(A) Recombinant human histone H3 and purified histones from S. cerevisiae (yeast) and HeLa cells (human) were probed with α-H3K56me1 antibody. Equivalent amounts of proteins were loaded as shown by Coomassie blue staining. (B) Histones were isolated from HeLa cells and analyzed by mass spectrometry. Full mass spectrum is shown for unmodified (pr-YQK_prSTELLIR, 681 m/z) and monomethylated (pr-YQK_prme1STELLIR, 688 m/z) peptides from HPLC separated elutions. Accurate mass on the peptide at 688.894 m/z clearly indicates that the modification is a monomethylation mark. (C) Tandem mass spectrum of the monomethylated peptide (pr-YQK_prme1STELLIR) shows that the monomethylation modification is localized to the K56 residue. See also Figure S1.

(A) HeLa cells were transfected with SET8, Ezh2, SETDB1/SETDB2, Suv39H1/Suv39H2 siRNAs or non-targeting siRNA (Control). Immunoblots showing that the expression of endogenous methyltransferases is decreased, while H3K56me1 level remains unchanged. (B) HeLa cells were transfected with non-targeting siRNA (Control) or G9a siRNAs. Immunoblot showing that the expression of endogenous G9a is decreased. Tubulin served as an internal control for protein loading. H3K56me1 is decreased in HeLa cells transfected with G9a siRNA compared to control, while endogenous histone H3 expression remains unchanged. (C) Nuclear extracts prepared from G9a^+/+ and G9a^−/− ES cells were blotted with α-H3K56me1 antibody. G9a^−/− ES cells showed a significant reduction of H3K56me1. Ponceau S staining of histones served as a loading control. (D) G9a^+/+ and G9a^−/− ES cells were pre-extracted with 0.5% Triton X-100 and fixed with methanol. Immunofluorescence staining was performed using H3K56me1 antibody. G9a^+/+ ES cells showed extensive H3K56me1 staining while the G9a^−/− ES cells exhibited a dramatic loss of H3K56me1 signal. Original magnification, ×64. (E) G9a HMTase activity for H3K56 was measured by in vitro HMTase assay using recombinant G9a. Substrate is recombinant H3 with tri-methyl lysine analog at K9 (H3K_C9me3). Immunoblot showing that H3 becomes monomethylated at K56 when incubated with G9a. Ponceau S staining indicates that equal amounts of histone H3 were used in each reaction. See also Figure S2.

(A) Nuclear extracts from human (HeLa cells) or mouse (Raw264.7 cells) were used in peptide pull-down assays with a series of biotinylated H3 peptides, either unmodified (unmod.) or methylated as indicated. PCNA was pulled down specifically by the H3K56me1 peptide (top and middle panels). None of the peptides associated with Asf1a/1b (bottom panels). (B) Peptide pull-down assay with purified recombinant PCNA and H3 peptides either unmodified or methylated as indicated. Recombinant PCNA was pulled down specifically by the H3K56me1 peptide. (C) Histone octamers with unmodified histone H3.1 (H3.1 unmod.) or H3.1 K_C56me1 were assembled for the pull-down experiment with purified PCNA. It should be noted that the octamers would dissociate to H3/H4 tetramers and H2A/H2B dimers under our binding buffer conditions (0.5M NaCl) (Eickbush and Moudrianakis, 1978). PCNA was associated with the H3.1 K_C56me1-H4 tetramer (top panel). Histone H3.1-H4 tetramer was equally immunoprecipitated by histone H3 antibody (bottom panel) and equal amounts of PCNA were used in the experiment (middle panel). (D) Nuclear fractions were prepared from synchronized HeLa cells and immunoprecipitated with α-PCNA antibody. Immunoblot showing that PCNA associated with H3K56me1 in G1, whereas the association decreased dramatically during S phase. Flow cytometry profiles are shown above the panels. (E) HeLa cells were incubated with EdU for 30 min, fixed with methanol without Triton X-100 pre-extraction and immunofluorescence confocal microscopy in combination with in situ PLA was performed. The incorporated EdU was stained using Click-iT^® EdU Alexa Fluor^® 488 Imaging Kit. Nuclei were stained with DAPI. Red dots in the confocal microscopic images indicate physically interacting H3K56me1/PCNA complexes. Note that H3K56me1/PCNA complex predominantly occurs in non-S phase cells. S phase cells incorporate EdU as indicated by arrowheads. Scale bar, 10 μm. Lower right panel: Quantification of the PLA signal in HeLa cells in two independent experiments (EdU-, n = 58; EdU+, n = 32; Mean ± s.e.m., ***, P < 0.001). (F) Quantification of the mean number of PLA signals in synchronized HeLa cells in at least two independent experiments (G1, n = 93; G1/S, n = 150; S, n = 193; G2/M, n = 79; Mean ± s.e.m., ***, P < 0.001). Flow cytometry profiles are shown below the bar graph panel. See also Figure S3.

