Fig 2: Inhibiting PARG triggers TOP1-DPC deubiquitylation by USP7.a His-tag pulldown-HPLC-MS/MS showing that His-TOP1 interacted USP7 under unperturbed condition. After transfection of 6×His-tagged TOP1 expression plasmid or empty vector control (pTrex), HCT116 cells were subjected to His-tag pulldown using Ni-NTA agarose, followed by HPLC-MS. b In vitro assay showing that USP7 reversed TOP1 ubiquitylation. Recombinant TOP1 was subjected to ubiquitylation with ubiquitin, Ube1 (E1), Ubc5Ha (E2), and RNF4 (E3) for 30 min, followed by termination with EDTA and incubation with increasing concentrations of recombinant USP7 for another 30 min. Samples were Western blotted with a-Ub antibody. c His-tag pulldown assay showing that PARGi enhanced TOP1-USP7 interaction. Following transfection of 6×His-tagged TOP1 expression plasmid and FLAG-USP7 expression plasmid, HEK293 cells were treated as indicated. His-tag pulldown was performed with Ni-NTA agarose under native conditions. Western blotting was performed with the indicated antibodies. d PLA assay showing TOP1-USP7 interaction in PARGi-treated cells. Following transfection of 6×His-tagged TOP1 expression plasmid and FLAG-USP7 expression plasmid, HEK293 cells were treated as indicated. PLA assays were performed using rabbit a-His-tag antibody and mouse a-FLAG tag antibody. The scale bar represents 10 µm. e Inhibiting USP7 restored TOP1-DPC ubiquitylation in the presence but not in the absence of PARGi. Upper panel: HEK293 cells were treated as indicated: CPT (20 µM, 1 h), CPT + FLAG-USP7 transfection, CPT + USP7i (10 µM, 1 h pre-treatment), CPT + PARGi (10 µM, 1 h pre-treatment), CPT + PARGi + FLAG-USP7 transfection, CPT + PARGi + USP7i. Following treatments, cells were subjected to the modified RADAR assay for detection of TOP1-DPCs and their ubiquitylation using a-TOP1 and a-Ub antibodies. Lower panel: densitometric quantitation of ubiquitylated TOP1-DPC signals generated from triplicate experiments including representative blots shown in (c) using ImageJ. n = 3 independent experiments. Data are presented as mean values +/- standard deviation (SD). P value was calculated by paired Student’s t-test (two-tailed distribution). *: p = . NS not significant. f Inhibiting USP7 did not impact the induction of ?H2AX upon exposure to CPT. U2OS cells were synchronized in the S phase by double thymidine and treated with CPT (1 µM) in the absence of presence of USP7i (10 µM, 1 h pre-treatment) and collected for IF by iSIM using an anti-?H2AX antibody. The scale bar represents 10 µm.

Fig 3: TRPT1 overexpression leads to ADP-ribosylation of different cellular RNA pools. (A) Scheme of the sample preparation to extract ADP-ribosylated RNA pools from cells. HeLa cells were reverse transfected in 6-well plates with siRNA to knock-down TARG1, PARG and ARH3. 48 hours after siRNA transfection, plasmids overexpressing transferases were transfected. 24–48 h later cells were harvested and total, large, and small RNA pools were isolated from cell extracts using a neutral buffer. mRNA was enriched from the total RNA fraction using oligo(dt) beads. (B) RNA samples from transiently TRPT1 overexpressing and wild-type cells in combination with hydrolase knock-downs were prepared as described in (A) and blotted onto a positively changed nylon membrane. ADP-ribosylation of the blotted RNAs was detected using a poly/mono ADP-ribose antibody (E6F6A0).

