PARP1 and PARP2 stabilise replication forks at base excision repair intermediates through Fbh1-dependent Rad51 regulation

Abstract

PARP1 regulates the repair of DNA single-strand breaks generated directly, or during base excision repair (BER). However, the role of PARP2 in these and other repair mechanisms is unknown. Here, we report a requirement for PARP2 in stabilising replication forks that encounter BER intermediates through Fbh1-dependent regulation of Rad51. Whereas PARP2 is dispensable for tolerance of cells to SSBs or homologous recombination dysfunction, it is redundant with PARP1 in BER. Therefore, combined disruption of PARP1 and PARP2 leads to defective BER, resulting in elevated levels of replication-associated DNA damage owing to an inability to stabilise Rad51 at damaged replication forks and prevent uncontrolled DNA resection. Together, our results demonstrate how PARP1 and PARP2 regulate two independent, but intrinsically linked aspects of DNA base damage tolerance by promoting BER directly, and by stabilising replication forks that encounter BER intermediates.

Fig. 1

PARP1 and PARP2 are redundant in the cellular response to MMS-induced DNA damage. a Whole cell extracts were prepared from U2OS or two independent parp1Δ cell lines and western blotting performed with the indicated antibodies (left panel). U2OS or parp1Δ cell lines were exposed to MMS and cell survival assessed by clonogenic assays (right panel). Error bars represent the standard error of the mean (SEM) from three independent experiments. Statistical significance was determined by two-way ANOVA (*** p < 0.001). b U2OS or parp1Δ cell lines, with or without exposure to Olaparib (PARPi), were exposed to MMS and cell survival assessed by clonogenic assays. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a. c Whole cell extracts were prepared from U2OS, parp1Δ, parp2Δ and parp1/2Δ cells and Western blotting performed using the indicated antibodies. d The indicated cell lines were left untreated (-) or exposed to 1.5 mM MMS ( + ) for 1 h and nuclear ADP-ribosylation analysed by immunofluorescence. Error bars represent the SEM from three independent experiments. Statistical significance was determined using a two-tailed Student’s t-test (** p < 0.01; *** p < 0.001). e U2OS, parp1Δ, parp2Δ or parp1/2Δ cell lines were exposed to MMS and cell survival assessed by clonogenic assays. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a. f Cells were treated with 250 µM MMS for 1 h, before recovery in fresh media. Samples were taken at the indicated times post treatment and the alkaline comet assay used to reveal strand breaks and alkali-labile sites. Comet data was normalised to the untreated sample. Error bars represent mean values ± SD from at least six independent experiments. Statistical significance was determined as in d

Fig. 2

Loss of PARP1 is synthetic lethal with HR inhibition. a U2OS cells were exposed to B02, with or without Olaparib (PARPi) or Nu7441 (DNA-PKi), and cell viability assessed by clonogenic survival assay. Error bars represent the SEM from three independent experiments. Statistical significance was determined by two-way ANOVA (NS, not significant; *** p < 0.001). b U2OS, parp1Δ, parp2Δ or parp1/2Δ cell lines were exposed to B02 and cell viability assessed by clonogenic survival assay. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a. c and d U2OS and parp1/2Δ cell lines, with or without siRNA depletion of PARP3 c, or U2OS cells with or without XRCC1 siRNA depletion d, were exposed to B02 and cell viability assessed by clonogenic survival assay. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a

Fig. 3

PARP1 and PARP2 function in the resolution of replication-associated damage independently of BER. a U2OS cells were treated with 1 μM EdU and 0.25 mM MMS for 1 h, before pre-extraction and fixation. EdU and γ-H2AX were detected by immunofluorescence. Representative images of EdU-positive and EdU-negative nuclei are shown. Scale bars represent 10 μm. b The indicated cell lines were treated as in a and following immunofluorescence γ-H2AX foci quantified in G1/G2 phase (EdU-negative) and S-phase (EdU-positive) cells. Error bars represent the SEM from four independent experiments. Statistical significance was determined using a two-tailed Student’s t-test (*** p < 0.001). c The indicated cell lines were treated with 1 μM EdU (Unt), or with 1 μM EdU and 0.25 mM MMS for 1 h before recovery in fresh media, and γ-H2AX foci quantified by immunofluorescence in EdU-positive cells. Times indicated show hours since addition of EdU and MMS. Error bars represent the SEM from four independent experiments. Statistical significance was determined as in b. d U2OS cells transfected with control or XRCC1 siRNA, or parp1/2Δ cells were treated as in c and γ-H2AX foci quantified in EdU-positive cells (lower panel). Knockdown of XRCC1 in U2OS cells was confirmed by Western blotting using the indicated antibodies (upper panel). Times indicated show hours since addition of EdU and MMS. Error bars represent the SEM from four independent experiments. Statistical significance was determined as in b. e The indicated cell lines transfected with XRCC1 siRNA were treated as in c and γ-H2AX foci quantified in EdU-positive cells. Times indicated show hours since addition of EdU and MMS. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in b

