Lysine methylation within the globular domain of histone H3 by Dot1 is important for telomeric silencing and Sir protein association

Abstract

The amino-terminal histone tails are subject to covalent post-translational modifications such as acetylation, methylation, and phosphorylation. In the histone code hypothesis, these exposed and unstructured histone tails are accessible to a repertoire of regulatory factors that specifically recognize the various modified histones, thereby generating altered chromatin structures that mediate specific biological responses. Here, we report that lysine (Lys) 79 of histone H3, which resides in the globular domain, is methylated in eukaryotic organisms. In the yeast Saccharomyces cerevisiae, Lys 79 of histone H3 is methylated by Dot1, a protein shown previously to play a role in telomeric silencing. Mutations of Lys 79 of histone H3 and mutations that abolish the catalytic activity of Dot1 impair telomeric silencing, suggesting that Dot1 mediates telomeric silencing largely through methylation of Lys 79. This defect in telomeric silencing might reflect an interaction between Sir proteins and Lys 79, because dot1 and Lys 79 mutations weaken the interaction of Sir2 and Sir3 with the telomeric region in vivo. Our results indicate that histone modifications in the core globular domain have important biological functions.

Figure 1

Identification of histone H3 Lys 79 methylation. (A) MALDI-TOF MS of histone H3. Calf thymus histone H3 was digested with trypsin and resulting peptides chromatographed on a Poros 50 R2 (Perseptive Biosystems) reversed-phase micro-tip and analyzed by matrix-assisted laser-desorption/ionization reflectron time-of-flight mass spectrometry (MALDI-reTOF MS) using a Reflex III instrument (Bruker Daltonics), as described (Erdjument-Bromage et al. 1998; Winkler et al. 2002). An ion pair differing by 13.99 amu is indicated; the lighter ion corresponds to an H3 tryptic peptide (sequence shown). (B) Nano-ES MS/MS. Tryptic digest mixture was analyzed by electropsray ionization (ESI) MS using an API 300 triple quadrupole instrument (Applied Biosystems/MDS SCIEX), modified with a continuous-flow NanoES source, as described (Geromanos et al. 2000; Winkler et al. 2002). The precursor ion (m/z = 675.4), corresponding to the doubly charged (z = 2) version of peptide ion marked 1349.72 in A was selected for collision-induced dissociation (CID)-based MS/MS analysis. The fragment ion spectrum was inspected for y” ions and the deduced sequence is indicated (note that y” ion series originate at the carboxyl terminus and, therefore, read backwards). (C) Location of Lys 79 of histone H3 relative to α-helices (α1 and α2) and loop 1 (L1) based on the crystal structure of nucleosome.

Figure 2

Loss of Dot1 abolishes histone H3 Lys 79 methylation in vivo. (A) Western blots of yeast whole-cell extracts and recombinant yeast histone H3 were probed with either antibodies against the histone H3 tail (residues 1–20) or against a peptide containing methylated Lys 79. (B) Specificity of the anti-methyl Lys 79 antibody. Antibodies (1/10,000 dilution) were preincubated with either no peptide, 25 μg nonmethylated peptide, or 25 μg of methylated Lys 79 peptide for 30 min at room temperature before probing yeast whole-cell extracts for histone H3. (C) Western blot screen for histone H3 Lys 79 methylase. Whole-cell extracts from the indicated deletion strains were probed with antibodies against methylated Lys 4, methylated Lys 79, and unmodified H3. (D) Restoration of Lys 79 methylation by constructs containing the DOT1 gene. Whole-cell extracts from the indicated strains were probed with either anti-methyl Lys 4, or anti-methyl Lys 79 antibodies.

Figure 2

Loss of Dot1 abolishes histone H3 Lys 79 methylation in vivo. (A) Western blots of yeast whole-cell extracts and recombinant yeast histone H3 were probed with either antibodies against the histone H3 tail (residues 1–20) or against a peptide containing methylated Lys 79. (B) Specificity of the anti-methyl Lys 79 antibody. Antibodies (1/10,000 dilution) were preincubated with either no peptide, 25 μg nonmethylated peptide, or 25 μg of methylated Lys 79 peptide for 30 min at room temperature before probing yeast whole-cell extracts for histone H3. (C) Western blot screen for histone H3 Lys 79 methylase. Whole-cell extracts from the indicated deletion strains were probed with antibodies against methylated Lys 4, methylated Lys 79, and unmodified H3. (D) Restoration of Lys 79 methylation by constructs containing the DOT1 gene. Whole-cell extracts from the indicated strains were probed with either anti-methyl Lys 4, or anti-methyl Lys 79 antibodies.

Figure 3

Dot1 methylates Lys 79 of histone H3 in vitro. (A) Alignment of S-adenosyl methionine (AdoMet) motifs (Dlakic 2001), with conserved residues shaded in gray and Gly 398 of DOT1 highlighted with an asterisk. G MTase is rat glycine N-methylase (accession no. P13255); HMT1 is yeast hnRNP methylase (accession no. NP_009590); L11 MTase is E. coli. L11 ribosomal protein methylase (accession no. P28637); IsoD MTase is human isoaspartate O-methylase (accession no. AAH07501). Amino acid positions of the indicated motifs are shown for Dot1. (B) Dot1 mutations abolish AdoMet-binding activity. GST derivatives of wild-type and mutant (G398R or ΔGVG400-402) Dot1 proteins were incubated with [³H] AdoMet and cross-linked by UV irradiation. The resultant proteins were resolved on SDS-PAGE (right) and tritium label was detected by fluorography (left). (C) Dot1 methylates nucleosomal histone H3. GST derivatives of Set2 and Dot1 were incubated with oligonucleosomes from HeLa cells, and histones were analyzed by autoradiography or Coomassie staining. (D) Dot1 methylation is specific to nucleosomal histone H3. GST–Dot1 was incubated with the indicated substrates and analyzed as described in C. (E) Dot1 mutant proteins are unable to methylate nucleosomal histone H3. Analysis was performed as described in C. (F) Dot1 methylates nucleosomal histone H3 at Lys 79 residue. GST–Dot1 was incubated with a crude nucleosomal preparation from dot1 cells, and the resulting material was analyzed by Western blotting using antibodies against metylated Lys 79 and unmodified H3.

