A single BRCA2 BRC repeat supports viability while multiple repeats ensure resilience under stress

BRCA2, BRC repeats, mESCs, DNA repair, homologous recombination, knock-in mice, CRISPR Cas9, oxidative stress, genomic stability 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 2

BRC repeats are integral to BRCA2 function and mediate RAD51 loading during homologous recombination (HR). Their number varies across species, but mammals, including mice and humans, harbor eight repeats. Prior studies have suggested functional redundancy among these repeats, but the biological significance of maintaining multiple repeats remains unresolved. Here, we demonstrate that the presence of a single BRC repeat, either BRC2 or BRC4, is sufficient for mouse embryonic stem cell (mESC) viability, RAD51 loading, PARP inhibitor resistance and protection of stalled replication forks. Consistent with these findings, we show that knock-in mice with a single BRC repeat 2 or 4 are viable and exhibit normal growth and fertility. In contrast, embryonic fibroblasts from these mice display genomic instability and impaired RAD51 recruitment. Notably, this defect is rescued under low oxygen culture conditions, which mimics physoxic levels, whereas exposure to oxidative stress impairs RAD51 recruitment in mESCs harboring a single BRC repeat. Together, our findings indicate that while a single BRC repeat is sufficient under physiological conditions, the evolutionary retention of multiple BRC repeats likely ensures robust genome stability under extreme oxidative stress. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 3

BRCA2 is a well-known tumor suppressor functioning as a genome caretaker by recruiting RAD51 to DNA double-strand breaks (DSB) facilitating their repair by homologous recombination (HR) 1,2. BRCA2 is essential for the survival of normal cells, including mouse embryonic stem cells3. One of the key functional domains of human BRCA2 are the eight BRC repeats, spanning residues 1002–2085 (Fig. 1a). Across metazoans, the BRC repeats have undergone significant evolutionary diversification in both number and sequence. The majority of mammalian BRCA2, including human and murine, contain 8 BRC repeats, each comprising of 35 non-identical (but conserved) 4 amino acids connected by a non-conserved linker. In contrast, BRCA2 protein from Trypanosoma species harbors between 1 and 15 BRC-like repeats (Suppl. Fig. 1a). The eight mammalian BRC repeats are known to physically interact with RAD51 at differing affinities, with BRC1-4 exhibiting higher RAD51 affinity than BRC5-8 5. Structural analysis elucidated two highly conserved hydrophobic regions (FxxA and LFDE) within BRC repeats responsible for RAD51 interaction 4,6. However, mutational analysis of human BRC4 in Ustilago maydis revealed the FxxA motif is functionally more important7. Germline inheritance of a pathogenic variant in BRCA2 significantly increases the lifetime risk of developing breast and ovarian cancers 8–10. To date, most pathogenic missense variants have been identified in exon3, encoding the N-terminal Partner and Localizer of BRCA2 (PALB2) binding domain, and exons 15-26, encoding the C-terminal DNA binding domain (DBD) of BRCA2 (Fig.1a, top) 11,12. While several frameshift and non-sense variants associated with increased cancer risk have been identified in exon11, that encodes the eight BRC repeats, no pathogenic missense variants have been reported in this exon 13. A recent functional study has identified two hypomorphic variants, p.Ser1221Pro in BRC2 and p.Thr1980Iso in BRC7, that 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 4 disrupt BRCA2’s interaction with RAD51 and render cells sensitive to chemotherapeutic agents14. The clinical significance of these variants, however, remains to be validated. Moreover, characterization of cisplatin and Poly(ADP-ribose) polymerase inhibitor (PARPi) resistant BRCA2-mutant breast (HCC1428) and pancreatic cancer cell line (Capan-1), along with patient tumors, has revealed revertant variants in BRCA2 that restores, in cis, the open reading frame encoding two BRC repeats 15. These studies hint towards inherent functional redundancy within BRC repeats. Moreover, a mini-BRCA2 carrying a single BRC repeat (repeats 1, 2, 3 or 4) exhibits HR proficiency, albeit two to five times lower than BRCA2 with BRC repeats 1-4 further corroborating their functional redundancy 16. Considering the inherent functional redundancy within BRC repeats, why species harbor multiple repeats is unknown. In this study we used mouse embryonic stem cells (mESC) and mouse models to investigate the relevance of presence of multiple BRC repeats. We demonstrate that the presence of a single BRC repeat is sufficient to support mESC viability and perform majority of canonical BRCA2 functions. Moreover, we were able to generate knock-in mice carrying BRCA2 with a single BRC repeat, BRC2 or BRC4. Both homoz ygous and hemizygous mutant mice are born at expected Mendelian ratios and are fully viable and fertile. However, contrary to the results observed in mESCs, mutant primary fibroblasts exhibited reduced RAD51 recruitment at DSBs. This RAD51 recruitment defect was alleviated under low oxygen (physoxic) conditions, whereas subjecting mESCs, harboring a single BRC repeat, to oxidative stress, by H 2O2 treatment, impaired RAD51 recruitment. Our results demonstrate that oxidative stress destabilizes RAD51 interaction when BRCA2 contains a single BRC repeat, whereas the presence of multiple repeats confers functional robustness. These findings suggest a mechanistic 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 5 basis by which multiple BRC repeats contribute to safeguarding the genome under oxidative stress.

Functional evaluation of BRCA2 BRC repeat mutants in mouse ES cells We evaluated the functional significance of BRCA2 BRC repeats using a well-established mESC-based functional assay 17. We used PL2F7 mESC line in which one of the endogenous Brca2 alleles is non-functional (knockout or KO) and the other is a conditional allele (CKO ) flanked with two halves of human HPRT minigene (Fig. 1a)17. We electroporated PL2F7 mESCs with recombineered BRCA2 cloned in a BAC (Bacterial artificial chromosome) having different BRC deletions, and deleted the endogenous conditional Brca2 allele, to examine their impact on cell viability23. We confirmed loss of the conditional allele in viable HA T resistant clones by Southern analysis (Suppl. Fig. 1b-d). BRC repeats 1-4 and 5-8 are known to be functionally distinct in RAD51 and ssDNA binding based on biochemical studies 18. Therefore, we generated BAC clones containing only BRCA2 with BRC repeats 1-4 or 5-8 (BRC5-8 and BRC1-4 deleted, respectively) to examine their functional relevance. We also generated BRCA2 with deletion of the well-studied BRC4 (BRCΔ 4) and the least conserved BRC repeat BRC6 (BRC Δ 6). We obtained HAT resistant Brca2-null colonies, confirmed by Southern blot analysis (Suppl. Fig. 1c), that were rescued by BAC expressing full length (WT) BRCA2 as well as by all other BRC mutants albeit with a reduced rescue rate (Fig. 1b). Interestingly, loss of all eight BRC repeats (BRC Δ 1–8) failed to

in any viable clone (Fig. 1b). 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 6 Further, we assessed the sensitivity of mESCs expressing BRC mutant BRCA2 towards DNA- damaging agents by an XTT-based cell proliferation assay. mESCs expressing BRCA2 BRC5-8 were hyper-sensitive to all the DNA-damaging agents examined similar to the cells expressing a known pathogenic BRCA2 variant (p.Leu2510Pro, Fig. 1c). However, the sensitivities of cells expressing BRCA2 with BRC1-4, BRC Δ 4 and BRCΔ 6 were comparable to that of WT BRCA2. Next, we examined the RAD51 recruitment ability of cells expressing BRCA2 with varying number of BRC repeats at the DSBs. We used γ− irradiation (IR) to induce DSBs and found mESCs expressing WT BRCA2 as well as BRC Δ 4, BRC Δ 6 and BRC1-4 mutants to have a comparable number of RAD51 foci (marker for HR) colocalizing with γH2AX (marker for DSBs) (Fig. 1d,e). However, the cells expressing BRCA2 with BRC5-8 exhibited a significantly reduced number of RAD51 foci that also appeared to be more diffused (Fig. 1d,e). As previously reported, BRCA2 L2510P mutant mESCs showed minimal or no IR-induced RAD51 foci 19. These results indicated that cells expressing BRCA2 with BRC1-4 are proficient in BRCA2’s canonical functions (cell survival, resistance to DNA-damaging drugs and RAD51 recruitment) but those expressing BRCA2 with BRC5-8, although proficient in cell survival, lack the ability of RAD51 recruitment thereby rendering the cells sensitive to genotoxins. A single BRC is proficient in BRCA2 canonical functions Since the cells expressing BRCA2 with BRC1-4 exhibited proficiency in BRCA2 functions, we next sought to determine which of these four BRC repeats is critical. We generated mESCs expressing BRCA2 with BRC1-3, BRC1-2 as well as those with a single BRC repeat 1, 2, 3 or 4. We found BRCA2 with a single BRC repeat rescued the mESC lethality due to Brca2 loss (Suppl. Fig. 1d) but there was around 66% reduction in the rescue rate compared to WT BRCA2 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 7 (Fig. 2a). These findings suggested that BRCA2 containing even a single BRC repeat can support mESC viability, albeit with reduced efficiency. Further, we examined RAD51 recruitment in these mESCs after IR and observed a significantly reduced number of RAD51 foci in cells expressing BRCA2 with BRC1-3, BRC3 and BRC1 (Fig. 2b,c) as compared to WT mESCs. Remarkably, cells with either BRC2 or BRC4 exhibited RAD51 foci comparable to those observed in WT BRCA2 expressing cells (Fig. 2b,c). We next performed clonogenic survival assay in the presence of varying doses of olaparib (PARP inhibitor) to determine the sensitivity of mESCs expressing mutant BRCA2 with various BRC repeats. Consistent with the observed reduction in RAD51 recruitment, cells expressing BRCA2 with BRC1-3, BRC1 and BRC3 resulted in significantly fewer colonies with olaparib treatment (Fig. 2d). Likewise, no significant difference in olaparib sensitivity was observed in cells expressing BRCA2 with BRC1-2, BRC2 and BRC4 compared to WT BRCA2. These findings suggest that BRCA2, with a single BRC repeat, BRC2 or BRC4, is proficient in RAD51 recruitment at the DSBs and provides resistance to olaparib. To directly examine the binding of BRC mutant BRCA2 with RAD51 we performed pull-down experiments. We overexpressed recombinant 2XMBP tagged BRCA2 (full-length or indicated BRC repeat mutant) in HEK-293T cells and pulled down the fusion protein using amylose beads to quantify the amount of bound endogenous RAD51. We did not detect any RAD51 bound to BRCA2 deleted for BRC1-8 (Fig 2e, lane 8) or BRC1-4 (Fig. 2e, lane 7) in agreement with prior