(A) Dual immunofluorescence staining of HeLa cells was performed using α-H3K56me1 and α-Pol II antibodies. H3K56me1 shows no obvious colocalization with Pol II. (B) Immunofluorescence staining showing that in HeLa cells H3K56me1 is distributed extensively throughout the nucleus, but is excluded from DAPI dense regions. (C) Dual immunofluorescence staining of HeLa cells was performed using α-H3K56me1 and α-HP1α antibodies. H3K56me1 shows no obvious colocalization with α-HP1α. Pearson’s correlation coefficients (Rr) are calculated using ImageJ software. Data represent mean ± s.e.m., (A, n = 105; B, n = 89; C, n = 72). Cells were fixed with methanol without Triton X-100 pre-extraction. Scale bar, 10 μm.

(A) Loss of G9a leads to delayed passage through S phase. G9a^+/+ and G9a^−/− ES cells were arrested in M phase by nocodazole treatment for 16 hrs, then released into the cell cycle. Cells were collected at indicated time points, fixed, stained with propidium iodide (PI), and DNA content was analyzed by FACS. Histograms of DNA content are shown (Left panels). The percentage of cells in G₁, S, and G₂/M phases of the cell cycle at each time point is shown (Right panel). (B) Inefficient DNA replication in G9a^−/− ES cells was detected by EdU incorporation. Cells were pulse labeled with EdU for 5 min, fixed, stained for EdU and PI, and analyzed by FACS. Left panels: Plots of EdU incorporation against DNA staining with PI are shown. Note that there is a partial “collapse” of the arc of S phase population of G9a^−/− ES cells. Right panels: Quantitation analysis showed that the median intensity of the incorporated EdU was decreased in G9a^−/− ES cells compared to G9a^+/+ ES cells. Data are presented as mean±SEM from 6 experiments, **, P < 0.01. (C) Immunoblots show that the phosphorylation of Chk1, H2AX is not activated in G9a^−/− ES cells. HEK293 cells treated with 0.01% MMS served as a positive control for checkpoint activation. (D) Lack of H3K56me1 leads to S-phase accumulation. HeLa cells were transfected with FLAG-tagged histone H3.1 or H3.3 wild type (WT) or K56 mutants (K56A, K56R, K56Q) or H3.1 K9A. The cells were fixed, incubated with anti-FLAG antibody, followed by secondary antibody conjugated to Alexa Fluor 647. The DNA content against PI staining of FLAG-tagged positive cells was analyzed by FACS and the percentage of cells in each stage of the cell cycle showing an accumulation in S phase when expressing H3.1K56 mutants. Data are presented as mean±SEM from at least 3 experiments, *, P < 0.05, **, P < 0.01. (E) Lack of H3K56me1 impairs PCNA binding to chromatin. The chromatin fraction was extracted from HeLa cells transfected with FLAG-tagged H3.1 WT or H3.1 K56A mutant and the levels of PCNA, RPA70 and RFC1 were measured by western blot. Immunoblot showing that the chromatin-bound PCNA is decreased in cells expressing H3.1K56A mutant. Neither chromatin-associated RPA70 nor RFC1 was decreased (left upper panels). Ponceau S staining showing that both wild type H3.1 and H3.1K56A incorporated into chromatin equally (left bottom panel). The levels of PCNA, RFC1, RPA70 in the whole cell lysate were unchanged (right panels). (F) Immunoblots showing that chromatin-bound PCNA is decreased in HeLa cells expressing H3.1K56 mutants but not H3.1K9A or H3.3K56 mutants. The level of PCNA in the whole cell lysate was unchanged (middle panel). Ponceau S staining showing that both wild type H3 and mutant H3 incorporated into chromatin equally (bottom panel).