Single-molecule tools elucidate H2A.Z nucleosome composition

Abstract

Although distinct epigenetic marks correlate with different chromatin states, how they are integrated within single nucleosomes to generate combinatorial signals remains largely unknown. We report the successful implementation of single molecule tools constituting fluorescence correlation spectroscopy (FCS), pulse interleave excitation-based Förster resonance energy transfer (PIE-FRET) and fluorescence lifetime imaging-based FRET (FLIM-FRET) to elucidate the composition of single nucleosomes containing histone variant H2A.Z (Htz1p in yeast) in vitro and in vivo. We demonstrate that yeast nucleosomes containing Htz1p are primarily composed of H4 K12ac and H3 K4me3 but not H3 K36me3 and that these patterns are conserved in mammalian cells. Quantification of epigenetic modifications in nucleosomes will provide a new dimension to epigenetics research and lead to a better understanding of how these patterns contribute to the targeting of chromatin-binding proteins and chromatin structure during gene regulation.

Fig. 1.

Isolation of mononucleosomes from yeast. (A) Chromatin fractions from wild-type, H3 K4R, htz1Δ or htz1Δ strains expressing HA–Htz1p. Histones in chromatin were analyzed by immunoblots probed with anti-HA antibodies (top), and then stripped and reprobed with anti-H3 antibodies (bottom). (B) Mononucleosome preparation. DNA extracted from nuclear extracts from 1×10⁸ cell equivalents of yeast expressing HA–Htz1p were digested with 0 or 2.0 U MNase/mg of solid and analyzed by agarose gel electrophoresis. (C–E) Analyses of antibody specificity and genotype of yeast used for single molecule studies. Immunoblots of whole cell extracts from the indicated yeast strains were probed with anti-HA antibodies (top panels) to evaluate HA–Hzt1p expression, then stripped and reprobed with anti-H4 K12ac (C), anti-H3 K4me3 (D) or anti-H3 K36me3 (E) antibodies (middle panels) to evaluate histone modifications. Blots were then restripped and reprobed with anti-H3 antibodies because H3 served as a loading control (bottom panels).

Fig. 2.

H4 K12ac and H3 K4me3 are present in Htz1p-containing mononucleosomes, as shown by PIE-FRET and FCS. (A) PIE-FRET efficiency histograms of interactions of Htz1–HA plus H4 K12ac of mononucleosomes isolated from wild-type yeast (left panel) or H4 K12R mutants (middle panel) expressing HA–Htz1p. Fluorescence cross-correlation analysis of Htz1–HA plus H4 K12ac of HA–Htz1p mononucleosomes (right panel). (B) PIE-FRET efficiency histograms of interactions of Htz1–HA plus H3 K4me3 of mononucleosomes isolated from wild-type yeast (left panel) or set1Δ mutants (middle panel) expressing HA–Htz1p. Fluorescence cross-correlation of Htz1–HA plus H3 K4me3 of HA–Htz1p mononucleosomes (right panel). Donor: anti-H4 K12ac–FAM-X (A) or anti-H3 K4me3–Alexa-Fluor-488 (B). Acceptor: anti-HA–Alexa-Fluor-647. (C,D) H3 K36me3 is present in H2A mononucleosomes but undetectable in Htz1p mononucleosomes. PIE-FRET efficiency histograms of interactions of H2A (C) or HA–Htz1p (D) with H3 K36me3 in mononucleosomes isolated from wild-type yeast (left panels) or set2Δ mutants (middle panels) expressing HA–Htz1p. Fluorescence cross-correlation analysis of H3 K36me3 plus H2A of wild-type yeast mononucleosome (C, right panel). Donor: anti-H3 K36me3–Alexa-Fluor-488. Acceptor: anti-H2A–Alexa-Fluor-647 (C) or anti-HA–Alexa-Fluor-647 (D). (E) H4 K12ac and H3 K4me3 are present in the same nucleosome. PIE-FRET efficiency histogram of interactions between H4 K12ac and H3 K4me3 in mononucleosomes isolated from wild-type yeast (left panel) or set1Δ mutants (middle panel). Fluorescence cross-correlation of H4 K12ac plus H3 K4me3 of wild-type yeast mononucleosomes (right panel). Donor: anti-H3 K4me3–Alexa-Fluor-488. Acceptor: anti-H4 K12ac–Alexa-Fluor-647.

Fig. 3.

Htz1p is associated with H4 K12ac and H3 K4me3 but not H3 K36me3 in yeast. (A,B) Analysis of H4 K12ac in yeast expressing (A) or lacking (B) HA–Htz1p. (C,D) H3 K4me3 in yeast expressing (C) or lacking (D) HA–Htz1p. (E,F) H3 K36me3 in yeast expressing (E) or lacking (F) HA–Htz1p. Yeast were fixed with methanol:acetic acid and incubated with the indicated fluorescently labeled antibodies and analyzed by FLIM-FRET as described in the Materials and Methods. Left column: Fluorescence lifetime distribution of donor (anti-H4 K12ac–FAM-X, anti-H3 K4me3–Alexa-Fluor-488 or anti-H3 K36me3–Alexa-Fluor-488) in the presence of acceptor (anti-HA–Alexa-Fluor-647). Middle column: FLIM from donor channel. Right column: FLIM from acceptor channel. Scale bars: 10 µm. FLIM scale: 1 nanosecond, blue; 4.5 nanoseconds, red. See also Table 1.

Fig. 4.

H2A.Z is associated with H3 K4me3 in mammalian cells. (A–D) MDA-MB-468 cells were fixed with methanol:acetic acid and incubated with anti-H3 K4me3–Alexa-Fluor-488 antibodies only (donor) (A), with anti-H3 K4me3–Alexa-Fluor-488 plus anti-H2A.Z–Alexa-Fluor-647 antibodies (donor and acceptor in left and right panels, respectively) (B), with anti-H3 K36me3–Alexa-Fluor-488 antibodies only (donor) (C) or with anti-H3 K36me3–Alexa-Fluor-488 plus anti-H2A.Z–Alexa-Fluor-647 antibodies (donor and acceptor in left and right panels, respectively) (D), and then analyzed by FLIM-FRET. (E) Fluorescence lifetime distribution of samples shown in A–D. (F) Sequence alignments between yeast and human H2A and Hta1p/H2A.Z are shown. Ten amino acid intervals on H2A and Htz1p/H2A.Z are noted by filled and open circles, respectively. Mutation of residues highlighted in red in yeast H2A result in defects in H3 K36me3 by Set2p. The docking domain is shown with a solid line. (G,H) Composition of nucleosomes containing the histone variants. Scale bars: 10 µm. FLIM scale: 0 nanoseconds, blue; 3.9 nanoseconds, red. See also Table 1.