18. Notably, significant amount of bound RAD51 was detected by full-length BRCA2 and BRCA2 containing BRC1-4 (Fig. 2e). The amount of RAD51 pulled down by BRCA2 with single BRC repeats 1, 2, 3 or 4 was reduced by at least 50% compared to the that with full-length protein (Fig. 2e,f, lanes 2-6, 9) with BRC3 showing the least RAD51 binding. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 8 We next assessed the level of HR proficiency in the mESCs expressing WT as well as single BRC repeats (1,2,3 or 4). We utilized a previously described HR assay in mESCs carrying a 29bp deletion in the blasticidin resistance gene 19 and repaired it with a donor DNA using CRISPR/Cas9. We quantified the number of blasticidin resistant colonies to assess HR efficiency. Surprisingly, despite normal RAD51 foci formation and olaparib resistance observed in BRC2/4 expressing cells, they exhibited significantly reduced HR efficiency. The level of HR in mESC with BRC3 was most reduced and comparable to the known BRCA2 hypomorph L2510P (Fig. 2g). We also assessed the role of different BRC mutants in protection of stalled replication forks (RF) by performing DNA fiber assay 20. The significance of BRC repeats in protecting stalled RFs is unknown as the C-terminal RAD51 binding (CTRB) domain of BRCA2, encoded by exon27 (Fig 1a, top), is known to play a crucial role 21. We induced replicative stress by hydroxyurea (HU) to stall RF and found that all BRC mutants of BRCA2 are proficient in protecting stalled RFs (IdU/CldU ratio>0.9) suggesting that BRC repeats had no impact on this function (Suppl. Fig. 1e). A BRCA2 hypomorphic variant (R2336H) used as control exhibited defect in RF protection (IdU/CldU ratio <0.5) 22. The FYSA residue is critical for BRC2-Mediated BRCA2 Function We employed our previously developed CRISPR-based Saturation Genome Editing (SGE) approach in mESCs 23 where we engineered a Brca2-null mESC line by stably expressing a transgene encoding human BRCA2 with a single BRC2 repeat [ Brca2–/–; tg(BRCA2BRC2)]24. We used this approach (Fig. 3a) to confirm the functional redundancy among BRC repeats by individually integrating each BRC repeat (BRC1–8) in place of BRC2. In line with our previous 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 9 results, we found BRC1/2/4 supported cell viability along with BRC8, whereas BRC3/5/6/7 alone failed to do so (Fig. 3b). Since, BRC repeats harbor a conserved FxxA motif, which is essential for RAD51 binding (Fig. 3c)4, we next interrogated its functional importance by substituting the FYSA residues of BRC2 with all possible 20 amino acids at each position. We performed SGE in Brca2–/–; tgBRCA2BRC2 mESC line (Fig. 3a). Analysis revealed that substitutions at F, S and A residues were largely intolerable, indicating their critical role in BRC2 function. Interestingly, conservative substitutions of F to W, I, L, Y, or V were partially tolerated in the absence of any drug treatment (Fig. 3d), consistent with findings in Ustilago maydis 7. However, in the presence of cisplatin or olaparib, only W exhibited partial cell viability. In contrast, all substitutions of the tyrosine residue of FYSA, had only a moderate impact on cell viability. Our results reiterate that within the FxxA motif of a BRC repeat, F and A residues are critical for BRCA2 function. Generation of knock-in mice expressing BRCA2 with a single BRC repeat Our studies in mESCs provide strong evidence that BRCA2 mutants with a single BRC repeat, especially BRC2 or BRC4, are fully functional. Encouraged by these findings, we tested whether BRCA2 with a single BRC repeat can support viability in mice, as loss of BRCA2 is embryonic lethal 3. We targeted exon11 of mouse Brca2, which encodes eight BRC repeats, using CRISPR/Cas9 system with two different gRNAs, upstream and downstream of the eight BRC repeats (Suppl. Fig. 2a). Using two different donor oligos to replace BRC repeats 1-8 with either BRC2 or BRC4 alone, we obtained multiple correctly targeted knock-in mice (Suppl. Fig. 2b, c; Fig. 4a). We obtained heterozygous mice for the targeted alleles and will be referred to as Brca2 BRC2/+ ( BRC2/+, for simplicity) and Brca2BRC4/+ ( BRC4/+). We intercrossed BRC2/+ as 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 10 well as BRC4/+ mice to test whether mice expressing a single BRC repeat are viable in homozygous state. Remarkably, we obtained viable homozygous mice of both genotypes (BRC2/BRC2 and BRC4/BRC4 ) in expected Mendelian ratios (Table 1). We also crossed the BRC2/+ and BRC4/+ mice with mice heterozygous for the Brca2-null allele (Brca2KO/+ or KO/+ for simplicity) to examine the impact of having a single BRC repeat in hemizygous state 3. Notably, we also obtained hemizygous animals ( BRC2/KO and BRC4/KO ) in expected Mendelian ratios (Table1). In addition to the BRC2 and BRC4 knock-in alleles, we obtained a mouse that had an in-frame deletion of 3393bp (Suppl. Fig. 2c, Fig. 4a) in exon11 predicted to skip 1131aa of BRCA2 lacking all eight BRC repeats ( Brca2BRCΔ 1-8, referred to as BRC Δ 1-8). When we intercrossed BRCΔ 1-8/+ mice or crossed them with KO/+ mice, we failed to obtain any viable BRC Δ 1- 8/BRCΔ 1-8 or BRCΔ 1-8/KO mice suggesting that loss of all the BRC repeats is embryonic lethal and at least one BRC repeat is required for embryonic development (Table 1). These results corroborate our in vitro findings showing that at least one BRC repeat of BRCA2 required to sustain viability. Mice harboring single BRC repeat of BRCA2 exhibit moderate developmental defects We observed no overt phenotype in BRC2 and BRC4 homozygous as well as hemizygous mutant mice. However, starting five weeks postpartum, the BRC4/BRC4 and BRC4/KO mice consistently displayed slightly reduced body weight compared with control mice, while the BRC2/BRC2 and BRC2/KO mice maintained their body weight (Fig. 4b). To evaluate this growth defect, we examined mammary gland branching and quantified the number of terminal end buds (TEBs) as a proxy of developmental progression. We harvested mammary glands from 5-week- old females of various genotypes, stained them with carmine alum. While the number of TEBs in 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 11 BRC2/BRC2 and BRC2/KO females were comparable to the control genotypes, they were significantly reduced in BRC4/BRC4 and BRC4/KO genotype (Fig. 4c, d). The reduction in mammary TEBs could reflect a proliferation defect in mammary adult stem cells, and potentially other adult stem cells. Therefore, we examined fetal liver cells, the site of embryonic hematopoiesis, to assess defects in hematopoietic stem and progenitor cells (HSPCs). We isolated fetal liver cells from 16.5dpc embryos of each genotype and cultured them with growth factors to promote colony formation. Equivalent number of colony forming units (CFUs) were observed in control genotypes along with BRC2 and BRC4 homozygous mutants while BRC2/KO and BRC4/KO fetal liver cells exhibited mildly reduced number of CFUs (Fig. 4e, Suppl. Fig. 3a). Additionally, the hemizygous HSPCs were hypersensitive to olaparib treatment comparable to LP/LP genotype (a known HR-deficient p.Leu2431Pro BRCA2 variant expressing mouse) 19 (Fig. 4e, Suppl. Fig. 3a). To examine the regenerative ability of HSPCs in vivo , we injected multiple (weekly) sub-lethal doses of 5-fluorouracil (5FU), to eradicate all dividing hematopoietic cells and stimulate the proliferation and differentiation of HSPCs 25, in mice of all genotypes. As expected, repeated 5FU administration promoted exhaustion of proliferating HSPCs and mice become moribund after 100 days. However, all mice displayed a comparable pattern of mortality, and the mutants did not exhibit increased sensitivity to this proliferative stress (Fig. 4f). This result shows that despite a mild impairment in CFU formation in the in vitro assay, the hemizygous hematopoietic cells are functionally normal in vivo. The minor developmental defects observed in BRC4 mutants did not affect their fertility and fecundity. Both male and female mutant mice were fully fertile, and we observed normal morphology in their testes and ovaries based on H&E staining (Suppl. Fig. 3b). Furthermore, 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 12 spermatocyte spreads revealed normal meiotic progression as well as normal RAD51 recruitment at leptotene/zygotene stages (Suppl. Fig. 3c, d). Primary fibroblasts from BRC mutant mice display impaired IR-induced RAD51 recruitment To compare the results observed in mESCs with somatic cells, we isolated embryonic fibroblasts (MEFs) of all genotypes and exposed them to IR to examine RAD51recruitment to DSBs. Unlike the observations made in mESCs, the homozygous and hemizygous mutant MEFs displayed significantly reduced number of RAD51 foci positive nuclei as well as reduced number of foci per nucleus, compared to control genotypes (Fig. 5a-c). To validate this result, we generated adult fibroblasts from ear punch and performed similar experiment. While the mutant adult fibroblasts showed an increase in RAD51 foci positive cells relative to the MEFs, these numbers remained significantly lower (Suppl. Fig. 4a, b). A defect in RAD51 recruitment indicates a defect in DNA repair. To further assess this defect, we analyzed the MEFs for the presence of chromosomal aberrations with or without MMC treatment using Brca1 Δ 11 (Brca1Δ 11/Δ 11) fibroblasts