A combination of H2A.Z and H4 acetylation recruits Brd2 to chromatin during transcriptional activation

Abstract

H2A.Z is an essential histone variant that has been implicated to have multiple chromosomal functions. To understand how H2A.Z participates in such diverse activities, we sought to identify downstream effector proteins that are recruited to chromatin via H2A.Z. For this purpose, we developed a nucleosome purification method to isolate H2A.Z-containing nucleosomes from human cells and used mass spectrometry to identify the co-purified nuclear proteins. Through stringent filtering, we identified the top 21 candidates, many of which have conserved structural motifs that bind post-translationally modified histones. We further validated the biological significance of one such candidate, Brd2, which is a double-bromodomain-containing protein known to function in transcriptional activation. We found that Brd2's preference for H2A.Z nucleosomes is mediated through a combination of hyperacetylated H4 on these nucleosomes, as well as additional features on H2A.Z itself. In addition, comparison of nucleosomes containing either H2A.Z-1 or H2A.Z-2 isoforms showed that significantly more Brd2 co-purifies with the former, suggesting these two isoforms engage different downstream effector proteins. Consistent with these biochemical analyses, we found that Brd2 is recruited to AR-regulated genes in an H2A.Z-dependent manner and that chemical inhibition of Brd2 recruitment greatly inhibits AR-regulated gene expression. Taken together, we propose that Brd2 is a key downstream mediator that links H2A.Z and transcriptional activation of AR-regulated genes. Moreover, this study validates the approach of using proteomics to identify nucleosome-interacting proteins in order to elucidate downstream mechanistic functions associated with the histone variant H2A.Z.

Figure 1. Nucleosome IP/Mass spectrometry approach for the identification of H2A.Z-nucleosome interacting proteins.

A. Schematic of the nucleosome IP & mass spectrometry protocol used. B. Heat map of spectral counts from amalgamated data showing the top 21 protein hits. C. Venn diagram summarizing the number of unique and overlapping hits between H2A.Z nucleosomes, H2A nucleosomes, and the Flag-tagged GFP control. D. Gene ontology analysis of the proteins identified in our MS analysis.

Figure 2. Interaction between Brd2 and H2A.Z nucleosomes increases following TSA treatment and is dependent on the bromodomains of Brd2.

A. H2A.Z and H2A nucleosomes were purified as described in Materials and Methods and eluted material was used for Western blot analysis. Prior to harvest cells were treated with TSA or DMSO (vehicle). Brd2 preferentially interacts with H2A.Z nucleosomes and this interaction increases in TSA-treated cells. Brd2 antibody detects both the long and short isoforms (upper and lower bands, respectively) in the INPUT fraction. H3 Westerns show that comparable levels of nucleosomes are loaded between samples and Flag blots show the relative expression/levels of Flag-H2A and Flag-H2A.Z. H4Ac antibody detects the tetra-acetylated form of H4 and is used to show hyperacetylation in TSA-treated samples. B. Mononucleosome IPs were performed using material from cells expressing Flag-H2A.Z and either HA-VPS72 or YFP-Brd2. Anti-HA and -GFP antibodies were used to detect HA-VPS72 and YFP-Brd2 proteins, respectively. Whereas hyperacetylation increases interaction of YFP-Brd2 with H2A.Z nucleosomes, it has no effect on the interaction with HA-VPS72. Total lysates from 293T cells expressing HA-VPS72 and YFP-Brd2 were used to show total expression of the tagged proteins and acetylation status of H4. C. Mononucleosome IPs were prepared from cells expressing Flag-H2A.Z alone (UT) or either wild type (WT) YFP-Brd2 or a mutant version of YFP-Brd2, which contains a single point mutation in each of the bromodomains (BD). Levels of the tagged proteins were detected using both anti-GFP antibody as well as anti-Brd2 antibody (note that the anti-Brd2 antibody detects both the endogenous and exogenous forms of Brd2). Despite the higher relative expression of BD mutant compared to the wild type version, the mutant version does not interact with H2A.Z nucleosomes at detectable levels.

Figure 3. Acetylated H4 lysines are the primary binding sites for Brd2 and are enriched on H2A.Z nucleosomes.

A. A schematic depicting the experimental design of the peptide competition assay. B. Western blots of eluted material from H2A.Z mononucleosome IPs performed in the presence of various competing peptides—sites of acetylation of the various peptides are indicated (UN, unacetylated peptide). H4 peptides acetylated at K12 alone or at K5, K8, K12 and K16 (Tetra) were able to efficiently compete away Brd2 binding to H2A.Z nucleosomes. C. H2A and H2A.Z nucleosomes were immunoprecipitated following treatment with DMSO or TSA—eluted material was analyzed by Western blot for various acetylation marks on H4 and H3. H2A.Z nucleosomes contain higher levels of acetylated H3 and H4 under both basal and hyperacetylated conditions. AcLys, an anti-pan-acetyl lysine antibody, which detects both acetyl H4 and H3 bands in our Western blot, was used as indicated.

Figure 4. Additional elements of the H2A.Z nucleosome contribute to the interaction with Brd2.

