A highly conserved region within H2B is important for FACT to act on nucleosomes

Abstract

Histone N-terminal tails play crucial roles in chromatin-related processes. The tails of histones H3 and H4 are highly conserved and well characterized, but much less is known about the functions of the tails of histones H2A and H2B and their sequences are more divergent among eukaryotes. Here we characterized the function of the only highly conserved region in the H2B tail, the H2B repression (HBR) domain. Once thought to play a role only in repression, it also has an uncharacterized function in gene activation and DNA damage responses. We report that deletion of the HBR domain impairs the eviction of nucleosomes at the promoters and open reading frames of genes. A closer examination of the HBR domain mutants revealed that they displayed phenotypes similar to those of histone chaperone complex FACT mutants, including an increase in intragenic transcription and the accumulation of free histones in cells. Biochemical characterization of recombinant nucleosomes indicates that deletion of the HBR domain impairs FACT-dependent removal of H2A-H2B from nucleosomes, suggesting that the HBR domain plays an important role in allowing FACT to disrupt dimer-DNA interactions. We have uncovered a previously unappreciated role for the HBR domain in regulating chromatin structure and have provided insight into how FACT acts on nucleosomes.

FIG 1

The HBR domain is important for GAL1 expression. (A) The uninduced level of GAL1 mRNA in wild-type and H2B Δ1-32, H2B Δ30-37, and H2B Δ3-37 mutant cells was measured by RT-PCR. Cultures were grown to log phase in medium containing 3% raffinose. The amount of GAL1 mRNA was normalized to that of scR1, a polymerase III-transcribed RNA. (B) GAL1 induction in wild-type and H2B mutant cells was measured after shifting the cells from 3% raffinose to 2% galactose for the times indicated. ChIP analysis of RNAP II (C) and TBP (D) recruitment to the GAL1 promoter upon activation. IP, immunoprecipitation.

FIG 2

The HBR promotes nucleosome disassembly during transcriptional activation. Nucleosome densities under repressed and activated conditions are shown. Cross-linking of H2A (A) and H3 (B) was measured at the GAL1 promoter in cells grown in medium containing 2% dextrose (Dex) or 2% galactose (Gal). (C) Cross-linking of H3 within the ORF of GAL1 in cells grown in medium containing 2% dextrose or 2% galactose. (D) Nucleosome density at the promoters of PMA1 and PYK1 was measured by ChIP with antibodies to H3. The time course of H2A (E) and H3 (F) eviction at the GAL1 promoter during activation is shown. The experiment was performed as described in the legend to Fig. 1B. IP, immunoprecipitation.

FIG 3

Deletion of the HBR domain does not affect SWI/SNF recruitment or ATP-dependent chromatin remodeling. (A) Recruitment of SWI/SNF to the GAL1 promoter in cells grown in medium containing dextrose (Dex) or galactose (Gal) was analyzed by ChIP. A polyclonal antibody raised against Swi2 was used (66, 67). IP, immunoprecipitation. (B) Nucleosome reconstitution. Chromatin was assembled onto a DNA fragment consisting of the 601 nucleosome-positioning sequence flanked by 24- and 69-bp linker DNA sequences. The resolution of free DNA (left), wild-type nucleosomes (Nuc, middle), and ΔHBR mutant nucleosomes (Δ30-37, right) on native gels is shown. DNA was visualized by ethidium bromide staining. (C) SWI/SNF remodeling of wild-type or ΔHBR mutant yeast nucleosomes (nuc). Fifty femtomoles of nucleosomes was incubated with 50 or 150 fmol of SWI/SNF, and the remodeled species were resolved on native gels. The migration of nucleosomes positioned in the middle or at the ends of the DNA fragment is indicated on the left. The arrow marks the location of the fully slid nucleosome. (D) Quantification of the fraction of the total amount of nucleosomes that is remodeled nucleosomes. The signal in the region where the remodeled species migrates (arrow) in each lane of the nucleosome-alone sample (no SWI/SNF) was subtracted from the total. The results represent the means and standard deviations of three independent experiments.

FIG 4

HBR mutants display FACT mutant phenotypes. (A) Northern blot analysis of FLO8 mRNA. Blots were probed with sequences complementary to the 3′ region of FLO8. The full-length (F.L.) and short cryptic transcripts produced from FLO8 are indicated. The scR1 signal was used as a loading control. The histone tail mutant strains (first five lanes) were grown at 30°C. All samples were analyzed on the same gel, but redundant samples were removed from the image by splicing, indicated by the black lines. (B) Analysis of the amount of histone H3 associated with Asf1. Asf1-myc/H3-H4 complexes were immunoprecipitated (IP) with 9E10 antibody, and the amounts of Asf1 and H3 were detected by Western blotting. Two amounts of each sample were loaded, 1× and 2×. (C) Spot tests for growth on HU. Serial dilutions of wild-type and mutant strains were spotted onto YPAD medium (YP medium supplemented with adenine sulfate and 2% dextrose) or YPAD medium plus 50 mM HU and incubated at 30°C for 48 h.

FIG 5

FACT binds ΔHBR mutant nucleosomes. (A) FACT (Spt16-myc) recruitment to the GAL1 promoter in cells grown in medium containing dextrose (Dex) or galactose (Gal). IP, immunoprecipitation. (B) FACT binding to nucleosomes. Spt16-TAP bound to IgG or IgG resin only (control) was incubated with wild-type or ΔHBR mutant nucleosomes prepared from cells. The bound proteins were analyzed by SDS-PAGE. The amount of Spt16/Pob3 in the upper panel was visualized by silver staining; the lower panel is a Western blot of the same samples probed with anti-H3 antibody. The band marked by an asterisk in the upper panel is the eluted IgG heavy chain. (C). Histones were reconstituted into NCPs onto a radiolabeled 147-bp 601 sequence. FACT, Nhp6, and SPN (FACT and Nhp6) were added, and the complexes were resolved on native gels. The weak-intensity retarded species observed with FACT alone (lane 2) most likely resulted from FACT binding to the naked DNA remaining in the nucleosome reconstitution. (D) Wild-type and ΔHBR mutant nucleosomes were incubated with increasing amounts of FACT and Nhp6 at a 1:1 ratio.

FIG 6

The HBR domain is important for FACT chaperone activity. (A) FACT binding to recombinant H2A-H2B dimers. TAP-purified FACT was incubated with recombinant wild-type or ΔHBR mutant dimers in solution and then captured on calmodulin-Sepharose beads. The control lanes contained dimer but no FACT. The bound proteins were resolved by SDS-PAGE and visualized by Coomassie blue staining. An unloaded lane was removed, which is indicated by the location of the black line. (B) Binding of FACT and recombinant dimers by the HI-FI assay. FACT was titrated into binding assays containing 1 nM dimer containing Alexa 488-labled H2A. The results are presented as the average and standard deviations from six determinations. (C) FACT evicts H2A-H2B dimers from recombinant yeast nucleosomes. Immobilized wild-type and ΔHBR mutant nucleosomes were incubated with purified FACT. After 60 min, the immobilized nucleosomes were collected and the supernatant was immunoblotted with antibodies to H2B and H3. The asterisk marks the location of the residual endogenous wild-type H2B that copurified with FACT. (D) Quantification of H2B eviction. The amount of FACT-stimulated removal of H2B is plotted as the means and standard deviations from three different experiments. The amount of endogenous H2B copurifying with FACT (asterisk in panel C) was subtracted from the signals of wild-type H2B.