Heterozygous mutations in the BRCA1 gene predispose women to breast and ovarian cancer, while biallelic BRCA1 mutations are a cause of Fanconi anemia (FA), a rare genetic disorder characterized by developmental abnormalities, early-onset bone marrow failure, increased risk of cancers, and hypersensitivity to DNA-crosslinking agents. BRCA1 is critical for homologous recombination of DNA double-strand breaks (DSB). Through its coiled-coil domain, BRCA1 interacts with an essential partner, PALB2, recruiting BRCA2 and RAD51 to sites of DNA damage. Missense mutations within the coiled-coil domain of BRCA1 (e.g., L1407P) that affect the interaction with PALB2 have been reported in familial breast cancer. We hypothesized that if PALB2 regulates or mediates BRCA1 tumor suppressor function, ablation of the BRCA1–PALB2 interaction may also elicit genomic instability and tumor susceptibility. We generated mice defective for the Brca1–Palb2 interaction (Brca1 L1363P in mice) and established MEF cells from these mice. Brca1L1363P/L1363P MEF exhibited hypersensitivity to DNA-damaging agents and failed to recruit Rad51 to DSB. Brca1L1363P/L1363P mice were viable but exhibited various FA symptoms including growth retardation, hyperpigmentation, skeletal abnormalities, and male/female infertility. Furthermore, all Brca1L1363P/L1363P mice exhibited macrocytosis and died due to bone marrow failure or lymphoblastic lymphoma/leukemia with activating Notch1 mutations. These phenotypes closely recapitulate clinical features observed in patients with FA. Collectively, this model effectively demonstrates the significance of the BRCA1–PALB2 interaction in genome integrity and provides an FA model to investigate hematopoietic stem cells for mechanisms underlying progressive failure of hematopoiesis and associated development of leukemia/lymphoma, and other FA phenotypes.

Significance:

A new Brca1 mouse model for Fanconi anemia (FA) complementation group S provides a system in which to study phenotypes observed in human FA patients including bone marrow failure.

See related commentary by Her and Bunting, p. 4044

Fanconi anemia (FA) is a rare autosomal recessive or X-linked genetic disease caused by biallelic mutations of one of the FA genes. FA patients develop congenital abnormalities including short stature, abnormal thumbs, microcephaly, hyper/hypopigmentation, and exhibit increased cancer risk (1–3). To date, FA-associated mutations have been identified in 22 different genes that encode proteins involved in DNA repair and replication (3). Hypersensitivity to interstrand cross-linking (ICL) agents such as mitomycin C (MMC), diepoxybutane, and cisplatin suggests that defects in ICL repair are critical. The subsequent accumulation of unrepaired DNA and chromosomal fragility frequently leads to early-onset bone marrow failure (BMF). This is likely due to abnormal activation of apoptosis inducing hematopoietic stem cell (HSC) depletion.

Given the central role of BRCA1 in DNA break repair through homologous recombination (HR) and ICL repair, it is not surprising that mutations in BRCA1 can produce FA-like features such as increased predisposition to solid cancers and hypersensitivity to ICL agents. In the few reported patients with biallelic germline mutations in BRCA1, there have been variable FA dysmorphology and increased cancer risk, establishing defective BRCA1 (FA complementation group S or FANCS) as an uncommon cause of FA (4–7). However, none of these FANCS patients developed BMF. In mice, conditional deletion of Brca1 in bone marrow using Mx1-Cre causes variable BMF and hematologic malignancies (8), and mice expressing human BRCA1 5382insC mutation, a common C-terminal frameshift mutation in Ashkenazi Jews, showed severe depletion of hematopoietic stem cells (HSC; ref. 9). In Brca1 knock-in mouse models, introduction of mutation within the first BRCT repeat (S1598F) or a truncating mutation in exon11 does not produce BMF despite having an increased predisposition to tumor development (10, 11). Bunting and colleagues (12) have suggested an HR-independent role for BRCA1 in ICL repair and there are studies noting that mutations in RAD51 can lead to FA-like phenotypes without BMF or leukemia (13, 14). However, the reasons for the variable BMF phenotypes depending on the mutation site have not been fully explained.

The effect of acquired cancer-associated mutations in BRCA1 has been well studied, with tumor-associated missense mutations in the BRCA1 coiled-coil domain (i.e., M1400V, L1407P, and M1411T) known to ablate the BRCA1–PALB2 interaction (15, 16). The functional BRCT domain of BRCA1 is also critical for genome integrity as well as tumor suppression (10). The BRCA1-BRCT domain recognizes a phospho-SPxF motif (17–19), and directly binds three different proteins, Abraxas, Bach1, and CtIP, in a phosphorylation-dependent manner, forming three distinct multiprotein complexes (18, 20, 21). Among these three complexes, the BRCA1–Abraxas–RAP80 complex plays an important role in accumulation of BRCA1 to DSB sites, with recruitment of PALB2 and BRCA2 sequentially (15, 20, 22–25). At DSB foci, PALB2 acts as the adaptor protein between BRCA1 and BRCA2, and the BRCA1–PALB2–BRCA2 complex functions as a scaffold to recruit the RAD51 recombinase, a central protein for HR.

We hypothesized that PALB2 might regulate BRCA1 tumor suppressor activity and designed a knock-in strategy to generate mice defective for Brca1–Palb2 interaction using the Brca1 L1363P mutation, which corresponds to the BRCA1 L1407P in humans. We show here that this model nearly fully recapitulates the human FA phenotype and can be used as a model for studying the effects of BRCA1 complex in producing the range of features associated with FA.

Targeted mutagenesis and mouse generation

The homology arms of the Brca1 L1363P targeting constructs were derived from subcloned fragments of murine 129/Sv genomic DNA. The leucine at position 1363 in exon 13 was changed to proline by site directed mutagenesis. In the final constructs, the homology fragment was interrupted in intron 12 by insertion of a loxP-flanked neomycin selection marker cassette that contains a SpeI restriction site. The targeting construct was linearized with NotI and electroporated into 129/Sv embryonic stem (ES) cells. After drug selection, genomic DNA from drug-resistant ES clones was digested with SpeI and correctly targeted heterozygous ES subclones (Brca1L1363P-neo/+) were identified by Southern blot assay. To produce chimeric mice, the targeted ES cells were injected into C57BL/6 blastocysts. Chimeric males were crossed to wild-type females to obtain heterozygous animals (Brca1L1363P-neo/+). The loxP-flanked neomycin cassette was excised from the targeted allele by mating Brca1L1363P-neo/+ animals with ROSA-Cre transgenic mice to obtain Brca1L1363P/+ mice.

All animal studies were approved by the Ohio State University Institutional Animal Care and Use Committee and performed in compliance with the Guide for the Care and Use of Laboratory under protocols 2012A00000063 (PI-TL).

Establishment of mouse embryonic fibroblasts

Heterozygous Brca1L1363P/+ animals were intercrossed and pregnant females euthanized at E.13.5 to dissect the embryos. After removal of the head and evisceration, the remainder of the embryo was finely minced, trypsinized, and neutralized. Primary MEFs were cultured in DMEM supplemented with 10% FBS, 100 units penicillin/100 μg/mL streptomycin, 2 mmol/L l-glutamine, and 0.25 μg/mL Plasmocin. To establish immortalized MEFs, early passage pMEFs (passage # 2 or #3) were transfected with SV40 large T antigen plasmid using Lipofectamine 2000 (Invitrogen). All MEFs were tested for Mycoplasma (Sigma, D9307) to certify the cells as Mycoplasma-free prior to use for experiments (primary MEFs, passage 1–3; immortalized MEFs, passage 10–15).