as control for genomic instability 26. Interestingly, the hemizygous MEFs exhibited significantly higher chromosomal aberrations per nuclei compared to the WT MEFs in untreated condition (Fig. 5d, Suppl. Fig. 4c). However, all the mutant MEFs exhibited a significant increase in chromosomal aberrations after MMC treatment. The increased genomic instability in mutant MEFs further supports that BRCA2 containing BRC repeats 2 or 4 are not fully proficient in DNA repair. Considering the defect in RAD51 recruitment in MEFs, we examined whether there was any difference in the protection of stalled RFs between mESCs and MEFs. We performed DNA fiber 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 13 assay using Brca1Δ 11/Δ 11 fibroblasts, known to have a defect in RF protection 26, as a control. Similar to the results observed in mESCs, mutant MEFs did not exhibit defect in stalled RF protection (Suppl. Fig. 4d). Next, we assessed the sensitivity of the whole animal to DNA damage by injecting them with MMC. Studies show that mice with impaired HR die within three weeks of MMC injection 27. Surprisingly, we observed that the MMC injection was well tolerated by mice of all genotypes except BRC4/KO. While half of BRC4/KO mice died within 3weeks of injection, the remaining survived until the end of the study (Fig. 5e). Such MMC sensitivity observed in BRC4/KO mice can be attributed to their lower body weight at the time of treatment (Fig. 4b). The in vivo results suggest that BRCA2 with single BRC repeat is not fully functional contrary to

observed in mESCs expressing BRCA2 with BRC2 or BRC4 (Fig. 2). Especially the mutant mESCs were fully functional in RAD51 recruitment (Fig. 2b, c) as opposed to the mutant MEFs. We hypothesized that the differential culture conditions used to maintain mESCs and MEFs maybe responsible for the observed phenotype. mESCs are cultured in media consisting of β -mercaptoethanol (antioxidant), which provides a minimal oxidative stress 28. However, fibroblasts are maintained in ambient oxygen levels (~21%), without any antioxidant in the media, which is far greater than the in vivo physoxic environment (3-7.4% pO2) 29,30. This difference in culture conditions may induce high oxidative stress in fibroblasts. To test this hypothesis, we cultured the MEFs under low oxygen (3%) conditions and examined IR-induced RAD51 recruitment. We observed a remarkable recovery in RAD51 recruitment in the mutant MEFs (Fig. 5f, g). In a complementary approach, we subjected the mESCs to oxidative stress by exposing them to varying concentrations of H 2O2 along with IR. We observed a significant dose- dependent reduction in RAD51 recruitment in the mESCs expressing BRCA2 with single BRC 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 14 repeats (BRC2/BRC4) (Fig. 5h, i). These results suggest that BRCA2 with a single BRC repeat is insufficient to cope with an additional oxidative stress, revealing the need to have multiple repeats. Residual HR has a dominant role relative to replication fork stability for mouse survival Our previous study has shown that Brca2L2431P knock-in mice, expressing HR-deficient Brca2 variant, do not survive in an Mlh1-null background 31. Mlh1-null mice are defective in DNA mismatch repair as well as DNA2-mediated stalled RF protection 31. We decided to examine the survival of our BRC2 and BRC4 mutant mice under this genetic stress by crossing them with Mlh1 mutant mice (Mlh1-/+). Surprisingly, we observed survival of mice homozygous for BRC2 and BRC4 on Mlh1 -/- background ( Brca2BRC2/BRC2;Mlh1-/- and Brca2BRC4/BRC4;Mlh1-/-, referred as double mutant) in expected Mendelian ratios (Table 2). Additionally, primary fibroblasts from the ear punches were generated and examined for IR-induced RAD51 recruitment. Consistent with our previous results (Suppl. Fig. 4a, b), double mutant fibroblasts displayed reduced percentage of RAD51 foci positive nuclei compared to the controls (Fig. 6a, b). We also examined stalled RF protection in these fibroblasts and the double mutant fibroblasts displayed RF degradation after HU treatment as observed in Mlh1 -/- fibroblasts (Fig. 6c). These findings confirm that residual HR is sufficient to support mouse viability even under defect in DNA mismatch repair and protection of stalled RFs. The critical function of BRCA2 in protecting MRE11-mediated stalled RF degradation is by its CTRB (residues 3260-3314) domain encoded by exon27 21. Moreover, a study reported CTRB is required for the RAD51 nucleation at the DSBs32. To ascertain which of the two RAD51 binding motifs, the BRC repeats or the CTRB, is critical we deleted the CTRB domain in a BAC containing full length BRCA2 ( Δ CTRB) and BRCA2 with BRC4 (BRC4 Δ CTRB). Expressing 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 15 these BACs in our mESCs revealed higher rescue rate (11.5%) in Δ CTRB as compared to that in BRC4Δ CTRB (3.43%) (Fig. 6d). Interestingly, the BRC4Δ CTRB rescue rate was comparable to the BRCA2 with BRC4 alone (Fig. 2a). Furthermore, mESCs expressing Δ CTRB exhibited comparable ability of RAD51 recruitment with WT and BRC4 cells. However, a slight reduction in RAD51 recruitment was observed in cells expressing BRC4 Δ CTRB (Fig. 6e, f). Next, we treated these cells with olaparib and found no difference in the sensitivity of mESCs expressing Δ CTRB or BRC4 Δ CTRB compared to WT BRCA2 (Fig. 6g). Furthermore, examining the stalled RF protection ability resulted in consistent fork degradation in cells expressing either Δ CTRB or BRC4 Δ CTRB, which was suppressed in the presence of mirin (MRE11 inhibitor) (Fig. 6h). These findings unequivocally demonstrate the division of functions among the two RAD51 binding domains of BRCA2. Furthermore, the failure to obtain viable mESCs or mice upon deletion of all BRC repeats demonstrates that BRC repeats are critical for survival in contrast to the CTRB domain. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 16

BRCA2 functions as a tumor suppressor because of its role in error free DNA repair by HR 1. It recruits RAD51 to the DSBs, via the eight BRC repeats in the middle of the protein and the CTRB near the C-terminal end 33. Most species have multiple BRC repeats, including mice and humans (eight BRC repeats), and several previous studies have revealed functional redundancy in them 5,16,33. Yet, the functional significance of retaining multiple repeats remains largely unknown. Our study addresses this long-standing question by examining, for the first time in vivo, the requirement for multiple BRC repeats within BRCA2. By using humanized mESCs, expressing BRCA2, we observed the BRC repeats 5-8 were dispensable for specific BRCA2 functions. We identified the most crucial BRC repeat for BRCA2 function and confirmed that BRCA2 harboring a single BRC repeat (BRC2 or BRC4) is fully functional in the mESCs, albeit with reduced HR efficiency. We failed to obtain viable mice lacking all the BRC repeats in hemizygous or homozygous states confirming our in vitro results. However, knock-in mouse models with BRCA2 harboring a single BRC repeat (BRC2 or BRC4) in homozygous as well as hemizygous state are viable and do not show any overt phenotype. This suggests that approximately one third of BRCA2 is dispensable for survival. Additionally, BRC2 mutant mice performed better across several developmental parameters compared to BRC4 mutants, suggesting superior in vivo performance of BRC2 relative to BRC4. We observed reduced HR efficiency in mESCs expressing BRCA2 with BRC2 or BRC4, but this did not hamper the BRC mutant mice survival in Mlh1-null background, defective in DNA mismatch repair and replication fork protection. We were puzzled by the severe defect in RAD51 recruitment and increased chromosomal aberrations in the mutant MEFs. This was in sharp contrast to the efficient RAD51 recruitment 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 17 observed in mESCs and spermatocytes (during meiosis). We hypothesized that the inconsistencies observed in the RAD51 recruitment abilities between fibroblasts and mESCs can be attributed to the different culture conditions of these cell lines. Regular culture conditions expose MEFs to ambient oxygen (21%), which is much higher than the 3-7.4% pO 2 in tissues inducing oxidative stress 29,30. However, mESCs are cultured in the presence of a feeder layer (along with β -mercaptoethanol in the media) that provides a highly supportive microenvironment ensuring pluripotency in them under minimal oxidative stress 28. Remarkably, the mutant MEFs exhibited a significant increase in RAD51 recruitment when cultured in low oxygen (3%) conditions. Likewise, we observed a decline in RAD51 recruitment ability of mESCs expressing BRCA2 with BRC2 or BRC4 in presence of H 2O2. These results along with the reduced binding of RAD51 with single BRC repeat containing BRCA2 (Fig. 2e,f), suggest that the oxidative stress might further destabilize this interaction and the presence of multiple BRC repeats provide robustness. Similar phenomena can also explain the reduction in mutant HSPC colonies and their hypersensitivity to olaparib, yet the mice were not sensitive to proliferative stress induced by 5FU. In conclusion, our functional studies demonstrate that BRC repeats are essential for the survival of cells and mice. Despite the presence of eight BRC repeats in human and mouse BRCA2, a single BRC repeat is sufficient for RAD51 recruitment to DSBs even in the absence of the CTRB domain. We have shown that 1093 amino acids (out of 3329) of mouse BRCA2, containing 7 BRC repeats, are dispensable for its critical function in HR and RF protection. We conclude that the residual HR present in the mutant (homozygous and hemizygous) mice is sufficient for their physiological growth and development but inefficient when the mutant cells are subjected to oxidative stress. We predict that inducing oxidative stress may resens itize certain HR-proficient 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 18 cancers, with large in-frame deletion in exon11 retaining few BRC repeats, that are resistant to olaparib15,34.

