Live Cell Detection of Poly(ADP-Ribose) for Use in Genetic and Genotoxic Compound Screens

Abstract

Poly(ADP-ribose) (PAR) is a molecular scaffold that aids in the formation of DNA repair protein complexes. Tools to sensitively quantify PAR in live cells have been lacking. We recently described the LivePAR probe (EGFP fused to the RNF146-encoded WWE PAR binding domain) to measure PAR formation at sites of laser micro-irradiation in live cells. Here, we present two methods that expand on the use of LivePAR and its WWE domain. First, LivePAR enriches in the nucleus of cells following genotoxic challenge. Image quantitation can identify single-cell PAR formation following genotoxic stress at concentrations lower than PAR ELISA or PAR immunoblot, with greater sensitivity to genotoxic stress than CometChip. In a second approach, we used the RNF146-encoded WWE domain to develop a split luciferase probe for analysis in a 96-well plate assay. We then applied these PAR analysis tools to demonstrate their broad applicability. First, we show that both approaches can identify genetic modifications that alter PARylation levels, such as hyper-PARylation in BRCA2-deficient cancer cells. Second, we demonstrate the utility of the WWE split luciferase assay to characterize the cellular response of genotoxins, PARP inhibitors, and PARG inhibitors, thereby providing a screening method to identify PAR modulating compounds.

Keywords: BER; BRCA2; DNA damage; LivePAR; PAR; Poly(ADP-ribose); WWE; split luciferase.

Figure 1

LivePAR as a molecular probe to investigate PARylation in live cells. (A) LivePAR is a fusion of the PAR-binding WWE domain from RNF146 (blue; PDB 3V3L) and EGFP (green; PDB 2Y0G) that preferentially binds to the iso-ADP-ribose moiety (red) of PAR chains [36,40]. (B) Expression of LivePAR in A549 cells. White scale bar denotes 20 µm. (C) Recruitment of LivePAR to sites of laser micro-irradiation (405 nm laser). Mutations in amino acids that eliminate PAR binding in the WWE domain (Y107A, Y144A, or R163A) prevent LivePAR recruitment; N ≥ 40 cells. (D) LivePAR recruitment to sites of laser-induced DNA damage is attenuated following PARPi treatment, while PARGi enhances and prolongs LivePAR recruitment; N ≥ 40 cells. (E) Images of LivePAR-expressing A549 cells following laser micro-irradiation. PARPi prevents foci formation while PARGi strengthens and prolongs foci formation compared to control. White scale bar denotes 20 µm.

Figure 2

Detection of PARylation using LivePAR nuclear enrichment in live cells. (A) Diagram of mechanism: Prior to stress, LivePAR (green circles) is uniformly observed throughout a cell. H₂O₂ generates DNA damage in the nucleus (red stars) and stimulates PAR formation and LivePAR nuclear enrichment. (B) LivePAR nuclear enrichment following H₂O₂ (250 µM) exposure. PARPi (ABT-888; 10 µM) prevented LivePAR nuclear enrichment, while PARGi (PDD00017273; 10 µM) enhanced LivePAR nuclear enrichment. White scale bar denotes 20 µm. (C) Time course of LivePAR nuclear enrichment following H₂O₂ dose–response over 30 min. Graph shows mean ± SEM. (D) LivePAR nuclear enrichment 30 min following H₂O₂ dose–response. Each point represents a single cell nucleus. Graph shows mean ± SEM (in red). (E) Time course of LivePAR nuclear enrichment following MMS dose–response. Graph shows mean ± SEM. (F) LivePAR nuclear enrichment 30 min following MMS dose–response. Each point represents a single cell nucleus. Graph shows mean ± SEM (in red). (G) Time course of LivePAR nuclear enrichment following H₂O₂ (250 µM) treatment, with and without PARPi or PARGi. Graph shows mean ± SEM. (H) LivePAR nuclear enrichment 30 min following H₂O₂ (250 µM) treatment, with and without PARPi or PARGi. Each point represents a single cell nucleus. Graph shows mean ± SEM (in red). NS = no significance, * p < 0.05, ** p < 0.01, **** p < 0.0001; a two-sample t-test (compared to the untreated controls) for panels D and F, and a one-way ANOVA with Tukey post-hoc test (significance compared to untreated controls) for panel H. For panels C-H, each data point represents the mean ± SEM for at least 100 individual nuclei.

