Progressive methylation of ageing histones by Dot1 functions as a timer

Abstract

Post-translational modifications of histone proteins have a crucial role in regulating gene expression. If efficiently re-established after chromosome duplication, histone modifications could help propagate gene expression patterns in dividing cells by epigenetic mechanisms. We used an integrated approach to investigate the dynamics of the conserved methylation of histone H3 Lys 79 (H3K79) by Dot1. Our results show that methylation of H3K79 progressively changes after histone deposition, which is incompatible with a rapid copy mechanism. Instead, methylation accumulates on ageing histones, providing the cell with a timer mechanism to directly couple cell-cycle length to changes in chromatin modification on the nucleosome core.

Figure 1

A model for H3K79 methylation in dividing cells. (A) Dot1 mono-, di- and trimethylates H3K79 with rate constants k₀, k₁ and k₂. The constant k can be interpreted as the specificity constant (pseudo second-order rate constant k_cat/K_m). Unmodified histones (K79me0) are synthesized at a rate γ and all forms of H3K79 are diluted with a first order rate constant μ because of an increase in cell volume caused by cell growth. The histone synthesis rate γ was assumed to be constant, reflecting an average over the cell cycle, and specific for each strain or growth condition. (B) Observed (average MS measurement of two independent clonal experimental cultures±s.e.m.) and predicted methylated (me1, me2 and me3) and unmethylated (me0) H3K79 fractions in yeast strains expressing different amounts of a galactose-inducible DOT1 gene (Frederiks et al, 2008). (C) Validation of the model using WT strain BY4742 cultured at a different growth rate than that of the strain used to build the model in B (average MS measurement of two independent clonal experimental cultures±s.e.m.). (D) Simulation of H3K79 methylation over a range of Dot1 concentrations. The dashed line indicates the approximate WT methylation level. (E) H3K79 methylation in a yeast strain lacking H2Bub (bre1Δ). Predicted values were on the basis of three sets of kinetic parameters: (i) those of WT cells; (ii) k₂ was allowed to vary but k₀ and k₁ were set to the values determined for WT cells (best fit k₂); or (iii) k₀, k₁ and k₂ were all allowed to vary and fitted independently. Similar H3K79 methylation levels were observed in a strain lacking H2BK123ub due to a H2BK123R mutation (see Supplementary Fig S2A online). (F) Predicted rate constants (μM⁻¹ min⁻¹) for the three methylation reactions. (G) Immunoblots of H3K79 methylation in dot1Δ and dot1Δbre1Δ strains carrying an empty vector or expression vectors for WT Dot1 or Dot1Δ2–136. MS, mass spectrometry; SS, sums of squares; WT, wild type.

Figure 2

Cell-cycle progression affects the accumulation of H3K79 methylation. (A) Simulated methylation patterns of three wild-type strains with different growth rates using WT kinetic parameters. (B) Methylation in a WT strain growing at 30°C in glucose (same sample as in Fig 1C) and growing slowly at 23°C in galactose media (average MS measurement of two independent clonal experimental cultures±s.e.m.). (C) H3K79 methylation in cells in log phase and arrested for 3 h in G1 phase (average MS measurement of two independent clonal experimental cultures±s.e.m.). (D) Immunoblot analysis of Dot1 expression in log phase and G1 arrest. MS, mass spectrometry; WT, wild type.

Figure 3

Time-dependent changes in H3K79 methylation. (A) Simulation of H3K79 methylation in a single cell starting at S-phase with a cell-cycle length of 120 min (left) or 600 min (right). Methylation levels at the start of S-phase (t=0) are identical to the levels at the end of G1 (last time point). (B) Histone H3 was tagged with a RITE cassette. Induction of recombination between LoxP sites (L) by Cre results in a tag switch. (C) Following induction of Cre in log phase, replicating cells were harvested after 0–4 divisions and histone H3–HA–6xHis of different average age was purified under denaturing conditions. By purification of old H3–HA–6xHIS after the tag switch, new histones are eliminated, thereby increasing the average age of H3 in the pool. (D) Immunoblot analysis of epitope-tag-switched histone H3 in whole-cell extracts. (E) MS analysis of H3K79 methylation of purified histones of different average age (average of two independent clonal experimental cultures±s.e.m.). Zero divisions represent the mix of old and new H3 found in non-switched cells. MS, mass spectrometry; RITE, recombination-induced tag exchange.

Figure 4

Role of demethylase activity in accumulation of methyl groups. (A) The inclusion of a demethylase parameter (compare 0.0005 with 0.05 min⁻¹ for all demethylation events) has only small effects on the steady-state model (compare with Fig 1B). (B) Demethylase activity can affect the kinetics of methylation in the single-cell model, but the effect depends on the rate of demethylation and cell-cycle length (compare with Fig 3A; supplementary Fig S4 online). (C) H3K4 methylation in a WT strain growing at 30°C in glucose, growing slowly at 23°C in galactose media and arrested for 2 h in G1. (D) H3K4 methylation on purified histones of increasing age (Fig 3B–E). Strains expressing H3–HA–6xHis, H3–T7 or a mix thereof were used as controls. Left and right panels represent two independent purifications. gal, galactose; gluc, glucose; WT, wild type.

Figure 5

Links between cell-cycle progression, H3K79 methylation and gene regulation. (A) Cells in which URA3 is silenced can grow on media containing FOA. Fraction of FOA-resistant colonies indicates the degree of silencing of a telomeric URA3 reporter gene in dot1Δ strains carrying Dot1 expression vectors at 23°C (slow growth) and 30°C (average of two independent clonal experimental cultures±s.e.m.). (B) Immunoblot analysis of Dot1-G401A expression. (C) Model for regulation of H3K79 methylation. All three methylation reactions are stimulated by H2Bub. H3K79 methylation does not reach a steady state but accumulates over time and is counteracted by dilution of methylated histones by unmodified histones. Dynamic H3K79 methylation can link changes in cell-cycle length (replication-dependent histone dilution) to changes in chromatin structure. FOA, 5-fluoroorotic acid; MT, methyltransferase; TEL, telomere; WT, wild type.