Fig 2: Characterization of the interaction between MCM8IP and MCM8-9.a Venn diagram of the number of proteins identified by mass spectrometry that were enriched in BirA*-MCM8IP and MCM8IP-BirA* streptavidin pulldowns from HEK293T T-REx cells or MCM8IP-HA immunoprecipitates from HEK293T cells relative to their respective controls. Proteins with multiple enriched isoforms were considered a single hit in this diagram. A list of proteins commonly identified by all three experiments is presented. See also Supplementary Data 2. b Detection by western blot of MCM8 and MCM9 co-immunoprecipitated by HA-GFP, HA-MCM8IP WT, or RBM from HEK293T cells. c Schematic representation of MCM8IP truncation mutants used to identify the region of interaction with MCM8-9. d Detection by western blot of MCM8, MCM9, and RPA1 co-immunoprecipitated from HEK293T cells by HA-GFP, HA-MCM8IP WT, and mutants presented in (c). e Alignment from various species of the minimal region of human MCM8IP required for MCM8-9 interaction. MCM8IP MBM #1 mutant carries alanine substitutions of the indicated residues. MCM8IP MBM #2 mutant carries a deletion of the indicated 18 amino acid residues. Sequence alignments were conducted using Clustal Omega and processed using ESPript. f Detection by western blot of MCM8 and MCM9 co-immunoprecipitated by HA-GFP, HA-MCM8IP WT, MBM #1 or MBM #2 from HEK293T cells.

Fig 3: Identification of MCM8IP by proximity-dependent biotin-identification (BioID) technology.a Schematic of the protocol used to identify RPA1 interactors by BioID. HEK293T T-REx cells expressing doxycycline-inducible BirA*- or BirA*-RPA1 were treated with HU in the presence of exogenous biotin. Biotinylated proteins were captured in denaturing conditions from cell lysates by streptavidin pulldown and subjected to mass spectrometry for protein identification. b List of selected proteins identified by mass spectrometry that were enriched in streptavidin pulldowns conducted from BirA*-RPA1-expressing HEK293T T-REx cells relative to pulldowns performed from control BirA*-expressing cells. See also Supplementary Data 1. c Detection by western blot of MCM8IP in streptavidin pulldowns from HEK293T T-REx cells expressing BirA* or BirA*-RPA1. Cells were treated with HU (1 mM), cisplatin (20 µM), or olaparib (20 µM) in the presence of exogenous biotin for 24 h prior to lysis. Vinculin is shown as a loading control. d Detection by western blot of RPA1 and RPA2 following immunoprecipitation of HA-GFP or HA-MCM8IP from HEK293T cells. e Detection by western blot of RPA1 and MCM8IP (short exposure, s.e.; long exposure, l.e.) following subcellular fractionation of HCT116 cell lysates upon treatment with HU (1 mM), cisplatin (10 µM), or olaparib (10 µM) for 24 h. Tubulin and histone H3 are shown as loading and fractionation controls. f Representative images of FLAG-MCM8IP recruitment to sites of UV laser microirradiation in U2OS cells. DNA damage tracts are indicated with ?H2AX staining. Scale bar = 20 µm. g Graphical representation of the percentage of FLAG-MCM8IP co-localizing with ?H2AX following UV laser microirradiation in U2OS cells transfected with control or CtIP siRNA. The mean values ± SD of three independent experiments are presented. Statistical analysis relative to control siRNA was conducted using Student’s t-test (***p < 0.001, two-tailed).

Fig 4: Analysis of replication fork progression in MCM8IP-deficient cells in response to cisplatin treatment.a Schematic representation of the CldU/IdU pulse-labeling assay (top panel). IdU labeling was performed in the absence or presence of cisplatin (30 µM). Representative images of fibers analyzed in untreated or cisplatin-treated HCT116 control cells or cells expressing the indicated MCM8IP sgRNAs (bottom panel). b Dot plot of IdU tract length for individual replication forks in untreated or cisplatin-treated HCT116 control cells or cells expressing the indicated MCM8IP sgRNAs. Experiments were conducted as shown in (a). The median values are indicated by red lines. Statistical analysis was conducted using a Mann–Whitney test (****p < 0.0001, two-tailed). Data are representative of two independent experiments. c Schematic representation of the CldU/IdU pulse-labeling assay as in (a, top panel). Representative images of fibers analyzed in cisplatin-treated HCT116 control cells or MCM8IP KO cells reconstituted with MCM8IP WT, MBM #2, RBM, or an empty vector (EV) control (bottom panel), as shown in Supplementary Fig. 9c. d Dot plot of IdU tract length for individual replication forks in cisplatin-treated HCT116 control cells or MCM8IP KO cells reconstituted with MCM8IP WT, MBM #2, RBM, or EV. Experiments were conducted as shown in (c). Data are shown and analyzed using a Mann–Whitney test as in (b) (****p < 0.0001, two-tailed), and are representative of two independent experiments. e Schematic representation of MCM8 with POI-associated mutations indicated. The DNA binding domain is indicated in blue and the ATPase domain is indicated in green. f Detection by western blot of MCM8IP and MCM9 co-immunoprecipitated from HEK293T cells by GFP-HA, MCM8-HA WT, or carrying the POI-associated mutations indicated in (e).

Fig 5: Homologous recombination frequency in MCM8IP-deficient cells.a Schematic representation of the DR-GFP reporter of gene conversion, as described80. b Graphical representation of the percentage of I-SceI-induced HR events in U2OS DR-GFP cells expressing the indicated MCM8IP sgRNAs relative to control cells. The relative level of background events in control cells transfected with an empty vector is additionally shown. The mean ± SD of three independent experiments is presented. Statistical analysis relative to the control was conducted using one-way ANOVA (**p < 0.01, ***p < 0.001, ****p < 0.0001). c Graphical representation of the percentage of I-SceI-induced HR events in an U2OS DR-GFP MCM8IP KO clone reconstituted with the indicated cDNAs or an empty vector (EV) control, as shown in Supplementary Fig. 6d, e, relative to WT MCM8IP-expressing KO cells. The relative level of background events in EV-complemented cells transfected with an empty vector is additionally shown. The mean ± SD of four independent experiments is presented. Statistical analysis relative to WT MCM8IP-expressing KO cells was conducted as in (b) (**p < 0.01, ****p < 0.0001). d Dot plot of the number of cisplatin- or ionizing radiation-induced RAD51 foci in cyclin A-positive HCT116 control cells or cells expressing the indicated MCM8IP sgRNAs. The median values are indicated by red lines within the dot plot. Data are representative of two independent experiments. Statistical analysis was conducted using a Mann–Whitney test (****p < 0.0001, two-tailed). e Schematic representation of the SCR/RFP reporter of gene conversion, as described44. Blocks A and B are artificial RFP exons and arrows are promoters. Short-tract gene conversion (STGC) results in GFP expression alone, while long-tract gene conversion (LTGC) results in the additional expression of RFP. f Graphical representation of the fold change in I-SceI-induced STGC (left panel), LTGC (middle panel), and LTGC as a fraction of total gene conversion events (right panel) in U2OS 35S cells expressing the indicated MCM8IP sgRNAs relative to control cells. Relative levels of background STGC and LTGC events in U2OS 35S control cells transfected with an empty vector are additionally shown. The mean ± SD of five independent experiments is presented. Statistical analysis was conducted as in (b) (**p < 0.01, ****p < 0.0001).