Fig 2: REV3L is essential to protect and restart stalled replication forks.a Quantification of survival assays in control (sgCTRL) and REV3L-depleted HeLa cells, untreated or treated with 0.5 µ? MMC or 1 µ? CIS. Bars represent the mean ± SD. Each dot represents one of three independent experiments. b Schematic and quantification of EdU incorporation intensity in control and REV3L-depleted HeLa cells. Cells were incubated with or without 5 µ? MMC for 24 h, followed by EdU labeling for 2 h before harvesting. White bars represent the median of three independent experiments. c Schematic and quantification of fork progression assays in control and REV3L-depleted HeLa cells, untreated or treated with 4 mM HU for 1 h. A representative fiber experiment from two independent biological replicates is shown. d Schematic and quantification of fork degradation assays. Control and REV3L-depleted HeLa cells were treated with 4 mM HU ± 50 µ? mirin for 4 h. Open circles, no treatment; closed circles, treatment. A representative fiber experiment from three independent biological replicates is shown. e Schematic and quantification of fork restart, fork stalling, and origin firing in control and REV3L-depleted HeLa cells treated with 4 mM HU for 1 h as indicated. Data are represented as mean ± SD. Each dot represents one of three independent experiments. f Quantification of the percentage of cells with REV3L/biotin PLA foci in control and MAD2L2-depleted HeLa cells, untreated or treated with 4 mM HU for 2 h. Bars represent the mean ± SD. Each dot represents one of three independent experiments. Statistical analysis in (a, e) was performed according to one-way ANOVA with Dunnett’s multiple comparisons test. Statistical analysis in (b–d) was performed according to two-tailed Mann-Whitney test. Pink bars in the fiber plots represent the mean. Additional replicates and combined fiber plots are provided in the Supplementary Information. Statistical analysis in (f) was performed according to two-way ANOVA with Tukey’s multiple comparisons test.

Fig 3: MAD2L2, REV1 and REV3L polymerase activity are required for replication fork protection and genome maintenance.a Schematic and quantification of fork degradation assays in control (sgCTRL) and REV1-depleted HeLa cells complemented with either empty vector (EV), GFP-tagged wild-type REV1 (WT), or GFP-tagged catalytic inactive mutant REV1 (D570A/E571A). A representative fiber experiment from three independent biological replicates is shown. b Schematic and quantification of fork degradation assays in control and REV3L-depleted HeLa cells complemented with either empty vector (EV), FLAG-tagged wild-type REV3L (WT), or FLAG-tagged catalytic inactive mutant REV3L (D2781A/E2783A). A representative fiber experiment from three independent biological replicates is shown. Pink bars in fiber plots represent the mean. Statistical analysis in (a, b) was performed according to two-tailed Mann-Whitney test. Additional replicates and combined fiber plots are provided in the Supplementary Information. c Quantification of chromosomal aberrations in control, REV1- and REV3L-depleted HeLa cells transduced with a control shRNA (shSCR, open circles) or MAD2L2 shRNA (closed circles). Cells were harvested following treatment with 4 mM HU for 4 h and recovery for 24 h. Bars represent the mean ± SD of three independent biological replicates. d Schematic model depicting the role of MAD2L2/REV3L/REV1 in fork protection and genome maintenance. Upon replication fork stalling due to e.g. nucleotide depletion, remodeling of stalled forks can generate a regressed, four-stranded DNA structure that is susceptible to the action of cellular nucleases such as MRE11. MAD2L2, in collaboration with REV1 and REV3L, stabilizes and protects reversed replication forks, thereby allowing efficient restart of DNA synthesis. In the absence of MAD2L2/REV3L/REV1, reversed forks undergo extensive MRE11-dependent nucleolytic degradation, which compromises fork restart and results in ssDNA accumulation and genome instability. Created with BioRender.com.

Fig 4: Defects in ssDNA gap filling result in double-strand breaks and cellular sensitivity due to nucleolytic processing of gaps(A) Neutral comet assay showing that BPDE exposure causes DSBs in HeLa cells, which are specifically increased by Polκ depletion. At least 30 comets were quantified for each sample. The median values are marked on the graph, and asterisks indicate statistical significance (Mann-Whitney, two-tailed).(B) Neutral comet assay showing that depletion of REV1 and REV3 increases DSB formation upon BPDE exposure in HeLa cells. At least 50 comets were quantified for each sample. The median values are marked on the graph, and asterisks indicate statistical significance (Mann-Whitney, two-tailed).(C) Neutral comet assay showing that co-depletion of PrimPol suppresses the increase in DSB formation upon Polκ knockdown in BPDE-treated HeLa cells. At least 55 comets were quantified for each sample. The median values are marked on the graph, and asterisks indicate statistical significance (Mann-Whitney, two-tailed). Western blots showing the co-depletion are presented in Figure S1H.(D) Neutral comet assay showing that treatment with the MRE11 endonuclease inhibitor PFM01 suppresses the increase in DSB formation upon Polκ knockdown in BPDE-treated HeLa cells. At least 60 comets were quantified for each sample. The median values are marked on the graph, and asterisks indicate statistical significance (Mann-Whitney, two-tailed).(E) Neutral comet assay showing that deletion of EXO1 suppresses the increase in DSB formation upon Polκ knockdown in BPDE-treated HeLa cells. At least 70 comets were quantified for each sample. The median values are marked on the graph, and asterisks indicate statistical significance (Mann-Whitney, two-tailed).(F) Cellular viability assays showing that deletion of EXO1 suppresses the BPDE sensitivity of Polκ-depleted HeLa cells. The average of three independent experiments, with standard deviations indicated as error bars, is shown. Asterisks indicate statistical significance (two-way ANOVA).See also Figures S1 and S5 and Table S1.

Fig 5: Recruitment of extension polymerases REV1 and REV3 to BPDE adducts on nascent DNA is PrimPol dependent(A–D) SIRF experiments showing that PrimPol depletion reduces the recruitment of REV1 to nascent DNA upon treatment with BPDE in HeLa cells. Knockdown of REV1 is used as control to confirm the specificity of the SIRF signals observed. Quantifications (A, B, and D) and representative micrographs (C), with scale bars representing 10 μm, are shown. At least 75 cells were quantified for each condition. Bars indicate the mean values, error bars represent standard errors of the mean, and asterisks indicate statistical significance (t test, two-tailed, unpaired). Schematic representations of the assay conditions are shown at the top. Western blots confirming REV1 knockdown are shown in Figure S1G.(E–H) SIRF experiments showing that PrimPol depletion reduces the recruitment of REV3 to nascent DNA upon treatment with BPDE in HeLa-BRCA2KO cells. Knockdown of REV3 is used as control to confirm the specificity of the SIRF signals observed. Quantifications (E–G) and representative micrographs (H), with scale bars representing 10 μm, are shown. At least 70 cells were quantified for each condition. Bars indicate the mean values, error bars represent standard errors of the mean, and asterisks indicate statistical significance (t test, two-tailed, unpaired). Schematic representations of the assay conditions are shown at the top.See also Figures S1–S3 and Table S1.