Immunofluorescence staining

Cells were fixed in 4% ice cold paraformaldehyde/PBS for 15 minutes at 1 hour after irradiation, permeabilized in 0.2% Triton X-100/PBS for 10 minutes, and blocked in 5% BSA/PBS for 30 minutes. For pericentrin and α-tubulin double immunostaining, ice-cold 100% methanol was used for fixation. Cells were then stained with primary antibodies, Abraxas 1:3,000, Brca1 1:500 (57J), Rad51 (Novus Biological, NBP1-90983, 1:300) and γH2AX (Millipore, Cat# 05-636, 1:5000), Pericentrin (Abcam, Cat#ab4448, 1:250), and α-tubulin (Sigma, Cat# T6199, 1:1,000) for an hour at room temperature. The stained cells were washed 3 times in 1× PBS-T, incubated with Alexa Fluor 594 goat anti-rabbit or Alexa Fluor 488 goat anti-mouse (Invitrogen, 1:400), stained with Hoechst 33342 (Life Technologies, REF H3570), and mounted onto a glass slide with Aqua-Poly/Mount medium (Polysciences Inc.).

Coimmunoprecipitation

pOZ-mouse Palb2 FLAG-HA or empty vector was transfected into MEFs using Jet-pei transfection reagents (Polyplus). Cells were lysed in low salt NP40 lysis buffer. 500 μg lysates were incubated with Brca1 antibody or HA-magnetic beads overnight and the precipitates resolved on 6% SDS-polyacrylamide gels. Transferred blots were probed with HA (Roche, Cat# 11-867-423-001, 1:1,000) or Brca1 (57J, 1:2,000) antibody.

Cytogenetic analysis

Cells were incubated in medium with or without DNA damaging agents (MMC 40 ng/mL or olaparib 1 μmol/L) for 16 hours and treated with 0.05 μg/mL KaryoMax colcemid (GIBCO) for 2 hours. Cells were harvested, incubated in prewarmed 0.56% KCL solution for 30 minutes at 37°C, and fixed in Carnoy's solution. Metaphase spreads were prepared and stained in 0.5% Giemsa solution and analyzed on a Zeiss microscope with a 100× objective under oil.

DR-GFP assay

To generate primary LP/LP MEFs carrying the DR-GFP reporter, we obtained Brca1L1363P/+;PimDRGFP/+ mice by crossing Brca1L1363P/L1363P mice with Brca1+/+;PimDRGFP/DRGFP. Then, Brca1L1363P/+;PimDRGFP/+ mice were bred to obtain Brca1+/+;PimDRGFP/+, Brca1L1363P/+;PimDRGFP/+, and Brca1L1363P/L1363P;PimDRGFP/+ embryos. Primary MEFs were generated at E13.5. To induce clean DSBs, the I-SceI or empty vector was electroporated at 230 V, 930 μF using a Bio-Rad Gene PulsarII. After 48 hours of transfection, cells were harvested and GFP-positive cells were counted by flow cytometry.

Histologic analysis

Tissue samples were collected from euthanized animals, fixed in 10% formalin for 24 to 48 hours, and embedded in paraffin. Tissues were cut into 4-μm sections and stained with hematoxylin and eosin.

Annexin V analysis

Thymus tissue were collected from mice at 1 month of age and were placed on 70-μm cell strainer and crushed using a syringe plunger. The flow through cells were washed and resuspended in PBS. Cells (1 × 106) were stained with Annexin V-FITC (Trevigen) and Annexin-V–positive cells were analyzed by flow cytometry, using a BD LSR II.

Colony-forming assay

Bone marrow cells were flushed from hind limbs of mice at 1 month of age and 1 × 105 cells were seeded in 30-mm dishes in complete Methocult medium (Stem Cell Technologies, M3434). Cells were cultured for 7 to 10 days and the number of colonies were counted.

IHC

Formalin-fixed paraffin-embedded (FFPE) sections were stained using a Bond Rx autostainer (Leica). Briefly, slides were baked at 65°C for 15 minutes and software automatically performed dewaxing, rehydration, antigen retrieval, blocking, primary antibody incubation, post primary antibody incubation, detection, and counterstaining using Bond reagents (Leica). Samples were then manually dehydrated through a graded alcohol series and xylene and mounted. CD3 (Abcam, cat#ab16669, 1:100) and GP100 (Abcam, cat#ab137078, 1:100) antibodies were diluted in antibody diluents (Leica).

For Plzf (Santa Cruz Biotechnology, cat#sc-28319, 1:300), Vasa (Abcam, cat#ab13840, 1:200), and Cyclin D1 (Abcam, cat#ab16663, 1:100) staining, the manufacturer's recommended protocol (Vector laboratories, VECTASTAIN ABC HRP Kit, PK-6101 for Vasa and Cyclin D1 and MOM on mouse basic kit, BMK-2202 for Plzf) was followed.

Quantitative real-time PCR

Total RNA was isolated from bone marrow of 1-month-old mice and cDNA was synthesized via SuperScript III Reverse Transcriptase (Invitrogen). TaqMan gene expression assay was used for qRT-PCR (Cdkn1a Mm00432448_m1 and GAPDH. M99999915_g1).

Bone marrow and thymic mass analysis

Thymic tissue and bone marrow from femur were collected in Eppendorf tubes containing RPMI medium. Thymic tissue was mechanically dissociated through a 70-μm cell strainer to obtain single-cell suspensions. Bone marrow cells were physically separated from the femur using PBS wash and were passed through a 70-μm cell strainer to obtain single-cell suspensions. Red blood cell (RBC) lysing buffer was used to lyse RBCs. After washing with PBS, cells were counted by hematocytometer. Cells (2 × 106) were pipetted for a flow panel. Conjugated antibodies for CD3-FITC (Cat. #553062), CD4-FITC (Cat. #553047), TER119-FITC (Cat. #557915), Gr1-FITC (Cat. #553127), CD117-PE-Cy7 (Cat. #558163), CD4-APC (Cat. #553051), CD4-APC-Cy7 (Cat. #552051), and CD44-PE (Cat. #553134) were from BD Biosciences. Conjugated antibodies for CD11b-PerCP/Cy5.5 (Cat. #101228), Sca1-BV421 (Cat. #108128), CD69-FITC (Cat. #104506), Gr1-AF700 (Cat. #108422), CD25-APC (Cat. #102012), CD8-BV421 (Cat. #100738), TCRβ-PE-Cy7 (Cat. #109222), and NK1.1-PerCP/Cy5.5 (Cat. #108728) were from BioLegend. Conjugated antibodies for CD5-FITC (REF. #11-0051-83) and CD34-AF700 (REF. #56-0341-82) were from eBioscience. All flow cytometry was performed using a BD LSR II. The results were analyzed using FlowJo software.

Gene sequencing

Gene sequencing was performed on DNA isolated from FFPE sections of mouse thymic lymphomas and on blood samples. An amplicon-based panel was custom-design that targeted 29 genes previously known to be mutated in human T-lymphoblastic lymphoma/leukemia (T-LBL/ALL) or in mouse models of precursor T-LBL, including the entire coding regions of Notch1, Phf6, Pten, the Ras genes and p53 as well as the BRCT and Palb2-binding domains of Brca1 and the corresponding interacting domain on Palb2 (26). Sequencing libraries were constructed using the TruSeq Custom Amplicon Low Input kit (Illumina) and sequencing performed on the MiSeq platform using Reagent kit v2/500 cycles (Illumina). A mean read depths of 600× was achieved with a demonstrated sensitivity of 3% variant allele fraction (VAF) established by comparison with nontumor samples and by bioinformatics criteria. Analysis was performed using MiSeq reporter (Illumina) and NextGEne software (SoftGenetics).

Mammary gland analysis

Inguinal mammary glands were harvested as described previously (27). Ductal branch was determined by counting the number of branch points per 3.4 mm of primary branch. Mammary stem cells were analyzed as described previously (28).

Statistical analysis

Statistical analyses were performed using unpaired two-tailed Student t test to compare sets of results from independent groups, and values of P < 0.05 were considered statistically significant. For Kaplan–Meier survival curves, significance was estimated with the log-rank test using Graph-Pad Prism 7 software.