Fig 4: DePARylation of TOP1-DPCs is required for their proteasomal degradation.a. Scheme for biotinylated TOP1cc preparation, PARylation, and proteasome digestion assays. b TOP1cc proteasomal digestion assay showing that unmodified TOP1ccs (without NAD+) are fully degraded by the 26S proteasome and partially degraded by the 20S proteasome after 45 min incubation whereas PARylated TOP1ccs (+ NAD+) are refractory to proteasomal degradation. c Inhibiting PARG blocked the interactions between TOP1-DPC and the proteasome subunit PSMD14. Following transfection of PSMD14-FLAG expression plasmid, HEK293 cells were treated with the indicated drugs (1 h, 10 µM PARGi pre-treatment followed by 30 min co-treatment with 20 µM CPT) and subjected to IP using an a-FLAG antibody. The immunoprecipitates and input samples were Western blotted with the indicated antibodies. d Inhibiting PARG blocked the interactions between TOP1-DPC and the proteasome subunit PSMB5. Following transfection of PSMB5-FLAG expression plasmid, HEK293 cells were treated with the indicated drugs (1 h, 10 µM PARGi pre-treatment followed by 30 min co-treatment with 20 µM CPT) and subjected to IP using an a-FLAG antibody. The immunoprecipitates and input samples were Western blotted with the indicated antibodies. e. Model for 26 S proteasome-dependent degradation of TOP1-DPCs. f Inhibiting PARG blocked the liberation of TOP1-induced DNA breaks. Upper panels: representative images of alkaline comet assay in HEK293 cells treated with DMSO, CPT (10 µM, 2 h), PARGi (10 µM, 3 h), and CPT + PARGi (pre-treatment with PARGi for 1 h then co-treatment with CPT and PARGi for 2 h). Cells were subjected to alkaline comet assay for detection of DNA breaks. Lower panel: quantitation of tail moments using OpenComet. n = 50 biologically independent cells. P value was calculated by paired Student’s t-test (two-tailed distribution). The scale bar represents 100 µm. g Inhibiting PARG attenuated TOP1-DPC-induced DNA damage response (DDR). HEK293 cells were treated with CPT in absence of the presence of PARGi (pre-treatment for 1 h). Cells were collected at the indicated time points and subjected to non-denaturing lysis and benzonase treatment, followed by Western blotting with the indicated antibodies. h Inhibiting PARG reduced CPT-induced ?H2AX foci. Upper panels: representative images of IF of ?H2AX and IdU foci by iSIM. U2OS cells were synchronized in the S phase by double thymidine block, followed by IdU incorporation and PARGi treatment for 1 h. Cells were then treated with CPT (1 µM) and collected at indicated time points for IF using anti-?H2AX and anti-BrdU antibodies. Lower panels: quantitation of ?H2AX foci (left) and quantitation of IdU foci (right). n = 10 biologically independent samples. Data are presented as mean values +/- standard deviation (SD). P value was calculated by paired Student’s t-test (two-tailed distribution). ***: p = 0.000015. NS not significant. The scale bar represents 10 µm.

Fig 5: Inhibiting PARG reveals the otherwise transient PARylation of TOP1-DPC and stabilizes TOP1-DPCs.a His-tag pulldown assay showing reversible PARylation of transfected His-TOP1. Following transfection of 6×His-tagged TOP1 expression construct, HEK293 cells were treated with the indicated drugs for His-tag pulldown using Ni-NTA agarose under denaturing conditions. The pulldown samples and input samples were subjected to IB using a-TOP1 and a-PAR antibodies. CPT: 20 µM, 30 min; PARPi: 10 µM, 1 h pre-treatment; PARGi: 10 µM, 1 h pre-treatment. b Proximity ligation assay (PLA) showing TOP1-PAR interaction in cells treated with CPT and PARGi. Following transfection of 6×His-tagged TOP1 expression construct, U2OS cells were treated with the indicated drugs for PLA using rabbit a-His-tag antibody and mouse a-PAR antibody (10H). The scale bar represents 3 µm. c PARGi revealed CPT-induced PAR polymers on chromatin. After pre-extraction, U2OS cells treated as indicated were subjected to IF using a-PAR antibody (10H). CPT: 20 µM; PARGi: 10 µM, 1 h pre-treatment. The scale bar represents 10 µm. d Scheme of immunodetection of TOP1-DPC PARylation in vivo by modification of RADAR assay. TOP1-DPC PARylation was detected with a-PAR antibody (10H). e PARGi induced the hyper-PARylation and stabilization of TOP1-DPCs. HEK293 cells were treated with CPT (20 µM) in the absence or presence of PARGi (10 µM, 1 h pre-treatment). Cells were collected at the indicated time points for the modified RADAR for detection of TOP1-DPCs and their PARylation and ubiquitylation using a-TOP1, a-PAR, and a-Ub antibodies. 2 µg of digested DNA from each sample was subjected for slot-blotting using a-dsDNA antibody as a loading control. f ICE assay confirming that PARG deficiency blocked the removal of TOP1-DPCs. HEK293 cells transfected with control siRNA (siCtrl) or siRNA targeting PARG were treated with CPT (2 µM). Cells were collected at the indicated time points for ICE assay. 2 µg of digested DNA from each ICE assay sample was subjected for slot-blotting using a-TOP1 antibody or a-dsDNA antibody. g The modified RADAR assay showing that in vivo TOP1-DPC PARylation by PARP1 was counteracted by PARG. After transfection of the PARP1-FLAG expression plasmid, HEK293 cells were pre-treated with PARPi (10 µM, 1 h), PARGi (10 µM, 1 h), or PARPi + PARGi, followed by CPT treatment (20 µM, 1 h). The cells were then subjected to the modified RADAR assay for detection of TOP1-DPCs and their PARylation using a-TOP1 and a-PAR antibodies. h Inhibiting replication or transcription did not affect TOP1-DPC PARylation. HEK263 cells were pre-treated PARGi (10 µM, 1 h) in the absence or presence of the replication inhibitor aphidicolin (APH, 10 µM, 1 h) or the transcription inhibitor DRB (100 µM, 1 h). Cells were co-treated with 20 µM CPT for 0, 1, or 4 h, followed by detection of TOP1-DPCs and their PARylation by the modified RADAR assay using a-TOP1 and a-PAR antibodies.