Generation of mESCs harboring BRCA2 with different deleted BRC repeats Different BRC repeat deletions in BRCA2 were generated using BRCA2 cloned in a Bacterial artificial chromosome (BAC). We used the oligonucleotide-based 'hit and fix' BAC recombineering based approach as described earlier 35. Correctly targeted bacterial clones were confirmed by PCR and sequencing. We electroporated 25 μ g BAC DNA into 10 7 actively dividing Pl2F7 mESCs and selected the colonies with geneticin (180 μ g/ml, 10131-035, Gibco). Geneticin resistant colonies were picked on a 96-well plate, and the total RNA was isolated. Expression of the integrated BAC was confirmed by one step reverse transcriptase PCR [G597, ABM) primer sequence in Extended Table 1]. The RT-PCR positive colonies were expanded and subjected to Cre-HA T selection as described earlier 17. Colonies were picked in 96-well plate and genomic DNA was isolated. Surviving colonies were identified using Southern blot analysis on these colonies as explained earlier 17. The percentage of rescued colonies (showing only the KO band on the southern blot, Suppl. Fig. 1B) was multiplied by the proportion of colonies acquired on a HAT plate as opposed to that on a no-HAT plate to get the rescue percentage. Assay to test sensitivity for different DNA damaging drugs Actively dividing mESCs were plated in 96-well plates (10,000 cells/well). Required concentrations of DNA damaging drugs (cisplatin, MMC, olaparib and IR) were added to each well and incubated at 37°C. After 72hrs the plates were washed with PBS and 100 μ l of XTT solution (J61726, Sigma) solution (1mg/ml in phenol red free DMEM, 21041025, Gibco) was 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 19 added. After incubation at 37°C for one hour the colorimetric readings were taken and the graph was plotted on GraphPad Prism. IR-induced RAD51 foci formation Cells were seeded (105 cells /well) on poly-d-Lysine coverslips (GG-12, Neuvitro) and irradiated 6Gy for mESCs and 10Gy for fibroblasts). After recovery (6hrs for mESCs and 3hrs for fibroblasts) the cells were treated with a hypotonic solution (85.5 mM NaCl, 5 mM MgCl2, pH 7) for 10mins and then fixed (4% paraformaldehyde and 0.03% SDS in PBS) for 10mins at room temperature. Cells were incubated overnight at 4°C with primary antibodies [ γ H2AX (1:500, JBW301, Millipore), RAD51 (1:250, PC130, Millipore)] diluted in antibody dilution buffer (1% BSA, 0.3% TritonX100, 5% goat serum in PBS). Next morning, the cells were washed 3X with PBST (PBS containing 0.2% Triton X-100) and incubated with secondary antibodies [Alexa- fluor anti-mouse 594 (1:1000, A11005, Invitrogen) and anti-rabbit 488 (1:500, A11034, Invitrogen)] diluted in PBS at 37°C for one hour. The cells were washed 3X with PBST and stained for one minute with DAPI (1:50,000, 11190301, Sigma). Clean, labeled slides were used to mount the coverslips using anti-fade mount (P36930, Invitrogen). The slides are examined using Zeiss AXIO imager M2 (63X). RAD51 foci formation in mESCs under oxidative stress mESCs expressing different BRCA2 constructs were cultures in M15 media. 105 cells were plated on coverslips. Next day the media was changed to varying doses of H2O2 dissolved in non β -mercaptoethanol containing DMEM media. After 3 hrs, these were given 6GY of IR and the media was changed to fresh H2O2 containing DMEM media. After 6 hrs, cells were stained for RAD51 and γ-H2AX as described above. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 20 HR assay We used mESCs (F7A10) with 29 /i5 bp deleted blasticidin resistance gene 19. 2×10 6 actively dividing F7A10 cells (harboring WT BRCA2 or with different BRC repeats) were plated. After 24/i5 h cells were nucleofected using nucleofector kit (VPH-1001, Lonza) with 2 /i5 µg of gRNA- Cas9 expressing plasmid px330 (Addgene plasmid #158973), and 1 /i5 µg of linearized donor blasticidin sequence (supplemented with the 29bp, PAM mutated) cloned in Topo-TA vector (Invitrogen 450641) as per manufacturer’s guidelines. Cells were resuspended in 5 /i5 ml M15 media (Knockout DMEM, /i5 15% FBS, 1X β -mercaptoethanol) and plated on 60 /i5 mm culture dish with blasticidin-resistant feeders. 48hrs post plating the cells were treated with blasticidin (15 /i5 µg/ml in M15 media, 11139-03 Gibco) for five days. After the treatment, cells were incubated in M15 media until the colonies started to appear. Assessment of HR was performed by staining the plates with methylene blue (0.05% in 70% ethanol) and colonies were counted. Amylose Pull-downs Transient transfection of 1µg of each construct (phCMV1 mammalian expression vector including a 2XMBP fusion to BRCA2 containing the BRC repeat(s) indicated) into actively dividing 5x10 5 293T cells was done in 6-well plates using TurboFect (R0531, ThermoFisher Scientific). After 36hrs the cells were harvested in 500µL of the lysis buffer [50mM HEPES (pH7.5), 250mM NaCl, 1% Igepal CA-630, 1mM MgCl2, 1mM DTT, 250 Units/mL Benzonase (EMD Millipore), and 1X EDTA-free protease inhibitor cocktail (C762Q72, Roche)]. Total cellular lysate aliquots were taken before batch binding for protein expression analysis. Cell lysates were batch bound to 20µL of amylose resin for 2hrs to capture the 2XMBP tagged BRCA2 proteins, washed 3X in wash buffer [50mM HEPES (pH7.5), 250mM NaCl, 0.5mM EDTA, and 1mM DTT]. Proteins were then eluted in the same buffer containing 10mM maltose 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 21 and 10% glycerol. Samples were run on 4-15% gradient SDS-PAGE TGX stain-free gels (456- 8086, Bio-Rad). The 2XMBP-BRCA2 proteins (amylose pull-downs) were visualized by Stain- Free imaging on a ChemiDoc MP imaging system (Bio-Rad). Total cell lysate gels were also visualized by Stain-Free imaging to ensure equal loading. BRCA2 and RAD51 proteins were detected by transferring gels to PVDF (IPVH00010, Millipore) membranes overnight at 4°C, blocking for 30 minutes with 5% milk in 1XTBS-T [50 mM Tris (pH7.5), 150 mM NaCl, 0.05% Tween20], incubating the membranes overnight with either anti-MBP (1:5000, E8032L, NEB) or anti-RAD51 antibody (1:1000, PC130 Millipore) in 1XTBS-T. Membranes were washed 3 times with TBS-T and then incubated with secondary mouse and rabbit antibodies (HRP-conjugated, sc-516102 and sc-2004, respectively, Santa Cruz Biotechnology). The Western blots were visualized using Clarity Western ECL substrate (170-5061, Bio-Rad) for five minutes and visualized on a ChemiDocMP system. Band densitometry was performed using ImageLab (Bio- Rad). DNA fiber assay In a 6-well plate, 5 × 10 5 cells were plated and treated with thymidine analogues 8 μ g/ml CldU (8µg/ml) and IdU (90 μ g/ml) for 30 minutes each, followed by a 4-hour treatment with 4 mM hydroxyurea (HU). After treatments, the cells were trypsinized and resuspended in PBS. Cell suspension and lysis buffer were added to the slide to accomplish cell lysis. The slides were tilted to allow the fibers to spread and air dry after incubation for roughly ten minutes. The fibers were fixed overnight using a methanol:acetic acid (3:1) mixture, then rehydrated using PBS and denatured for one hour in 2.5M HCl. After washing with PBS, the slides are blocked for 40 minutes using 5% BSA. The primary mouse anti-BrdU antibody (1:500, 347580, Becton Dickinson) and the rat anti-BrdU antibody (1:500, ab6326, Abcam) were incubated on the fibers 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 22 for two hours after blocking. After rinsing with PBST, the slides were treated for one hour at room temperature with secondary anti-mouse AlexaFluor488 (1:500, A21202, Invitrogen) and anti-rat AlexaFluor594 (1:500, A11007, Invitrogen). The slides were washed three times with PBST and were mounted (P36930, Invitrogen) and examined using Zeiss AXIO imager M2 (63X). Green and red fiber lengths (at least 100 fibers per sample) were measured using Fiji software, and their ratios were calculated and shown 36. Saturation Genome Editing in BRC repeats We used an mESC line expressing a single copy of human BRCA2 ( Brca2−/−; Tg[BRCA2] ) [Clone: F7/F7]24. Two sgRNAs targeting regions upstream of the BRC1 repeat and downstream of the BRC8 repeat were cloned into the pX458-Cas9-GFP plasmid. Three million cells were nucleofected with these plasmids and a ssODN containing only the BRC2 repeat using the Lonza Nucleofector 2B (program A030). GFP-positive cells were sorted and plated, and 96 colonies screened to isolate a clonal line containing only the BRC2 repeat ( Brca2 −/−; Tg[BRCA2BRC2]). This line was used for subsequent saturation genome editing (SGE) experiments, where oligo donor pools containing either individual BRC