H2AZ is enriched at polycomb complex target genes in ES cells and is necessary for lineage commitment

Abstract

Elucidating how chromatin influences gene expression patterns and ultimately cell fate is fundamental to understanding development and disease. The histone variant H2AZ has emerged as a key regulator of chromatin function and plays an essential but unknown role during mammalian development. Here, genome-wide analysis reveals that H2AZ occupies the promoters of developmentally important genes in a manner that is remarkably similar to that of the Polycomb group (PcG) protein Suz12. By using RNAi, we demonstrate a role for H2AZ in regulating target gene expression, find that H2AZ and PcG protein occupancy is interdependent at promoters, and further show that H2AZ is necessary for ES cell differentiation. Notably, H2AZ occupies a different subset of genes in lineage-committed cells, suggesting that its dynamic redistribution is necessary for cell fate transitions. Thus, H2AZ, together with PcG proteins, may establish specialized chromatin states in ES cells necessary for the proper execution of developmental gene expression programs.

Figure 1. H2AZ occupies promoter regions in ES cells

(A) Representative examples of DNA sequences occupied by H2AZ isolated using chromatin immunoprecipitation (ChIP) and promoter microarrays. ChIP with core histone H3 was hybridized together with H2AZ to control for nucleosome density. Note that hybridization of H2AZ ChIP-DNA with bulk chromatin as input yielded highly similar results in a replicate set of experiments. The plots show unprocessed enrichment ratios (blue) for all probes within a genomic region. Chromosomal positions are from NCBI build 34 (mm6) of the mouse genome. Genes are shown to scale below plots. The start and direction of transcription are both indicated by an arrow. (B) Distribution of the distance between bound probes and the closest transcription start site (TSS). Data are the average unprocessed enrichment ratios for each oligonucleotide probe within the –4 kb to +4 kb genomic region for all enriched genes. (C) Representative example of DNA sequences occupied by H2AZ using ChIP and tiled chromosome 19 microarrays. The plots show unprocessed enrichment ratios (blue) for all probes within a genomic region (ChIP versus histone H3). Genes are shown to scale below plots as in (A). (D) Gene ontology analysis for biological process of H2AZ-enriched genes. Ontology terms are represented on the y-axis and the p-value for enrichment of bound genes relative to all genes represented on the microarray is shown for each category on the x-axis.

Figure 2. H2AZ and Suz12 occupy a highly similar set of genes in ES cells

(A) Venn diagram showing the overlap of genes enriched with H2AZ and occupied by Suz12 at high confidence in ES cells as determined by ChIP combined with promoter arrays. 93% of H2AZ-enriched genes were also occupied by Suz12 using high-confidence threshold criteria. (B) H2AZ and Suz12 enriched regions were binned according to length of the bound region as determined by promoter microarrays. The majority of bound regions are less then 2 kb whereas a small fraction (~15%) exhibit an extended domain of binding > 4 kb. (C) H2AZ and Suz12 display the same spatial patterning at target genes. The plots show unprocessed enrichment ratios for H2AZ (blue) and Suz12 (green) for all probes within a genomic region as in Figure 1A. (D) H2AZ and Suz12 occupy large domains (> 100kb) that span multiple contiguous genes encompassing the HOX gene clusters. Data are derived from the promoter microarrays that were also designed to contain probes tiling the entire HoxA locus. Unprocessed enrichment ratios for H2AZ (blue) and Suz12 (green) for all probes within the region are shown. The grey bars represent the approximate location of the gene cluster. (E) H2AZ and Suz12 display a highly similar defined spatial pattering across target genes in ES cells. K-means clustering was performed using the Cluster algorithm (http://rana.standord.edu/software) and the set of 1655 H2AZ-enriched genes. Each horizontal line represents an individual gene and the enrichment values for H2AZ (blue) and Suz12 (green) for each probe within the region –3.5 kb to + 3.5 kb relative to the TSS are represented by color intensity (darker color represents higher enrichment ratio).