Brca1L1363P phenotype in mice

To determine whether the Brca1–Palb2 interaction is required for Brca1-mediated HR and tumor suppression, we generated a mouse model expressing Brca1 L1363P, the ortholog of the breast cancer–associated BRCA1 L1407P mutation in humans that was shown to ablate the binding of BRCA1 to PALB2 (15). The Brca1 L1363P-neo construct contained this exon 13 mutation and a loxP-flanked selection marker cassette in intron 12 (Supplementary Fig. S1A). The targeted ES cells (Brca1L1363P-neo/+) were identified by Southern blot analysis (Supplementary Fig. S1B) and injected into C57BL/6 blastocysts to obtain chimeric mice, which were bred to produce heterozygous animals. Genotyping for Brca1 L1363P was performed by PCR and the single point mutation was confirmed by direct sequencing (Supplementary Fig. S1C and S1D). Brca1L1363P/L1363P were born with the expected Mendelian frequency. Both Brca1L1363P/L1363P embryos and placentas at E.13.5 were smaller than in Brca1+/+ or Brca1L1363P/+ littermates, indicating that the Brca1 L1363P mutation adversely affected embryonic development (Supplementary Fig. S1E). Henceforth, Brca1+/+, Brca1L1363P/+, and Brca1L1363P/L1363P are simply referred as +/+, LP/+, and LP/LP, respectively.

All LP/LP animals were smaller than littermates (+/+ or LP/+) and exhibited kinked tails (Fig. 1A and B). LP/LP mice exhibited dark pigmentation in the footpads due to increased intraepithelial melanocytes, as demonstrated by GP100 immunostaining (Fig. 1C–E). In addition, mutant LP/LP animals had prominent midbrain areas, likely due to hypoplasia of the cerebellum and the cortex (Fig. 1F).

Figure 1.

Brca1L1363P/L1363P mice show a variety of phenotypic abnormalities. A and B, LP/LP mice exhibit smaller body size, darker skin, and kinky tail. Twenty 4- to 5-week old mice for each genotype were weighed. C, LP/LP mice have darker pigmentation on footpads. D, Hematoxylin and eosin and IHC staining for GP100 in tail skin sections show increased melanocyte numbers producing more melanin in LP/LP epidermis. E, Bar graphs showing percentage of melanin+ cells (on hematoxylin and eosin) and GP100+ melanocytes (immunostain) per field, respectively; >60 fields per slide were analyzed. F, LP/LP mice exhibit prominent colliculus area (yellow dotted line). Error bars, SEM and P values were calculated by unpaired t test. ****, P < 0.0001.

Figure 1.

Brca1L1363P/L1363P mice show a variety of phenotypic abnormalities. A and B, LP/LP mice exhibit smaller body size, darker skin, and kinky tail. Twenty 4- to 5-week old mice for each genotype were weighed. C, LP/LP mice have darker pigmentation on footpads. D, Hematoxylin and eosin and IHC staining for GP100 in tail skin sections show increased melanocyte numbers producing more melanin in LP/LP epidermis. E, Bar graphs showing percentage of melanin+ cells (on hematoxylin and eosin) and GP100+ melanocytes (immunostain) per field, respectively; >60 fields per slide were analyzed. F, LP/LP mice exhibit prominent colliculus area (yellow dotted line). Error bars, SEM and P values were calculated by unpaired t test. ****, P < 0.0001.

Close modal

Homozygous mutant animals were sterile, with smaller reproductive organs compared to age-matched controls (Fig. 2A). The ovary in adult LP/LP animals showed no follicle formation (Fig. 2B). LP/LP testes showed smaller seminiferous tubules with a Sertoli-cell-only phenotype in which all spermatogenic cells were missing (Fig. 2B) with intact maturation in +/+ mice. The levels of spermatogonial stem cells (SSC), as assessed by Plzf (undifferentiated SSC) and Vasa (germ cells) immunostains, were also significantly different at postnatal day 6 (P6) and 10 (P10), with few intratubular Plzf or Vasa positive cells in LP/LP, compared to controls (Fig. 2C and D). In addition, expression of cyclin D1, a proliferation marker for SSC, was reduced in LP/LP testes (Supplementary Fig. S2). The breast tissue in LP/LP females also exhibited markedly reduced glandular branch formation and decreased mammary stem cell populations (Fig. 2E and F).

Figure 2.

Dramatic reduction of stem cell populations in testes and mammary gland was observed in Brca1L1363P/L1363P mice. A, Testes or ovaries were smaller in LP/LP animals. B, Hematoxylin and eosin staining on testes or ovaries from 1-month old mice. Seminiferous tubules were almost empty in LP/LP testes and no follicle formation was observed in LP/LP ovaries. C and D, Plzf and Vasa IHC staining on testes from perinatal days 6 and 10 (P6, P10) showed dramatic reduction of spermatogonial stem cells in LP/LP mice. E, Carmine-stained whole mounts of mammary glands exhibit decreased branch formation in LP/LP female animals. F, Flow cytometry analysis for mammary epithelial stem cells count from 2-month old mice showed reduced counts in LP/LP female animals (P2, luminal; P3, mammary epithelial stem cells). Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001.

Figure 2.

Dramatic reduction of stem cell populations in testes and mammary gland was observed in Brca1L1363P/L1363P mice. A, Testes or ovaries were smaller in LP/LP animals. B, Hematoxylin and eosin staining on testes or ovaries from 1-month old mice. Seminiferous tubules were almost empty in LP/LP testes and no follicle formation was observed in LP/LP ovaries. C and D, Plzf and Vasa IHC staining on testes from perinatal days 6 and 10 (P6, P10) showed dramatic reduction of spermatogonial stem cells in LP/LP mice. E, Carmine-stained whole mounts of mammary glands exhibit decreased branch formation in LP/LP female animals. F, Flow cytometry analysis for mammary epithelial stem cells count from 2-month old mice showed reduced counts in LP/LP female animals (P2, luminal; P3, mammary epithelial stem cells). Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001.

Close modal

Thus, LP/LP-mutant mice develop many FA phenotypes, including growth retardation, hyperpigmentation, cerebellar-hypoplasia and infertility, supporting a distinct role for the Brca1 coiled-coil domain.

Brca1L1363P/L1363P mice develop bone marrow failure, indicative of FA-like phenotype

All LP/LP-mutant animals (n = 63) died within four months, but showed 3 distinct mortality groups. Five mice died perinatally, seven died within a month (T50 = 27), and the remainder died within 40 to 140 days (T50 = 88, as shown later in Figs. 4B and 3A). The five mice that died within three days after birth, showed extremely decreased extramedullary hematopoiesis (EMH) in liver versus controls (Fig. 3B). The seven animals (∼13%) that died within a month were extremely small from birth, developed spontaneous bone marrow failure with bone tissue sections showing variably hypocellular marrow with depletion of all cell types (Fig. 3C). Blood counts from mutant mice displayed progressive macrocytosis and decreased hemoglobin levels over time, indicating that LP/LP mice develop macrocytic anemia, another feature of FA (Fig. 3D).

Figure 3.

A subset of Brca1L1363P/L1363P animals (13%) developed bone marrow failure. A, Kaplan–Meier survival curve of LP/LP mice that died perinatally (yellow line) or of bone marrow failure (blue line) versus overall survival of LP/LP mice (red line; P < 0.0001). B, Liver hematoxylin and eosin staining at postnatal day one exhibited decreased EMH in LP/LP mice that died perinatal. C, Hematoxylin and eosin–stained bone tissue showed depletion of marrow cells in LP/LP mice. D, LP/LP mice develop macrocytic anemia. Macrocytosis was detected in mutant animals from 1 week of age and maintained throughout life. No differences in hemoglobin (Hb) levels were detected from 1 to 4 weeks of age in control and mutant group. However, moribund mutant animals at age 12 weeks showed very low Hb counts versus control (top). Mean corpuscular volume (MCV) and Hb levels in moribund LP/LP animals that survived longer than 3 weeks were compared with that in control group (bottom). E, Flow cytometry analysis for LSK cells count at 4 weeks of age showed reduced HSCs in LP/LP mice. F, Quantitative real-time PCR analysis showed inverse correlation of p21 mRNA level with LSK numbers. RNA was isolated from total bone marrow cell pellets. G, CFU capacity of mouse bone marrow at 4 weeks of age is correlated to the LSK numbers from each animal. Total bone marrow cells (1 × 105) were seeded in Methocult medium and cultured for 7 to 14 days. Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001; N.S., not significant.

Figure 3.

