Inhibition of DNMT1 methyltransferase activity via glucose-regulated O-GlcNAcylation alters the epigenome
- Department of Diabetes Complications and Metabolism, Beckman Research Institute, City of Hope, United States
- Irell and Manella Graduate School of Biological Sciences, City of Hope, United States
- Integrated Mass Spectrometry Shared Resource, City of Hope Comprehensive Cancer Center Duarte, United States
- Cancer & Cell Biology Division, Translational Genomics Research Institute, United States
Figures
Figure 1 with 5 supplements
High glucose increases O-GlcNAcylation of DNMT1 in cell lines and primary cells.
(A) Hep3B cells were treated with glucose (5 mM or 25 mM) with or without Thiamet-G (TMG). Shown are immunoblots of collected lysates using antibody targeting O-GlcNAc and GAPDH (n = 3). (B) Lysates …
-
Figure 1—source data 1
Uncropped blot files of Figure 1A–D.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig1-data1-v2.zip
Figure 1—figure supplement 1
DNMT1 can be O-GlcNAcylated in Hep3B cells.
Hep3B cells were treated with Thiamet-G (TMG) or OSMI-4 (OSMI). Shown are representative immunoblots of treated Hep3B lysates performed with antibodies targeting O-GlcNAc and GAPDH and bar graphs of …
-
Figure 1—figure supplement 1—source data 1
Uncropped blot files of Figure 1—figure supplement 1.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig1-figsupp1-data1-v2.zip
Figure 1—figure supplement 2
DNMT1 can be O-GlcNAcylated in HepG2 cells and B-cells-derived lymphocytes.
(A) HepG2 cells were treated with Thiamet-G or OSMI. Shown are immunoblots of treated HepG2 lysates performed with immunoblots of immunoprecipitates performed with antibodies targeting O-GlcNAc (n = …
-
Figure 1—figure supplement 2—source data 1
Uncropped blot files of Figure 1—figure supplement 2A and B.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig1-figsupp2-data1-v2.zip
Figure 1—figure supplement 3
Global protein O-GlcNAcylation was induced with high concentrations of sucrose.
(A) Hep3B cells were treated with sucrose (5 mM or 25 mM). Shown are immunoblots of collected lysates using antibody targeting O-GlcNAc and GAPDH (n = 3). Lysates of Hep3B treated with sucrose were …
-
Figure 1—figure supplement 3—source data 1
Uncropped blot files of Figure 1—figure supplement 3A and B.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig1-figsupp3-data1-v2.zip
Figure 1—figure supplement 4
The enzymatic activity of OGT or OGA was not significantly changed by glucose treatment.
(A) OGT activity was measured with low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc) using the UDP-Glo Glycosyltransferase activity kit (Promega) (n = 3). (B) OGA activity was …
Figure 1—figure supplement 5
DNMT1 can be O-GlcNAcylated in primary cells (peripheral blood mononuclear cells [PBMCs]).
Pooled PBMCs treated with low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc) were immunoprecipitated with antibody targeting DNMT1 and immunoblotted for O-GlcNAc (n = 3). ***p<0.0001 …
-
Figure 1—figure supplement 5—source data 1
Uncropped blot files of Figure 1—figure supplement 5.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig1-figsupp5-data1-v2.zip
Figure 2 with 2 supplements
Identification of O-GlcNAcylated sites within DNMT1 by LC-MS/MS.
(A) Schematic drawing of the DNMT1 O-GlcNAc-modified region enriched from Hep3B cells based on mass spectrometry (MS) data and tandem MS (MS/MS) peaks. FTMS+ p NSI full MS (400.0000–1600.0000). …
-
Figure 2—source data 1
Uncropped blot files of Figure 2D.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig2-data1-v2.zip
Figure 2—figure supplement 1
Myc-DNMT1-WT in Hep3B cells can be O-GlcNAcylated.
Myc-DNMT1-WT were transfected into Hep3B cells. Shown are immunoblots of treated DNMT1-WT lysates performed with antibodies targeting Myc, DNMT1, Tubulin, H3, and O-GlcNAc (n = 3). Lysates from …
-
Figure 2—figure supplement 1—source data 1
Uncropped blot files of Figure 2—figure supplement 1.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig2-figsupp1-data1-v2.zip
Figure 2—figure supplement 2
Tandem MS/MS peaks of O-GlcNAcylated DNMT1 peptides.