Fig. 4

Assembly of Rad51 at damaged replication forks is compromised in the absence of PARP1 and PARP2. a DNA fibre analysis was performed in the indicated cells after MMS exposure and recovery for the times shown. The number of stalled forks (red-only tracts) was determined as a ratio of all red-labelled replication structures. These values were subsequently normalised to the equivalent ratio in the corresponding non-treated sample. Error bars represent the SEM from at least three independent experiments. Statistical significance from the normalised ratio of stalled forks in parental U2OS cells was determined using a two-tailed Student’s t-test (NS, not significant; * p < 0.05; ** p < 0.01; *** p < 0.001). b The indicated cell lines transfected with control or XRCC1 siRNA were treated with 1 μM EdU (Unt), or with 1 μM EdU and 0.5 mM MMS for 1 h before recovery in fresh media, and Rad51 foci quantified in EdU-positive cells. Times indicated show hours since addition of EdU and MMS. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a. c U2OS cells transfected with control or XRCC1 siRNA were treated as in b and Rad51 nuclear foci analysed in EdU-positive cells. Cells were fixed 12 h after addition of EdU and MMS. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a. d U2OS or parp1/2Δ cells, in the presence or absence of 50 μM B02 (Rad51i), were treated as in (3 C) and γ-H2AX foci quantified in EdU-positive cells. Times indicated show hours since addition of EdU and MMS. Error bars represent the SEM from three independent experiments. Statistical significance was determined as in a

Fig. 5

PARP1 and PARP2 are required to stabilise Rad51 at stalled and/or damaged replication forks. a The indicated cell lines were treated with 1 μM EdU (Unt), or with 1 μM EdU and 0.5 mM MMS for 1 h before recovery in fresh media, and RPA foci quantified in EdU-positive cells. Times indicated show hours since addition of EdU and MMS. Representative images of EdU-positive nuclei 12 h after addition of MMS are shown (upper panel). Scale bars represent 10 μm. Error bars represent the SEM from three independent experiments. b The indicated cell lines transfected with control or XRCC1 siRNA were treated as in a and RPA nuclear foci analysed in EdU-positive cells. Cells were fixed 12 h after addition of EdU and MMS. Error bars represent the SEM from three independent experiments. c The indicated cell lines, transfected with control or XRCC1 siRNA, were left untreated (–) or exposed to 0.5 mM MMS for 1 h before recovery in fresh media for 12 h (+). Western blotting of whole cell extracts was performed with the indicated antibodies. d U2OS cells transfected with control or Fbh1 siRNA –/ + olaparib (PARPi), were treated as in a and Rad51 nuclear foci analysed in EdU-positive cells. Cells were fixed 12 h after addition of EdU and MMS. Knockdown of Fbh1 was confirmed by western blotting. Error bars represent the SEM from three independent experiments. e U2OS or parp1/2Δ cells transfected with control or Fbh1 siRNA were treated as in a and Rad51 nuclear foci analysed in EdU-positive cells. Times indicated show hours since addition of EdU and MMS. Representative images of EdU-positive nuclei 12 h after MMS addition are shown. Scale bars represent 10 μm. Error bars represent the SEM from three independent experiments. f Flp-In T-Rex HT1080 cells expressing FE-PALB2, with or without Olaparib (PARPi), were left untreated (–) or exposed to 0.5 mM MMS for 1 h before recovery in fresh media (+). Extracts were prepared 6 h after MMS addition. Western blotting of whole cell and chromatin extracts was performed with the indicated antibodies. In all cases, statistical significance was determined using a two-tailed Student’s t-test (NS, not significant; *** p < 0.001)

Fig. 6

PARP1 and PARP2 contribute to the resolution of DNA damage caused by base alkylation through two mechanisms. Alkylated bases are processed into DNA single-strand breaks during BER and their repair is accelerated by XRCC1, PARP1 and PARP2. During S-phase, active replication forks can collide with unrepaired SSBs, or SSB repair intermediates, generating replication-associated DNA damage. This damage can be resolved through homologous recombination, which requires the formation of Rad51 filaments. PARP1 and PARP2 also contribute to this mechanism by antagonising the anti-recombinogenic activity of Fbh1, thus stabilising Rad51 nucleofilaments