A basic domain in the histone H2B N-terminal tail is important for nucleosome assembly by FACT

Abstract

Nucleosome assembly in vivo requires assembly factors, such as histone chaperones, to bind to histones and mediate their deposition onto DNA. In yeast, the essential histone chaperone FACT (FAcilitates Chromatin Transcription) functions in nucleosome assembly and H2A-H2B deposition during transcription elongation and DNA replication. Recent studies have identified candidate histone residues that mediate FACT binding to histones, but it is not known which histone residues are important for FACT to deposit histones onto DNA during nucleosome assembly. In this study, we report that the histone H2B repression (HBR) domain within the H2B N-terminal tail is important for histone deposition by FACT. Deletion of the HBR domain causes significant defects in histone occupancy in the yeast genome, particularly at HBR-repressed genes, and a pronounced increase in H2A-H2B dimers that remain bound to FACT in vivo Moreover, the HBR domain is required for purified FACT to efficiently assemble recombinant nucleosomes in vitro We propose that the interaction between the highly basic HBR domain and DNA plays an important role in stabilizing the nascent nucleosome during the process of histone H2A-H2B deposition by FACT.

Figure 1.

HBR domain affects histone occupancy at HBR-repressed genes and elsewhere in the genome. (A) Plot showing average histone H2B occupancy in HBR-repressed genes in wt and HBRΔ mutant. ChIP-chip probe data for HBR-repressed genes (18) with transcript coordinates (471 genes total) were divided into bins representing the promoter region (2 bins), coding region (6 bins), and downstream region (2 bins). ChIP-chip data was averaged for each bin. TSS: transcription start site; TTS; transcription termination site. (B) H2B occupancy in genes that are not repressed by HBR domain (5300 genes total).

Figure 2.

HBR domain directly regulates histone H2B occupancy. (A) Snapshot of ChIP-chip data showing decrease in H2B occupancy in the HBRΔ mutant at the bidirectional promoter of HBR-repressed SNO1 and SNZ1 genes. Wt-1 and wt-2 and hbr-1 and hbr-2 represent independent ChIP-chip IP data sets from wild-type and HBRΔ mutant cells, respectively. Image was generated using IGV (28). (B) ChIP-qPCR confirmation of H2B occupancy defect in HBRΔ mutant at the SNO1/SNZ1 promoter shown in A. ‘FLAG only’ indicates the only source for H2B protein in cells is FLAG-tagged wild-type H2B or HBRΔ mutant. ‘FLAG+wt H2B’ indicates a second plasmid expressing untagged wild-type H2B protein is present in addition to the FLAG-tagged H2B (wild-type or HBRΔ mutant). (C) ChIP-qPCR analysis of FLAG-H2B occupancy at three different chromatin loci. ‘FLAG only’ indicates the only source for H2B protein in cells is FLAG-tagged wild-type H2B or HBRΔ mutant. In the ChIP-chip data, H2B occupancy in IME1/RPL43B and YAR035W is decreased in the HBRΔ mutant, while H2B occupancy at PMA1 is not affected. These findings were confirmed by ChIP-qPCR. (D) Same as C, except ‘FLAG + H2B wt’ indicates a second plasmid expressing untagged wild-type H2B is present in all the yeast strains.

Figure 3.

Co-Immunoprecipitation (Co-IP) of wild-type or HBRΔ mutant histones bound to the histone chaperones FACT (Spt16) or Nap1. (A) Whole cell extracts from Myc-tagged Spt16 or Nap1 strains were immunoprecipitated (IP) with anti-Myc agarose beads (Thermo Scientific). The abundance of co-immunoprecipitated histones was detected by western blotting using specific antibodies against H2B (Abcam), H2A (Active Motif), and H2AZ (Active Motif). A yeast strain lacking the Myc-tag was used as a negative control. H2B lacking the HBR domain (HBRΔ) migrates faster than the wild-type (WT) H2B protein on the15% SDS gel. (B) Same as A, except two H2B expression plasmids (either both expressing wild-type H2B or one expressing wild-type and one expressing HBRΔ) were co-expressed in each yeast strain.

Figure 4.

In vitro binding of FACT to wild-type H2A/H2B and mutant H2A/HBRΔ dimers. (A) Western blots showing the binding of FACT to wild type (H2A/H2B) and HBR-deleted mutant (H2A/HBR) histone dimers. Wild-type or mutant dimers were incubated separately with FACT (Spt16-Myc) bound to anti-Myc agarose beads. Two histone dimer concentrations, 100 nM (+) and 200 nM (++) were used in this assay. Western blot analysis was used to detect FACT-bound histones (‘Beads’). 'Dimer input’ shows the level of wild-type or mutant dimers present in the reaction before incubating with FACT. (B) Competition between wild-type and mutant histone dimers for FACT binding. Wild-type and mutant histone dimers were mixed at different molar ratios and subsequently incubated with FACT bound to anti-myc agarose beads. The concentration of wild-type dimer was kept at 100 nM while the mutant dimer concentrations were 120, 360 and 1080 nM, respectively.

Figure 5.

In vitro nucleosome assembly using purified yeast FACT complex. (A) Mono-nucleosome assembly on radiolabeled 208-bp 5S rDNA. ‘NCP’ indicates the position of nucleosome core particle formed by FACT or salt dialysis. ‘Free DNA’ indicates the position of 5S DNA that was not assembled into nucleosomes. ‘5S NUC*’ indicates NCPs that are formed through standard salt dialysis. (B) Oligo-nucleosome assembly on the 5S-12 DNA substrate.

Figure 6.

Model of the mechanism by which the HBR domain affects nucleosome assembly and disassembly by FACT. Histone H2A–H2B dimers are in blue; and H3–H4 tetramer is in pink. DNA is represented as a black line, and the FACT heterodimer is in green. In this model, the (A) wild-type HBR domain stabilizes a partially assembled nucleosome intermediate that is required for both assembly and disassembly pathways. (B) In the HBR mutant (dotted outline of H2A–H2B dimers), the nucleosome intermediate is destabilized, potentially due to weakened histone–DNA interactions, thus inhibiting both nucleosome assembly and disassembly pathways.