Cla4 phosphorylates histone methyltransferase Set1 to prevent its degradation by the APC/CCdh1 complex

Abstract

H3K4 trimethylation (H3K4me3) is a conserved histone modification catalyzed by histone methyltransferase Set1, and its dysregulation is associated with pathologies. Here, we show that Set1 is intrinsically unstable and elucidate how its protein levels are controlled within cell cycle and during gene transcription. Specifically, Set1 contains a destruction box (D-box) that is recognized by E3 ligase APC/C^Cdh1 and degraded by the ubiquitin-proteasome pathway. Cla4 phosphorylates serine 228 (S228) within Set1 D-box, which inhibits APC/C^Cdh1-mediated Set1 proteolysis. During gene transcription, PAF complex facilitates Cla4 to phosphorylate Set1-S228 and protect chromatin-bound Set1 from degradation. By modulating Set1 stability and its binding to chromatin, Cla4 and APC/C^Cdh1 control H3K4me3 levels, which then regulate gene transcription, cell cycle progression, and chronological aging. In addition, there are 141 proteins containing the D-box that can be potentially phosphorylated by Cla4 to prevent their degradation by APC/C^Cdh1. We addressed the long-standing question about how Set1 stability is controlled and uncovered a new mechanism to regulate protein stability.

Fig. 1.. Set1 is degraded by the APC/C^Cdh1 complex.

(A) Set1 protein levels were cell cycle regulated. Yeast cells were arrested in G₁ phase using α-factor for 3 hours and then released into fresh YPD medium. Cells were harvested at indicated time points and then subjected to immunoblot analysis (left panel) and flow cytometry analysis (right panel). (B) Set1 protein levels were reduced in G₁ phase. Cells were arrested in G₁, S, and G₂-M using α-factor (10 μg/ml), 10 μM HU, and 8 μM nocodazole (Noco) for 3 hours, respectively. (C) Immunoblot analysis of Set1 in cells treated with 5 mM MG132 or 5 mM PMSF for 2 hours. Cells were treated with dimethyl sulfoxide (DMSO) and isopropanol as controls. (D) Diagram showing the D-box (226-RNSL-229) within WT Set1 and mutated D-box (226-KNSA-229) within Set1-KNSA mutant. (E) Immunoblot analysis of Set1 in WT and Set1-KNSA mutant. (F) Immunoblot analysis of Set1 in WT, cdh1Δ, and Cdh1 overexpression (CDH1 OE) cells. Cells transfected with empty vector (EV) were used as controls. (G) Immunoblot analysis of Set1 in WT and cdh1Δ mutant synchronized at G₁, S, and G₂-M phases by α-factor, HU, or Noco for 3 hours, respectively. (H) Analysis of Set1 ubiquitination in WT, cdh1Δ, and Set1-KNSA mutants. (I) In vitro ubiquitination assay showing that Cdh1 can mediate Set1 polyubiquitination. (J and K) Set1 interacted with Cdh1 as determined by Co-IP and reciprocal IP. Cells expressing untagged endogenous Cdh1 were used as a negative control. (L) In vitro Co-IP showing that Cdh1 preferentially bound WT Set1 but not Set1-KNSA mutant. The purified FLAG-Set1 and FLAG-Set1-KNSA were incubated with the purified recombinant His-Cdh1. Set1 was immunoprecipitated with anti-FLAG agarose. For (B), (C), and (E), data represent the mean ± SE of three biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 2.. Cla4 phosphorylates Set1-S228 to stabilize Set1.

(A) Set1 was phosphorylated at S228 inside cells. Left panel: Diagram showing the predicted Set1-S228 phosphorylation site within the D-box. Right panel: Immunoblot analysis of Set1-S228 phosphorylation in WT, Set1-S228A, and Set1-S228E mutants with anti–Set1-S228p antibody. (B) Immunoblot analysis of Set1 protein levels in WT, Set1-S228A, and Set1-S228E mutants. (C) Immunoblot analysis of Set1 stability in WT and Set1-S228A mutant. WT and Set1-S228A mutant were treated with cycloheximide (CHX) (0.1 μg/ml) for 0 to 40 min. (D) Liquid chromatography–mass spectrometry analysis of proteins copurified with untagged FLAG-Set1. The identified unique peptides and sequence coverage were listed. (E) Immunoblot analysis of Set1 in WT, cla4Δ, cka1Δ, skm1Δ, and sky1Δ mutants. (F and G) Set1 interacted with Cla4 as determined by Co-IP assay (F) and reciprocal IP (G). Cells expressing untagged endogenous Cla4 were used as a negative control in (F). (H) Analysis of the effect of Cla4 on Set1-S228 phosphorylation. (I) In vitro kinase assay showing the purified recombinant His-Cla4 phosphorylated Set1 peptide. (J) Set1 was phosphorylated at S228 by the purified recombinant His-Cla4 in vitro. (K and L) Immunoblot analysis of Cla4, Set1, and Set1-S228p in cells that were arrested by α-factor, HU, or Noco of 3 hours. For (B), (C), and (E), data represent the mean ± SE of three biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 3.. Set1-S228 phosphorylation prevents Cdh1-mediated Set1 degradation.

