Structural basis of nucleosomal histone H4 lysine 20 methylation by SET8 methyltransferase

Abstract

SET8 is solely responsible for histone H4 lysine-20 (H4K20) monomethylation, which preferentially occurs in nucleosomal H4. However, the underlying mechanism by which SET8 specifically promotes the H4K20 monomethylation in the nucleosome has not been elucidated. Here, we report the cryo-EM structures of the human SET8-nucleosome complexes with histone H3 and the centromeric H3 variant, CENP-A. Surprisingly, we found that the overall cryo-EM structures of the SET8-nucleosome complexes are substantially different from the previous crystal structure models. In the complexes with H3 and CENP-A nucleosomes, SET8 specifically binds the nucleosomal acidic patch via an arginine anchor, composed of the Arg188 and Arg192 residues. Mutational analyses revealed that the interaction between the SET8 arginine anchor and the nucleosomal acidic patch plays an essential role in the H4K20 monomethylation activity. These results provide the groundwork for understanding the mechanism by which SET8 specifically accomplishes the H4K20 monomethylation in the nucleosome.

Figure 1.. Cryo-EM structures of the SET8–NCP and SET8–NCP^CENP-A complexes.

(A) Schematic representation of the SET8 domain structures. (B) Digital micrograph of the SET8–NCP complex. Scale bar, 100 nm. (C) Representative 2D class averages of the SET8–NCP complex. Box size, 18.9 nm². (D, E) Cryo-EM reconstructions of the SET8–NCP complex (D) and the SET8–NCP^CENP-A complex (E). The cryo-EM maps were visualized by UCSF ChimeraX.

Figure S1.. Preparation of the SET8–NCP and SET8–NCP^CENP-A complexes.

(A) Purified SET8 was analyzed by 16% SDS–PAGE with Coomassie Brilliant Blue staining. (B) The reconstituted NCP with 145-base pair Widom 601 DNA was analyzed by 18% SDS–PAGE with Coomassie Brilliant Blue staining. (C) The reconstituted NCP with the 145-base pair Widom 601 DNA was analyzed by 6% non-denaturing PAGE with ethidium bromide staining. (D) The SET8–NCP complex purified by GraFix was analyzed by 6% non-denaturing PAGE with ethidium bromide staining. (E) The reconstituted NCP^CENP-A with 145-base pair Widom 601 DNA was analyzed by 18% SDS–PAGE with Coomassie Brilliant Blue staining. (F) The reconstituted NCP^CENP-A with the 145-base pair Widom 601 DNA was analyzed by 6% non-denaturing PAGE with ethidium bromide staining. (G) The SET8–NCP^CENP-A complex purified by GraFix was analyzed by 6% non-denaturing PAGE with ethidium bromide staining.

Figure S2.. Cryo-EM processing of the SET8–NCP and SET8–NCP^CENP-A complexes.

(A, D) Fourier Shell Correlation curves of the SET8–NCP complex (A) and the SET8–NCP^CENP-A complex (D). The overall resolutions of the SET8–NCP and SET8–NCP^CENP-A complexes are 3.15 and 3.00 Å at the 0.143 criteria, respectively. (B, E) Euler angle distribution maps of the SET8–NCP complex (B) and the SET8–NCP^CENP-A complex (E). (C, F) Local resolution maps of the SET8–NCP complex (C) and the SET8–NCP^CENP-A complex (F), calculated by RELION. The resolution ranges of the final maps are between 2 and 7 Å.

Figure S3.. Comparison between the cryo-EM structure of the SET8–NCP complex (magenta) and the model of the crystal structure of the SET8–NCP complex (blue, PDB: 5HQ2).

The H4 N-terminal tail is shown as a space filling model (green).

Figure S4.. Comparison between the SET8–NCP complex and the SET8–NCP^CENP-A complex for the nucleosomal DNA ends and the L1 loops of the H3 and CENP-A molecules.

(A, B) Atomic models of the SET8–NCP complex (A) and the SET8–NCP^CENP-A complex (B) are docked into the cryo-EM maps. Red circles indicate the nucleosomal DNA ends. (C, D) Close-up views of the L1 loops (arrowheads) of H3 (C) and CENP-A (D).

Figure 2.. Arginine anchor of SET8 bound to the acidic patches of the nucleosome core particle (NCP) and the NCP^CENP-A.