repeats or NNN degenerate codons for all possible amino acid substitutions at defined residues were nucleofected into mESCs. Post-nucleofection, cells were pooled, treated with cisplatin, olaparib, or DMSO, and harvested for DNA extraction and deep sequencing, with data analyzed as previously described 23. BRC repeats counts and frequencies were extracted from FASTQ files using a custom Python script (v 3.10.12). Heatmaps were generated in R Studio (v4.4.0) using the ggplot2 (v 4.0.1) and ComplexHeatmap packages (2.20.0) 37,38. Generation of Brca2BRC2/+ and Brca2BRC4/+ Knock-in mice 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 23 Fertilized zygotes (0.5dpc) were isolated from C57Bl/6 female mouse. These zygotes were microinjected with ribonucleoprotein complex (RNP complex) containing pure Cas9 protein, upstream and downstream gRNAs for exon11, and donor DNA (containing either for BRC2 or BRC4) (for sequences see supporting document). As explained in Suppl. Fig. 2A, both the gRNAs help in excising out all the eight BRC repeats from exon11 from one of the Brca2 alleles and are repaired by homologous recombination using the flanking sequences of the donor DNA provided. The injected zygotes were implanted in pseudo-pregnant females for embryo development. Live pups obtained from these females are weaned at 3 weeks of age and tail clips are obtained for genotyping and sequencing (Suppl. Fig. 2b,c, Suppl. Table 1). Sequencing confirmed animals are back crossed to C57Bl/6NCR mice (for at least 10 generations) and maintained as separate mouse lines of BRC2 and BRC4. Ethics statement The Guide for the Care and usage of Laboratory Animals (The National Academies Press; 8th edition) was followed in the housing, breeding, and study usage of all mice. The NCI-Frederick Animal Care and Usage Committee (ACUC) approved the study protocol (Animal Study# 24- 471). The animals were kept in a 12-hour cycle of light and dark. Temperatures between 20ºC- 27ºC and relative humidity levels between 30 and 70% were maintained in the rooms. All animal studies were performed in compliance with ARRIVE (Animal Research: Reporting In Vivo Experiments) guidelines (https://arriveguidelines.org/arrive-guidelines). Mammary gland isolation and Carmine Alum staining Five-week-old female mice were used to harvest their mammary glands, which were then preserved in Carnoy's solution (6:3:1 of ethanol, chloroform, and glacial acetic acid). Carmine 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 24 Alum (C1022, Sigma) was used to stain the glands overnight after they had been rehydrated using progressively lower grades of ethanol wash. After being dried in progressively higher grades of ethanol, the glands were incubated for 2-3 days in xylene replacement to clean the tissue. The whole-mount glands were evaluated after they were photographed with a Zeiss Axiocam brightfield microscope. Colony forming assay from fetal liver cells Embryos were collected at E16.5 from pregnant females and euthanized. These embryos' livers were meticulously removed and tweezed in IMDM. A 40µm cell strainer was used to filter the cell suspension. MethoCult (M3231, Stem Cell Technologies) media with growth factors (10% FBS, 100ng/ml muSCF, 100ng/ml huTpo, 100ng/ml huFlt3L, 50ng/ml muIL6, 30ng/ml muIL3) were used to plate 25,000 live cells. These were incubated at 37ºC for 7-10 days. Iodonitrotetrazolium chloride (1mg/ml, I10406, Sigma) was used to stain the colonies. Exhaustion of hematopoietic cells by multiple 5-Flurouracil injection Six to eight weeks old mice from each genotype were intraperitoneally injected (weekly) with 5- Flurouracil (5FU) (135mg/kg, NDC68001-525-27, USP grade, BluePoint Laboratories) for 3 weeks. After 50 days from first injection, the animals were again injected with 5FU continuously every week until they started to show mortality. In accordance with ACUC procedures, mice exhibiting symptoms of distress, such as weight loss, were euthanized. Generation of embryonic mouse fibroblasts E13.5 embryos were obtained from timed mating of different genotypes. After genotyping the embryos, they were minced in 0.5% trypsin (15400-054, Gibco) and incubated at 37°C for 30 minutes. After incubation the slurry was plated on 100mm cell-culture plates in DMEM+10% 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 25 FBS in 37°C, 5% CO 2 incubator. After they became confluent, small aliquots of mouse embryonic fibroblasts (MEFs) were frozen in liquid nitrogen as P0 (passage zero). Generation of adult fibroblasts Ear punch from the animals of required genotype were obtained. After a brief wash in 70% ethanol, they were rinsed in sterile Hank’s balanced salt solution (HBSS). The ear punches were finely minced in collagenase solution (2000U/ml in HBSS, C7657, Sigma) and incubated at room temperature (on a roller mixer) for 3hr. Digested ear punches were centrifuged, and supernatant was removed. Rest of the procedure is same as discussed above for embryonic fibroblasts. Analyses of chromosomal aberrations MEFs of desired genotypes were incubated with MMC (100nM, 11435, Cayman Chemicals) for 12hr and released in normal media for 12hr. After release, the cells were arrested in metaphase using colcemid (10ug/ml, 15210-016, KaryoMAX) dissolved in DMEM and incubation for 12hr. The cells were fixed in methanol: acetic acid (3:1) and metaphase spreads were made on a clean labeled slide. Slides were stained with Giemsa, and the chromosomal aberrations were quantified. Animal survival study post MMC injection Six to eight weeks old animals from each genotype were intraperitoneally injected with MMC (3.5mg/kg, single dose, NDC 55150-451-01, USP grade, Eugia). Animals were observed every day till they started to show mortality. In accordance with ACUC procedures, mice exhibiting symptoms of distress, such as weight loss, were euthanized. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 26 Histology Tissues (ovaries and testes) were overnight fixed in 10% formalin, paraffin-embedded, sectioned (5µm) and stained for H&E (hematoxylin and eosin). The stained slides were visualized under bright field microscope. Meiotic chromosomes spread and analysis Male mice aged 4-6 weeks had their testes used to prepare meiotic spreads. The testes were placed in hypo-extraction buffer PBS (30 mM Tris pH8.2, 50 mM sucrose, 17 mM citric acid, 5 mM EDTA, 0.5 mM DTT, and 0.1 mM PMSF) after being quickly rinsed in. After carefully removing the tunica from the testes and removing the tubules with fine forceps, the mixture was incubated at room temperature for half an hour. 25µl of 0.1M sucrose solution and a tiny piece of the digested tissue segment from the previous stage were put on a clean labeled slide that had been prerinsed with PFA (4%, pH 9.2 adjusted with 50 mM boric acid). This tissue was shredded and spread out on the slide using a fine needle. For slow drying the slides were kept in a humid chamber overnight. The immunofluorescence staining was performed using primary antibodies: mouse anti-SYCP3 (1:500, sc-74568, Santa Cruz); rabbit anti-RAD51 (1:250, PC130, Millipore). Secondary antibody staining and further processes were performed as described in the RAD51 foci formation assay section. Generation of Brca2 BRC2 and Brca2BRC4 animals in Mlh1-/- background We mated the Brca2 BRC2/BRC2 and Brca2BRC4/BRC4 mice with Mlh1-/+ mice ( Mlh1-/- mice are infertile)39 to obtain heterozygous mice for both the genes ( Brca2BRC2/+;Mlh1-/+ and Brca2BRC4/+;Mlh1-/+). We intercrossed these double heterozygous animals to obtain double mutants (Brca2BRC2/BRC2;Mlh1-/- and Brca2BRC4/BRC4;Mlh1-/-). 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 27 Statistical analysis Microsoft Excel and GraphPad Prism version 6.0 were used for all statistical tests. Figure legends for all experiments explain the exact statistical test and p-values along with the error bars. Biological material availability All unique materials used in the study are readily available from the authors upon request as indicated in the Methods section. Data availability All relevant data are provided as supplementary information. Any additional information will be provided upon request. Code availability https://github.com/MelissaGall/BRC_saturation_heatmap

