The fission yeast HIRA histone chaperone is required for promoter silencing and the suppression of cryptic antisense transcripts

Abstract

The assembly of nucleosomes by histone chaperones is an important component of transcriptional regulation. Here, we have assessed the global roles of the HIRA histone chaperone in Schizosaccharomyces pombe. Microarray analysis indicates that inactivation of the HIRA complex results in increased expression of at least 4% of fission yeast genes. HIRA-regulated genes overlap with those which are normally repressed in vegetatively growing cells, such as targets of the Clr6 histone deacetylase and silenced genes located in subtelomeric regions. HIRA is also required for silencing of all 13 intact copies of the Tf2 long terminal repeat (LTR) retrotransposon. However, the role of HIRA is not restricted to bona fide promoters, because HIRA also suppresses noncoding transcripts from solo LTR elements and spurious antisense transcripts from cryptic promoters associated with transcribed regions. Furthermore, the HIRA complex is essential in the absence of the quality control provided by nuclear exosome-mediated degradation of illegitimate transcripts. This suggests that HIRA restricts genomic accessibility, and consistent with this, the chromosomes of cells lacking HIRA are more susceptible to genotoxic agents that cause double-strand breaks. Thus, the HIRA histone chaperone is required to maintain the protective functions of chromatin.

FIG. 1.

Transcriptomes of HIRA mutants. (A) Venn diagrams showing overlap between genes upregulated (≥1.5-fold) in HIRA mutants with genes upregulated under the indicated condition. The P values indicate the probability that the observed overlap happened by chance. (B) Graphs showing mutant versus wild-type (wt) expression levels of subtelomeric genes of the right (R) and left (L) arms of chromosomes I and II. Lines indicating the 1.5-fold threshold are shown. Genes whose expression was confirmed by quantitative RT-PCR are marked by asterisks.

FIG. 2.

Influence of HIRA on telomere length. DNA samples purified from the indicated strains were digested with ApaI and analyzed by Southern blotting with a telomere repeat probe (lower panel) and loading control probe (upper panel) as described in Materials and Methods. w.t., wild type.

FIG. 3.

HIRA silences the expression of all 13 Tf2 LTR retrotransposons. (A) Schematic of the lacZ reporter used to tag individual Tf2 elements. UTR, untranslated region. (B) Expression of Tf2-lacZ reporters in wild-type cells. Strains harboring a tagged Tf2 element were grown in rich medium (YE5S) at 30°C to mid-log phase. Cells were then harvested and processed for liquid β-galactosidase assays. Shown are the mean values of results from three experiments. Error bars indicate standard deviations. (C) Expression of Tf2-lacZ reporters in a hip1Δ background. Strains were processed as described for panel B. (D) HIRA-dependent repression is maintained at a novel site in the genome. Wild-type or hip1Δ strains harboring either the Tf2-1 or the LTR lacZ reporter integrated into chromosome II were processed as described for panel B.

FIG. 4.

HIRA represses expression from Solo LTRs. RNA was prepared from wild-type (wt), hip1Δ, and slm9Δ cells. Quantitative real-time RT-PCR was then used to quantify the levels of noncoding transcripts ∼100 bp downstream of selected LTR elements. Change in expression (mutant versus wild type) is shown. Error bars indicate standard deviations. For each element, the location, type, and homology to the Tf2-1 LTR is shown.

FIG. 5.

Loss of HIRA leads to the accumulation of spurious antisense transcripts. (A) RNA was purified from wild-type (w.t.), hip1Δ, slm9Δ, and hip3Δ cells and analyzed by strand-specific RT-PCR. One primer, complementary to either the forward or the reverse transcripts, was included during the reverse transcription step, and the second primer was then added during PCR amplification. Control reactions omitting the reverse transcription step (−RT) were included to demonstrate the absence of DNA. (B) RNA from wild-type, hip1Δ, rrp6Δ, clr6-1, and hip1Δ clr6-1 strains was analyzed as described for panel A. (C) The nuclear exosome is essential in the absence of HIRA. The left hand panel shows an example of a tetrad resulting from a cross between hip1Δ and rrp6Δ strains, and the right hand panel shows the terminal morphology of the hip1Δ rrp6Δ double mutant.

FIG. 6.

Antisense transcription is correlated with sense transcription in wild-type but not HIRA mutant cells. (A) Schematic diagram showing regulation of thi4⁺ expression. (B) Wild-type and hip1Δ cells were grown to log phase in minimal (EMM) medium, thiamine was then added to give a final concentration of 2 μM, and samples of cells were harvested at the indicated time points. RNA, prepared from the resulting cell pellets, was analyzed by strand-specific RT-PCR as described in the legend to Fig. 5. −RT, control reactions omitting the reverse transcription step. (C) Antisense transcripts persist under conditions of constant thi4⁺ repression. slm9Δ cells were grown to log phase either in EMM medium or in EMM medium supplemented with 2 μM thiamine. RNA preparation and RT-PCR analysis were performed as described in the legend to Fig. 5.

FIG. 7.

HIRA is required for protection against genotoxic agents. (A) The indicated strains were grown to log phase in YE5S medium, subjected to fivefold serial dilutions, and spotted on YE5S agar and YE5S agar supplemented with the indicated genotoxic agent. wt, wild type; HU, hydroxyurea; MMS, methyl methanesulfonate. (B) The indicated strains were treated with bleomycin at the indicated concentrations for 90 min. Chromosomal DNA was then analyzed using PFGE. (C) The indicated strains were grown to log phase and treated as described for panel A.