A subset of Brca1L1363P/L1363P animals (13%) developed bone marrow failure. A, Kaplan–Meier survival curve of LP/LP mice that died perinatally (yellow line) or of bone marrow failure (blue line) versus overall survival of LP/LP mice (red line; P < 0.0001). B, Liver hematoxylin and eosin staining at postnatal day one exhibited decreased EMH in LP/LP mice that died perinatal. C, Hematoxylin and eosin–stained bone tissue showed depletion of marrow cells in LP/LP mice. D, LP/LP mice develop macrocytic anemia. Macrocytosis was detected in mutant animals from 1 week of age and maintained throughout life. No differences in hemoglobin (Hb) levels were detected from 1 to 4 weeks of age in control and mutant group. However, moribund mutant animals at age 12 weeks showed very low Hb counts versus control (top). Mean corpuscular volume (MCV) and Hb levels in moribund LP/LP animals that survived longer than 3 weeks were compared with that in control group (bottom). E, Flow cytometry analysis for LSK cells count at 4 weeks of age showed reduced HSCs in LP/LP mice. F, Quantitative real-time PCR analysis showed inverse correlation of p21 mRNA level with LSK numbers. RNA was isolated from total bone marrow cell pellets. G, CFU capacity of mouse bone marrow at 4 weeks of age is correlated to the LSK numbers from each animal. Total bone marrow cells (1 × 105) were seeded in Methocult medium and cultured for 7 to 14 days. Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001; N.S., not significant.

Close modal

To assess HSC populations in bone marrow, we performed LSK (Lin Sca-1+c-kit+) flow analysis at 1 month of age (29). LP/LP mice showed a marked decrease in HSC population versus control mice (Fig. 3E). We also examined HSCs in the BRCT domain mutant Brca1S1655F/S1655F (SF/SF). This mutant Brca1 protein S1655F fails to interact with Abraxas, Bach1, and CtIP and SF/SF-mutant cells exhibit HR defects (10). These SF/SF animals develop solid tumors but they did not show HSC defects (Fig. 3E). The levels of HSCs are inversely correlated with mRNA levels of p21, a p53 target gene, in total bone marrow cells, indicating activation of p53 in LP/LP HSCs but not in SF/SF cells (Fig. 3F).

We also measured progenitor activity of HSCs by the colony-forming unit assay (CFU). Plating of 1 × 105 total LP/LP bone marrow cells rarely yielded colonies in culture and the few colonies obtained were very small and contained few cells, compared to +/+ or LP/+ control HSCs (Fig. 3G). Altogether, these data indicate that a functional BRCA1 coiled-coil domain is essential for proliferation and expansion of the HSC population, whereas functional BRCT repeats appear dispensable.

Brca1L1363P/L1363P mice develop T-LBL that are driven by Notch1 mutations

At four weeks of age, the thymus in LP/LP animals was much smaller than age-matched controls, and the basal number of apoptotic thymocytes in mutant LP/LP mice was 2-fold higher than in control mice (Fig. 4A). T-cell progenitors are derived from HSCs and migrate to the thymus for subsequent T-cell development, proceeding from CD4 CD8 (double negative) to CD4+ CD8+ (double positive) and then single positive (CD4+ CD8 or CD4 CD8+) forms (30). To examine whether Brca1L1363P mutation affects this maturation sequence, T cells in thymus were analyzed in 1-week-old animals by flow cytometry. The early double-negative T-cell subset was significantly proportionally reduced, which is likely is a consequence of the decreased HSCs population in LP/LP animals (Supplementary Fig. S3A and S3B). The proportion of double positive T cells were similar in LP/LP vs. control thymus (Supplementary Fig. S3C), and single positive cells were increased in LP/LP thymus (Supplementary Fig. S3D and S3E), indicating that the homozygous mutant hematopoietic progenitor cells in thymus have intact, qualitative T-cell maturation, at a gross level.

Figure 4.

Brca1L1363P/L1363P animals developed precursor T-lymphoblastic lymphoma/leukemia at 2 to 4 months of age. A, Representative image of the thymus extracted from 4 weeks old +/+ and LP/LP mice. Annexin V assay showed significant increases in apoptotic thymocytes in LP/LP mice at 4 weeks of age. B, Kaplan–Meier survival curve of LP/LP mice that developed massive thymic expansion (green line) with overall survival (red line; P < 0.0001). C, Hematoxylin and eosin staining on normal thymus (C, cortex; M, medullar) and thymic tumor. Cells of the thymic mass were completely replaced with large lymphoblasts with large nuclei. The yellow dotted line indicates normal thymus in +/+ and thymic mass in LP/LP mice. D, Hematoxylin and eosin–stained lung sections showed invasion of thymic tumor into lung. The invading blast cells are CD3-positive T cells. The average percent CD3-positive area per field from three mice/group is shown at the top of the CD3-stained images. E and F, Flow cytometry analysis from total bone marrow revealed that the tumor cells were CD34, CD4+, and CD8+. G, Wright-Giemsa staining of blood slides showed that fewer erythrocytes and large numbers of lymphoblast cells were observed in the moribund LP/LP mice. White blood cells (WBC) from control and mutant mice were counted and graphed. H, Flow cytometry analysis from blood showed that the tumor cells are CD4+ and CD8+. I, Notch1 mutations were present in 26 of 30 tumors analyzed by sequencing. Mutation patterns were similar to those found in human T-lymphoblastic lymphoma/leukemia. Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001.

Figure 4.

Brca1L1363P/L1363P animals developed precursor T-lymphoblastic lymphoma/leukemia at 2 to 4 months of age. A, Representative image of the thymus extracted from 4 weeks old +/+ and LP/LP mice. Annexin V assay showed significant increases in apoptotic thymocytes in LP/LP mice at 4 weeks of age. B, Kaplan–Meier survival curve of LP/LP mice that developed massive thymic expansion (green line) with overall survival (red line; P < 0.0001). C, Hematoxylin and eosin staining on normal thymus (C, cortex; M, medullar) and thymic tumor. Cells of the thymic mass were completely replaced with large lymphoblasts with large nuclei. The yellow dotted line indicates normal thymus in +/+ and thymic mass in LP/LP mice. D, Hematoxylin and eosin–stained lung sections showed invasion of thymic tumor into lung. The invading blast cells are CD3-positive T cells. The average percent CD3-positive area per field from three mice/group is shown at the top of the CD3-stained images. E and F, Flow cytometry analysis from total bone marrow revealed that the tumor cells were CD34, CD4+, and CD8+. G, Wright-Giemsa staining of blood slides showed that fewer erythrocytes and large numbers of lymphoblast cells were observed in the moribund LP/LP mice. White blood cells (WBC) from control and mutant mice were counted and graphed. H, Flow cytometry analysis from blood showed that the tumor cells are CD4+ and CD8+. I, Notch1 mutations were present in 26 of 30 tumors analyzed by sequencing. Mutation patterns were similar to those found in human T-lymphoblastic lymphoma/leukemia. Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001.

Close modal

However, between 2 and 4 months of age, thymic expansion occurred in the LP/LP mice (51/63) that had not died of BMF (7/63) or defective EMH (5/63) due to the expansion of thymocytes terminating in T-LBL. During this period, most LP/LP mice (51/63) developed progressive wasting and upon necropsy displayed massive thymic lymphomatous expansion (Fig. 4B and C). CD3+ T-LBL also invaded the lung and pleura (Fig. 4D). Flow cytometry performed in three LP/LP cases showed that the leukemic blast cells were immature T cells, double positive for CD4 and CD8 and negative for CD34 (Fig. 4E and F). Before 4 weeks of age, no circulating lymphoblasts were detected in blood of mutant animals. By 2 to 4 months, however, blood from LP/LP animals with thymic tumors showed high white blood cell count due to blood and bone marrow leukemic dissemination of the CD4+CD8+ T-LBL (Fig. 4G and H).

