PMC

Senescence-associated heterochromatic foci (SAHF) are enriched for H4K20me3. a Immunofluorescent images of control (CON) and OIS cells co-stained with antibodies against H4K20me3 and PML. b Cells from a stained with antibodies to H4K20me3 and 53BP1. c Cells from a stained with antibodies to H4K20me3 and γ-H2AX. d Cells from a co-stained with 4′,6-diamidino-2-phenylindole (DAPI) and an antibody against H4K20me3. e Linescan intensity analysis of H4K20me3 and DAPI fluorescence intensity profiles along the arrows indicated in d. f Cells from a stained with antibodies to H4K20me3 and H3K9me3 and DAPI. g Linescan intensity analysis of H4K20me3 and H3K9me3 and DAPI fluorescence intensity profiles along the arrow indicated in f

Elevated H4K20me3 levels reinforce a stable proliferative arrest in senescent cells. a Western blot of indicated proteins from whole cell extracts of IMR90 cells 12 days after infection with vector control (CON), Myc-tagged SUV420H1 (H1), Myc-tagged SUV420H2 (H2) or H-RASG12V (OIS). b Western blot of histone H4 and indicated H4K20 modifications from whole cell extracts described in a, normalized for total histone H4 content. c Immunofluorescent images of Myc-tagged SUV420H1 or SUV420H2 and H4K20me3 in CON, H1, and H2 IMR90 cells from a. d Growth curves expressed as cumulative population doublings for CON, H1, and H2 IMR90 cells measured for 38 days after infection. e Western blot of indicated proteins from whole cell extracts of CON, H1, and H2 IMR90 cells from d 3 and 38 days after infection. f Western blot of indicated proteins from whole cell extracts of CON and H2 IMR90 cells collected 5, 10, or 15 days after infection with either vector control (CON) or H-RASG12V (OIS); (s) and (l) denote short and long autoradiographic exposures, respectively

H4K20me3 is frequently enriched at ZNF and repeat class genes in senescent cells. a Observed overlap/expected overlap (fold log2; enrichment compared with random) between base pairs in RS and OIS H4K20me3 DiffBind peaks, RS H4K16ac DiffBind peaks, DNA hypermethylated in RS regions, and DNA hypomethylated in RS regions and base pairs covered by specified genomic features. b Observed overlap/expected overlap (fold log2; enrichment compared with random) of RS H4K20me3 DiffBind peaks (in base pairs) and transposable elements (TEs). The x-axis shows TE evolutionary order ranked from most ancient to most recent, as defined previously []. c As in b but using OIS H4K20me3 DiffBind peaks. d All coding genes ranked by RS and OIS H4K20me3 enrichment (read count) at gene body normalized to histone H4. e Gene families represented among the 500 gene bodies most highly enriched for H4K20me3 in RS cells. f As in c but for H4K20me3 in OIS cells. g Gene families represented among all genes in the genome. h UCSC Genome Browser view of histone H4 and H4K20me3 ChIP-seq reads aligned along a 500-kb segment of chromosome 19 in proliferating (PRO) and RS cells

H4K20me3 genic enrichment in senescent cells occurs at lowly expressed and unexpressed genes. a Mean H4K20me3 enrichment (read count) normalized to histone H4 in proliferating (PRO) cells across the gene body (transcription start site to transcription end site) of all non-repeat class coding genes divided into quartiles on the basis of expression level (Q1 = highest expression, Q4 = lowest expression, UN = unexpressed (FPKM = 0)). b As in a but evaluating H4K20me3 enrichment in control (CON) cells. c As in a but evaluating H4K20me3 enrichment in RS cells. d As in a but evaluating H4K20me3 enrichment in OIS cells. e Scatter plot according to the R function “densCols” comparing H4K20me3 ChIP enrichment difference (RS-PRO) at gene bodies normalized to histone H4 and expression level (FPKM) for all genes in PRO cells (control for RS cells). Colors are determined by the density of data points at each point of the chart, moving from low density (blue) to high density (red). f As in e but comparing H4K20me3 ChIP enrichment difference (OIS-CON) and expression level (FPKM) for all genes in CON cells (control for OIS cells)

ChIP-seq confirms enrichment of H4K20me3 at SAHF in senescent cells. a H4K20me3 enrichment at telomeric repeat sequences relative to DNA input (left) and histone H4 (right) in proliferating (PRO; blue) and RS (red) cells. The % mapped H4K20me3/% mapped control was calculated separately for each antibody and control. Mean value (n = 2) was plotted with standard error of the mean (SEM). b Quantitative PCR of H4K20me3 ChIP enrichment at 17p telomeres normalized to H4K20me3 ChIP enrichment at the β-globin locus in PRO and RS cells; error bars represent SEM of three experiments (two experiments with the Millipore 04–079 antibody and one experiment with the Cell Signalling 5737 antibody). c Quantitative PCR of H4K20me3 ChIP enrichment at 18q telomeres normalized to H4K20me3 ChIP enrichment at the β-globin locus in PRO and RS cells; error bars represent SEM of three experiments (as in panel b). d Total number of overlapping H4K20me3 peaks identified with both antibodies (intersection) in PRO and RS cells and significantly different (false discovery rate (FDR) <0.01) peaks between PRO and RS cells determined by DiffBind. e Total number of base pairs comprising H4K20me3 peaks identified with both antibodies (intersection) in PRO and RS cells and significantly different (FDR <0.01) peaks between PRO and RS determined by DiffBind. f Number of H4K20me3 DiffBind peaks from d that increase and decrease in RS cells relative to PRO cells. g Observed overlap and expected overlap (enrichment compared to random) between base pairs covered by RS H4K20me3 DiffBind peaks with base pairs covered by H3K9me3 peaks in senescent cells (empirical p < 0.001). h Observed overlap and expected (enrichment compared with random) overlap between base pairs covered by RS H4K20me3 peaks (intersection of both antibodies) with base pairs covered by H3K9me3 in senescent cells (empirical p < 0.001). i Mean RS and OIS H4K20me3 (normalized to histone H4) and OIS H3K9me3 (normalized to input) enrichment profiles (read count) at a composite H3K9me3 peak. j Observed/expected overlap (log2 fold enrichment compared with random) between base pairs covered by RS and OIS H4K20me3 peaks, RS H4K16ac peaks, DNA hypermethylated in RS regions, and DNA hypomethylated in RS regions with H3K9me3-marked late-replicating regions, H3K9me3-marked not late-replicating regions, and late-replicating regions not marked by H3K9me3. k Mean difference (RS − PRO) in H4K20me3 enrichment, H4K16ac enrichment, and percentage of methylated CpGs at a composite H3K9me3-marked late replicating region