Figure 3 with 3 supplements
Site-specific O-GlcNAcylation inhibits DNMT1 methyltransferase function.
For (A–D), bar graphs are of relative activity of DNA methyltransferase activity measured as absorbance from a DNMT activity/Inhibition ELISA kit and representative immunoblots of immunoprecipitates …
Figure 3—figure supplement 1
Site-specific O-GlcNAcylation at DNMT1 sites abrogate the function of methyltransferase and DNA loss of methylation at CpG island under high glucose/Thiamet-G (TMG) conditions.
(A) HepG2 cells were treated with low (5 mM, CTRL) or high glucose/TMG (25 mM, O-GlcNAc). Shown are absorbance of DNA methyltransferase activity performed with DNA methyltransferase activity kit (n …
Figure 3—figure supplement 2
The methylation loss by high glucose/Thiamet-G (TMG) conditions was not apparent in the DNMT1-S878A mutant.
(A) Each Hep3B and Myc-DNMT1-overexpressed mutant (DNMT1-WT or DNMT1-S878A) was treated with low (5 mM, CTRL) or high glucose/TMG (25 mM, O-GlcNAc), 5-aza or GSK-3484862 (negative control). Shown …
Figure 3—figure supplement 3
High glucose/Thiamet-G (TMG) conditions do not impact localization of DNMT1-WT or DNMT1-S878A.
(A) Hep3B cells expressed with DNMT1-WT and DNMT1-S878 cells were treated with low (5 mM, CTRL) or high glucose/TMG (25 mM, O-GlcNAc). (B) HepG2 cells expressed with DNMT1-WT and DNMT1-S878 cells …
-
Figure 3—figure supplement 3—source data 1
Raw fluorescence image files of Figure 3.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig3-figsupp3-data1-v2.zip
Figure 4 with 8 supplements
High glucose leads to loss of DNA methylation at cancer-specific partially methylated domains (PMDs).
(A) Density plot of DNA methylation for DNMT1-WT and DNMT1-S878A cells with either low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc). (B) Genome browser screenshot of DNA …
Figure 4—figure supplement 1
DNA loss of methylation by increased global O-GlcNAcylation decreases.
Density plot of DNA methylation for DNMT1-WT and DNMT1-S878A cells with either low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc). (A) Partially methylated domain (PMD) regions, (B) …
Figure 4—figure supplement 2
Global DNA methylation of wild type and DNMT1 mutants between low Fragments Per Kilobase of transcript per Million mapped reads (FPKM) regions and high FPKM regions (DNMT1-WT or DNMT1-S878A) that treated low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc) were determined by Nanopolish call methylation.
These are defined 'low FPKM' as containing <25% of FPKM regions per Mb window and 'high FPKM' as containing >75% of FPKM regions per Mb window.
Figure 4—figure supplement 3
Methylation changes from O-GlcNAcylation of DNMT1 in DNMT1-S878A mutant.
The loss of methylation was not observed in S878A mutant cells. (A, B) The distribution of each DNA methylation was divided by DNA replication timing.
Figure 4—figure supplement 4
DNA loss of methylation by increased global O-GlcNAcylation decreases around the transposable element (TE) regions.
(A) Boxplot represents the levels of DNA methylation on the TE regions or non-TE regions of each Myc-DNMT1-overexpressed mutants (DNMT1-WT or DNMT1-S878A) were treated with low (5 mM, CTRL) or high …
Figure 4—figure supplement 5
Loss of DNA methylation by increased global O-GlcNAcylation decreases.
Evolutionarily recent TEs are more likely to lose methylation than older elements in a variety of systems. Methylation percentages are shown around (A) LTR12C and (B) HERVH-int regions. The …
Figure 4—figure supplement 6
ZFP57 and ZNF605 demonstrate binding to a significant number of LTR12C elements present in liver cancer partially methylated domains (PMDs).
Genome browser screenshot of DNA methylation data LTR12C elements (blue) that demonstrate binding with ZFP57 and ZNF605 by glucose concentration.
Figure 4—figure supplement 7
Evolutionarily recent elements are less likely to lose methylation induced by O-GlcNAcylation of DNMT1.
The evolutionary distance between each group is from Perelman et al., 2011. Boxplot represents the DNA methylation by clades of the human genome (Homo sapiens to Haplorhini). CTRL, control; O-Glc, O-…
Figure 4—figure supplement 8
Only loci with >5× coverage were retained for analysis, comprising 90% of CpGs in the genome.
