Fig 2: Depletion of EHMT1, EHMT2, and LMNB1 in an age-dependent manner AWestern blot analysis for EHMT1, EHMT2, and GAPDH in HDF cell lysates derived from various age groups.BQuantification of EHMT1 protein in HDFs of indicated age groups. One sample t-test (n = 3). For fetal versus 65, there was significant downregulation of EHMT1 protein (*P < 0.05); however, Fetal versus 18Y and Fetal versus 31Y did not show significant difference in EHMT1 levels. Error bars are represented as ±SD.CQuantification of EHMT2 protein expression in HDFs of indicated age groups. One sample t-test (n = 6). For Fetal versus 18Y (***P = 0.0004) and Fetal versus 65Y (***P = 0.0002). Error bars are represented as ±SD.DImmunostaining for EHMT1 (scale bar: 20 µm), EHMT2 and Pan H3 (scale bar: 40 µm) in fibroblasts from indicated age groups.E, FWestern blot and immunostaining (scale bar: 40 µm) analysis for H3K9me2 in indicated age groups.GRepresentative TEM images for nuclei of 18Y, 31Y, 40Y, and 65Y HDFs. Arrows indicate peripheral heterochromatin. Inset is the zoomed version of the same images (scale bar: 1 µm).HWestern blot analysis for LMNA/C and LMNB1 and in various age group HDFs. GAPDH was used as an internal control.IQuantification of LMNB1 protein expression in HDFs of indicated age groups. One-sample t-test (n = 6). For Fetal versus 18Y (**P = 0.0069) and Fetal versus 65Y (**P = 0.0016). Error bars are represented as ±SD.J–LReduced levels of methylated LMNB1 during physiological aging. Nuclear extracts from indicated age groups were subjected to IP using LMNB1 antibody. IPed material was divided into halves, blot 1 (left, J) was probed with anti-LMNB1 antibody, blot 2 was probed with anti-Methyl-K (middle, K) antibody. Anti-Methyl-K blot was reprobed with LMNB1 antibody to demonstrate the methyl band indicated is indeed LMNB1 (right, L). 80 µg of fetal HDFs derived nuclear lysate was used as input control. Dotted line indicates the spliced lane from the different exposure of the same gel (see Source data for this figure). Source data are available online for this figure.

Fig 3: EHMT1, EHMT2, and LMNB1 are members of the same complex AEHMT1 interacting proteins identified by mass spectrometric analysis with details indicating coverage and peptide score.BSequential IP in HEK293 cells demonstrating EHMT1, EHMT2, and LMNB1 are a part of the same complex. The dotted line indicates spliced lane from two different exposures of the same gel (source data file linked).CLMNB1 interacts with EHMT1 via SET domain. Recombinant GST or GST-LMNB1 was incubated with Ni-NTA bound His-EHMT1 SET protein. Postwashing eluents were loaded for immunoblotting using GST or His antibody. Recombinant pure proteins GST-LMNB1 (lane 1), GST (lane 2), and EHMT1-SET (lane 3) were used as controls.DVenn diagram showing unique and overlapping reads obtained from EHMT1 and LMNB1 ChIP sequencing.EComposite profile of EHMT1 and LMNB1 read density around the transcription start site (TSS).FGenomic distribution of EHMT1 and LMNB1 peaks. The majority of binding sites obtained were enriched in an intronic region or distal regions from a gene.G, HRepresentative figure showing normalized ChIP-seq read density (above 1.5-fold over expected) of EHMT1 and LMNB1 in 1 MB bin for chromosomes 1 and 9. Source data are available online for this figure.

Fig 4: GLP/G9a mediate SPOP mutation-induced hypermethylation independent of its HMTase activity.a Pearson correlation of 5mC and GLP/G9a IHC scores in 84 PCa patient specimens. b–d Western blots of WCL from 22Rv1 cells infected with lentivirus expressing indicated constructs and shRNAs (b). Representative IFC images of 5mC and HA-SPOP F133V staining are shown in (c), and ImageJ was used to quantify the optical density (OD)/nuclear area (pixel) of 5mC signals (d). Data shown means ± SD (n = 50 cells/group). Scale bar, 10 μm. e, f 22Rv1 cells infected with lentivirus expressing indicated constructs and shRNAs were harvested for Western blots with indicated antibodies (e) or used for MTS assay to measure cell proliferation (f). Data shown means ± SD (n = 5 replicates/group). g-i 22Rv1 cells infected with lentivirus expressing EV and SPOP F133V and treated with 2 µM UNC0642 for 24 h for western blots (g). Representative IFC images of 5mC and HA-SPOP F133V staining are shown in (h) and 5mC signals in each group were quantified using ImageJ (i). Data shown means ± SD (n = 50 cells/group). Scale bar, 10 μm. Statistical significance was determined by unpaired two-tailed Student’s t test in (a, d, f, i). Experiments in (b, e, g) were repeated twice. Source data are provided as a Source Data file.

Fig 5: SPOP-GLP axis regulates DNA methylation and TSG expression.a A Jitter plot comparison of overall beta value in 22Rv1 cells expressing empty vector (EV) or SPOP F102C mutant with four replicates per group. b Venn diagram showing the overlap between hypermethylated CpGs from EPIC 850K array (F102C vs EV) in 22Rv1 cells (n = 514056, FDR < 0.05) and hypermethylated CpGs from 450K (SPOP mutated vs WT) in the TCGA database (n = 32016, FDR < 0.05). Two-tailed Hypergeometric test, P = 1.07E−07. c Venn diagram showing the overlap among genes with increased DNA methylation in SPOP F102C expressing 22Rv1 (EPIC 850K) and SPOP mutant patient sample (450K), genes downregulated in SPOP-mutated prostate tumors in the TCGA cohort, and tumor suppressor genes reported previously (https://bioinfo.uth.edu/TSGene/index.html?csrt=17897587749110957416). Two-tailed Hypergeometric test, P = 9.01E−05. d Heat map shows the differential expression of 115 overlapped genes identified in (c) in SPOP-mutated compared to SPOP WT tumors in the TCGA cohort. e Heat map shows the methylation alterations in the CpGs mapped to FOXO3, GATA5 and NDRG1 gene loci in F102C-expressing compared to EV control 22Rv1 cells. f MeDIP-qPCR analysis of DNA methylation in the indicated gene promoter in EV or SPOP mutant expressing 22Rv1 cells. Data shown means ± SD (n = 3 replicates/group). g, h Analysis of DNA methylation in the indicated gene promoters in 22Rv1 cells infected with lentivirus expressing indicated plasmids and shRNAs using MeDIP-qPCR (g) and methylation-specific PCR (h). Data shown means ± SD (n = 3 replicates/group). i, j Measurement of FOXO3, GATA5, NDRG1 mRNA and protein expression in 22Rv1 cells expressing indicated plasmids and shRNAs using RT-qPCR (i) and western blot (j). Data shown means ± SD (n = 3 replicates/group). k, l Analysis of FOXO3, GATA5 and NDRG1 mRNA and protein expression by RT-qPCR (k) and western blot (l), respectively in EV and F133V-expressing 22Rv1 cells treated with or without 2 μM 5-AzaC for 48 h. Data shown means ± SD (n = 3 replicates/group). Statistical significance was determined by unpaired two-tailed Student’s t test in (f, g, i, k). Experiments in (j, l) were repeated twice. Source data are provided as a Source Data file.