A protein complex containing the conserved Swi2/Snf2-related ATPase Swr1p deposits histone variant H2A.Z into euchromatin

Abstract

The conserved histone variant H2A.Z functions in euchromatin to antagonize the spread of heterochromatin. The mechanism by which histone H2A is replaced by H2A.Z in the nucleosome is unknown. We identified a complex containing 13 different polypeptides associated with a soluble pool of H2A.Z in Saccharomyces cerevisiae. This complex was designated SWR1-Com in reference to the Swr1p subunit, a Swi2/Snf2-paralog. Swr1p and six other subunits were found only in SWR1-Com, whereas six other subunits were also found in the NuA4 histone acetyltransferase and/or the Ino80 chromatin remodeling complex. H2A.Z and SWR1 were essential for viability of cells lacking the EAF1 component of NuA4, pointing to a close functional connection between these two complexes. Strikingly, chromatin immunoprecipitation analysis of cells lacking Swr1p, the presumed ATPase of the complex, revealed a profound defect in the deposition of H2A.Z at euchromatic regions that flank the silent mating type cassette HMR and at 12 other chromosomal sites tested. Consistent with a specialized role for Swr1p in H2A.Z deposition, the majority of the genome-wide transcriptional defects seen in swr1Delta cells were also found in htz1Delta cells. These studies revealed a novel role for a member of the ATP-dependent chromatin remodeling enzyme family in determining the region-specific histone subunit composition of chromatin in vivo and controlling the epigenetic state of chromatin. Metazoan orthologs of Swr1p (Drosophila Domino; human SRCAP and p400) may have analogous functions.

Figure 1. Subunit Architecture of SWR1-Com and Overlap with NuA4 and Ino80-C Complexes

Venn diagram showing proposed subunit compositions of the SWR1, NuA4, Ino80p-C, and Nap1p/Kap114p complexes. Assignments were based on the data shown in Table 1 and Figure 2 and Figure 3. Proteins used in TAP purifications are indicated by “*” and proteins encoded by essential genes are underlined.

Figure 2. SWR1-Com Shared Subunits with NuA4 and Contained Proteins with Motifs Involved in Chromatin Biology

(A) Protein complex overlap. Purifications were performed under high stringency conditions (see Materials and Methods) from Swr1-TAP, Yaf9-TAP, and untagged control strains, resolved by SDS-PAGE and stained with silver. Due to the relatively low efficiency of the Swr1-TAP purification, the wt and Swr1-TAP purifications were performed from twice the amount of starting material compared to Yaf9-TAP. Not all proteins identified by mass spectrometry were clearly visible on the gel. Arrows point to proteins that were common to the Swr1-TAP and Yaf9-TAP purifications, whereas stars point to proteins that were found only in the Yaf9-TAP purifications as judged by visual inspection and comparison of protein sizes with the data deduced from mass spectrometry. The vertical bar indicates that proteins in that area of the gel could not be clearly resolved. (B) Domain structure of SWR1-Com. Shown are SMART domain representations of individual proteins assigned to the SWR1-Com taken from the SMART database (http://smart.embl-heidelberg.de/). Domain names are included, green bars indicate coil-coiled regions, and magenta bars indicate regions of low complexity. The amino-terminal part of Swr1p is not to scale.

Figure 3. Swc4p and Bdf1p Were Components of SWR1-Com

This figure shows immunoblots of analytical-scale TAP purifications. The captured TAP-tagged protein is indicated above the gels, and the protein that was tested for association is indicated at the right side. (A) Association of Swc4p and Tra1p. Swc4-HA was present in purifications from Yaf9-TAP, Esa1-TAP, Rvb2-TAP, and Swr1-TAP but not Ino80-TAP. NuA4 was only present in the Yaf9-TAP and Esa1-TAP material. (B) Reciprocal confirmation of Swc4p being part of NuA4. Swc4-TAP and Yaf9-TAP purified material contained NuA4 components Esa1p and Tra1p. (C) Association of Bdf1p. Bdf1p was present in purifications from Swr1-TAP, Yaf9-TAP, and Swc4-TAP but not Esa1-TAP.

