Acetylation Mimics Within a Single Nucleosome Alter Local DNA Accessibility In Compacted Nucleosome Arrays

Abstract

The activation of a silent gene locus is thought to involve pioneering transcription factors that initiate changes in the local chromatin structure to increase promoter accessibility and binding of downstream effectors. To better understand the molecular requirements for the first steps of locus activation, we investigated whether acetylation of a single nucleosome is sufficient to alter DNA accessibility within a condensed 25-nucleosome array. We found that acetylation mimics within the histone H4 tail domain increased accessibility of the surrounding linker DNA, with the increased accessibility localized to the immediate vicinity of the modified nucleosome. In contrast, acetylation mimics within the H3 tail had little effect, but were able to synergize with H4 tail acetylation mimics to further increase accessibility. Moreover, replacement of the central nucleosome with a nucleosome free region also resulted in increased local, but not global DNA accessibility. Our results indicate that modification or disruption of only a single target nucleosome results in significant changes in local chromatin architecture and suggest that very localized chromatin modifications imparted by pioneer transcription factors are sufficient to initiate a cascade of events leading to promoter activation.

Figure 1. Histones and nucleosome arrays.

(A) Positions of lysine acetylation mimics and ligation strategy. Top, Schematic showing arrangement of 12-mer arrays and DC nucleosome upon ligation to the 25-mer nucleosome array. Directional ligation sites are indicted by arrowheads, t-bars indicate unligatable DNA ends. Bottom. The locations of lysine to glutamine substitutions within H3 and H4 used in this study are shown. (B) Increasing amounts of the 4.5 kb control DNA (1, 2, 3, and 4 μg) are loaded in lanes 1–4. Ligated WT, H4 4KQ, H3 6KQ and H3 6KQ/H4 4KQ arrays are shown (lanes 5–8). Some unligated 12-mer array remains after ligation. (C) Ligated arrays are saturated with nucleosomes and completely self-associated in NEB4 digestion buffer. Self-association assays with the arrays (right) were performed as described with increasing MgCl²⁺ (lanes 1–4), as indicated. Self-association of arrays in the absence (−) or presence (+) of NEB4 buffer. (D) Self-association and analytical ultracentrifugation (AU) analysis of arrays in NEB4 buffer. 12-mer nucleosome array templates were reconstituted to increasing saturations with either chicken or recombinant Xenopus core histones and the level of saturation determined by AU. The fraction of each array remaining in solution as determined by self-association assays was plotted vs number of nucleosomes per template. Duplicates of assays are shown, 12 nucleosomes per template is considered saturation.

Figure 2. Accessibility of the central nucleosome linker DNA within a 25-nucleosome array.

(A) Example of digestion time course for arrays containing wt H4, H4 K16Q or H4 4KQ within the central DC nucleosome. Time points (min.) of digestion are shown at the top of the gel. Digestion products were analyzed on 0.7% SDS-agarose gels followed by EtBr staining. Bands corresponding to the 5.3 kb 25-mer template, the 4.5 kb naked DNA control, and residual 12/13 mer arrays and other digestion products are indicated. Lane 1 contains molecular size markers (M). (B) Example plots of quantified digestion data as described in Materials and Methods. Lines represent linear regression fits with R² for each fit ranging from 0.96 to 0.98. The rate of the chromatin digestion relative to the naked DNA internal control (k_rel) is indicated below each plot. (C) Effects of acetylation mimics are localized to the DC nucleosome in the 25-mer arrays. The 25-mer arrays were digested with Eco RI and products run on an SDS-agarose gel. Note that EcoRI sites lie between each nucleosome in the 25-mer array. Relative rate of the loss of the uncut 25-mer normalized to WT is indicated below the gel.

Figure 3. Effect of a nucleosome free region on DC Linker DNA accessibility.

(A) A 25-mer array with a nucleosome-free region is maximally self-associated in DraIII digestion buffer. Lanes 1 and 2 show arrays soluble in the absence and presence of NEB4 buffer. Lanes 3–9 show soluble arrays in NEB 4 buffer with increasing concentrations of additional MgCl₂, as indicated. (B) A Nucleosome free region enhances accessibility of DraIII sites in the center of the 25-nucleosome array. DraIII digestions were performed with 25-mer arrays in which the central nucleosome contained native histones (WT, lanes 1–8) or was ligated in as a naked template representing a nucleosome-free region (DC NFR, lanes 9–16). Digestions were as described for 0, 2, 4, 6, 8, 10, 15, 30, 45, and 60 minutes. k_rel calculated for each of the digests is shown below the gels.

Figure 4. Effect of reduced chromatin compaction on DC Linker DNA accessibility.

25-mer Arrays were assembled with either unmodified histones (WT) or H3 6KQ/H4 4KQ in the central DC nucleosome. (A) Arrays are not self-associated in digestion buffer containing 0.5 mM magnesium acetate. DraIII digestion buffer was prepared containing either 0.5 mM or 10 mM magnesium acetate (lanes 2 and 5 or 3 and 6, respectively) and the extent of self-association determined by a centrifuge assay. Arrays in TE buffer are shown for comparison (lanes 1 and 4). (B) Effect of reduced self-association on DC nucleosome linker DNA accessibility. The relative rate (krel) of DraIII digestion in buffer containing 0.5 mM magnesium acetate was determined. krels are indicated below the gels. (C) DC nucleosomes were ligated with naked p12 templates and the relative rate of digestion in NEB4 buffer determined. Lanes 1–10 show digestion time points taken over 60 minutes, as indicated.