Figure 3. H2AZ is enriched at a distinct set of genes in neural precursors

(A) Representative examples of DNA sequences occupied by H2AZ from a replicate set of experiments isolated using ChIP and promoter microarrays. ChIP with core histone H3 was hybridized together with H2AZ to control for nucleosome density. The plots show unprocessed enrichment ratios for H2AZ in ES cells (blue) and NPs (red) for all probes within a genomic region as shown in Figure 1A. (B) The spatial distribution of H2AZ differs in NPs as compared to ES cells. H2AZ enriched regions in ES cells (blue) and NPs (red) were binned according to length of the bound region based on data from the promoter microarrays. (C) H2AZ is enriched at active genes in NPs. Cumulative distribution plot of the expression of genes associated with H2AZ or H3K27me3 enrichment in ES cells and NPs. Cumulative distribution of all genes in ES cells (black) and NPs (gray) are shown as a reference. Affymetrix expression and H3K27me3 enrichment data were derived from Mikkelsen et al., 2007.

Figure 4. RNAi-mediated H2AZ depletion in ES cells

(A) Schematic representation of H2AZ hairpin targeting sequences (red) and the homology between H2A and H2AZ. (B) mRNA levels were significantly reduced upon RNAi-mediated suppression of H2AZ as measured by quantitative real-time PCR in H2AZ-depleted ES cell lines relative to the control ES cell line. Data were normalized to Gapdh and error bars represent 2 standard deviations. (C) H2AZ protein abundance s determined by Immunoblot analysis is specifically reduced in depleted ES cell lines whereas no reduction in histone H2A or Oct4 levels is observed in the control cell line. Actin serves as a loading control. (D) ES cell colony morphology and Oct4 levels are similar between control and H2AZ-deficient ES cell lines.

Figure 5. H2AZ is necessary for control of target gene expression

Target genes display a propensity to become de-repressed upon RNAi-mediated suppression of H2AZ in ES cells. Real-time PCR analysis of mRNA levels in H2AZ-depleted ES cells for genes that are enriched (green) or un-enriched (orange). Data were normalized to Gapdh and are shown relative to control ES cell lines. Reactions were performed in triplicate and error bars represent 2 standard deviations of the mean.

Figure 6. Interdependent localization of H2AZ and Polycomb complexes at target promoters in ES cells

ChIP combined with real time PCR for (A) Suz12 in control (green bars) or H2AZ-depleted ES cells (purple bars), (B) H2AZ in control (green bars) or suz12 null ES cells (purple bars) and (C) Rnf2/Ring1b in control (green bars) or H2AZ-depleted ES cells (purple bars) for the promoter regions of the indicated genes. All ChIP-qPCR reactions were performed in triplicate and error bars represent 2 standard deviations of the mean. Primers were designed to amplify a region within 1 kb of the TSS and are listed in Table S13. Tcf3 and Oct4 (Pou5f1) represent negative controls. Similar results were obtained in both H2AZ-depleted ES cell lines.

Figure 7. H2AZ is necessary for ES cell differentiation

(A) Hematoxylin and eosin staining of sectioned day 10 embryoid bodies (EBs) derived from control (upper panel) or H2AZ-depleted (bottom panel) ES cells. (B) Real-time PCR on mRNA isolated from control (black bars) or H2AZ-depleted EBs (grey bars) for the indicated genes. Relative mRNA levels are normalized to Gapdh. Days shown indicate the time EBs were maintained in culture prior to RNA isolation. Far right panel shows end-point comparison of Oct4 and Nanog levels. Reactions were performed in triplicate and error bars represent 2 standard deviations. Similar results were obtained using both H2AZ-depleted ES cells lines. (C) Panels show EB images cultured for 7 days in the absence (left panels) or presence (right panels) of retinoic acid. Upper panels display control cells while the lower panels represent EBs derived from H2AZ-depleted ES cells. Arrows indicate formation of neuron like structures typical for retinoic acid induced differentiation. (D) Real-time PCR of mRNA levels for neural differentiation markers Sox1 and Nestin in EBs as shown in (C). Reactions were performed in triplicate and error bars represent 2 standard deviations. Similar results were obtained using both H2AZ-depleted ES cells lines.