Single-molecule tracking in living cells reveals dynamics of DNMT1 through the cell cycle and its redistribution upon drug treatment

DNMT1 is a methyltransferase that restores 5-methylcytidine marks on newly replicated DNA and is required for maintaining epigenetic inheritance. Using Halo-tagged DNMT1 and highly inclined thin illumination (HiLo) microscopy, we show that DNMT1 mobility in living human cells changes under a variety of conditions. Consistent with its role in methylating DNA, DNMT1 molecules become increasingly bound to chromatin in the S-phase of the cell cycle, but surprisingly a 4% increase in chromatin bound DNMT1 is sufficient to maintain DNA methylation. In G2 phase, however, DNMT1 is released from DNA and becomes even more dynamic. Upon treatment with small molecule inhibitors, GSK-3484862 (GSK) and 5-azacytidine (5-azaC), the in vivo DNMT1 dynamics are greatly altered. Unexpectedly, treatment of cells with GSK, a non-covalent inhibitor, causes binding of DNMT1 to chromatin similar to that observed upon treatment with 5-azaC, a covalent inhibitor. 5-azaC inhibition of DNMT1 dynamics occurs during the S phase of the cell cycle. Collectively, our work using live cell single molecule imaging quantifies the molecular dynamics of DNMT1 and how this relates to its function under physiological (different phases of the cell cycle) and non-physiological (drug inhibition) conditions. Understanding the dynamics of DNMT1 in vivo provides a framework for developing better therapeutics that target DNMT1. Overall design: Nanopore sequencing genome skimming was used to monitor DNMT1 inhibition in U2OS cells following drug treatment with either DMSO (vehicle), 5-azacytidine, or GSK-3484862 . Following drug treatment, cells were trypsinized and genomic DNA was extracted. Genomic DNA was quantified using ThermoFisher Qubit 1xdsDNA High Sensitivity Assay Kit per manufacturer's instructions. DNA quality was measured by A260/280 and A260/230 ratios using a ThermoScientific NanoDrop ONEc. DNA size distribution was measured using Agilent Bioanalyzer 2100 DNA 12000 chip per manufacturer's instructions. 400 ng genomic DNA was carried forward to library preparation. Libraries were prepared using Oxford Nanopore Technologies Native Barcoding Kit 24 V14 (SQK-NBD114.24) per manufacturer's instructions. Libraries were barcoded, pooled, and loaded onto an Oxford Nanopore Technologies MinION Flow Cell (R10.4.1, FLO-MIN114). Libraries were sequenced for 48 h at a rate of 400-450 bases per second with live basecalling off. Approximately 500 Mb of data per genome were collected, representing ~0.1X genome coverage for an accurate and precise measure of global DNA methylation. For post-sequencing analysis, data were basecalled on the command-line interface of the CU Boulder BioFrontiers Institute FIJI Compute Cluster. Basecalling was performed using Dorado v0.7.4. Reads were aligned to the GRCh38 human genome assembly using Samtools v1.3.1. Methylation data were extracted using the Dorado v0.7.4 modkit pileup command. Downstream analyses were conducted in R version 2024.12.0+467.