriptional regulation is largely unknown. Here, we determined the genome-wide localization and dynamics of Htz1, the yeast histone H2A variant. Htz1 localizes to hundreds of repressed/basal Pol II promoters and prefers TATA-less promoters. Specific Htz1 deposition requires the SWR1 complex, which largely colocalizes with Htz1. Htz1 occupancy correlates with particular histone modifications, and Htz1 deposition is partially reliant on Gcn5 (a histone acetyltransferase) and Bdf1, an SWR1 complex member that binds acetylated histones. Changes in growth conditions cause a striking redistribution of Htz1 from activated to repressed/basal promoters. Furthermore, Htz1 promotes full gene activation but does not generally impact repression. Importantly, Htz1 releases from purified chromatin in vitro under conditions where H2A and H3 remain associated. We suggest that Htz1-bearing nucleosomes are deposited at repressed/basal promoters but facilitate activation through their susceptibility to loss, thereby helping to expose promoter DNA.

Figure 1

Genome-Wide Localization of Htz1 and SWR1 (A) Htz1 occupancy is reproducible. Htz1 ChIP enrichment at IGRs from Replicate 1 (sorted by log₂ ratio, x axis) versus Htz1 ChIP enrichment in Replicates 2 and 3, plotted as a moving average (window size 80, step 1) of the log₂ ratios (y axis). ORF replicates yielded similar results (not shown). Strain: YBC1867. (B) Htz1 occupancy is specific. Plotted as in (A), versus the untagged control (YBC1895). (C and D) Htz1 ChIP is efficient. Distribution of the log₂ median ratios of Htz1 ChIP enrichment at IGRs (C) or ORFs (D). (E) Htz1 occupancy is correlated with Swr1 occupancy. Htz1 ChIP enrichment at IGRs (sorted by percentile rank, x axis) versus Swr1 ChIP enrichment (YBC2170, or the untagged control strain YBC1895), plotted as the moving average (window size 40, step 1) of the percentile ranks (y axis). (F) Venn diagrams depicting the overlap of IGRs with high levels of Htz1 and Swr1. The full dataset consisted of IGRs available in all three (Htz1, Swr1, and control) individual datasets (2457 IGRs total), from which we compared the top 10% of each. (G) Swr1 is required for specific Htz1 deposition genome-wide. Htz1 ChIP enrichment in wt cells (sorted by percentile rank, x axis) versus Htz1 occupancy in swr1Δ mutants (YBC2162), plotted as the moving average (window size 40, step 1) of the percentile ranks (y axis).

Figure 2

Htz1 Prefers Promoters (A and B) Htz1 strongly prefers promoters genome-wide. (C and D) H2A weakly prefers nonpromoters. IGRs were assigned to one of three promoter-type classes (see text). Htz1 or H2A occupancy in tagged (A and C) or untagged control strains (B and D). Htz1 enrichment (log₂ median ratio, x axis) versus the percent of IGRs in each promoter class (y axis). Strains for H2A ChIP: YBC2200 and YBC1894. (E) Htz1 prefers promoters at individual genes. ChIP enrichment was determined by qPCR (note: RRP43 and RBK1 are divergent). Values are the average of three independent ChIPs with qPCR determination performed twice. Error bars: SD. The primer sets used for each amplicon are listed in Table S2; format, (GeneID: promoter, ORF): YOR285W: D, F; YDL218W: I, K; YNL092W: O, Q; PRP12: U, V; YDC1: Y, AA; NUP159: AC, AE; MRK1: AF, AG; YNL116W: AH, AI; RIM11: AJ, AK; RRP43: AM, AL; RBK1: AM, AN. Values are normalized to an amplicon within iYMR325W, using primer set A.

Figure 3

Htz1 Occupancy Is Correlated with Bdf1 and Shows a Preference for TATA-less Promoters (A) Bdf1 prefers TATA-less promoters. (B) Spt3 weakly prefers TATA-containing promoters. (C) Occupancy correlation between Htz1 and Bdf1. Bdf1 occupancy at IGRs (sorted by log₂ ratio, x axis) versus Htz1 occupancy, plotted as the moving average (window size 40, step 1) of the log₂ ratio (y axis). (D) Occupancy correlation between Htz1 and Spt3. As in (C), IGRs were sorted by log₂ ratio of Spt3 ChIP (x axis). (E) Htz1 prefers TATA-less promoters. Plots for (A), (B) and (E): factor occupancy (single promoter class, sorted by percentile rank, x axis) compared to the fraction of promoters containing a TATA element (in a sliding window of 80 genes, as a percent of total, y axis). Bdf1 occupancy data is from Kurdistani et al. (2004), and Spt3 occupancy data is from Zanton and Pugh (2004).