We thank Drs. Ira Daar and Jonathan Keller for helpful discussions and critical review of the manuscript. This research was supported in part by the Intramural Research Program of the National Institutes of Health (NIH) (L.T. and S.K. Sharan). The contributions of the NIH author(s) were made as part of their official duties as NIH federal employees, are in compliance with agency policy requirements, and are considered Works of the United States Government. However, the findings and conclusions presented in this paper are those of the author(s) and do not necessarily reflect the views of the NIH or the U.S. Department of Health and Human Services. R.B.J. was supported by an NIH grant (R01 CA270788). 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 28 Author Contribution APM conceived, designed and performed most of the experiments and analyzed the results; SS, ES, TS, DC performed CRISPR/cas9-based SGE; GM, JRJ, RBJ performed in vitro RAD51 binding assay; SKSengodan performed DNA fiber assays, NO generated BRC mutant BAC constructs, SP performed mammary gland studies, MG performed the bioinformatics analysis; M.A. helped with mice studies. PPA generated the knock-in mice, FTA and L T helped with gRNAs and donor DNA design for knock-in mice. SB performed cytogenetic analysis, APM and SS prepared the figures; SKSharan conceived and supervised the study. APM and SKSharan wrote the manuscript, and all authors reviewed and edited the manuscript. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 29

1. Sung, P. & Klein, H. Mechanism of homologous recombination: Mediators and helicases take on regulatory functions. Nature Reviews Molecular Cell Biology vol. 7 739–750 Preprint at https://doi.org/10.1038/nrm2008 (2006). 2. Scully, R., Panday, A., Elango, R. & Willis, N. A. DNA double-strand break repair-pathway choice in somatic mammalian cells. Nat Rev Mol Cell Biol 20, 698–714 (2019). 3. Sharan, S. K. et al. Embryonic lethality and radiation hypersens itivity mediated by Rad51 in mice lacking Brca2. Nature 386, 804–810 (1997). 4. Bignell, G., Micklem, G., Stratton, M. R., Ashworth, A. & Wooster, R. The BRC Repeats Are Conserved in Mammalian BRCA2 Proteins. Human Molecular Genetics vol. 6 https://academic.oup.com/hmg/article-abstract/6/1/53/2356685 (1997). 5. Carreira, A. & Kowalczykowski, S. C. Two classes of BRC repeats in BRCA2 promote RAD51 nucleoprotein filament function by distinct mechanisms. 108, (2011). 6. Rajendra, E. & Venkitaraman, A. R. Two modules in the BRC repeats of BRCA2 mediate structural and functional interactions with the RAD51 recombinase. Nucleic Acids Res 38, 82–96 (2010). 7. Sutherland, J. H. & Holloman, W. K. Determinants governing BRC function evaluated by mutational analysis of Brh2 in Ustilago maydis. DNA Repair (Amst) 127, 103511 (2023). 8. Finch, A. et al. Salpingo-oophorectomy and the Risk of Ovarian, Fallopian Tube, and Peritoneal Cancers in Women With a BRCA1 or BRCA2 Mutation. JAMA 296, 185–192 (2006). 9. Cavanagh, H. & Rogers, K. M. A. The role of BRCA1 and BRCA2 mutations in prostate, pancreatic and stomach cancers. Hered Cancer Clin Pract 13, 16 (2015). 10. Levine, D. A. et al. Fallopian Tube and Primary Peritoneal Carcinomas Associated With BRCA Mutations. Journal of Clinical Oncology 21, 4222–4227 (2003). 11. Guidugli, L. et al. Functional assays for analysis of variants of uncertain significance in BRCA2. Hum Mutat 35, 151–164 (2014). 12. Karchin, R., Agarwal, M., Sali, A., Couch, F. & Beattie, M. S. Classifying V ariants of Undetermined Significance in BRCA2 with Protein Likelihood Ratios. Cancer Inform 6, CIN.S618 (2008). 13. Gayther, S. A. et al. Variation of risks of breast and ovarian cancer associated with different germline mutations of the BRCA2 gene. Nat Genet 15, 103–105 (1997). 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 30 14. Jimenez-Sainz, J. et al. BRCA2 BRC missense variants disrupt RAD51-dependent DNA repair. Elife 11, e79183 (2022). 15. Sakai, W. et al. Secondary mutations as a mechanism of cisplatin resistance in BRCA2- mutated cancers. Nature 451, 1116–1120 (2008). 16. Siaud, N. et al. Plasticity of BRCA2 function in homologous recombination: Genetic interactions of the PALB2 and DNA binding domains. PLoS Genet 7, (2011). 17. Kuznetsov, S. G., Liu, P. & Sharan, S. K. Mouse embryonic stem cell-based functional assay to evaluate mutations in BRCA2. Nat Med 14, 875–81 (2008). 18. Chatterjee, G., Jimenez-Sainz, J., Presti, T., Nguyen, T. & Jensen, R. B. Distinct binding of BRCA2 BRC repeats to RAD51 generates differential DNA damage sensitivity. Nucleic Acids Res 44, 5256–5270 (2016). 19. Mishra, A. P. et al. BRCA2-DSS1 interaction is dispensable for RAD51 recruitment at replication-induced and meiotic DNA double strand breaks. Nat Commun 13, 1751 (2022). 20. S., K. S. et al. Degradation of 5hmC-marked stalled replication forks by APE1 causes genomic instability. Sci Signal 13, eaba8091 (2020). 21. Schlacher, K. et al. Double-strand break repair-independent role for BRCA2 in blocking stalled replication fork degradation by MRE11. Cell 145, 529–542 (2011). 22. Kharat, S. S. et al. Degradation of 5hmC-marked stalled replication forks by APE1 causes genomic instability. Sci Signal 13, (2020). 23. Sahu, S. et al. Saturation genome editing-based clinical classification of BRCA2 variants. Nature 638, 538–545 (2025). 24. Sahu, S. et al. Saturation genome editing of 11 codons and exon13 of BRCA2 coupled with chemotherapeutic drug response accurately determines pathogenicity of variants. PLoS Genet 19, e1010940 (2023). 25. Ganguly, R., Anand, S., Metkari, S. & Bhartiya, D. Effect of Aging and 5-Fluorouracil Treatment on Bone Marrow Stem Cell Dynamics. Stem Cell Rev Rep 16, 909–921 (2020). 26. Nacson, J. et al. BRCA1 Mutational Complementation Induces Synthetic Viability. Mol Cell 78, 951-959.e6 (2020). 27. Lim, P. X., Zaman, M., Feng, W. & Jasin, M. BRCA2 promotes genomic integrity and therapy resistance primarily through its role in homology-directed repair. Mol Cell 84, 447- 462.e10 (2024). 28. Joddar, B. & Ito, Y . Artificial niche substrates for embryonic and induced pluripotent stem cell cultures. J Biotechnol 168, 218–228 (2013). 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 31 29. McKeown, S. R. Defining normoxia, physoxia and hypoxia in tumours—implications for treatment response. Br J Radiol 87, 20130676 (2014). 30. Keeley, T. P. & Mann, G. E. Defining Physiological Normoxia for Improved Translation of Cell Physiology to Animal Models and Humans. Physiol Rev 99, 161–234 (2019). 31. Sengodan, S. K. et al. Mismatch repair protein MLH1 suppresses replicative stress in BRCA2-deficient breast tumors. Journal of Clinical Investigation 134, (2024). 32. Kwon, Y . et al. DNA binding and RAD51 engagement by the BRCA2 C-terminus orchestrate DNA repair and replication fork preservation. Nat Commun 14, 432 (2023). 33. Filippo, J. S. et al. Recombination Mediator and Rad51 Targeting Activities of a Human BRCA2 Polypeptide*. Journal of Biological Chemistry 281, 11649–11657 (2006). 34. Tao, H. et al. Acquired multiple secondary BRCA2 mutations upon PARPi resistance in a metastatic pancreatic cancer patient harboring a BRCA2 germline mutation. Am J Transl Res 12, 612–617 (2020). 35. Sharan, S. K., Thomason, L. C., Kuznetsov, S. G. & Court, D. L. Recombineering: a homologous recombination-based method of genetic engineering. Nat Protoc 4, 206–223 (2009). 36. Mishra, A. P. et al. Characterization of BRCA2 R3052Q variant in mice supports its functional impact as a low-risk variant. Cell Death Dis 14, 753 (2023). 37. Gu, Z., Eils, R. & Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–2849 (2016). 38. Wickham, H. Ggplot2. (Springer International Publishing, Cham, 2016). doi:10.1007/978-3- 319-24277-4. 39. Baker, S. M. et al. Involvement of mouse Mlh1 in DNA mismatch repair and meiotic crossing over. Nat Genet 13, 336–342 (1996). 