Shown are overall CpGs sites that detected with >5× coverage DNA methylation analysis using Nanopore technology PromethION sequencer. Each condition is biologically replicated.
Figure 5 with 4 supplements
High glucose-induced reactive oxygen species (ROS) and DNA damage cause apoptotic cell death in DNMT1-WT cells.
(A) Quantitative fluorescence image of ROS in DNMT1-WT and DNMT1-S878A cells with either low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc). (B) Quantitative fluorescence image of …
-
Figure 5—source data 1
Raw fluorescence image files of Figure 5A–C.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig5-data1-v2.zip
Figure 5—figure supplement 1
Heatmap representation of promoter DNA methylation for DNMT1-WT and DNMT1-S878A cells under low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc) at gene-poor and gene-rich regions.
(A) Whole promoters and (B) differentially methylated promoters.
Figure 5—figure supplement 2
DNA loss of methylation within promoter region by increased global O-GlcNAcylation impact different gene pathways.
(A) Gene Ontology of the top 11 pathways of hypomethylated promoter DNA by high glucose treatment. (B) Gene interaction map of hypomethylated promoter DNA of DNMT1-WT by high glucose treatment.
Figure 5—figure supplement 3
Quantitative analysis of human apoptosis related proteins in DNMT1-WT and DNMT1-S878A with either low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc) using Proteome profiler (n = 3).
PC, positive control; NC, negative control; 1. cleaved-caspase3; 2, cIAP1; 3, claspin; 4, phospho-p53(S15); 5, survivin. ***p<0.0001 by Student’s t-test; Data are represented as mean ± SD from three …
-
Figure 5—figure supplement 3—source data 1
Uncropped blot files of Figure 5—figure supplement 3.
- https://cdn.elifesciences.org/articles/85595/elife-85595-fig5-figsupp3-data1-v2.zip
Figure 5—figure supplement 4
Each Hep3B (Mock) and Myc-DNMT1-overexpressed mutant (DNMT1-WT or DNMT1-S878A) was treated with low (5 mM, CTRL) or high glucose/Thiamet-G (TMG) (25 mM, O-GlcNAc), 5-aza, or GSK-3484862 (negative control).
Shown are absorbance of 8-OHdG performed with DNA damage quantification kit (n = 3). **p<0.0005; ***p<0.0001 by Student’s t-test; ns, not significant; data are represented as mean ± SD from three …
Tables
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Gene (Homo sapiens) | DNMT1 | HUGO Gene Nomenclature Committee | HGNC:2976 | - |
| Cell line (H. sapiens) | Hep 3B2.1–7 | ATCC | HB-8064 | - |
| Cell line (H. sapiens) | Hep G2 | ATCC | HB-8065 | - |
| Transfected construct (H. sapiens) | pcDNA3/Myc-DNMT1 | Addgene | Plasmid #36939 | |
| Antibody | Anti-beta-actin (D6A8) (rabbit monoclonal) | Cell Signaling Technology | Cat# 8457 | WB (1:1000) |
| Antibody | Anti-alpha-tubulin (11H10) (rabbit monoclonal) | Cell Signaling Technology | Cat# 2125 | WB (1:1000) |
| Antibody | Anti-DNMT1 (60B1220.1) (mouse monoclonal) | Novus Biologicals | Cat# NB100-56519 | IP (1:250) WB (1:1000) |
| Antibody | Anti-DNMT1 (H-12) (mouse monoclonal) | Santa Cruz Biotechnology | Cat# sc-271729 | WB (1:1000) |
| Antibody | Anti-gamma H2A.X (rabbit polyclonal) | Abcam | Cat# ab11174 | IF (1:1000) |
| Antibody | Anti-GAPDH (rabbit monoclonal) | Abcam | Cat# ab181602 | WB (1:1000) |
| Antibody | Anti-H3 (rabbit polyclonal) | Abcam | Cat# ab1791 | WB (1:1000) |
| Antibody | Anti-Myc [Myc.A7] (mouse monoclonal) | Abcam | Cat# ab18185 | IP (1:250) WB (1:1000) |