Senescent cells accumulate elevated levels of H4K20me3 in vitro and in vivo. a Quantification of SA β-galactosidase-positive (SA β-gal+) IMR90 cells 3–12 days after infection with either empty vector control (CON) or H-RASG12V virus. b Cells from a were pulse labeled with 5-ethynyl-2′-deoxyuridine (EdU) and positive cells scored. c Western blot of indicated proteins in whole cell extracts of cells from a. d Western blot of H4K20me3 and histone H4 in whole cell extracts from c, normalized for total histone H4 content. e Western blot of histone H4 and indicated H4K20 modifications from whole cell extracts of proliferating (PRO), replicative senescent (RS), control-infected proliferating (CON) and H-RASG12V-infected senescent (OIS) IMR90 cells, normalized for total histone H4 content. f Western blot of H4K20me3 and histone H4 from whole cell extracts of proliferating (PRO), RS and quiescent (QUI) IMR90 cells, normalized for total histone H4 content; (s) and (l) denote short and long autoradiographic exposures, respectively. Experiments in a–f are representative of at least five similar experiments. g Western blot of SUV420H2 and GAPDH from whole cell extracts of PRO, RS, CON, and OIS cells. h Immunofluorescent images of H4K20me3 staining in CON and OIS cells 12 days after infection. i Quantitative image analysis of H4K20me3 immunofluorescence in CON and OIS cells (181 CON and 129 OIS cells were scored). j Relative percentages of the different methylation states of H4K20 in PRO and RS cells as determined by quantitative mass spectrometry; error bars represent standard error of the mean. k Immunohistochemical images of human melanocytic nevus (N) and overlaying epidermis (E) stained with antibodies against Melan-A and H4K20me3. The arrow indicates a non-nevus epidermal melanocyte. Data are representative of at least ten different human nevi

Reintroduction of SUV420H2/H4K20me3 attenuates the proliferative capacity of SUV420H2/H4K20me3-deficent HT1080 tumor cells. a Western blot of SUV420H2 and β-actin from whole cell extracts of proliferating (PRO) and RS IMR90 cells and HT1080 cells. b Western blot of H4K20me3 and histone H4 from whole cell extracts of PRO and RS IMR90 cells and HT1080 cells. c SUV420H2 expression in various human cancers relative to the reference population (either all tumors that are diploid for the gene in question or, when available, normal adjacent tissue). Data obtained from the cBioPortal for Cancer Genomics. X-axis, cancer type: (1) acute myeloid leukemia, (2) acute myeloid leukemia, (3) bladder urothelial carcinoma, (4) bladder urothelial carcinoma, (5) brain lower grade glioma, (6) breast invasive carcinoma, (7) cervical squamous cell carcinoma and endocervical adenocarcinoma, (8) colon and rectum adenocarcinoma, (9) glioblastoma multiforme, (10) glioblastoma, (11) head and neck squamous cell carcinoma, (12) head and neck squamous cell carcinoma, (13) kidney chromophobe, (14) kidney renal clear cell carcinoma, (15) kidney renal clear cell carcinoma, (16) kidney renal papillary cell carcinoma, (17) liver hepatocellular carcinoma, (18) lung adenocarcinoma, (19) lung adenocarcinoma, (20) lung squamous cell carcinoma, (21) ovarian serous cystadenocarcinoma, (22) pancreatic adenocarcinoma, (23) prostate adenocarcinoma, (24) sarcoma, (25) skin cutaneous melanoma, (26) stomach adenocarcinoma, (27) thyroid carcinoma, (28) uterine corpus endometrioid carcinoma. Y-axis, difference in SUV420H2 expression (Z score, normal/cancer). d Western blot of indicated proteins from whole cell extracts of HT1080 cells infected with vector control (CON) or MYC-tagged SUV420H2 (H2). e CON and H2 HT1080 cells were pulse labeled with 5-BrdU, fixed, and stained with propidium iodide. Fluorescence activated cell sorting (FACS) analysis to determine cell cycle distribution based on propidium iodide. f FACS analysis of cells from e to determine proportion of cells in G1, S, and G2/M phases based on 5-BrdU and propidium iodide; error bars represent standard deviation (SD; n = 2). g Growth curves expressed as log cumulative cell number for CON and H2 HT1080 cells measured for 32 days after infection. h Mean volumes of tumors formed after subcutaneous injection of CON or H2 HT1080 cells into CD-1 nude mice (Crl:NU-Foxn1 ^nu); n = 3 mice/group, error bars represent SD (representative of two independent experiments). i Maximum growth rates for tumors formed by CON or H2 HT1080 cells in h expressed as mm³/day; n = 3 mice/group, error bars represent SD (representative of two independent experiments)