Identification of histone mutants that are defective for transcription-coupled nucleosome occupancy

Abstract

Our previous studies of Saccharomyces cerevisiae described a gene repression mechanism where the transcription of intergenic noncoding DNA (ncDNA) (SRG1) assembles nucleosomes across the promoter of the adjacent SER3 gene that interfere with the binding of transcription factors. To investigate the role of histones in this mechanism, we screened a comprehensive library of histone H3 and H4 mutants for those that derepress SER3. We identified mutations altering eight histone residues (H3 residues V46, R49, V117, Q120, and K122 and H4 residues R36, I46, and S47) that strongly increase SER3 expression without reducing the transcription of the intergenic SRG1 ncDNA. We detected reduced nucleosome occupancy across SRG1 in these mutants to degrees that correlate well with the level of SER3 derepression. The histone chromatin immunoprecipitation experiments on several other genes suggest that the loss of nucleosomes in these mutants is specific to highly transcribed regions. Interestingly, two of these histone mutants, H3 R49A and H3 V46A, reduce Set2-dependent methylation of lysine 36 of histone H3 and allow transcription initiation from cryptic intragenic promoters. Taken together, our data identify a new class of histone mutants that is defective for transcription-dependent nucleosome occupancy.

Fig. 1.

Detection of SER3 derepression from an ectopically expressed SER3pr-lacZ reporter. (A) Diagram of SER3pr-lacZ reporter. The LYP1 ORF was replaced by SER3 5′ UTR sequence from −713 to −1, including SRG1 and its promoter, fused to the lacZ ORF. Block arrows beneath the diagram indicate the expected SRG1 and SER3-lacZ transcripts in wild-type and mutant strains grown in serine-rich medium (YPD). The table on the right indicates the expected results for an X-Gal overlay assay for wild-type and mutant strains. (B) X-Gal overlay detects SER3pr-lacZ derepression in snf2Δ (YJ924, YJ982, and YJ983), spt6-1004 (YJ977, YJ978, and YJ979), and spt16-197 (YJ974, YJ975, and YJ976) strains that is not seen in wild-type strains (YJ921, YJ980, and YJ981). Cells were grown on YPD medium and incubated with X-Gal for 32 min.

Fig. 2.

Single-amino-acid substitutions in histones H3 and H4 strongly derepress SER3. (A) Northern blot analysis examining the effect of histone mutants on SER3, SRG1, and SCR1 (loading control). Total RNA was isolated from a wild-type strain (FY4) and derivatives of JDY86 expressing either synthetic or wild-type copies of histone H3 and H4 (HHTS/HHFS) or mutants hhts-K122A, hhts-K122R, hhts-K122Q, hhts-Q120A, hhts-V117A, hhts-R49A, hhts-V46A, hhfs-I46A, hhfs-R36A, and hhfs-S47D that were grown to a density of 1 × 10⁷ to 2 × 10⁷ cells/ml in YPD at 30°C. (B) Quantitation of Northern analyses. SRG1 and SER3 RNA levels for the histone mutants are normalized to the SCR1 loading control and are relative to the SRG1 and SER3 RNA levels measured in control HHTS-HHFS strains (arbitrarily set to 1). Each bar represents the means ± standard errors of the means (SEM) from three independent experiments involving JDY86 derivatives (which are shown in panel A) and related strains generated by genetic crosses (YJ925 to YJ946). (C) Western analysis examining the effect of histone mutants on total histone H3 and histone H4 protein levels. Strains expressing the indicated histone alleles were grown to ∼3 × 10⁷ cells/ml in YPD at 30°C. Proteins were extracted with trichloroacetic acid and subjected to Western analysis using anti-H3, anti-H4, and anti-G6PDH (loading control). Similar results were obtained for three independent experiments using the strains listed in panel B.

Fig. 3.

Mapping of the eight H3/H4 histone residues that strongly derepress SER3 onto the yeast nucleosome crystal structure. (A) A surface representation of the yeast nucleosome core particle viewed down the DNA superhelical axis. Histone proteins are color coded as follows: H3 in white, H4 in gray, and H2A/H2B in purple. The DNA helix is shown in yellow. The five histone H3 and three histone H4 residues required for SER3 repression are highlighted in red (first H3-H4 dimer) and blue (second H3-H4 dimer). (B) Rotation of the view in panel A by 90° around the horizontal axis revealing the lateral surface surrounding the nucleosome dyad. These images were generated by Pymol (Protein Data Bank number 1ID3).

Fig. 4.

Analysis of histone mutants for sin phenotype. Wild-type (FY4), snf2Δ (YJ112), HHTS-HHFS snf2Δ (YJ1049), hhts-T1181 snf2Δ (YJ1081), hhts-K122A snf2Δ (YJ1051), hhts-K122R snf2Δ (YJ1054), hhts-K122Q snf2Δ (YJ1057), hhts-Q120A snf2Δ (YJ1060), hhts-V117A snf2Δ (YJ1063), hhts-R49A snf2Δ (YJ1066), hhts-V46A snf2Δ (YJ1069), hhfs-R36A snf2Δ (YJ1072), hhfs-S47D snf2Δ (YJ1075), and hhfs-I46A snf2Δ (YJ1078) were grown to saturation in YPD at 30°C. Three μl of 10-fold serial dilutions were spotted onto solid YPD, YPgal, and YPraff media and incubated for 3 days. A representative growth assay of three biological replicates that produced equivalent results is shown.