| Antibody | Anti-O-GlcNAc (RL2) (mouse monoclonal) | Abcam | Cat# ab2739 | WB (1:1000) |
| Antibody | Aoat anti-rabbit IgG H&L (HRP) (goat polyclonal) | Abcam | Cat# ab6721 | WB (1:5000) |
| Antibody | Goat anti-mouse IgG H&L (HRP) (goat polyclonal) | Abcam | Cat# ab6789 | WB (1:5000) |
| Antibody | Goat anti-mouse IgG (H+L), Alexa 488 (goat polyclonal) | Invitrogen | Cat# A32723 | WB (1:1000) |
| Sequence-based reagent | DNMT1-T158A | This paper | PCR primers | agccccaggatt CGA aggaaaagcacc |
| Sequence-based reagent | DNMT1-T158A | This paper | PCR primers | ggtgcttttcct TCG aatcctggggct |
| Sequence-based reagent | DNMT1-T616A | This paper | PCR primers | gacaggggaccc GCG aaagccaccacc |
| Sequence-based reagent | DNMT1-T616A | This paper | PCR primers | ggtggtggcttt CGC gggtcccctgtc |
| Sequence-based reagent | DNMT1-S878A | This paper | PCR primers | gcgagattcgag GAG cctccaaaaacc |
| Sequence-based reagent | DNMT1-S878A | This paper | PCR primers | ggtttttggagg CTC ctcgaatctcgc |
| Sequence-based reagent | DNMT1-S878D | This paper | PCR primers | gcgagattcgag GAC cctccaaaaacc |
| Sequence-based reagent | DNMT1-S878D | This paper | PCR primers | ggtttttggagg GTC ctcgaatctcgc |
| Sequence-based reagent | DNMT1-T882A | This paper | PCR primers | tcccctccaaaa GCC cagccaacagag |
| Sequence-based reagent | DNMT1-T882A | This paper | PCR primers | ctctgttggctg GGC ttttggagggga |
| Commercial assay or kit | Q5 Site-Directed Mutagenesis kit | NEB | Cat# E0554S | - |
| Commercial assay or kit | EpiQuik DNMT Activity/Inhibition ELISA Easy Kit | EpiGentek | Cat# P-3139 | - |
| Commercial assay or kit | Global DNA methylation LINE-1 | Active Motif | Cat# 55017 | - |
| Commercial assay or kit | EpiQuik 8-OHdG DNA Damage Quantification Direct kit | EpiGentek | Cat# P-6003 | - |
| Chemical compound, drug | OSMI-4 | Selleck Chem | Cat# S8910 | - |
| Chemical compound, drug | Thiamet-G (TMG) | Cayman Chemical | Cat# 13237 | - |
| Software, algorithm | bedGraphToBigWig | Kent et al., 2010 | - | - |
| Software, algorithm | Clustal Omega | Sievers et al., 2011 | - | Version 1.2.4 |
| Software, algorithm | GraphPad Prism 9 | GraphPad | - | Version: 9.3.1 |
| Software, algorithm | Minimap2 | Li and Birol, 2018 | RRID:SCR_018550 | Version: 2.17 |
| Software, algorithm | Nanopolish | Loman et al., 2015 | RRID:SCR_016157 | Version: 0.11.1 |
| Software, algorithm | Python | Python Core Team | Version: 3.8.2 | |
| Software, algorithm | R | R Core Team | - | Version: 3.4.3 |
| Software, algorithm | Samtools | Lister et al., 2009 | RRID:SCR_002105 | Version: 1.10 |
| Other | UniProt | The UniProt Consortium | - | Database of protein information (https://www.uniprot.org/) |
| Other | DAPI stain | Invitrogen | D1306 | 1 µg/ml; marker for nuclear DNA |
Additional files
-
Supplementary file 1
Prediction of O-GlcNAcylated sites within DNMT1 using OGTSite.
- https://cdn.elifesciences.org/articles/85595/elife-85595-supp1-v2.docx
-
Supplementary file 2
List of total identified proteins.
- https://cdn.elifesciences.org/articles/85595/elife-85595-supp2-v2.docx
-
Supplementary file 3
List of post-translational modification (PTM) sites of human DNMT1.
- https://cdn.elifesciences.org/articles/85595/elife-85595-supp3-v2.docx
-
Supplementary file 4
List of deposited data in this study.
- https://cdn.elifesciences.org/articles/85595/elife-85595-supp4-v2.docx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/85595/elife-85595-mdarchecklist1-v2.docx
