Expression and purification of recombinant proteins

The recombinant, histidine-tagged, proteins -catalytic domains (CD) of murine TET1 (1367–2038 aa) and TET2 (1044–1920 aa with the unstructured region replaced by a flexible 15aa linker) and the full-length murine PARP-1, were expressed in Escherichia coli BL21 (DE3) CodonPlus RIL (Novagen) cultivated in Luria–Bertani media. Protein expression was induced by 0.5 mM isopropyl β-D-1-thiogalactopyranoside and cultivation proceeded for 14–15 h at 20 °C. Harvested cells were washed with sodium chloride–Tris–EDTA buffer and resuspended in sonication/wash buffer (50 mM HEPES pH 6.8, 35 mM imidazole, 1 mM DTT, 500 mM NaCl, 10% glycerol) supplemented with protease inhibitor cocktail. Crude cell lysates were prepared by sonification, proteins were purified by Ni–NTA affinity chromatography (Genaxxon) and eluted with a buffer containing high imidazole concentration (50 mM Hepes pH 6.8, 300 mM imidazole, 1 mM DTT, 500 mM NaCl, 10% glycerol). Elution fractions with the highest protein concentrations were pulled together and dialysed for 3 h in dialysis buffer (50 mM Hepes pH 6.8, 1 mM DTT, 300 mM NaCl, 10% glycerol). Aliquots of purified recombinant proteins were flash-frozen in liquid nitrogen and stored at −80 °C until use.

In vitro PARylation

In vitro PARylation was achieved by incubating 5 μM purified recombinant murine TET1-CD or TET2-CD with 0.18 μg of commercial human recombinant PARP-1 or PARP-2 (Enzo Life Sciences, Inc.) in 50 mM Tris–HCl pH 8, 1.5 mM DTT, 1 mM MgCl2, 0.25 μg/μl DNK salmon sperm, 200 μM NAD+ at room temperature (RT). The reaction was stopped after 5 min, 15 min, 30 min or 60 min by adding sample buffer. In the case of TET2 in vitro PARylation, one of each PARP-1 and PARP-2 reactions were stopped after 60 min by 8 μM niraparib (MedChemExpress) and then PARG enzyme (0.043 ng/ml) (Trevigen) was added and incubated at 37 °C for 1 h.

All samples were separated by Tris–glycine SDS-PAGE on 10% polyacrylamide gels and electro-transferred onto nitrocellulose membranes. Membranes were stained by Ponceau-S, photographed and probed by primary murine anti-PAR (1:1000, Enzo Life Sciences, Inc., H10) or a murine anti-HIS (1:2000, Roche) antibody and secondary anti-mouse horseradish peroxidase-conjugated secondary antibody (1:10,000, GE Health care). Signal was visualised on X-ray film with the enhanced chemiluminescence solution reagent (Thermo Scientific).

ATP docking (SwissDock)

To identify possible adenine binding sits on the surface of the TET2 protein, we have used SwissDock platform. The structure of human TET2 catalytic domain was used as the template to dock ATP models. The resulting docking model was visualised in PyMol.

ELISA-based plate assay for DNA hydroxymethylation analysis

Kinetics of TET activity was measured as previously described [27, 29], with minor modifications, by an ELISA-based assay that allows detection of 5hmC. Prior to the assay recombinant TET1-CD (1 μM) was in vitro PARylated for 5 min at RT with increasing concentrations of recombinant purified PARP-1 in a reaction mixture with biotin-labelled methylated DNA substrate. After TET activation reaction temperature was raised to 37 °C. At specific time points (0 s, 30 s, 1 min, 2 min, 3 min, 5 min, 7.5 min, 10 min) 2-μl aliquots were transferred into avidin-coated (Sigma-Aldrich) ELISA plate wells filled with NaOH to stop the reaction. After blocking with 2% BSA (Roth), primary rabbit anti-5hmC (1: 10,000, Active Motif) and subsequently secondary goat anti-rabbit HRP-conjugated (1: 5000, GE Healthcare) antibodies were added. The signal was developed using an ECL reagent (Thermo Scientific) and detected on a 2300 EnSpire Multimode ELISA reader (Perkin Elmer). Detailed protocol is presented in Additional file 3: materials and methods.

Cell culture and treatment

Murine embryonic fibroblasts NIH3T3 (ATCC-CRL-1658) and embryonic fibroblasts isolated from PARP-1 knock-out mice[42] were cultivated in modified Dulbecco's modified Eagle medium (DMEM) (Biological Industries) to which penicillin, streptomycin and 10% fetal bovine serum were added. The cells were cultivated at 37 °C with 5% CO2 in the atmosphere and after reaching about 80% confluency, were subcultured by trypsinisation.

NIH3T3 cells were seeded in sterile 6-well plates and treated with 10 μM niraparib for 72 h (with treatment renewal every 24 h). Since niraparib was dissolved in dimethyl sulfoxide (DMSO), cells were also treated with DMSO in the final concentration of 0.1% as respective control.