Figure 3

Enhanced PARylation in BRCA2-deficient cell lines. (A) Validation of A549/BRCA2-deficient cell lines. CRISPR/Cas9-mediated BRCA2 deficiencies were generated using 2 unique gRNAs (i.e., g1, g2). (B) Laser micro-irradiation of A549/Cas9 or A549/BRCA2-deficient cells; N ≥ 40 cells. (C) Time course of LivePAR nuclear enrichment in BRCA2-deficient cells following H₂O₂ (250 µM) dose–response over 30 min. (D) Time course of LivePAR nuclear enrichment in BRCA2-deficient cells following H₂O₂ dose–response over 30 min in PARGi pretreated cells. (E) LivePAR nuclear enrichment 30 min following H₂O₂ dose–response in BRCA2-deficient cells. Each point represents a single cell nucleus. Graph shows mean ± SEM (in red). (F) LivePAR nuclear enrichment with 1 h PARGi pre-treatment followed by 30 min H₂O₂ exposure in BRCA2-deficient cells. Each point represents a single cell nucleus. Graph shows mean ± SEM (in red). (G) Laser micro-irradiation of PEO1/LivePAR or C4-2/LivePAR cells; N ≥ 40 cells. Graph shows mean ± SEM. (H) LivePAR nuclear enrichment in PEO1 or C4-2 cells following 1 h PARGi pre-treatment and 30 min H₂O₂ exposure. Each point represents a single cell nucleus. Graph shows mean ± SEM (in red). For statistical comparisons from ANOVA with Tukey post hoc test: a: p < 0.01 in comparison to Cas9 or untreated controls; b: p < 0.01 in comparison to Cas9-H₂O₂ cells; c: p < 0.01 in comparison to Cas9-H₂O₂-PARGi cells; NS = no significance.

Figure 4

Development of WWE split luciferase (Luc) assay to measure PARylation. (A) Vector designs to assess the effect of WWE and split-Luc orientation. WWE was fused to either the N-terminal side of nLuc or cLuc, or to the C-terminal side of nLuc or cLuc. Four combinations were tested. (B) PAR formation following 30 min H₂O₂ dose–response. Values are normalized to untreated or DMSO controls. (C) PAR formation following 30 min MMS dose–response. (D) PAR formation following 30 min etoposide dose–response. Each dot represents a single plate reading, with mean and SEM (red bars) shown; N ≥ 8 reads. * p < 0.05 using a one-sample t-test.

Figure 5

Sensitivity of the single vector WWE split luciferase (Luc) assay. (A) Expression of WWE-nLuc and WWE-cLuc in U2OS cells using a single lentiviral vector. WWE(R163A)-nLuc was alternatively expressed to control for luciferase binding effects and serve as a non-PAR binding control. (B) PAR formation following 30 min MMS dose–response. Values are normalized to untreated controls. (C) PAR formation following 30 min H₂O₂ dose–response. (D) CometChip results following 30 min MMS dose–response. (E) CometChip results following 30 min H₂O₂ dose–response. (F) Immunoblot for PAR formation in U2OS cells following 30 min MMS dose–response. (G) Immunoblot for PAR formation in U2OS cells following 30 min MMS (1000 μM) with or without PARGi pre-treatment (PDD00017273; 10 μM, 1 h). (H) PAR ELISA results following 30 min MMS (1000 μM) with or without PARGi pre-treatment (PDD00017273; 10 μM, 1 h). For (B,C), each dot represents a single plate reading, with mean and SEM (red bars) shown; N ≥ 8 reads. * p < 0.05 using a one-sample t-test. For (D,E), each dot represents a single cell, with mean and SEM (red bars) shown; * p < 0.05 using an ANOVA with Tukey post hoc test when compared to untreated control. For (H): * p < 0.05 using an ANOVA with Tukey post hoc test when compared to untreated control.

Figure 6

Enhanced PARylation in BRCA2-deficient cells identified using the WWE split luciferase assay. WWE-nLuc [or WWE(R163A)-nLuc] and WWE-cLuc were expressed in BRCA2-deficient PEO1 or BRCA2-resurgent C4-2 cells using a single lentiviral vector. (A) PAR formation following 30 min MMS dose–response. Values are normalized to untreated controls. (B) PAR formation following 30 min H₂O₂ dose–response. For (A,B), each dot represents a single plate reading, with mean and SEM (red bars) shown; N ≥ 6 reads. * p < 0.05 using a one-sample t-test.

Figure 7

Genotoxin screening for PARylation using the WWE split luciferase assay. U2OS cells expressing WWE-nLuc (or WWE(R163A)-nLuc) and WWE-cLuc were treated with individual compounds at the specified concentration. The vertical dotted line denotes a DMSO-normalized value. (A) PAR formation following 30 min of compound exposure. Values are normalized to untreated controls. (B) PAR formation following 6 h of compound exposure. For (A,B), each dot represents a single plate reading, with mean and SEM (red bars) shown; N ≥ 8 reads. * p < 0.05 using a two-sample t-test in comparison between WWE-nLuc and WWE(R163A)-nLuc for each test condition.