Genomic sequencing with a panel of T-cell oncogenes showed recurrent, acquired Notch1 mutations in 25 of 30 cases (83.3%), with 19 of the T-LBL having known or presumed pathogenic Notch1 variants by analogy to human oncogenic variants (26, 31, 32). Most significantly, 9 T-LBL had mutations in the NOD/heterodimerization domain (HD) the domain that is ubiquitously mutated in human T-LBL. These included the most common change in human neoplasms (L1668P in 5) as well as L1574Q, V1578M, I1670S, C1682R, A1686D, and A1691P variants in 1 tumor each (Fig. 4I; Supplementary Tables S1–S3). Truncating or frameshift mutations in the C-terminal TAD and PEST domains of Notch1 were also common, similar to findings in most human T-ALL (33). On the basis of a comparison of VAF and tumor percentage, presumed subclonal/oligoclonal or multiclonal patterns of Notch1 mutations were found in 14 T-LBL samples. There was a higher frequency of pathogenic HD mutations in LP/LP tumors on a p53+/+ wild-type background (5/7) compared with p53+/− heterozygous (2/14) and p53−/− null backgrounds (2/7, P = 0.01).

Hypofunctional p53 rescues much of the Brca1L1363P/L1363P phenotype

In prior mouse models with hyperactivated p53, phenotypes have included small body size, dark footpads and tail skin, as well as elevated levels of thymocyte apoptosis, and cerebellar hypoplasia (34, 35), which resemble many findings seen in LP/LP mice. By crossing these Brca1 mutants with p53 deletion mutants, we sought to assess the degree to which the LP/LP phenotype could be rescued by p53 downmodulation.

Deletion of p53 did not affect the decreased body weight phenotype of LP/LP animals. Both LP/LP; p53+/− or LP/LP; p53−/− mice were still consistently smaller than control mice (Fig. 5A). McGowan and colleagues reported that stabilization of p53 stimulates the proliferation of melanocytes in the epidermis (34). We then determined whether the increase of melanocytes in the epidermis of LP/LP mice is caused by stabilization and hyperactivation of p53 melanocytes by generating LP/LP; p53+/− and LP/LP; p53−/– mice. The pigmentary phenotype was reduced in LP/LP; p53+/− animals and fully reversed in LP/LP; p53−/− animals (Fig. 5B). Likewise, the prominent midbrain area in LP/LP mice due to the hypoplasia of the brain cortex and cerebellum could also be reversed by deletion of one copy of the p53 gene in LP/LP; p53+/− animals, indicating that this phenotype is also caused by elevation of p53 level (Fig. 5C). More importantly, exhaustion of HSCs and scanty CFU formation seen in LP/LP mice were partially rescued by p53 gene deletion, suggesting that reduced HSCs in LP/LP bone marrow are caused by p53 hyperactivation (Fig. 5D–F). Consequently, none of the LP/LP; p53+/− and LP/LP; p53−/– mice died of BMF. However, all of them developed T-LBL with similar latency (Fig. 5G).

Figure 5.

Deletion of p53 rescues several phenotypes of Brca1L1363P/L1363P mice. A, Body weight of control (+/+ or LP/+) and LP/LP mice in p53 +/+ [postnatal day 26–31 (P26–P31)] p53 +/− [postnatal day 28–31 (P28–P31)] or p53 −/− background (P25–P29). p53 deletion did not rescue smaller body size of LP/LP mice. B, LP/LP mice have darker pigmentation on footpads, a phenotype rescued by deletion of the p53 gene. C, Prominent midbrain area in LP/LP mice was partially rescued by loss of one p53 allele. D, Flow cytometry analysis for LSK cell counts from bone marrow of 4-week-old mice. p53 deletion increased the number of LSK cells in LP/LP animals. E, CFU counting from 1 × 105 total bone marrow cells of 4-week-old mice. p53 deletion partially rescue colony-forming capacity of homozygous mutant cells. F, Real-time PCR analysis of p21 from total bone marrow at 4 weeks of age. p53 deletion leads to decreased p21 level. G, Kaplan–Meier survival curve of LP/LP; p53+/− (blue line) and LP/LP; p53−/− (green line) mice that developed massive thymic expansion with overall survival (red line). Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001; N.S., not significant.

Figure 5.

Deletion of p53 rescues several phenotypes of Brca1L1363P/L1363P mice. A, Body weight of control (+/+ or LP/+) and LP/LP mice in p53 +/+ [postnatal day 26–31 (P26–P31)] p53 +/− [postnatal day 28–31 (P28–P31)] or p53 −/− background (P25–P29). p53 deletion did not rescue smaller body size of LP/LP mice. B, LP/LP mice have darker pigmentation on footpads, a phenotype rescued by deletion of the p53 gene. C, Prominent midbrain area in LP/LP mice was partially rescued by loss of one p53 allele. D, Flow cytometry analysis for LSK cell counts from bone marrow of 4-week-old mice. p53 deletion increased the number of LSK cells in LP/LP animals. E, CFU counting from 1 × 105 total bone marrow cells of 4-week-old mice. p53 deletion partially rescue colony-forming capacity of homozygous mutant cells. F, Real-time PCR analysis of p21 from total bone marrow at 4 weeks of age. p53 deletion leads to decreased p21 level. G, Kaplan–Meier survival curve of LP/LP; p53+/− (blue line) and LP/LP; p53−/− (green line) mice that developed massive thymic expansion with overall survival (red line). Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001; N.S., not significant.

Close modal

Brca1L1363P/L1363P selectively ablates Palb2 interactions and promotes HR deficiency and hypersensitivity to ICL

The steady-state levels of the mutant Brca1 L1363P protein were similar to wild-type Brca1 protein in immortalized mouse embryonic fibroblasts (MEF; Fig. 6A). However, LP/LP primary MEFs exhibited impaired proliferation (Fig. 6B), premature senescence and centrosome amplification with hyper-activation of p53 (Supplementary Fig. S4A–S4C).

Figure 6.

Brca1L1363P/L1363P MEFs exhibit defective HR and hypersensitivity to ICL. A, Immunoblot analysis of Brca1 and Bard1 proteins in LP/LP immortalized MEFs. Cells were harvested at 2 hours post-IR with 10 Gy. B, LP/LP primary MEFs show impaired proliferation. The proliferation rate was measured in cells (passage 1) by MTT assay. C, Mutant Brca1 LP protein fails to interact with Palb2. Whole-cell lysates overexpressing empty vector (E) or pOZ-mouse Palb2 FLAG-HA (FH) were subjected to immunoprecipitation (IP) with HA beads or Brca1 antibody followed by immunoblot analyses. D–F, Recruitment of Abraxas (ABX), Brca1, and Rad51 to sites of DNA damage in LP/LP immortalized MEFs. Cells were treated with 10 Gy IR and fixed 2 hours post-IR. G, LP/LP MEFs are defective in HR. Control (+/+ and LP/+), LP/LP, SF/SF primary MEFs (passage 1) containing DR-GFP reporter gene were electroporated with either empty vector or I-SceI expression vector. GFP-positive cells were quantified by flow cytometry. H, Coiled-coil mutant LP/LP and BRCT domain mutant SF/SF immortalized MEFs exhibited hypersensitivity to olaparib and MMC versus +/+ immortalized MEFs. I, Kaplan–Meier survival curve of LP/LP mice treated with MMC or IR. Homozygous LP/LP mice were extremely sensitive to MMC and IR. Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001; N.S., not significant.

Figure 6.

Brca1L1363P/L1363P MEFs exhibit defective HR and hypersensitivity to ICL. A, Immunoblot analysis of Brca1 and Bard1 proteins in LP/LP immortalized MEFs. Cells were harvested at 2 hours post-IR with 10 Gy. B, LP/LP primary MEFs show impaired proliferation. The proliferation rate was measured in cells (passage 1) by MTT assay. C, Mutant Brca1 LP protein fails to interact with Palb2. Whole-cell lysates overexpressing empty vector (E) or pOZ-mouse Palb2 FLAG-HA (FH) were subjected to immunoprecipitation (IP) with HA beads or Brca1 antibody followed by immunoblot analyses. D–F, Recruitment of Abraxas (ABX), Brca1, and Rad51 to sites of DNA damage in LP/LP immortalized MEFs. Cells were treated with 10 Gy IR and fixed 2 hours post-IR. G, LP/LP MEFs are defective in HR. Control (+/+ and LP/+), LP/LP, SF/SF primary MEFs (passage 1) containing DR-GFP reporter gene were electroporated with either empty vector or I-SceI expression vector. GFP-positive cells were quantified by flow cytometry. H, Coiled-coil mutant LP/LP and BRCT domain mutant SF/SF immortalized MEFs exhibited hypersensitivity to olaparib and MMC versus +/+ immortalized MEFs. I, Kaplan–Meier survival curve of LP/LP mice treated with MMC or IR. Homozygous LP/LP mice were extremely sensitive to MMC and IR. Error bars, SEM. P values were calculated by unpaired t test. ****, P < 0.0001; N.S., not significant.