(A) Cdh1 preferentially interacted with Set1-S228A mutant as determined by Co-IP. FLAG-Set1 and FLAG-Set1-S228A were immunoprecipitated with anti-FLAG agarose and then incubated with purified recombinant His-Cdh1. (B) Isothermal titration calorimetry (ITC) assay showing that Cdh1 had a higher binding affinity toward Set1-S228 peptide than Set1-S228p peptide. (C) Cla4-catalyzed Set1-S228 phosphorylation reduced the interaction between Set1 and Cdh1. FLAG-Set1 and FLAG-Set1-S228A were immobilized with anti-FLAG agarose. The purified recombinant His-Cla4 was added to phosphorylate Set1 for 2 hours. After His-Cla4 was washed away, the immobilized Set1 was incubated with the purified recombinant His-Cdh1 for 1 hour. (D) Immunoblot analysis of Set1 in WT, cla4Δ, cdh1Δ, and cla4Δ cdh1Δ mutants. (E) Immunoblot analysis of Set1 in WT, Set1-S228A, Set1-S228E, cdh1Δ, cdh1Δ Set1-S228A, and cdh1Δ Set1-S228E mutants. (F) Analysis of the ubiquitination of Set1 in WT, Set1-S228A, and Set1-S228E mutants. (G) In vitro ubiquitination assay showing that Cla4-catalyzed Set1-S228 phosphorylation inhibited Cdh1-mediated Set1 polyubiquitination. The immobilized FLAG-Set1 was first phosphorylated by His-Cla4. After His-Cla4 was washed away, phosphorylated Set1 was then incubated with Cdh1, Ube1, Ubc4, adenosine triphosphate (ATP), and free ubiquitin. The unphosphorylated Set1 was used as a control. For (D) and (E), data represent the mean ± SE of three biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 4.. The PAF complex facilitates Cla4 to phosphorylate Set1-S228 and stabilize Set1.

(A) Immunoblot analysis of Set1 in WT and PAF complex mutants. (B) Immunoblot analysis of Set1-S228 phosphorylation in WT, paf1Δ, and rtf1Δ mutants. (C) IP–mass spectrometry analysis of proteins copurified with Set1 and Cla4. (D) Cla4 interacted with PAF complex as determined by Co-IP assay. (E) Loss of Paf1 reduced Set1-Cla4 interaction. (F) In vitro Co-IP assay showing that PAF complex enhances Set1-Cla4 interaction. The purified FLAG-Set1 was incubated with His-Cla4 with or without purified PAF complex (Ctr9-TAP). (G) PAF complex facilitates Cla4 to phosphorylate Set1-S228. The purified His-Set1 was incubated with His-Cla4 with or without purified PAF complex (Ctr9-TAP). (H) Immunoblot analysis of Set1 ubiquitination in WT and paf1Δ mutant. (I) Immunoblot analysis of Set1 in WT, paf1Δ, cdh1Δ, and paf1Δ cdh1Δ mutants. (J) Immunoblot analysis of Set1 in WT and cdh1Δ mutant when treated with 6AU for 0 to 45 min. (K) Box plots showing the normalized read intensity of Set1 occupancy at chromatin in cdh1Δ WT Set1, cdh1Δ Set1-S228A, and cdh1Δ Set1-S228E mutants. Centerlines denote medians, box limits denote 25th and 75th percentiles, and whiskers denote maximum and minimum values. Two-sided Wilcoxon test in R (package ggpval) was used for statistical analysis. (L) ChIP-seq tracks of Set1 occupancy at indicated genes in cdh1Δ WT Set1, cdh1Δ Set1-S228A, and cdh1Δ Set1-S228E mutants. (M) ChIP-qPCR analysis of Set1 occupancy at indicated genes in cdh1Δ WT Set1, cdh1Δ Set1-S228A, and cdh1Δ Set1-S228E mutants. (N) Nucleosomes facilitates Cla4 to phosphorylate Set1-S228. The purified His-Set1 was incubated with His-Cla4 and PAF complex with or without nucleosomes. (O) Diagram showing that Cla4-mediated Set1-S228 phosphorylation was facilitated by PAF complex and RNA Pol II. For (M), data represent the mean ± SE of three biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 5.. Cla4-catalyzed Set1-S228 phosphorylation promotes Set1-catalyzed H3K4me3.