Fig. 5.

Effect of histone mutants on nucleosome positions at SER3. (A) Diagram of the SER3 locus. The gray ovals mark the position of nucleosomes when wild-type cells are grown in SER3-repressing conditions (YPD). The block arrow indicates SRG1 transcription. (B to J) Nucleosome scanning assays were performed on (hht1-hhf1)Δ strains expressing either synthetic wild-type copies of histone H3 and H4 (HHTS and HHFS) or the indicated histone mutant alleles. Cells were grown in YPD medium at 30°C. Each experiment was done in triplicate using one set of strains from the original histone mutation library (JDY86 derivatives) and two additional sets of strains generated by genetic crosses (YJ925 to YJ946). MNase protection across the SER3 locus relative to a positioned nucleosome within the GAL1 promoter was determined by qPCR, and the means ± SEM for the three replicates is plotted at the midpoint for each PCR product. Shown below each graph is a diagram of the SER3 locus indicating the positions of nucleosomes (gray ovals) extrapolated from the MNase protection data for each histone mutant. The light gray ovals are indicative of less dramatic reductions in MNase protections compared to that of the wild-type control shown in panel A.

Fig. 6.

Correlation between MNase protection of SRG1 and SER3 expression. The extent of MNase protection across the SRG1-transcribed unit for wild-type and histone mutant strains (Fig. 4) was plotted against the relative level of SER3 expression in these strains as determined by Northern analysis (Fig. 1). Change in MNase protection was calculated by taking the area under the curve over the SRG1 transcription unit in the histone mutant strain and subtracting this from the area under the curve over the SRG1 transcription unit in the wild-type control. All values were normalized to those of strains expressing synthetic copies of wild-type histone H3 and H4 genes (HHTS-HHFS), where the MNase protection across SRG1 was set to 100% and SER3 expression was set to 1. The line of best fit and R² values were determined by linear regression.

Fig. 7.

Relative occupancy of histone H3 in histone mutants over SER3. Histone H3 ChIP was performed on chromatin isolated from (hht1-hhf1)Δ strains expressing HHTS-HHFS alleles (JDY86, YJ927, and YJ928) or the indicated histone mutant alleles (JDY86 derivative, YJ925, YJ926, and YJ930 to YJ946) that were grown in YPD at 30°C. The amount of immunoprecipitated DNA was determined by qPCR as a percentage of the input material normalized to a control region in chromosome V and represent the means ± SEM of three experiments. Histone H3 occupancy at each genomic location, determined for the strains expressing wild-type histone H3 and H4, was arbitrarily set to 1. Below the graph is a schematic of SER3, with black bars corresponding to the regions amplified by qPCR.

Fig. 8.

Effect of histone mutants on cryptic intragenic transcription and posttranslational histone modifications. (A) Northern analysis of FLO8, STE11, and SYF1 for cryptic intragenic transcription. Total RNA was isolated from (hht1-hhf1)Δ strains that express either synthetic wild-type copies of histone H3 and H4 or the indicated histone mutant alleles (JDY86 derivatives). Strains wild type for both copies of histone H3 and H4 (WT) and expressing either a normal copy of SPT6 (WT) or the spt6-1004 mutant allele were included as negative and positive controls for cryptic transcription. All strains were grown in YPD at 30°C except for the spt6-1004 mutant, which also was shifted to 37°C for 60 min as indicated. Cryptic transcripts for each gene are marked with an asterisk. SCR1 serves as a loading control. (B to D) Western analyses of posttranslational histone modifications. Whole-cell extracts were prepared from wild-type (FY4), set2Δ (KY1716), set1Δ (KY1755), and dot1Δ (KY1907) strains and (hht1-hhf1)Δ strains expressing either synthetic wild-type copies of histone H3 and H4 or the indicated histone mutant alleles (JDY86 derivatives) that were grown in YPD at 30°C. Immunoblots of WCEs were probed with H3 K36 (B), H3 K4 (C), or H3 K79 (D) methyl-specific antibodies. Immunoblots of total H3 and G6PDH are provided as loading controls. Similar results were observed for two distinct sets of strains (YJ925 to YJ946).

Fig. 9.

Effect of histone mutants on histone H3 occupancy over the coding regions of a subset of yeast genes. (A) Histone H3 ChIP analysis was performed on chromatin prepared from (hht1-hhf1)Δ strains expressing HHTS-HHFS alleles (JDY86, YJ927, and YJ928) or the indicated histone mutant alleles (JDY86 derivative, YJ925, YJ926, and YJ930 to YJ946) that were grown in YPD at 30°C. Histone H3 occupancy was measured within the coding region of three highly transcribed genes: PMA1, PYK1, and ADH1. The regions assayed by qPCR are marked with the black bars in the diagram provided for each gene. All values are normalized to a control region in chromosome V and represent the means ± SEM of three experiments. Histone H3 occupancy at each genomic location determined for the strains expressing wild-type histone H3 and H4 was arbitrarily set to 1. (B) Histone H3 occupancy at three lowly transcribed genes, GAL1, TUB2, and CYC1, was determined as described for panel A.