Close modal

To determine whether the Brca1 L1363P mutation ablates its interaction with Palb2, we performed coimmunoprecipitation (co-IP). Wild-type Brca1 was efficiently coimmunoprecipitated with HA-Palb2, whereas mutant Brca1 L1363P failed to do so, indicating that the mutant Brca1 L1363P does not bind to Palb2 (top IP blot in Fig. 6C). Reciprocal co-IP confirmed the ablation of Brca1–Palb2 interaction in LP/LP cells (bottom IP blot in Fig. 6C).

Brca1 and Bard1 proteins form a heterodimer through their N-terminal RING domain, which stabilizes both proteins (36–38). Both Brca1 and Bard1 proteins from +/+, LP/+, or LP/LP MEFs showed electrophoretic mobility shifts upon irradiation (IR), suggesting that damage-induced phosphorylation of Brca1 and Bard1 is independent of the Brca1–Palb2 interaction (Fig. 6A).

Brca1 accumulation at sites of DNA damage is mediated by the RAP80–Abraxas complex through binding of pAbraxas to the tandem repeats of BRCT of Brca1 (20, 39–41). Thus, MEFs with homozygous S1598F mutation in the Brca1-BRCT repeat domain (SF/SF) fail to assemble Brca1 foci after IR (10). LP/LP MEFs can still form Abraxas and Brca1 IR induced foci (IRIF) (Fig. 6D and E), suggesting that. phospho-ligand binding to the BRCT repeats occurs independent of the Brca1–Palb2 interaction. We confirmed the binding of Bach1 to Brca1 in both +/+ and LP/LP MEFs expressing FLAG-HA tagged Bach1 (given lack of a suitable Bach1 antibody; Supplementary Fig. S5).

Brca1 associates with Brca2 through Palb2, and the Brca1–Palb2 interaction is required for Brca2-mediated Rad51 localization, which is an essential step in HR (15, 25, 42). Previously, our laboratory reported HR defects in Brca1-BRCT domain mutant cells (SF/SF) cells by Rad51 IRIF formation and the DR-GFP assay (10). Similarly, Rad51 focus formation was markedly impaired in LP/LP MEFs (Fig. 6F), although there was no difference in recruitment of Brca1 proteins to sites of DNA damage (Fig. 6E). HR was measured using a DR-GFP reporter assay using control, LP/LP, and SF/SF MEFs. In control MEFs, approximately 1.1% of cells were GFP positive following I-SceI expression. However, in both LP/LP and SF/SF cells, GFP-positive cells were dramatically reduced, indicating significantly impaired HR (Fig. 6G). SF/SF cells had consistently more GFP-positive cells than LP/LP; DR-GFP cells.

Next, we compared sensitivity with olaparib, a PARP1 inhibitor, or MMC, an ICL-inducing agent, using chromosomal analysis. Our results showed that coiled-coil domain mutant LP/LP and BRCT domain mutant SF/SF MEFs both had an elevated basal level of chromosomal aberrations and that both mutants were extremely sensitive to olaparib and MMC, indicating that both the Brca1-Palb2 interaction and functional Brca1 BRCT repeats are critical for HR and ICL repair (Fig. 6H). We also compared IR and ICL sensitivity in control and LP/LP-mutant groups in vivo. After single injection of MMC into the peritoneal cavity, no effect was observed in either +/+ or LP/+ animals, while all LP/LP animals were moribund within a few days (Fig. 6I). Histology upon death showed complete depletion of bone marrow cells in LP/LP mice. LP/LP animals were also hypersensitive to IR, as all mutant animals died within 16 days following whole body IR (Fig. 6I). Together, these data demonstrate that both coiled-coil domain and BRCT domains of Brca1 are essential for HR and for ICL repair. Of note, cellular hypersensitivity to ICL inducing agents is a cellular hallmark of FA.

In this study, we analyzed a knock-in mouse model expressing the patient-derived Brca1 L1363P mutation (L1407P in human). Surprisingly, the phenotypes observed in LP/LP mutant cells and mice closely resemble some clinical features of FA patients not seen in other Brca1 mutant models. Many Brca1-mutant animals develop solid tumors but not dysmorphology, whereas LP/LP mice showed congenital abnormalities including smaller body size, pigmentation in skin, hypoplasia in brain and reproductive organs, and developed hematopoietic defects with macrocytic anemia. More importantly, LP/LP animals display both spontaneous BMF and early development of T-LBL with Notch1 mutations. Our novel LP/LP mouse model not only shows an HR defect and hypersensitivity to ICLs but also hematologic defects, implying distinct functions of the coiled-coil domain (Brca1-Palb2 interaction) of Brca1 (Fig. 7).

Figure 7.

Phenotypes of LP/LP mouse model in comparison with SF/SF mouse.

Figure 7.

Phenotypes of LP/LP mouse model in comparison with SF/SF mouse.

Close modal

The BRCA1 gene, classified as the FANCS gene, has been regarded as a FA-like gene, and none of individuals with biallelic BRCA1 mutations displayed the clinical “hallmarks” of FA, namely hematopoietic defects (4–7). However, the role of BRCA1 in hematopoiesis has not been fully defined. Several studies reported development of BMF or hematopoietic malignancy in Brca1 deleted or mutated mouse models, in accordance with our finding that Brca1 contributes to hematopoietic function (8, 9). The initial reports of biallelic germline BRCA1 mutation concerned two patients who presented with multiple FA-like congenital abnormities, subsequently developing ovarian cancer and ductal breast carcinoma. Importantly, sequence analysis confirmed biallelic mutations in the BRCA1 gene; one allele had a deleterious mutation and the second allele had a pathogenic point mutation in the BRCT repeat (5). Previously, our lab studied hypomorphic variants within the BRCT domain of Brca1, using genetically engineered mice, and found that nonfunctional BRCT repeats caused HR defects and enhanced solid tumor formation (mainly breast cancer) in mice (10). Also, as in the two human patients, our results showed that the BRCT mutant SF/SF mice did not exhibit hematopoietic defects. We observed some residual HR in SF/SF mouse embryonic stem cells and MEFs by Rad51 IRIF formation and DR-GFP assay (10). In contrast, LP/LP MEFs barely showed Rad51 IRIF and showed greater reduction of HR compared with SF/SF MEFs. The residual HR in SF/SF mice may be sufficient to maintain hematopoiesis, and the near complete lack of HR in LP/LP mice possibly disrupts the hematopoietic system. Mice and cellular phenotypes of Brca1-mutant LP/LP and SF/SF mutants are summarized in Fig. 7 (Fig. 7).

In prenatal mice, HSCs actively cycle in the fetal liver to generate blood cells for oxygen transport and for development of the immune system (43). Rapidly growing cells including HSCs can be more severely affected by DNA damage repair defects than other cell types. Unrepaired DNA damages cause activation of the p53/p21 pathway, arresting cells at the G0–G1 phase of the cell cycle, which ultimately impairs HSC pool expansion (44). In addition, mice with increased activation of p53 developed aplastic anemia with dramatically reduced HSCs population (35). Our LP/LP mouse showed dramatic decrease of HSCs with elevated p53 activation, and scant HSC that could be rescued by deletion of p53, implying an important relationship between p53 activation and FA-like bone marrow failure.

In this regard, many of the FA phenotypes observed in LP/LP-mutant mice, including growth retardation, hyperpigmentation and cerebellar-hypoplasia, may also be a consequence of p53 hyperactivation triggered by DNA repair defects. Similarly, cultured cells from patients with FA show constitutive activation of p53 due to accumulation of unresolved DNA damage and endogenous stress (44).