Figure 4. SWR1-Com Associated Selectively with H2A.Z and Contained H2B

(A) Analytical-scale TAP purifications from H2A.Z-TAP, Yaf9-TAP, and H2A-TAP were analyzed by immunoblotting for the components indicated on the right. SWR1-Com preferentially associated with H2A.Z-TAP, whereas Kap114-HA associated equally with H2A.Z-TAP and H2A-TAP but not Yaf9-TAP. (B) SWR1-Com was purified from strains with HA-tagged versions of either H2A.Z or H2B and analyzed by immunoblotting for the presence of these histones as well as the SWR1-Com subunit Act1p.

Figure 5. Chromosomal Distribution of Swr1p-Activated Genes

(A) Histogram showing the number of Swr1p-activated genes as a function of their distance to the nearest chromosome end. (B) The statistical significance of the enrichment of Swr1p-activated genes as a function of distance to the nearest telomere, and the significance of the depletion of Swr1p-activated genes in regions greater than 40 kb from a telomere, were determined using the hypergeometric function (Tavazoie et al. 1999).

Figure 6. Relationship of Genes Activated by Swr1p or H2A.Z

(A) The Venn diagram of number of genes that exhibited a significant decrease in expression in swr1Δ cells (this work) or htz1Δ cells (Meneghini et al. 2003), revealing a large overlap. Shown on the top is the relationship for the genome overall and on the bottom for genes within 20 kB of a telomere. H2A.Z-dependent genes whose expression could not be determined in swr1Δ cells were omitted. (B) A color representation of all genes that were significantly reduced in expression in swr1Δ cells only, htz1Δ cells only, or both, grouped according to (A). Each column represents data from an independent microarray experiment that compared genome-wide expression in mutant cells of the indicated genotype to wt cells. Each row represents the changes in expression of a single gene across the eight experiments. Change in expression measured as the log₂ of the mutant/wt expression ratio is indicated according to the color scale shown. Red cells refer to genes found to have increased expression in either swr1Δ cells or htz1Δ cells that decreased in expression in the other mutant. Excluded from representation are genes that increased expression in both mutants.

Figure 7. ChIP Analysis of HA3-H2A.Z Deposition in the HMR Region and Near the Right Telomere of Chromosome XIV

(A) Location of PCR primers. (B) ChIP results in wild type (bars indicate relative enrichment versus a probe in the PRP8 ORF; standard errors are shown). The ChIP enrichment signal at HMR relative to PRP8 being less than 1.0 indicated some H2A.Z deposition occurred at the PRP8 control region. (C) ChIP results in swr1Δ cells.

Figure 8. ChIP Analysis of H2A.Z Deposition at Nontelomeric Euchromatic Sites

(A) ChIP results in wild type. (B) ChIP results in swr1Δ cells. We detected a reduced enrichment of H2A.Z at all these loci when we estimated the absolute H2A.Z abundance by dividing the amount of immunoprecipitated DNA by the amount of total input DNA for each locus.

Figure 9. Eaf1p Was a Subunit of the NuA4 HAT

(A) Eaf1-HA associated with NuA4 subunits. Immunoblots of analytical-scale TAP purifications are shown. The captured TAP-tagged protein is indicated above the gels, and the protein that was tested for association is indicated at the right. (B) Strains lacking EAF1 have defects in histone H4 acetylation. Whole cell extracts from mutant strains indicated on the top were tested for global histone acetylation using antibodies directed against different forms of acetylated H4 and H2A as indicated on the right.

Figure 10. NuA4 and SWR1-Com Shared Similar Phenotypes and Interacted Genetically

(A) SWR1-Com and Eaf1p were required for resistance to DNA damage and genotoxic stress. Ten-fold serial dilutions of strains from a stationary overnight culture with the indicated deletions of SWR1-Com subunits and of EAF1 were plated and incubated at 30 °C for 2–3 d. YPD plates with the following concentrations of chemicals were used: 100 mM HU, 10 μg/ml of benomyl, 2% formamide, or 3 mM caffeine. (B) SWR1-Com and NuA4 interacted genetically. Double mutants, deduced from genetic analysis of the viable spore clones, are circled, with the two mutations of interest in each cross indicated at the side. All double mutants were inviable.