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 32 Figure legends Fig. 1: BRC repeats 5-8 are dispensable for BRCA2 functions. a) Pictorial representation of BRCA2 with different functional domains including N-terminal PALB2 binding domain, eight BRC repeat, C-terminal DNA binding domain (DBD) and the C-terminus RAD51 binding (CTRB) domain. Schematic representation of mESC-based BRCA2 functional complementation assay for cell survival using HA T selection. b) HA T rescue percentage of mESCs harboring BRCA2 with different BRC repeat deletions. BRCA2, with all the BRC repeats deleted, failed to rescue any mESC colony (n=3 independent clones, error bar-SEM, one way ANOV A, compared with WT). c) XTT based drug sensitivity assay using different DNA damaging agents. mESCs harboring BRCA2 with BRC5-8 are hypersensitive to all the drugs comparable to the known hypomorphic BRCA2 variant L2510P (n=3 independent clones, error bar-SEM, Students t-test, two tailed, for each point compared with WT). d) Representative images depicting RAD51 foci at the DSBs generated by 6Gy IR to mESCs harboring BRCA2 with different BRC repeat deletions. e) Quantification of percentage RAD51 positive nuclei observed in (d). mESCs with BRC5-8 exhibited significantly lower number of RAD51 positive nuclei compared to WT (n=4 independent clones, error bar-SEM, one way ANOVA, compared with WT). **p<0.01, ***p<0.001 Fig. 2: BRCA2 with single BRC repeat is proficient in canonical functions. a) HA T rescue percentage of mESCs with deletions in BRC repeat 1-4 (pictorial representation of mutant BRCA2 on left). Rescue percentage of BRCA2, with single BRC repeats, fell by one-third as compared to full length (n=3 independent clones, error bar-SEM, one way ANOV A, compared with WT). b) Representative images depicting RAD51 foci at the DSBs generated by 6Gy IR to mESCs harboring BRCA2 with single BRC repeat. c) Quantification of percentage RAD51 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 33 positive nuclei observed in (b). mESCs with BRC3 exhibited significantly lower number of RAD51 positive nuclei compared to WT and BRC2 and BRC4 were equivalent to that in WT (n=4 independent clones, error bar-SEM, one way ANOVA, compared with WT). d) Clonogenic survival assay to determine olaparib sensitivity of mESCs with deletions in BRC repeat 1-4. mESCs with BRC3 exhibited significant sensitivity towards olaparib (n=3 independent clones, error bar-SEM, Students t-test, two tailed). e) Western blots of total cellular lysates (TLC) from 293T cells transiently transfected with the indicated 2XMBP tagged BRCA2 BRC constructs (left panel, lanes 2-9). Lane 1 is a non-transfected control to visualize non-specific binding of RAD51 to the amylose beads. Western blots of amylose pull-downs from the same lysates depicted in first one. 36 hours post-transfection, cell lysates were detected by western blotting against MBP (BRCA2 constructs) or endogenous RAD51 (37 kDa). f) Quantification of band densitometry performed on amylose pull-down gel. RAD51 protein bound to each BRCA2 BRC construct was normalized to the amount of 2XMBP-BRCA2 fusion protein bound and eluted from amylose beads in the StainFree image. The percentage of RAD51 bound to full-length BRCA2 protein was set to 100% in the analysis (n=3 technical replicates, error bar-SEM, Students paired t-test). g) Schematic representation of blasticidin-resistance based HR assay in mESCs with single BRC repeats (top). Bottom panel shows quantification of HR levels relative to WT. Significant reduction in total HR is observed in all the samples with single BRC repeat. mESCs with BRC3 exhibited HR values comparable to hypomorphic BRCA2 mutation (L2510P) (n=3 independent clones, error bar-SEM, Students t-test, two tailed). *p<0.05, p<0.01, ***p<0.001 Fig. 3: Functional analysis of BRC repeats in BRCA2. a) Schematic representation of the saturation mutagenesis approach using an ES cell line expressing BRCA2 only the BRC2 repeat. 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 34 This approach was used to generate all 8 possible BRC repeat variants. WT BRC2 was targeted with a PAM modified BRC2 repeat. Cells were selected for 14 days, followed by deep sequencing to assess the impact of each BRC repeat on cell fitness and response to DNA- damaging agents. b) Heatmap showing cell survival in DMSO, cisplatin and olaparib for each of the eight BRC repeats. The shade of blue indicates relative survival, while white denotes lethality. Data is derived from three independent replicates. c) Sequence alignment showing the conserved FxxA motif across eight BRC repeats present in BRCA2. d) Heatmap showing saturation mutagenesis of the FYSA motif within the BRC2 repeat to all 20 amino acids. Data is derived from three independent replicates. Fig. 4: Phenotypic analysis of knock-in mice expressing BRCA2 with a single repeat, BRC2 and BRC4. a) Pictorial representation of BRCA2 protein expressing in the knock-in mice BRC2, BRC4 and BRC Δ 1-8. b) Body weights of males and females of all the genotypes from weaning (3-weeks) to 8-weeks old. BRC4/KO mice showed lower body weights from week 5 onwards (n=5, error bars-SEM, Students t-test for each point compared with WT). c) Representative images of Carmine-alum-stained mammary glands, isolated from 5-week-old females of each genotype. d) Quantification of visible TEBs in the mammary glands. BRC4 homozygous and hemizygous females have significantly less TEBS compared to WT (n=6 mammary glands, error bar-SEM, one way ANOVA, compared with WT). e) Quantification of fetal liver colonies. BRC2/KO and BRC4/KO exhibited significantly lower number of colonies compared to WT. BRCA2 hypomorphic variant L2431P ( LP/LP) is used as a negative control. BRC2/KO and BRC4/KO exhibited hypersensitivity to olaparib (n=3, error bar-SEM, one way ANOV A, compared with WT). f) Mouse survival plot after multiple 5FU injections at indicated time points 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 35 to accelerate hematopoietic aging. No significant difference in survival is observed in any genotype (sample size mentioned in the parenthesis). *p<0.05, p<0.01, ***p<0.001 Fig. 5: Primary fibroblasts from BRC mutant mice exhibit defect in RAD51 recruitment. a) Representative images depicting RAD51 foci at the DSBs generated by IR in MEFs isolated from mice of all genotypes. b) Quantification of percentage RAD51 positive nuclei observed in (a). Homozygous and hemizygous fibroblasts exhibited significantly lower number of RAD51 positive nuclei compared to WT (n=3, error bar-SEM, one way ANOV A, compared with WT). c) Quantification of RAD51 foci per nuclei in MEFs of all genotypes. Homozygous and hemizygous fibroblasts exhibited significantly lower number of RAD51 foci per nuclei (n=100, error bar-SD, one way ANOV A, compared with WT). d) Quantification of chromosomal aberrations per spread in MEFs of all genotypes in untreated and 100nM MMC treated conditions. BRC2/KO and BRC4/KO MEFs exhibited increased number of aberrations in untreated conditions as compared to WT (n=45, error bar-SD, one way ANOV A, compared with WT). MMC treatment exacerbates these aberrations in each genotype. Brca1 Δ 11/ Δ 11 MEFs are used as control for chromosomal aberrations. Representative images are shown in Suppl. fig. 4c. e) Mouse survival plot after single dose of MMC injections. BRC4/KO mice exhibited increased lethality and half of them dies within 3-weeks of injections (n=10, Log-rank Mantel–Cox test). f) Representative images depicting RAD51 foci at IR induced DSBs in MEFs of all genotypes cultured in low oxygen (3%). g) Quantification of percentage RAD51 positive nuclei observed in (f). Significantly increased percentages of RAD51 positive nuclei are visible in both homozygous and hemizygous fibroblasts compared to that in regular (21% oxygen) culture conditions (a, b) (n=3, error bar-SEM, one way ANOVA, compared with WT). h) Representative images depicting RAD51 foci at IR induced DSBs in mESCs under different concentrations of 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 36 H2O2. i) Quantification of percentage RAD51 positive nuclei observed in (h). A dose dependent decline in percentages of RAD51 positive nuclei are visible in both BRC2 and BRC4 expressing mESCs (n=3, error bar-SEM, Students t-test, compared with IR). *p<0.05, p<0.01, ***p<0.001 Fig. 6 : Replication fork defects do not hamper mouse and ES cell survival even with minimal HR. a) Representative images revealing RAD51 foci at the IR induced DSBs in adult fibroblasts isolated from mice of all genotypes in Mlh1-null background. b) Quantification of percentage RAD51 positive nuclei observed in (a). BRC2 and BRC4 homozygous fibroblasts (in Mlh1-null background) exhibited significantly lower number of RAD51 positive nuclei compared to WT (n=3, error bar-SEM, one way ANOV A, compared with WT). c) Replication fork stability measured by DNA fiber assay on fibroblasts of all genotypes in Mlh1-null