A. Schematic workflow of nucleosome preparation used to randomly re-assemble H2A–H2B and H2A.Z-H2B dimers with H3–H4 tetramers, generating H2A- and H2A.Z-nucleosomes with comparable levels of H3 and H4 PTMs. B. Western blot comparison of “scrambled” versus “non-scrambled” nucleosomes for H3 and H4 PMTs and Brd2 binding. Immunoprecipitated nucleosomes subjected to re-assembly via high-salt/low-salt dialysis (scrambled) show comparable levels of H4 acetylation and H3 methylation compared to non-scrambled control nucleosomes, which show an enrichment of these marks on H2A.Z nucleosomes. Consistent with previous experiments, Brd2 shows preferential enrichment on H2A.Z nucleosomes in the non-scrambled control samples. Brd2 still shows a slight preference for H2A.Z nucleosomes, compared to H2A nucleosomes even when levels of H4 acetylation are comparable in the scrambled nucleosomes. The dashed lined between the two Flag blots is to indicate that a single membrane was cut and therefore Flag-NLS-GFP and Flag-H2A/H2A.Z blots are shown as separate panels. C. Comparison of Brd2 interaction with mononucleosomes containing the different isoforms of H2A.Z. Mononucleosomes IPs were performed as described in Materials and Methods, using cells expressing either Flag-H2A.Z-1 or Flag-H2A.Z-2. Although comparable levels of H4 acetylation are present on H2A.Z-1- and H2A.Z-2-nucleosomes, Brd2 is preferentially enriched on H2A.Z-1 nucleosomes, under both DMSO- and TSA-treated conditions.

Figure 5. Brd2 is recruited to AR–regulated genes in hormone-stimulated cells.

A. A schematic of the prostate specific antigen (PSA) gene. Arrows indicate approximate positions of primers used for qPCR analysis. “(−)2 Kb” represents the region in between the enhancer and promoter and was used as a negative control region, which is approximately 2 kb upstream of the transcription start site. B. ChIP analysis of the PSA gene following a time-course of androgen stimulation (10 nM)—antibodies used for immunoprecipitation are shown on the left. H2A.Z and H4Ac (tetra-acetylated H4) ChIP data were normalized to total H3 signal (data not shown) to account for changes in nucleosome density and are therefore presented as “H2A.Z/H4Ac enrichment”. Each qPCR reaction was performed in triplicate with each experiment repeated at least three times independently. Values are presented as means, ± standard deviation.

Figure 6. H2A.Z influences H4 acetylation and Brd2 recruitment at AR–regulated genes.

A. Western blot analyses of whole-cell lysates from LNCaP cells stably expressing either a scrambled control shRNA or an shRNA targeting H2A.Z-1 mRNA. H2A.Z knockdown does not significantly affect the protein levels of AR or Brd2, nor does it reduce global levels of tetra-acetylated H4 (H4Ac). Tubulin and H3 are shown for the purpose of loading controls. B. ChIP analysis of Brd2 and tetra-acetylated H4 at the PSA and KLK2 genes. Knockdown of H2A.Z reduces the recruitment of Brd2 as well as the acetylation of H4 at AR-regulated genes following hormone stimulation for 60 minutes. The data represent the fold-enrichment in hormone-stimulated cells relative to respective ethanol-treated controls (vehicle control). H4Ac ChIP was also normalized to H3 to account for changes in nucleosome density. Each qPCR reaction was performed in triplicate with each experiment repeated at least three times independently. Values are presented as means, ± standard deviation.

Figure 7. In vitro and in vivo effects of bromodomain inhibition by JQ1 treatment.

A. H2A.Z mononucleosome IPs were performed using cells pre-treated with DMSO or the small molecule inhibitor JQ1. Starting input material (prior to adding beads) and eluted material (post IP) was analyzed by Western blotting. Both concentrations of JQ1 (125 nM and 250 nM) were able to reduce the amount of Brd2 immunoprecipitating with H2A.Z nucleosomes. A slight reduction of Brd2 in the INPUT fraction is also observed following JQ1 treatment, as well as a slight reduction in the levels of H4Ac in the IP fraction (see main text for discussion). H3 and Flag blots are shown for loading purposes. B. Whole-cell lysates of LNCaP cells treated with JQ1 (125 nM or 250 nM) or DMSO were analyzed by Western blot. JQ1 treatment does not reduce total levels of Brd2, AR or H2A.Z although a small decrease in total levels of H4Ac are observed (see main text). Alpha tubulin and H3 are shown as loading controls. C. PSA and KLK2 regulatory regions were analyzed by ChIP for Brd2 recruitment and H4 acetylation (tetra-acetylated) in cells stimulated with ethanol or DHT (10 nM) for 60 minutes. JQ1 treatment reduces the recruitment of Brd2 to both the enhancer and promoter regions of PSA and KLK2 and reduces the H4 acetylation levels to a lesser extent. The data represent the fold-enrichment in hormone-stimulated cells relative to respective ethanol-treated controls (vehicle control). H4Ac ChIP was also normalized to H3 to account for changes in nucleosome density. Each qPCR reaction was performed in triplicate with each experiment repeated at least three times independently. Values are presented as means, ± standard deviation. D. RT-qPCR analysis of LNCaP cells pre-treated with JQ1 or DMSO for 24 hrs then stimulated with 10 nM of androgen for 24 hrs. Analysis of both the PSA and KLK2 genes showed a dose-dependent decrease in expression in cells pre-treated with JQ1. E. MTS proliferation assay of LNCaP cells treated with various concentration of JQ1, or an equivalent volume of DMSO, for 24 hrs. Absorbance values were normalized to non-treated control wells and presented as “Proliferation Index”. Assay wells were prepared in triplicate and the experiment was repeated three times independently. Values are presented as means, ± standard deviation.

Figure 8. A model of Brd2 recruitment to AR–regulated genes.

Following hormone stimulation, Brd2 is recruited to H2A.Z nucleosomes containing acetylated H4 lysines. Association of Brd2 with histone acetyltransferase (HAT) activity promotes acetylation of neighboring nucleosomes. Recruitment of chromatin remodeling activity causes eviction of H2A.Z nucleosomes, which promotes the recruitment of DNA-binding transcription factors, such as AR, and spreading of the acetylated H4 mark promotes a subsequent spreading of Brd2.