Figure 3

Dot1 methylates Lys 79 of histone H3 in vitro. (A) Alignment of S-adenosyl methionine (AdoMet) motifs (Dlakic 2001), with conserved residues shaded in gray and Gly 398 of DOT1 highlighted with an asterisk. G MTase is rat glycine N-methylase (accession no. P13255); HMT1 is yeast hnRNP methylase (accession no. NP_009590); L11 MTase is E. coli. L11 ribosomal protein methylase (accession no. P28637); IsoD MTase is human isoaspartate O-methylase (accession no. AAH07501). Amino acid positions of the indicated motifs are shown for Dot1. (B) Dot1 mutations abolish AdoMet-binding activity. GST derivatives of wild-type and mutant (G398R or ΔGVG400-402) Dot1 proteins were incubated with [³H] AdoMet and cross-linked by UV irradiation. The resultant proteins were resolved on SDS-PAGE (right) and tritium label was detected by fluorography (left). (C) Dot1 methylates nucleosomal histone H3. GST derivatives of Set2 and Dot1 were incubated with oligonucleosomes from HeLa cells, and histones were analyzed by autoradiography or Coomassie staining. (D) Dot1 methylation is specific to nucleosomal histone H3. GST–Dot1 was incubated with the indicated substrates and analyzed as described in C. (E) Dot1 mutant proteins are unable to methylate nucleosomal histone H3. Analysis was performed as described in C. (F) Dot1 methylates nucleosomal histone H3 at Lys 79 residue. GST–Dot1 was incubated with a crude nucleosomal preparation from dot1 cells, and the resulting material was analyzed by Western blotting using antibodies against metylated Lys 79 and unmodified H3.

Figure 4

Dot1 methylation of Lys 79 of histone H3 is required for telomeric silencing. (A) URA3-based telomeric silencing assay in which equal numbers of wild-type (WT) and mutant (dot1, K79A, K79P, K79Q) strains were spotted at 10-fold dilution on synthetic complete (SC) medium in the presence or absence of 0.12% 5-fluoro-orotic acid (FOA) and/or Uracil (Ura), and incubated at 30°C for 3 d. (B) ADE2-based telomeric silencing assay in which the indicated wild-type and mutant strains were patched on YPD medium with or without 40 mg/L adenine for 2 d and transferred to 4°C for 2 d. These strains (WZY42 background) contain an ADE2 reporter integrated adjacent to the telomere of chromosome V.

Figure 5

Catalytic activity of Dot1 is required for telomeric silencing. (A) Equal numbers of yeast cells of wild-type (WT) and dot1 mutant strains expressing the indicated Dot1 proteins were spotted at 10-fold dilution on synthetic complete (SC) medium in the absence of tryptophan (Trp) and/or Uracil (Ura) in the presence or absence of 0.12% FOA, and incubated at 30°C for 3 d. These strains (UCC1111 background) contain a URA3 gene integrated near the telomere of chromosome VII. (B) Analysis of Dot1 protein levels and histone H3 methylation at Lys 79 and Lys 4 in the above strains, as assayed by Western blotting with the indicated antibodies.

Figure 6

Dot1 methylation of Lys 79 of histone H3 is important for association of Sir proteins with telomeric regions. (A) Formaldehyde cross-linked chromatin from the indicated yeast strains was immunoprecipitated with affinity-purified Sir2 antibodies. Immunoprecipitated and input DNAs were quantified by PCR using primers that probe regions around 300 and 3500 bp away from chromosome end and the POL1-coding regions, which serves as a control for background signal. (B) Quantitation of relative Sir2 and Sir3 occupancy at TEL300 and TEL3500; the background signal from the POL1 region has been subtracted in all cases. Results presented are based on three independent experiments (standard deviation is shown). (C) Analysis of Sir2 and Sir3 protein levels in the indicated strains as assayed by Western blots of whole-cell extracts. (D) Lys 79 histone H3 methylation near the vicinity of telomere (right arm of chromosome VI) depends on Dot1 and Lys 79. Formaldehyde cross-linked chromatin from the indicated strains was immunoprecipitated with antibodies against methylated Lys 79 or Lys 4, and the resulting material quantitated by PCR.

Figure 6

Dot1 methylation of Lys 79 of histone H3 is important for association of Sir proteins with telomeric regions. (A) Formaldehyde cross-linked chromatin from the indicated yeast strains was immunoprecipitated with affinity-purified Sir2 antibodies. Immunoprecipitated and input DNAs were quantified by PCR using primers that probe regions around 300 and 3500 bp away from chromosome end and the POL1-coding regions, which serves as a control for background signal. (B) Quantitation of relative Sir2 and Sir3 occupancy at TEL300 and TEL3500; the background signal from the POL1 region has been subtracted in all cases. Results presented are based on three independent experiments (standard deviation is shown). (C) Analysis of Sir2 and Sir3 protein levels in the indicated strains as assayed by Western blots of whole-cell extracts. (D) Lys 79 histone H3 methylation near the vicinity of telomere (right arm of chromosome VI) depends on Dot1 and Lys 79. Formaldehyde cross-linked chromatin from the indicated strains was immunoprecipitated with antibodies against methylated Lys 79 or Lys 4, and the resulting material quantitated by PCR.