Figure 4

Strains Lacking Gcn5 or Bdf1 Display Reductions in Htz1 Occupancy (A-D) Htz1 occupancy at IGRs in wt cells (sorted by percentile rank, x axis) was compared to Htz1 occupancy in mutant strains. Changes in Htz1 occupancy (ΔMPR; MPR in mutant - MPR in wt) in mutants were plotted as a moving average of 80 genes (window size 80, step 1, y axis). (E) Relative abundance of Htz1 at 10 promoters in wt (YBC1894), gcn5Δ (YBC1662), sas3Δ (YBC1911), bdf1Δ (YBC2512), and bdf2Δ (YBC2513). ChIPs utilized polyclonal αHtz1 antibody, and values are the average of three independent ChIPs quantified by qPCR. Error bars: SD. Primer sets as in Figure 2 (Z, for YDC1).

Figure 5

Htz1 Is Negatively Correlated with Transcription and Redistributes as Transcription Changes (A and B) Htz1 is negatively correlated with transcription and to a greater extent than H2A. Genes were sorted by transcription rate (x axis), and compared to Htz1 occupancy (A) or H2A occupancy (B), plotted as the moving average (window size 40, step 1) of ChIP enrichment (log₂ ratio, y axis). (C) During HS, Htz1 abandons activated promoters and occupies repressed promoters. Changes in gene expression resulting from HS (30 min) were quantified by microarray analysis and sorted according to their magnitude (log₂ ratios, x axis). Values are the average of three biological replicates. Htz1 occupancy during HS (30 min) or following recovery from HS (30 min after their return to 25°C). Plots depict the moving average (window size 40, step 1) of the change in Htz1 ChIP enrichment relative to time zero (no HS), either during HS or following the recovery from HS (y axis: log₂ [MPR ratio (HS or recovery/no HS)]). (D and E) Plot parameters as in (C). (D) H2A occupancy decreases slightly at activated promoters in response to HS. (E) Htz1 dynamics during diauxic shift. Conditions: 8hr growth to a final OD₆₀₀ of 6.2. Values are the average of two biological replicates. (F) Htz1 and TBP occupancy are negatively correlated. TBP occupancy (promoter IGRs, Pol III targets omitted) was sorted (by percentile rank, x axis) and plotted against a moving average (window size 80, step 1) of Htz1 occupancy (percentile rank, y axis).

Figure 6

Htz1 Is Susceptible to Loss, Which Both Accompanies and Facilitates Activation (A) During activation of YDC1, Htz1 is lost to a greater extent than H2A or H3. Histone ChIP enrichment at YDC1 promoter was determined by qPCR (primer set Y). Values are the average of three independent ChIPs with qPCR determination performed twice. Error bars: SD. Strains: YBC2128 (Htz1 and H3 ChIPs) and YBC2228 (H2A ChIPs). (B) Deletion of HTZ1 attenuates activation but not repression in response to HS. The change in Htz1 occupancy in wt upon HS (30 min; sorted by log₂ [MPR ratio (HS/no HS)], x axis) at each promoter was compared to the change in the expression of their linked ORFs in wt and htz1Δ strains, plotted as the moving average (window size 40, step 1) of expression changes upon HS (y axis). Boxes denote regions of the graph discussed in the text. (C) HTZ1 is required for full activation of YDC1 in response to HS. Changes in YDC1 expression during HS time course in wt, and htz1Δ strains were quantified by microarray analysis, and values are the average of three biological replicates. Error bars: SD. (D) Htz1 is more susceptible to loss from yeast chromatin than is H2A or H3. Chromatin was prepared from a strain (YBC2228) bearing HA-Htz1 and H2A-TAP alleles. Equal portions were treated with increasing levels of sodium chloride (indicated), supernatants were removed, and the resulting chromatin pellet was treated with MNase to generate mononucleosomes. Histones were separated by SDS-PAGE and immunoblotted (anti-HA, anti-Protein A, and polyclonal αH3 antibodies were used, respectively), or stained with Coomassie blue dye (bottom panel). Asterisk denotes full-length H3 and the lower band a common proteolytic product.

Figure 7

A Model for Htz1 in Transcriptional Regulation An Htz1-containing nucleosome (green disks, central nucleosome) occupies the promoters of certain repressed/basal genes, with a preference for TATA-less promoters. The deposition process requires SWR1 complex and is facilitated/targeted by Bdf1 and the acetyltransferase Gcn5. Acetylation of H3K14 and other residues by Gcn5 (and other HATs) likely underlies the observed correlations between histone acetylation and Htz1 occupancy. Although present during repression, Htz1 does not have a specialized function that promotes repression. Rather, biochemical experiments and occupancy dynamics establish the Htz1 nucleosome as susceptible to ejection (“fragile”), suggesting that the presence of Htz1 poises the repressed and basal states for full activation. Transition to the active state typically involves the action of chromatin remodeling factors (black oval) and the binding of activators to the enhancer (Enh). These factors likely collaborate to eject the Htz1 nucleosome, which facilitates activation by exposing promoter DNA to TFIID and other transcription factors.