(explained in the schematic, Replication Fork is protected if the ratio of IdU/CldU tracks is close to 1, and it is degraded if the ratio is significantly reduced). Mlh1-null fibroblasts along with BRC2 and BRC4 homozygous fibroblasts (in Mlh1-null background) exhibited significant fork degradation as compared to WT (n>100, error bar-SD, Students t-test). d) Representative Southern blot images showing rescue percentage of Δ CTRB and BRC4 Δ CTRB expressing mESCs. e) Representative images depicting RAD51 foci at DSBs generated by IR in Δ CTRB and BRC4 Δ CTRB expressing mESCs. f) Quantification of percentage RAD51 positive nuclei observed in (e). BRC4 Δ CTRB expressing mESCs exhibited significantly lower number of RAD51 positive nuclei compared to WT (n=3, error bar-SEM, one way ANOV A, compared with WT). g) Colony forming assay to determine olaparib sensitivity of mESCs expressing Δ CTRB and BRC4 Δ CTRB. Only the known hypomorphic BRCA2 variant L2510P mESCs exhibited significant sensitivity towards olaparib (n=3, error bar-SEM, Students t-test, two tailed). h) Replication fork stability measured by DNA fiber assay on Δ CTRB and BRC4Δ CTRB 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 37 expressing mESCs (in presence of HU and HU+Mirin). Δ CTRB and BRC4 Δ CTRB expressing mESCs exhibited significant fork degradation as compared to WT which was rescued in presence of MRE11 inhibitor Mirin (fork protection when IdU/CldU ≥ 1, n>100, error bar-SD, Students t- test). **p<0.01, ***p<0.001 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 38 Table 1- BRC2 and BRC4 mutant mice are obtained in expected Mendelian ratios in homozygous and hemizygous conditions BRC2/+ X BRC2/+ BRC2/+ X KO/+ WT BRC2/+ BRC2/BRC2 WT BRC2/+ KO/+ BRC2/KO Observed 56 151 64 Observed 75 73 76 79 Expected 67.75 135.5 67.75 Expected 75.75 75.75 75.75 75.75 χ 2 value P=0.134 χ 2 value P=0.969 BRC4/+ X BRC4/+ BRC4/+ X KO/+ WT BRC4/+ BRC4/BRC4 WT BRC4/+ KO/+ BRC4/KO Observed 28 38 37 Observed 75 50 70 60 Expected 25.75 51.5 25.75 Expected 63.75 63.75 63.75 63.75 χ 2 value P=0.013 χ 2 value P=0.122 BRCΔ 1-8/+ X BRCΔ 1-8/+ BRC Δ 1-8/+ X KO/+ WT BRC Δ 1-8/+ BRC Δ 1-8/ BRCΔ 1-8 WT BRC Δ 1-8/+ KO/+ BRC Δ 1-8/KO Observed 42 76 0 Observed 99 41 35 0 Expected 29.5 59 29.5 Expected 43.75 43.75 43.75 43.75 χ 2 value P=2.4E-09 χ 2 value P=7E-25 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint 39 Table 2- Mating of BRC2 and BRC4 mutant mice with Mlh1 mutant mice Brca2BRC2/+;Mlh1-/+ X Brca2BRC2/+;Mlh1-/+ Observed Expected Brca2BRC4/+;Mlh1-/ + X Brca2BRC4/+;Mlh1-/+ Observed Expected Brca2+/+;Mlh1+/+ 10 13 Brca2+/+;Mlh1+/+ 14 8.8125 Brca2BRC2/+;Mlh1+/+ 44 26 Brca2BRC4/+;Mlh1+/+ 26 17.625 Brca2BRC2/+;Mlh1-/+ 40 52 Brca2BRC4/+;Mlh1-/+ 30 35.25 Brca2BRC2/BRC2;Mlh1+/+ 19 13 Brca2BRC4/BRC4;Mlh1+/+ 16 8.8125 Brca2BRC2/BRC2;Mlh1-/+ 26 26 Brca2BRC4/BRC4;Mlh1-/+ 14 17.625 Brca2BRC2/BRC2;Mlh1-/- 12 13 Brca2BRC4/BRC4;Mlh1-/- 8 8.8125 Brca2+/+;Mlh1-/+ 29 26 Brca2+/+;Mlh1-/+ 14 17.625 Brca2+/+;Mlh1-/- 9 13 Brca2+/+;Mlh1-/- 8 8.8125 Brca2BRC2/+;Mlh1-/- 19 26 Brca2BRC4/+;Mlh1-/- 11 17.625 χ 2 value P=0.004 χ 2 value P=0.022 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint Mouse Embryonic Stem Cell Brca2 ko/ko Tg (BRCA2) Figure 1 a PALB2- Binding Domain BRC Repeats HD OB1 OB2 OB3 CTRB Domain 3418 aa WT BRC ∆4 BRC ∆6 BRC1-4BRC5-8L2510P 0 20 40 60 80% RAD51/ H2AX positive cells ** *** *** *** γ WT BRC Δ4 BRC Δ6 BRC1-4BRC5-8BRC Δ1-8 0 5 10 15 20Rescue percentage ** ** ****** *** γH2AX RAD51 Merge WT WT BRC 4 BRC 6 BRC1-4 BRC5-8 L2510P 0.0 0.2 0.4 0.6 1.0 1.2 1.5 0.0 0.5 1.0 Relative survival Conc. µM Cisplatin * ** *** *** *** *** 0.0 2.5 5.0 10.0 100.01000.010000.0 0.0 0.5 1.0 Conc. nM Relative survival *** *** ** Olaparib 0 5 10 20 40 60 80 0.0 0.5 1.0 Conc ng/ml Relative survival MMC ** ** *** *** *** *** 0 2 4 6 8 10 0.0 0.5 1.0 Gy Relative survival ** *** *** *** *** Irradiation BRC 4 BRC 6 BRC 1-4BRC 5-8L2510P b c d e HD OB1 OB2 OB3 WT BRCA2 HD OB1 OB2 OB3 HD OB1 OB2 OB3 HD OB1 OB2 OB3 HD OB1 OB2 OB3 HD OB1 OB2 OB3 BRC 4 BRC 6 BRC 1-8 BRC5-8 BRC1-4 DBD Figure 1 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint Figure 2 a b c WT BRC1-3BRC1-2BRC1BRC2BRC3BRC4 0 5 10 15 20Rescue percentage *** *** *** *** *** *** WT BRC1-3BRC1-2BRC1BRC2BRC3BRC4 0 20 40 60 80% RAD51/ H2AX positive cells *** BRC1BRC2BRC3BRC4BRC1-4BRC5-8 FLdelBRC1-8Full-Length 0.0 0.5 1.0 1.5% RAD51 Bound to BRCA2 0 5 10 100 1000 0.0 0.5 1.0 Relative survival γ d e f g γH2AX RAD51 Merge BRC 1-3BRC 1-2BRC 1BRC 2BRC 3BRC 4 WTBRC1BRC2BRC3BRC4L2510P 0.0 0.1 0.2 0.3 0.8 1.0 1.2 1.4Relative HR ns HD OB1 OB2 OB3 BRC1-3 HD OB1 OB2 OB3 BRC1-2 HD OB1 OB2 OB3 BRC1 HD OB1 OB2 OB3 BRC2 HD OB1 OB2 OB3 BRC3 HD OB1 OB2 OB3 BRC4 Olaparib (nM) WT BRC2 BRC3 BRC1-3 BRC1 BRC1-2 BRC4 L2510P 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 ** * * * *** *** *** *** *** *** ** 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint Figure 3 c BRC1 NHSFGGSFRTASNKEIKLSEHNIKKSKMFFKDIEE BRC2 NEVGFRGFYSAHGTKLNVSTEALQKAVKLFSDIEN BRC3 FETSDTFFQTASGKNISVAKESFNKIVNFFDQKPE BRC4 KEPTLLGFHTASGKKVKIAKESLDKVKNLFDEKEQ BRC5 IENSALAFYTSCSRKTSVSQTSLLEAKKWLREGIF BRC6 FEVGPPAFRIASGKIVCVSHETIKKVKDIFTDSFS BRC7 SANTCGIFSTASGKSVQVSDASLQNARQVFSEIED BRC8 NSSAFSGFSTASGKQVSILESSLHKVKGVLEEFDL Consensus FxxA DMSO CIS OLA PositiveNegativePolar neutralNon−polarAromatic H K R D E N Q S T A C G I L M P V F W Y Silent * F Y S A F Y S A F Y S A log2(ratio) 0 5 DMSO BRC-1 BRC-2 BRC-3 BRC-4 BRC-5 BRC-6 BRC-7 BRC-8 0 1 2 3 4 CisplatinOlaparib Survival ratio Day 14/Day 3 All BRC repeat variants generated by SGE DMSO Cisplatin Olaparib Cell growth Response to DNA damaging drugs Day-0Day-3Day-8-14 Mouse embryonic stem cells (Brca2-/-;Tg[BRCA2BRC2]) Cas9 sgRNA BRC ssODN library BRC-2 BRC-3 BRC-1 BRC-5 BRC-6 BRC-4 BRC-7 BRC-8 BRC2 NEVGFRGFYSAHGTKLNVSTEALQKAVKLFSDIEN Saturation into all possible amino acids DMSO Cisplatin Olaparib D3 NGS Survival ratio = Log2 Freq. Day 14 Freq. Day 3 D14 a b d Non-sense Survival ratio Day 14/Day 3 Maximum survival No survival Max survival (Resistant) No survival (Sensitive) 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint Figure 4 a 3 4 5 6 7 8 0 10 20 30 Age in weeks Body wt. in gramsWT BRC2/+ BRC4/+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO ** Males (n=5) 3 4 5 6 7 8 0 10 20 30 Age in weeks Body wt. in gramsWT BRC2/+ BRC4/+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO ** Females (n=5) WT BRC2/+ BRC4/+ WT BRC2/+BRC4/+ BRC2/BRC2BRC2/KOBRC4/BRC4BRC4/KO 0 10 20 30 40 * * BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO 0 40 80 100 120 140 0 50 100 Days Probability of Survival WT (n = 8) BRC2/+ (n = 6) BRC4/+ (n = 5) BRC2/BRC2 (n = 6) BRC2/KO (n = 5) BRC4/BRC4 (n = 5) BRC4/KO (n = 5) Multiple injection of 5FU (135 mg/kg) f d e HD OB1 OB2 OB3 HD OB1 OB2 OB3 2 4 2238 aa 2233 aa BRCA2 BRC2 BRCA2 BRC4 HD OB1 OB2 OB3 2198 aaBRCA2 BRC 1-8 Number of TEBs b c WT BRC2/+BRC4/+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO LP/LP 0 10 20 30 40 50Number of colonies UT 100nM Ola *** **** * * * * * * * 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint Figure 5 a b c WT BRC2/+BRC4/+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO 0 5 10 15 20 25Aberrations per nuclei Untreated MMC WTKO/+ BRC2+BRC4+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO 0 20 40 60% RAD51/ H2AX positive cellsγ γH2AX RAD51 Merge WTKO/+BRC2/+BRC4/+BRC2/BRC2BRC4/BRC4 BRC2/KOBRC4/KO γH2AXRAD51Merge WT BRC2/KO BRC4/KOBRC2/BRC2 BRC4/BRC4 d f g e MEFs MEFs (3% Oxygen) 0 50 100 0 50 100 Days post MMC Probability of Survival WT BRC2/+ BRC4/+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO n=10 p<0.01 Single 3.5mg/kg MMC injection Brca1 11/ 11 γH2AXRAD51DAPI WTBRC2BRC4 IR + 300µM H2O2 IR + 400µM H2O2 IR + 500µM H2O2 IR Mouse Embryonic stem cellsh WT KO/+ BRC2+BRC4+ BRC2/BRC2 BRC2/KO BRC4/BRC4 BRC4/KO 0 50 100RAD51 foci/nuclie WT BRC2 BRC4 0 20 40 60 80% RAD51/ H2AX positive cells IR IR+300µM H2O2 IR+400µM H2O2 IR+500µM H2O2 * * * *** ******* γ WT BRC2/BRC2BRC2/KOBRC4/BRC4BRC4/KO 0 20 40 60% RAD51/ H2AX positive cellsγ *** ****** ** *** *** 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint Figure 6 a CTRB BRC4 CTRB BRC Repeats HD OB1 OB2 OB3 CTRB Domain HD OB1 OB2 OB3 HD OB1 OB2 OB3 Wildtype CKO (4.8 Kb) KO (2.2 Kb) (11.6% rescue) (3.43% rescue) W T B R C 4ΔC T R B B R C 4 ΔC T R BL 2 5 1 0 P 0 20 40 60 80% RAD51/ H2AX positive cells WT BRC4 ΔCTRB BRC4 ΔCTRB R2336H 0.0 0.5 1.0 1.5 2.0 2.5 IdU/CldU Ratio HU HU+Mirin 0 10 100 1000 0.0 0.5 1.0 WT BRC4 ΔCTRB BRC4 ΔCTRB L2510P γH2AX RAD51 Merge γH2AX RAD51 Merge d b c γ e f WTBRC 4CTRBCTRB BRC 4 L2510P g h Brca2+/+ Mlh1+/+ Brca2+/+ Mlh1-/+ Brca2+/+ Mlh1-/- Brca2BRC2/+ Mlh1-/+ Brca2BRC2/BRC2 Mlh1-/- Brca2BRC4/BRC4 Mlh1-/- Brca2BRC4/+ Mlh1-/+ Adult fibroblasts Relative Survival CTRB BRC4 CTRB Olaparib (nM) Brca2 +/+;Mlh1 +/+ Brca2 +/+;Mlh1 -/+ Brca2 +/+;Mlh1 -/- Brca2 BRC2/+ ;Mlh1 -/+ Brca2 BRC4/+ ;Mlh1 -/+ Brca2 BRC2/BRC2 ;Mlh1 -/- Brca2 BRC4/BRC4 ;Mlh1 -/- 0 20 40 60% RAD51/ H2AX positive cells ** ** Brca2 +/+; Mlh1 +/+ Brca2 +/+; Mlh1 -/+ Brca2 +/+; Mlh1 -/- Brca2 BRC2/+ ; Mlh1 -/+ Brca2 BRC2/BRC2 ; Mlh1 -/- Brca2 BRC4/+ ; Mlh1 -/+ Brca2 BRC4/BRC4 ; Mlh1 -/- 0.0 0.5 1.0 1.5 2.0IdU/CldU Ratio Mouse Embryonic Stem Cells ****** *** ********* ** *** 105 and is also made available for use under a CC0 license. (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC The copyright holder for this preprintthis version posted January 8, 2026. ; https://doi.org/10.64898/2026.01.07.698272doi: bioRxiv preprint