(A, B, C, D) Close-up views of the SET8 Arg188 and Arg192 residues bound to the H2A Glu56 and H2B Glu113 residues (B, D, left) and the H2A Glu61 and Glu92 residues (B, D, right), respectively, in the acidic patches of the NCP (A, B) and the NCP^CENP-A (C, D). The atomic model of SET8 (PDB: 1ZKK) was docked into the EM density maps of the SET8–NCP and SET8–NCP^CENP-A complexes.

Figure S5.. Preparation of the NCP^apd and SET8 R188A/R192A mutant.

(A) The reconstituted NCP and NCP^apd with the 145-base pair Widom 601 DNA were analyzed by 20% SDS–PAGE with Coomassie Brilliant Blue staining. (B) The reconstituted NCP and NCP^apd with the 145-base pair Widom 601 DNA were analyzed by 6% non-denaturing PAGE with ethidium bromide staining. (C) Purified SET8 and SET8 R188A/R192A were analyzed by 18% SDS–PAGE with Coomassie Brilliant Blue staining.

Figure 3.. Mutational analyses for the SET8–NCP complex interaction.

(A) Gel shift assay of the NCP or the NCP^apd (the acidic patch-defective nucleosome) with SET8 and SET8 R188A/R192A. Double-stranded DNA and single-stranded DNA are denoted as dsDNA and ssDNA, respectively. The amount of SET8 was titrated. A double-stranded DNA 50-mer containing a trace amount of single-stranded 50-mer was included as competitor DNA. NCP (0.52 μM; lanes 1–3, 7–9) and NCP^apd (0.52 μM; lanes 4–6, 10–12) were mixed with SET8 (0, 1.0, and 2.1 μM; lanes 1 and 4, 2 and 5, and 3 and 6, respectively) or SET8 R188A/R192A (0, 1.0, and 2.1 μM; lanes 7 and 10, 8 and 11, and 9 and 12, respectively). (A, B) Quantification of the results in (A). The average % values of three independent experiments shown in Figs 3A and S6 are plotted against the SET8 concentration. (C) Time course methylation assay of the NCP with SET8 or SET8 R188A/R192A, and the NCP^apd with SET8. Lanes 1–3, 4–6, and 7–9 indicate results for the NCP with SET8, the NCP^apd with SET8, and the NCP with SET8 R188A/R192A, respectively. The experiments were repeated three times, and the reproducibility was confirmed (Fig S7).

Figure S6.. Replicated experiments of the gel shift assay in Fig 3A.

(A, B) Gel shift assay of the nucleosome core particle (NCP) or the NCP^apd with SET8 or SET8 R188A/R192A. Double-stranded DNA and single-stranded DNA are denoted as dsDNA and ssDNA, respectively. The amount of SET8 was titrated. The double-stranded DNA 50-mer containing a trace amount of single-stranded DNA 50-mer was included as competitor DNA. In addition to the experimental results presented in Fig 3A, the assay was independently repeated twice. NCP (0.52 μM; lanes 1–3, 7–9) and NCP^apd (0.52 μM; lanes 4–6, 10–12) were mixed with SET8 (0, 1.0, and 2.1 μM; lanes 1 and 4, 2 and 5, and 3 and 6, respectively) or SET8 R188A/R192A (0, 1.0, and 2.1 μM; lanes 7 and 10, 8 and 11, and 9 and 12, respectively). For comparison, lanes 13 and 14 indicate results for the dsDNA 50-mer with SET8, and lanes 15 and 16 indicate results for the 145 bp DNA with SET8.

Figure S7.. Replicated experiments of the methylation assay in Fig 3C.

(A, B) Methylation assay of the nucleosome core particle (NCP) with SET8 or SET8 R188A/R192A, and the NCP^apd with SET8. In addition to the experimental results presented in Fig 3C, the assay was independently repeated twice. Lanes 1–3, 4–6, and 7–9 indicate results for the NCP with SET8, the NCP^apd with SET8, and the NCP with SET8 R188A/R192A, respectively.

Figure 4.. Interaction of SET8 with the H4 N-terminal tails of the nucleosome core particle (NCP) and the NCP^CENP-A.

(A, B) Close-up views of the H4 N-terminal tail bound to the SET domain. (C) Comparison of the H4 N-terminal tail conformations in the SET8–NCP complex, the SET8–NCP^CENP-A complex, and the NCP (PDB: 5Y0C). Histone H3, CENP-A, and histone H4 are colored light blue, blue, and light green, respectively. Three structures are superimposed.