Experimental procedures for deep enzymology reactions with randomized substrates: For analysis of flanking sequence preferences of the TET enzymes, a similar approach as described for DNMTs (Emperle et al., 2019; Gao et al., 2020; Adam et al., 2020; Dukatz et al., 2020) was used. Briefly, the following single-stranded oligonucleotides containing a methylated or hydroxymethylated CpG or CpH site flanked by 10 randomized nucleotides on either side were obtained from IDT and primer extension was performed to obtain the double stranded DNA substrates. A CpN substrate was prepared as a mixture of CpG and CpH in a 1:3 ratio. For the randomized hydroxymethylated substrate, the single-stranded oligo was purchased coupled to Desthiobiotin-TEG. Primer extension was conducted and the substrate was purified via Streptavidin beads (Dynabeads M-280, ThermoFisher Scientific) and eluted with a biotin solution.

HM rand. GAGTGTGACTAGGCTCTCACTGCCNNNNNNNNNN mC GNNNNNNNNNNGAGAGGAGACCTAGTGAGAAG OH rand. GAGTGTGACTAGGCTCTCACTGCCNNNNNNNNNN hmC GNNNNNNNNNNGAGAGGAGACCTAGTGAGAAG CH rand. GAGTGTGACTAGGCTCTCACTGCCNNNNNNNNNN mC HNNNNNNNNNNGAGAGGAGACCTAGTGAGAAG

The randomized double stranded substrates were incubated with the TET enzyme at 37 °C for 45 min (CN context) or 1 h (CG context) using mixtures containing 1x reaction buffer (50 mM HEPES pH 6.8, 100 mM NaCl, 1 mM DTT, 1 mM alpha-ketoglutarate and 2 mM ascorbic acid), 100 µM ammonium iron(II) sulfate, using different enzyme concentrations and variable amounts of dialysis buffer to keep a fixed salt and glycerol concentration. Reactions were stopped by freezing in liquid nitrogen. Afterwards, Proteinase K (NEB) treatment was used for enzyme inactivation for 1 h at 50 °C, followed by purification with a PCR clean-up kit (MACHEREY-NAGEL). Hairpin ligation and bisulfite conversion was performed using EZ DNA Methylation-Lightning kit (ZYMO).

Library preparation for Illumina Next Generation Sequencing was conducted using a two-step PCR approach as described in (Gao et al., 2020). Unique combinations of barcode and index sequences were introduced to distinguish different samples and experiments. For bioinformatic analysis of the NGS datasets, a local instance of a Galaxy server (Afgan et al., 2018) was used. Sequence reads were trimmed with Trim Galore! (Galaxy Version 0.4.3.1, https://www.bioinformatics.babraham.ac.uk/projects/trim_galore/) keeping only the sequences with a quality score above 20 for further analysis, and filtered according to the expected DNA size using the Filter FASTQ tool (Blankenberg et al., 2010).

The data in this entry contain the Fastq sequence files and extracted DNA sequences obtained with the hemimethylated CpG substrate (HM CG), hemimethylated CpN substrate mixture (HM CN) and hemihydroxymethylated CpG substrate (OH CG). Enzyme kinetics were conducted with TET1 and two versions of TET2 (V1 and V2) as described in the accompanying paper. Individual repeats of experiments are indicated with R1-R5 as appropriate. Control reaction refer to samples treated identically but without enzyme.

The cited references are listed in the accompanying publication to this dataset.

PMID: 35075236