Because the majority of patients with FA have defective HSCs, they are at high risk of hematologic malignancies, mostly acute myeloid leukemia. Although lymphoid leukemia is rare in patients with FA, there is a case report of T-LBL in a BRCA2-mutant patients with FA (45). Moreover, there is a patient carrying homozygous BRCA1 mutation (c.3082C>T) who developed chronic lymphatic leukemia and breast cancer (7). The reasons for the predominance of T-cell neoplasms with Notch1 mutation in our LP/LP mouse model merits further investigation.

During early spermatogenesis, LP/LP mice failed to self-renew male germ stem cells, while SF/SF male mice did not show germ cell defect (10). Depletion of HSCs in LP/LP mice was observed, linking infertility in LP/LP mice to reduced SSCs population in the mutant mice. Oogenesis was normal in SF/SF females, resulting in normal fertility. However, LP/LP female mice were sterile, with a very small number of follicles at four weeks of age, and no follicles in adult. We also observed a decreased number of mammary stem cells in LP/LP females, which is possibly caused by steroid hormones not being secreted from the ovary (46). These results indicate that three stem cell populations; HSCs, germ stem cells and mammary stem cells, are likely affected in LP/LP mice. These observations underlie the importance of a functional coiled-coil domain of Brca1 and its interaction with Palb2 in different types of stem cell formation and maintenance. It is noteworthy that reduced fertility is a common phenotype of FA with defective primordial germ cell proliferation (1, 47).

This is the first mouse model for FA complementation group S (FANCS) that phenocopies FA disease. Prior models of FA, including cell lines and mouse models generated by biallelic deletion of other FA genes have recapitulated various aspects of the disease, including sensitivity to ICL agents but not many FA mouse models recapitulate the full range of disease phenotypes. Given the full range of FA features, this model provides an ideal system to further explore FA pathogenesis and test new therapeutic targets to could be used to treat FA.

No potential conflicts of interest were disclosed.

D. Park: Conceptualization, resources, data curation, formal analysis, validation, investigation, methodology, writing-original draft, writing-review and editing. S.M. Bergin: Resources, formal analysis, investigation, methodology, writing-review and editing. D. Jones: Resources, formal analysis, investigation, methodology, writing-original draft, writing-review and editing. P. Ru: Investigation, methodology, writing-review and editing. C.S. Koivisto: Investigation, writing-review and editing. Y.-J. Jeon: Resources, investigation, writing-review and editing. G.M. Sizemore: Resources, investigation, writing-review and editing. R.D. Kladney: Resources, investigation, writing-review and editing. A. Hadjis: Investigation, writing-review and editing. R. Shakya: Investigation, writing-review and editing. T. Ludwig: Conceptualization, resources, supervision, funding acquisition, investigation, methodology, writing-original draft, writing-review and editing.