MTT assay

Cell viability was examined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were seeded in 96-well plates. At 60–70% confluency cells were treated with niraparib (1.25 µM, 2.5 µM, 5 µM, 10 µM, 20 µM). After 72 h medium was replaced by 200 μl of 0.5 mg/ml MTT (Sigma, M5655) dissolved in DMEM. Following 2 h incubation at 37 °C, MTT was replaced by DMSO (100 µl/well). The absorbance was measured at 570 nm using an ELISA plate reader.

Detection of protein colocalisation by immunocytochemistry

NIH3T3 cells were seeded on glass cover-slips (2 × 104 cells per cover-slip) and were grown for 72 h. Cover-slips were washed with PBS, fixed with 2% paraformaldehyde (PFA) for 10 min at RT and cell permeabilisation was performed with 0.25% Triton X-100 at RT for 10 min. After blocking in 3% BSA cover-slips were incubated with rabbit anti-TET1 (1:10,000, Merck Millipore) followed by goat anti-rabbit Alexa Fluor 647 (1:400, Invitrogen) antibody. Thereafter, the cover-slips were incubated with rat anti-PARP-1 (1:25, R&D Systems) followed by secondary anti-rat FITC (1:400) antibody. After washes with PBST and water, the cover-slips were mounted on microscope slides with Mowiol reagent (Calbiochem).

Cover-slips were photographed on a Leica TCS SP5 II confocal microscope (laser excitation at wavelengths of 488 nm and 633 nm, × 63 magnification). Colocalisation analysis was done using LAS AF software, which sets thresholds for background and red and green fluorescent signals to determine points of colocalisation in the foreground area of the image. This software calculates colocalisation rate as the ratio of total image area where the overlapping of two colours (signals from two fluorophores) is detected, to the foreground area of the image. Same setup for signal and background thresholds was applied for analysis of all images.

Determination of 5hmC level by immunocytochemistry

NIH3T3 cells were seeded on glass cover-slips (2 × 104 cells per cover-slip) and were grown for 72 h. Cover-slips were washed with PBS, fixed with 2% PFA for 10 min at RT and cell permeabilisation was performed with 0.25% Triton X-100 at RT for 10 min. For DNA denaturation the cover-slips were incubated in 2 N HCl for 30 min at 37 °C. After blocking in 3% BSA cover-slips were incubated with a rabbit anti-5hmC (1:10,000, Active Motif) followed by secondary donkey anti-rabbit Alexa Fluor 555 (1:400, Invitrogen) antibody. After washes with PBST and water, the cover-slips were mounted on microscope slides with Mowiol reagent (Calbiochem).

Cover-slips were photographed on a Leica TCS SP5 II confocal microscope (laser excitation at the wavelength of 543 nm, × 63 magnification). Level of 5hmC was analysed using ImageJ 1.52p software [43, 44] with a macro written for the analysis (Additional file 3: Materials and methods).

PARP activity assay

PARP Universal Colorimetric Assay Kit (Trevigen) was used according to the manufacturer's instructions Histone-coated wells were filled with 25 μl of lysate containing 40 μg of protein and PARP buffer. This ELISA-based assay measures incorporation of biotinylated PAR onto histone proteins after addition of PARP cocktail with biotinylated NAD+. The absorbance was measured at 450 nm using an ELISA plate reader.

Genomic DNA isolation and 5mC DNA ELISA assay

Cells were lysed overnight at 55 °C with modified Bradley buffer (2 mM EDTA, 10 mM NaCl, 0.5% SDS, 10 mM Tris–HCl, pH 7.5) supplemented with proteinase K. DNA was isolated by ethanol precipitation (using absolute ethanol supplemented with 75 mM Na-acetate followed by washes in 70% ethanol) and dissolved in water.

Global DNA methylation levels were measured using a commercial 5mC DNA ELISA Kit (Zymo Research) according to the manufacturer's instructions. The wells were coated with denatured DNA (100 ng genomic DNA and methylated DNA standards) and methylation was detected after addition of anti-5mC and HRP-conjugated secondary antibody mixture. The absorbance was measured at 450 nm using an ELISA plate reader.

Statistical analysis

For PARP activity assay and 5mC DNA ELISA assay, statistical significance was analysed by the one-way ANOVA test with blocking, where the set of results of all groups of one independent experiment represented a block. The result of each experimental group was compared with the results of the control group of NIH3T3 cells by Dunnett’s test.

For determination of 5hmC level by immunocytochemistry results were log10 transformed and statistical significance was analysed by nested ANOVA followed by Tukey post hoc test.

Statistical analysis and graphical representation of results were done using IBM SPSS Statistics 20.0, GraphPad Prism 8.0.2 and Microsoft Excel 2013 (Microsoft Corp.).