The H4 tail domain participates in intra- and internucleosome interactions with protein and DNA during folding and oligomerization of nucleosome arrays

Abstract

The condensation of nucleosome arrays into higher-order secondary and tertiary chromatin structures likely involves long-range internucleosomal interactions mediated by the core histone tail domains. We have characterized interarray interactions mediated by the H4 tail domain, known to play a predominant role in the formation of such structures. We find that the N-terminal end of the H4 tail mediates interarray contacts with DNA during self-association of oligonucleosome arrays similar to that found previously for the H3 tail domain. However, a site near the histone fold domain of H4 participates in a distinct set of interactions, contacting both DNA and H2A in condensed structures. Moreover, we also find that H4-H2A interactions occur via an intra- as well as an internucleosomal fashion, supporting an additional intranucleosomal function for the tail. Interestingly, acetylation of the H4 tail has little effect on interarray interactions by itself but overrides the strong stimulation of interarray interactions induced by linker histones. Our results indicate that the H4 tail facilitates secondary and tertiary chromatin structure formation via a complex array of potentially exclusive interactions that are distinct from those of the H3 tail domain.

FIG. 1.

Detection of interarray H4 tail-DNA interactions from a position near the N terminus of the H4 tail domain. (A) Scheme of histone H4 showing the location of cysteine substitution (H4 G2C) and the site of APB-cross-linker attachment. The related site on the H3 tail domain (H3 T6C) investigated previously is also shown. (B) The 12-mer nucleosome arrays reconstituted with *H4 G2C-APB were mixed with 35-mer arrays, and the extent of the MgCl₂-dependent interarray H4 tail-DNA cross-linking was determined. Cross-linking reactions were carried out in 0, 2, 4, 6, and 8 mM MgCl₂ (lanes 2 to 6, respectively) as indicated. The sample in lane 1 was incubated in 10 mM Tris without UV irradiation. Upper panel, EtBr-stained gel; lower panel, autoradiograph of the same gel. (C) Interarray cross-linking of the H4 tail domain was quantified and plotted as the fraction of total cross-linking at each MgCl₂ concentration (diamonds). For comparison, data previously obtained with *H3 T6C-APB (squares) are also shown. (D) Chromatin compaction decreases intraarray cross-linking efficiency of H4 G2C-APB. The relative extent of cross-linking to the 12-mer template at various MgCl₂ concentrations (25) (C) was determined and normalized to cross-linking in the absence of salt.

FIG. 2.

Regions within the H3 and H4 tails near the histone fold domain exhibit distinctly different patterns of interactions. (A) A scheme of the H4 protein showing the site near the H4 histone fold domain (H4 V21C) modified by APB. The location of an analogous site in the H3 tail domain (H3 V35C) previously investigated is also shown. (B) The 12-mer nucleosome arrays were reconstituted with *H4 V21C-APB or *H3 V35C-APB and analyzed for interarray tail-DNA cross-linking as described in the legend to Fig. 1B. (C) Interarray cross-linking for regions near the histone fold domains in H4 (diamonds) and H3 (squares) was quantified and plotted as described in the legend to Fig. 1.

FIG. 3.

A region within the H4 tail interacts with H2A in both an intra- and internucleosomal fashion. The 12-mer nucleosome arrays reconstituted with H2A G2C-fluorescein and *H4 V21C-APB were incubated in 0, 1, 2, and 4 mM MgCl₂ and then UV-irradiated (lanes 2 to 5). The sample in lane 1 was incubated in 10 mM Tris without UV irradiation. Cross-linking products observed by Coomassie blue staining (A), fluorography (B), and autoradiography (C) are shown. (D) Quantification of H2A-H4 cross-linked species in nucleosome arrays (squares) and mononucleosomes (diamonds). (E) Specific competition of H4-H2A cross-linking with LANA peptide. Cross-linking of arrays was carried out in the presence of the LANA or LRS peptide (10), and the extent of cross-linking was quantified and plotted relative to that in the absence of peptide. Errors represent ±1 standard deviation. (F and G) H4-H2A cross-linking in mononucleosomes analyzed by Coomassie blue staining (E) and fluorography (F).

FIG. 4.

Effect of H4 acetylation on H4 tail-DNA interarray interactions. The 12-mer array was prepared with tetraacetylated *H4, and interarray cross-linking was analyzed as described in the legend to Fig. 1. (A) A scheme of H4 showing the four sites of acetylation in the tail domain. (B) A Triton-acid-urea gel showing H4 tetraacetylated and H3 monoacetylated by HAT Piccolo. Lane 1, nonacetylated H3/H4; lane 2, highly acetylated H3/H4; lane 3, moderately acetylated H3/H4 revealing various extents of H4 acetylation. (C and D) Analysis of interarray cross-linking mediated by control (squares) and tetraacetylated (diamonds) *H4 G2C-APB. Experimental conditions and data analysis were as described in the legend to Fig. 1.

FIG. 5.

Acetylation overrides H1 enhancement of interarray interactions of the H4 tail domain. (A) Arrays incubated in the absence (lanes 1 to 6) or presence (lanes 7 to 12) of H1 were assayed for interarray H4 tail-DNA cross-linking. The cross-linking assay was carried out as described in the legend to Fig. 1 except that each sample also contained 50 mM NaCl to facilitate H1 binding. (B) Interarray cross-linking in the absence (circles) and presence (diamonds) of H1 was quantified and plotted as indicated. Quantification of cross-linking for arrays in buffer without NaCl (squares) is shown for reference. Panel C is as described in panel B except that arrays contained acetylated H4.

FIG. 6.

Summary of H3 and H4 tail interactions. Intra-array and interarray interactions with DNA and histones detected for two positions in either tail domain are as indicated.