Role of the histone tails in histone octamer transfer

Abstract

The exceptionally high positive charge of the histones, concentrated in the N- and C-terminal tails, is believed to contribute to the stability of the nucleosome by neutralizing the negative charge of the nucleosomal DNA. We find, on the contrary, that the high positive charge contributes to instability, performing an essential function in chromatin remodeling. We show that the tails are required for removal of the histone octamer by the RSC chromatin remodeling complex, and this function is not due to direct RSC-tail interaction. We also show that the tails are required for histone octamer transfer from nucleosomes to DNA, and this activity of the tails is a consequence of their positive charge. Thus, the histone tails, intrinsically disordered protein regions, perform a critical role in chromatin structure and transcription, unrelated to their well-known role in regulation through posttranscriptional modification.

Figure 1.

Histone tails destabilize the nucleosome: importance of Mg²⁺. (A) Nucleosomes assembled with rat liver histones (6 μl of peak gradient fraction containing 0.9 ng DNA/μl) were digested with trypsin (0.7 μl of 100 ng/μl) or not (same except with 0.7 μl of water instead of trypsin) for 15 min at 30°C, and digestion was terminated by the addition of bovine pancreatic trypsin inhibitor (0.8 μl of 200 ng/μl). Digested nucleosomes (0.65 μl) or undigested control were incubated in 10 μl of 15 mM HEPES (pH 7.6), 0.1 mg/ml BSA, with or without 3 mM MgCl₂ and 0.1% NP-40 as indicated, and with NaCl at the concentrations indicated (mM), for 1 h at 30°, followed by gel electrophoresis. (B) Optical densities were integrated over scans of bands in panel (A) for nucleosomes and naked DNA.

Figure 2.

Histone tails are required for chromatin remodeling by RSC. (A) Nucleosomes (1 ng DNA) assembled with wild-type yeast histones, or with all four tailless histones, were treated with RSC (150 ng), Nap1 (240 ng) and 0.5 mM ATP in 15 mM HEPES (pH 8), 3 mM MgCl₂ and 100 mg/ml BSA in a total volume of 10 ml for the times indicated at 30°C and analyzed by gel electrophoresis. (B) Optical densities were integrated over scans of bands in panel (A) for nucleosomes and naked DNA. (C) Nucleosomes assembled with yeast histones, with tailless H2A, tailless H2B, both tailless H2A and tailless H2B, or with all four tailless histones, were treated as in panel (A) and analyzed as in panel (B). (D) Nucleosomes assembled with yeast histones, with tailless H3, tailless H4, or tailless H3 and tailless H4, were treated as in panel (A) and analyzed as in panel (B).

Figure 3.

Histone tails are required for histone octamer transfer to DNA. (A) Nucleosomes (1 ng in 0.7 μl of peak gradient fraction) assembled with rat liver histones were combined with salmon sperm DNA (0.5 μl of the concentrations indicated, in mg/ml) in 1.5 μl containing 10 mM sodium acetate, incubated at 30°C for the times indicated, diluted to 10 μl containing 25 mM HEPES (pH 8) and 150 mg/ml BSA, and analyzed by gel electrophoresis. (B) Optical densities were integrated over scans of bands in panel (A) for nucleosomes and naked DNA. Ordinate refers to ln(fraction nucleosomes remaining). (C) Nucleosomes (0.7 μl of peak gradient fraction containing 0.9 ng of DNA) assembled with rat liver histones (‘with tails’) or digested with trypsin and reisolated (‘without tails’) were combined with salmon sperm DNA (0.4 mg) in 1.5 μl containing the concentrations of sodium acetate indicated, kept for 45 min at 30°C and analyzed by gel electrophoresis. (D) Optical densities were integrated over scans of bands for nucleosomes and naked DNA. (E) Nucleosomes assembled with yeast histones with tailless H2A, with yeast histones with tailless H2B or all four wild-type yeast histones (wt) were treated as in panel (C) with the concentrations of NaCl indicated and analyzed as in panels (C) and (D). (F) Nucleosomes assembled with yeast histones with tailless H3, with yeast histones with tailless H4 or all four wild-type yeast histones (wt) were treated as in panel (C) with the concentrations of NaCl indicated and analyzed as in panels (C) and (D). (G) Dissociation rates (averages of five determinations, min⁻¹ × 10⁴) from experiments as in panel (A) as a function of DNA concentration (mg/0.3 ml). (H) Scatchard plot of data from panel (G) for the case of pre-equilibrium formation of an octamer–nucDNA:DNA* complex, with the use of rates at various DNA concentrations as measures of the fraction of nucleosomes bound (fraction of the value at saturation).