We thank Teresa Commisso for critical reading of the manuscript, Drs. Bin Wang (MD Anderson Cancer Center) for Abraxas antibody, Bing Xia (Rutgers Cancer Institute of New Jersey) for Palb2 plasmids, Christine Burd (OSU) for GP100 antibody, and Takesi Kurita (OSU) for Vasa antibody. We also thank the OSU Comprehensive Cancer Center Comparative Pathology, and Mouse Phenotyping Shared Resource for tissue processing for histologic specimens, the Genetically Engineered Mouse Modeling Core for chimeric mouse generation and the Analytical Cytometry Core for flow cytometry analysis. This work was supported by Pelotonia Graduate and undergraduate Fellowships (to D. Park, C.S. Koivisto, and A. Hadjis), OSU CCC funds (to T. Ludwig).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Auerbach
AD
. 
Fanconi anemia and its diagnosis
.
Mutat Res
2009
;
668
:
4
10
.
2.
Meetei
AR
,
Levitus
M
,
Xue
Y
,
Medhurst
AL
,
Zwaan
M
,
Ling
C
, et al
X-linked inheritance of Fanconi anemia complementation group B
.
Nat Genet
2004
;
36
:
1219
24
.
3.
Ceccaldi
R
,
Sarangi
P
,
D'Andrea
AD
. 
The Fanconi anaemia pathway: new players and new functions
.
Nat Rev Mol Cell Biol
2016
;
17
:
337
49
.
4.
Domchek
SM
,
Tang
J
,
Stopfer
J
,
Lilli
DR
,
Hamel
N
,
Tischkowitz
M
, et al
Biallelic deleterious BRCA1 mutations in a woman with early-onset ovarian cancer
.
Cancer Discov
2013
;
3
:
399
405
.
5.
Sawyer
SL
,
Tian
L
,
Kähkönen
M
,
Schwartzentruber
J
,
Kircher
M
,
Majewski
J
, et al
Biallelic mutations in BRCA1 cause a new Fanconi anemia subtype
.
Cancer Discov
2015
;
5
:
135
42
.
6.
Freire
BL
,
Homma
TK
,
Funari
MFA
,
Lerario
AM
,
Leal
AM
,
Velloso
EDRP
, et al
Homozygous loss of function BRCA1 variant causing a Fanconi-anemia-like phenotype, a clinical report and review of previous patients
.
Eur J Med Genet
2018
;
61
:
130
3
.
7.
Bondavalli
D
,
Malvestiti
F
,
Pensotti
V
,
Feroce
I
,
Bonanni
B
. 
Homozygous unclassified variant in a patient with non-Fanconi anemia: a case report
.
Oncol Lett
2018
;
15
:
3329
32
.
8.
Vasanthakumar
A
,
Arnovitz
S
,
Marquez
R
,
Lepore
J
,
Rafidi
G
,
Asom
A
, et al
Brca1 deficiency causes bone marrow failure and spontaneous hematologic malignancies in mice
.
Blood
2016
;
127
:
310
3
.
9.
Mgbemena
VE
,
Signer
RAJ
,
Wijayatunge
R
,
Laxson
T
,
Morrison
SJ
,
Ross
TS
. 
Distinct Brca1 mutations differentially reduce hematopoietic stem cell function
.
Cell Rep
2017
;
18
:
947
60
.
10.
Shakya
R
,
Reid
LJ
,
Reczek
CR
,
Cole
F
,
Egli
D
,
Lin
CS
, et al
BRCA1 tumor suppression depends on BRCT phosphoprotein binding, but not its E3 ligase activity
.
Science
2011
;
334
:
525
8
.
11.
Ludwig
T
,
Fisher
P
,
Ganesan
S
,
Efstratiadis
A
. 
Tumorigenesis in mice carrying a truncating Brca1 mutation
.
Genes Dev
2001
;
15
:
1188
93
.
12.
Bunting
SF
,
Callén
E
,
Kozak
ML
,
Kim
JM
,
Wong
N
,
López-Contreras
AJ
, et al
BRCA1 functions independently of homologous recombination in DNA interstrand crosslink repair
.
Mol Cell
2012
;
46
:
125
35
.
13.
Takenaka
S
,
Kuroda
Y
,
Ohta
S
,
Mizuno
Y
,
Hiwatari
M
,
Miyatake
S
, et al
A Japanese patient with RAD51-associated Fanconi anemia
.
Am J Med Genet A
2019
;
179
:
900
2
.
14.
Ameziane
N
,
May
P
,
Haitjema
A
,
van de Vrugt
HJ
,
van Rossum-Fikkert
SE
,
Ristic
D
, et al
A novel Fanconi anaemia subtype associated with a dominant-negative mutation in RAD51
.
Nat Commun
2015
;
6
:
8829
.
15.
Sy
SM
,
Huen
MS
,
Chen
J
. 
PALB2 is an integral component of the BRCA complex required for homologous recombination repair
.
Proc Natl Acad Sci U S A
2009
;
106
:
7155
60
.
16.
Woods
NT
,
Baskin
R
,
Golubeva
V
,
Jhuraney
A
,
De-Gregoriis
G
,
Vaclova
T
, et al
Functional assays provide a robust tool for the clinical annotation of genetic variants of uncertain significance
.
NPJ Genom Med
2016
;
1
:
16001
.
17.
Manke
IA
,
Lowery
DM
,
Nguyen
A
,
Yaffe
MB
. 
BRCT repeats as phosphopeptide-binding modules involved in protein targeting
.
Science
2003
;
302
:
636
9
.
18.
Yu
X
,
Chini
CC
,
He
M
,
Mer
G
,
Chen
J
. 
The BRCT domain is a phospho-protein binding domain
.
Science
2003
;
302
:
639
42
.
19.
Rodriguez
M
,
Yu
X
,
Chen
J
,
Songyang
Z
. 
Phosphopeptide binding specificities of BRCA1 COOH-terminal (BRCT) domains
.
J Biol Chem
2003
;
278
:
52914
8
.
20.
Wang
B
,
Matsuoka
S
,
Ballif
BA
,
Zhang
D
,
Smogorzewska
A
,
Gygi
SP
, et al
Abraxas and RAP80 form a BRCA1 protein complex required for the DNA damage response
.
Science
2007
;
316
:
1194
8
.
21.
Greenberg
RA
,
Sobhian
B
,
Pathania
S
,
Cantor
SB
,
Nakatani
Y
,
Livingston
DM
. 
Multifactorial contributions to an acute DNA damage response by BRCA1/BARD1-containing complexes
.
Genes Dev
2006
;
20
:
34
46
.
22.
Kim
H
,
Huang
J
,
Chen
J
. 
CCDC98 is a BRCA1-BRCT domain-binding protein involved in the DNA damage response
.
Nat Struct Mol Biol
2007
;
14
:
710
5
.
23.
Sobhian
B
,
Shao
G
,
Lilli
DR
,
Culhane
AC
,
Moreau
LA
,
Xia
B
, et al
RAP80 targets BRCA1 to specific ubiquitin structures at DNA damage sites
.
Science
2007
;
316
:
1198
202
.
24.
Zhang
F
,
Fan
Q
,
Ren
K
,
Andreassen
PR
. 
PALB2 functionally connects the breast cancer susceptibility proteins BRCA1 and BRCA2
.
Mol Cancer Res
2009
;
7
:
1110
8
.
25.
Zhang
F
,
Ma
J
,
Wu
J
,
Ye
L
,
Cai
H
,
Xia
B
, et al
PALB2 links BRCA1 and BRCA2 in the DNA-damage response
.
Curr Biol
2009
;
19
:
524
9
.
26.
Malecki
MJ
,
Sanchez-Irizarry
C
,
Mitchell
JL
,
Histen
G
,
Xu
ML
,
Aster
JC
, et al
Leukemia-associated mutations within the NOTCH1 heterodimerization domain fall into at least two distinct mechanistic classes
.
Mol Cell Biol
2006
;
26
:
4642
51
.
27.
Bernardo
GM
,
Lozada
KL
,
Miedler
JD
,
Harburg
G
,
Hewitt
SC
,
Mosley
JD
, et al
FOXA1 is an essential determinant of ERalpha expression and mammary ductal morphogenesis
.
Development
2010
;
137
:
2045
54
.
28.
Sizemore
GM
,
Balakrishnan
S
,
Hammer
AM
,
Thies
KA
,
Trimboli
AJ
,
Wallace
JA
, et al
Stromal PTEN inhibits the expansion of mammary epithelial stem cells through Jagged-1
.
Oncogene
2017
;
36
:
2297
308
.
29.
Guo
G
,
Luc
S
,
Marco
E
,
Lin
TW
,
Peng
C
,
Kerenyi
MA
, et al
Mapping cellular hierarchy by single-cell analysis of the cell surface repertoire
.
Cell Stem Cell
2013
;
13
:
492
505
.
30.
Rothenberg
EV
,
Moore
JE
,
Yui
MA
. 
Launching the T-cell-lineage developmental programme
.
Nat Rev Immunol
2008
;
8
:
9
21
.
31.
Liu
Y
,
Easton
J
,
Shao
Y
,
Maciaszek
J
,
Wang
Z
,
Wilkinson
MR
, et al
The genomic landscape of pediatric and young adult T-lineage acute lymphoblastic leukemia
.
Nat Genet
2017
;
49
:
1211
8
.
32.
Van Vlierberghe
P
,
Ambesi-Impiombato
A
,
De Keersmaecker
K
,
Hadler
M
,
Paietta
E
,
Tallman
MS
, et al
Prognostic relevance of integrated genetic profiling in adult T-cell acute lymphoblastic leukemia
.
Blood
2013
;
122
:
74
82
.
33.
Weng
AP
,
Ferrando
AA
,
Lee
W
,
Morris
JP
,
Silverman
LB
,
Sanchez-Irizarry
C
, et al
Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia
.
Science
2004
;
306
:
269
71
.
34.
McGowan
KA
,
Li
JZ
,
Park
CY
,
Beaudry
V
,
Tabor
HK
,
Sabnis
AJ
, et al
Ribosomal mutations cause p53-mediated dark skin and pleiotropic effects
.
Nat Genet
2008
;
40
:
963
70
.
35.
Simeonova
I
,
Jaber
S
,
Draskovic
I
,
Bardot
B
,
Fang
M
,
Bouarich-Bourimi
R
, et al
Mutant mice lacking the p53 C-terminal domain model telomere syndromes
.
Cell Rep
2013
;
3
:
2046
58
.
36.
Mallery
DL
,
Vandenberg
CJ
,
Hiom
K
. 
Activation of the E3 ligase function of the BRCA1/BARD1 complex by polyubiquitin chains
.
EMBO J
2002
;
21
:
6755
62
.
37.
Baer
R
,
Ludwig
T
. 
The BRCA1/BARD1 heterodimer, a tumor suppressor complex with ubiquitin E3 ligase activity
.
Curr Opin Genet Dev
2002
;
12
:
86
91
.
38.
Brzovic
PS
,
Rajagopal
P
,
Hoyt
DW
,
King
MC
,
Klevit
RE
. 
Structure of a BRCA1-BARD1 heterodimeric RING-RING complex
.
Nat Struct Biol
2001
;
8
:
833
7
.
39.
Wu
Q
,
Paul
A
,
Su
D
,
Mehmood
S
,
Foo
TK
,
Ochi
T
, et al
Structure of BRCA1-BRCT/Abraxas complex reveals phosphorylation-dependent BRCT dimerization at DNA damage sites
.
Mol Cell
2016
;
61
:
434
48
.
40.
Her
J
,
Soo Lee
N
,
Kim
Y
,
Kim
H
. 
Factors forming the BRCA1-A complex orchestrate BRCA1 recruitment to the sites of DNA damage
.
Acta Biochim Biophys Sin
2016
;
48
:
658
64
.
41.
Wang
B
,
Elledge
SJ
. 
Ubc13/Rnf8 ubiquitin ligases control foci formation of the Rap80/Abraxas/Brca1/Brcc36 complex in response to DNA damage
.
Proc Natl Acad Sci U S A
2007
;
104
:
20759
63
.
42.
Roy
R
,
Chun
J
,
Powell
SN
. 
BRCA1 and BRCA2: different roles in a common pathway of genome protection
.
Nat Rev Cancer
2011
;
12
:
68
78
.
43.
Pietras
EM
,
Warr
MR
,
Passegué
E
. 
Cell cycle regulation in hematopoietic stem cells
.
J Cell Biol
2011
;
195
:
709
20
.
44.
Ceccaldi
R
,
Parmar
K
,
Mouly
E
,
Delord
M
,
Kim
JM
,
Regairaz
M
, et al
Bone marrow failure in Fanconi anemia is triggered by an exacerbated p53/p21 DNA damage response that impairs hematopoietic stem and progenitor cells
.
Cell Stem Cell
2012
;
11
:
36
49
.
45.
Wagner
JE
,
Tolar
J
,
Levran
O
,
Scholl
T
,
Deffenbaugh
A
,
Satagopan
J
, et al
Germline mutations in BRCA2: shared genetic susceptibility to breast cancer, early onset leukemia, and Fanconi anemia
.
Blood
2004
;
103
:
3226
9
.
46.
Stingl
J
. 
Estrogen and progesterone in normal mammary gland development and in cancer
.
Horm Cancer
2011
;
2
:
85
90
.
47.
Tsui
V
,
Crismani
W
. 
The Fanconi anemia pathway and fertility
.
Trends Genet
2019
;
35
:
199
214
.