Histone H3 lysine 4 methylation is mediated by Set1 and required for cell growth and rDNA silencing in Saccharomyces cerevisiae

Abstract

Histone methylation is known to be associated with both transcriptionally active and repressive chromatin states. Recent studies have identified SET domain-containing proteins such as SUV39H1 and Clr4 as mediators of H3 lysine 9 (Lys9) methylation and heterochromatin formation. Interestingly, H3 Lys9 methylation is not observed from bulk histones isolated from asynchronous populations of Saccharomyces cerevisiae or Tetrahymena thermophila. In contrast, H3 lysine 4 (Lys4) methylation is a predominant modification in these smaller eukaryotes. To identify the responsible methyltransferase(s) and to gain insight into the function of H3 Lys4 methylation, we have developed a histone H3 Lys4 methyl-specific antiserum. With this antiserum, we show that deletion of SET1, but not of other putative SET domain-containing genes, in S. cerevisiae, results in the complete abolishment of H3 Lys4 methylation in vivo. Furthermore, loss of H3 Lys4 methylation in a set1 Delta strain can be rescued by SET1. Analysis of histone H3 mutations at Lys4 revealed a slow-growth defect similar to a set1 Delta strain. Chromatin immunoprecipitation assays show that H3 Lys4 methylation is present at the rDNA locus and that Set1-mediated H3 Lys4 methylation is required for repression of RNA polymerase II transcription within rDNA. Taken together, these data suggest that Set1-mediated H3 Lys4 methylation is required for normal cell growth and transcriptional silencing.

Figure 1

H3 lysine 4 (Lys4) methyl-specific antiserum recognizes H3 Lys4 methylation on peptides and histones. (A) Amino acid sequences of the different H3 peptides used in the ELISA analyses are shown in B. Underlined amino acids are artificial to the H3 sequence and used for coupling purposes. (B) enzyme-linked immunosorbent assay (ELISA) analyses of the α-H3(Lys4)Me antiserum for its recognition of and specificity toward unmodified (unmod_[1–8]) and modified H3 (K4Me_[1–8]) peptides. Peptide competitions were performed with 20 μg/mL H3 peptides K4Me_(1–20) or K9Me_(1–20). (C) Whole cell extracts were isolated from wild-type (Wt) and H3 K4R, K4A, and K9R yeast strains and immunoblotted with α-Me(lys4)H3 or α-acetyl H4 and α-H3 antiserum as a loading controls. Wt, K4R, K4A, and K9R were introduced into the MSY421 background.

Figure 2

Conservation and abundance of H3 lysine 4 (Lys4) and lysine 9 (Lys9) methylation; 1 μg of recombinant histone from Xenopus and 5 μg of total core histones from asynchronously growing Saccharomyces cerevisiae (S. cerevisiae), Tetrahymena thermophila (Tetrahymena), chicken, and human 293T cells (human) were resolved by 15% SDS-PAGE, transferred to PVDF membrane, and probed with α-Me(lys4)H3 or α-Me(Lys9)H3. Note that the Tetrahymena sample represents macronuclear histones. Identical samples were examined in parallel by Coomassie staining to show histone loading.

Figure 3

Set1 mediates H3 lysine 4 (Ly4) methylation. (A) Whole cell extracts were isolated from wild-type (Wt; MBY1198), MBY1217 (set1Δ [YHR119w]), Wt (BY4742), and yeast strains carrying individual disruptions in SET domain–containing genes: set2Δ (YJL168c), set3Δ (YKR029c), set4Δ (YJL105w), set5Δ (YPL165c), set6Δ (YHR207c), and set7Δ (YDR257c). MBY1198 is the isogenic Wt strain to set1Δ strain (MBY1217), and BY4742 is the isogenic Wt strain to set2Δ, set3Δ, set4Δ, set5Δ, set6Δ, and set7Δ strains. H3 Lys4 methylation was detected by immunoblotting with α-Me(lys4)H3 antiserum; H4 acetylation, with α-acetyl H4 antiserum; and H3, with α-H3 antiserum. (B) Schematic representation of the Set1 constructs used to rescue H3 Lys4 methylation shown in C. Numbers indicate the amino acids in Set1, and amino acids 938–1067 represent the SET domain of Set1. (C) Set1 constructs shown in B were transformed into a yeast set1Δ strain, MBY1217, and whole extracts of the transformed yeast were generated and probed with α-Me(Lys4)H3 or α-acetyl H4 and α-H3 antiserum as a loading controls.

Figure 4

Histone H3 mutations at lysine 4 (Lys4) and a set1Δ strain (MBY1587) show a slow-growth phenotype. (A) Cell growth between wild-type (Wt), and histone H3 K4R, K4A, and K9R yeast strains was analyzed by spot assays. Tenfold serial dilutions were spotted on YPD and grown for 3 d at 30°C or for 7 d at 14°C. (B) Fivefold serial dilutions of Wt, K9R, K4R, and set1Δ strains were spotted on YPD and grown for 2 d at 30°C. (C) Cell growth of set1Δ strains transformed with Set1 constructs described in Fig. 3B were determined by plating on SC-Ura medium for 3 d at 30°C. Wt, K4R, K4A, K9R, and set1Δ (MBY1587) were all generated in the MSY421 background.

Figure 5

Set1-mediated H3 lysine 4 (Lys4) methylation at rDNA is required for rDNA silencing. (A) Ty1 transposition patch assays were preformed to determine complementation of rDNA silencing within the set1Δ strain MBY1217 with Set1 constructs SET1_(1–1080), set1ΔSET_(1–900), set1_(780–1080), and set1_(900–1080). The ability of Set1 constructs to rescue (+) or not rescue (−) H3 Lys4 methylation in the set1Δ (MBY1217) strain is indicated by the patch assays (see Fig. 3C). (B) Chromatin from wild-type (Wt; MBY1198), set1Δ, and Wt (WZY42), and H3 K4R strains were isolated and immunoprecipitated with α-Me(Lys4)H3 or α-H3 di-acetyl antiserum. The set1Δ strain used was MBY1217, and the H3 K4R was introduced into the WZY42 background. The NTS region of rDNA present in the immunoprecipitated samples were amplified by PCR. Input PCR was shown for loading controls.