(A) Immunoblot analysis of H3K4me1, H3K4me2, and H3K4me3 in WT, cla4Δ, cdh1Δ, and cla4Δ cdh1Δ mutants. (B) Immunoblot analysis of H3K4me1, H3K4me2, and H3K4me3 in WT, Set1-S228A, and Set1-S228E mutants. (C) Immunoblot analysis of H3K4me3 in cdh1Δ WT Set1, cdh1Δ Set1-S228A, and cdh1Δ Set1-S228E mutants. (D) Immunoblot analysis of H3K4me1, H3K4me2, and H3K4me3 in WT, cla4Δ, Set1-S228E, and cla4Δ Set1-S228E mutants. (E) Immunoblot analysis of H3K4me1, H3K4me2, and H3K4me3 in WT, paf1Δ, cdh1Δ, and paf1Δ cdh1Δ mutants. (F) Average metagene profiles of H3K4me3 in WT, Set1-S228A, and Set1-S228E mutants. (G) Venn diagram of genes that overlap among the Set1-peak, Paf1-peak, and Cla4-peak gene sets as determined by ChIP-seq. The numbers of binding sites are represented. (H) Box plots showing the normalized read intensity of H3K4me3 in WT, Set1-S228A, and Set1-S228E mutants. Centerlines denote medians, box limits denote 25th and 75th percentiles, and whiskers denote maximum and minimum values. Two-sided Wilcoxon test in R (package ggpval) was used for statistical analysis. (I) ChIP-seq tracks of Set1, Cla4, and Paf1 occupancy in WT cells as well as H3K4me3/H3 enrichment in WT, Set1-S228A, and Set1-S228E mutants at indicated genes. For (A) to (C), data represent the mean ± SE of three biological independent experiments. For (D), data represent the mean ± SE of four biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 6.. Cla4 and Cdh1 regulate gene transcription and chronological life span by affecting Set1 stability.

(A) ChIP-seq tracks of Set1, Cla4, and Paf1 occupancy at eight core histone genes. (B) RT-qPCR analysis of core histone gene transcription in WT, Set1-S228A, and Set1-S228E mutants. HHT2 encodes histone H3; HHF2 encodes histone H4; HTA2 encodes histone H2A; HTB2 encodes histone H2B. (C) Immunoblot analysis of core histones in WT, Set1-S228A, and Set1-S228E mutants. (D) Volcano plots for genes differentially expressed in WT and Set1-S228A mutant. Differential expression levels of aligned sequences were calculated using significant thresholds set at log₂(FC) ≤ −0.75, log₂(FC) ≥ 0.75, P < 0.05. Red color designates significantly down-regulated genes, and blue color designates significantly up-regulated genes. (E) Heatmap showing the transcription of eight core histone genes in WT and Set1-S228A mutant. (F) GSEA analysis of genes involved in cell cycle, DNA replication, and mismatch repair in WT and Set1-S228A mutant. (G) Box plots showing the normalized read intensity of H3K4me3 at subtelomeric regions on 16 chromosomes in WT, Set1-S228A, and Set1-S228E mutants. The subtelomeric regions refer to 0 to 50 kb from the nearest left or right telomeres. (H) RT-qPCR analysis of the transcription of telomere-proximal genes in WT, Set1-S228A, and Set1-S228E mutants. (I) Heatmap showing the transcription of the longevity pathway genes in WT and Set1-S228A mutant. (J) Chronological life-span analysis of WT, Set1-S228A, and Set1-S228E mutants. (K) Chronological life-span analysis of WT, cla4Δ, Set1-S228E, and Set1-S228E cla4Δ mutants. For (B), (C), (H), (J), and (K), data represent the mean ± SE of three biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 7.. Cla4 and Cdh1 co-regulate the stability of yeast proteins at the proteome level.

(A) Venn diagram of proteins that were predicted to contain P-box, interact with Cla4, and phosphorylate at P-box. (B) Proteins containing phosphorylated P-box were categorized by gene ontology (GO) annotations. Proteins that were confirmed to be regulated by Cla4 and Cdh1 were labeled in red. (C) Immunoblot analysis of the expression of Gin4, Smc3, Chd1, Sin3, Sas4, Sln1, and Atg9 in cells that were arrested in G₁, S, and G₂-M using α-factor (10 μg/ml), 10 μM HU, and 8 μM Noco for 3 hours, respectively. (D) Immunoblot analysis of Gin4 in WT, cla4Δ, cdh1Δ, and cla4Δ cdh1Δ mutants. The two P-boxes within Gin4 were indicated with red color. (E) Immunoblot analysis of Smc3 protein levels in WT, cla4Δ, cdh1Δ, and cla4Δ cdh1Δ mutants. The P-box within Smc3 was indicated with red color. For (D) and (E), data represent the mean ± SE of three biological independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001.