53BP1 loss rescues embryonic lethality but not genomic instability of BRCA1 total knockout mice

Abstract

BRCA1 is critical for DNA double-strand break (DSB) repair by homologous recombination (HR). BRCA1 deficient mice are embryonic lethal. Previous studies have shown that 53BP1 knockout (KO) rescues embryonic lethality of BRCA1 hypomorphic mutant mice by restoring HR. Here, we show that 53BP1 KO can partially rescue embryonic lethality of BRCA1 total KO mice, but HR is not restored in BRCA1-53BP1 double knockout (DKO) mice. As a result, BRCA1-53BP1 DKO cells are extremely sensitive to PARP inhibitors (PARPi). In addition to HR deficiency, BRCA1-53BP1 DKO cells have elevated microhomology-mediated end joining (MMEJ) activity and G2/M cell cycle checkpoint defects, causing severe genomic instability in these cells. Interestingly, BRCA1-53BP1 DKO mice rapidly develop thymic lymphoma that is 100% penetrant, which is not observed in any BRCA1 mutant mice rescued by 53BP1 KO. Taken together, our study reveals that 53BP1 KO can partially rescue embryonic lethality caused by complete BRCA1 loss without rescuing HR-related defects. This finding suggests that loss of 53BP1 can support the development of cancers with silenced BRCA1 expression without causing PARPi resistance.

Fig. 1. 53BP1 KO rescues embryonic lethality of BRCA1 KO mice.

a Domain structures of BRCA1 WT, Δ11, Δ2, ΔC, and Δ5–13 proteins. CC coiled coil. b Summary of mice at 3 weeks or embryos at 13.5 dpc of indicated genotypes. c Western blotting analyses of BRCA1, 53BP1, and BARD1 protein in MEFs with indicated genotypes. α-tubulin was used as loading control. Asterisk, nonspecific band. Arrowhead, BRCA1 isoforms or degraded BRCA1 proteins. d Immunofluorescence staining analyses of BRCA1, 53BP1, and BARD1 foci in MEFs with indicated genotypes 6 h after 10 Gy IR exposure. γH2AX marks sites of DNA damage. Hoechst 33342 marks nucleus. Scale bars, 10 μm. e Quantification of BRCA1, 53BP1, and BARD1 foci. Error bars represent SEM from three independent experiments. *p < 0.05; **p < 0.01.

Fig. 2. HR efficiency is not restored in BRCA1-53BP1 DKO cells.

a Schematic representation of the DR-GFP reporter for analyzing efficiency of DSB repair by HR. b Summary of HR efficiency obtained using the DR-GFP reporter in MEFs with indicated genotypes. c Immunofluorescence staining analyses of RAD51 foci in MEFs with indicated genotypes 6 h after 10 Gy IR exposure. γH2AX marks sites of DNA damage. Hoechst 33342 marks nucleus. Scale bars, 10 μm. d Quantification of RAD51 foci. e Western blotting analyses of BRCA1 and 53BP1 protein in WT and 53BP1 KO U2OS DR-GFP cells after control or BRCA1 knockdown. α-tubulin was used as loading control. f Summary of HR efficiency in WT and 53BP1 KO U2OS DR-GFP cells after control or BRCA1 knockdown. g MEFs with indicated genotypes were labeled with 10 μM BrdU and 24 h later cells were treated with 10 Gy IR and recovered for 1 h. Representative BrdU foci were shown. Hoechst 33342 marks nucleus. Scale bars, 10 μm. h Quantification of BrdU foci. i Immunofluorescence staining analyses of RPA2 foci in MEFs with indicated genotypes 6 h after 10 Gy IR exposure. γH2AX marks site of DNA damage. Hoechst 33342 marks nucleus. Scale bars, 10 μm. j Quantification of RPA2 foci. k Western blotting analyses of RPA2 and RPA2 phosphorylation (S4/8) in MEFs with indicated genotypes 1 h after 10 Gy IR exposure. α-tubulin was used as loading control. Error bars represent SEM from three independent experiments. *p < 0.05; **p < 0.01; ***p < 0.001; ns not significant.

Fig. 3. Genomic instability is not restored in BRCA1-53BP1 DKO cells.

a Summary of chromosomal aberration numbers per metaphase in ES cells with indicated genotypes after overnight treatment of 1 μM PARPi olaparib. Arrows show examples of chromosome aberrations. Scale bars, 10 μm. b Cell viability analyses of ES cells with indicated genotypes after treatment with indicated doses of olaparib. c Cell viability analyses of WT and 53BP1 KO U2OS DR-GFP cells after control or BRCA1 knockdown and treatment with indicated doses of olaparib. Error bars represent SEM from three independent experiments.

Fig. 4. BRCA1-53BP1 DKO mice develop thymic lymphoma.

a Summary of lymphoma-free survival of mice with indicated genotypes over 7 months using the Kaplan–Meier method. b Representative images of thymic lymphomas from BRCA1-53BP1 DKO mice. Representative flow cytometry analyses of CD4/CD8 (c) and TCRβ (d) surface expression in lymphoma cells from BRCA1-53BP1 DKO mice using thymocytes from 8-week-old WT mice as control. Representative flow cytometry analyses of CD4/CD8 (e) and TCRβ (f) surface expression in thymocytes from 1-month-old WT and BRCA1-53BP1 DKO mice.

Fig. 5. BRCA1-53BP1 DKO lymphomas have unique characteristics of genomic instability.

a Representative metaphase spreads of lymphoma cells from 2 BRCA1-53BP1 DKO mice. For each lymphoma, chromosome numbers were counted in 20 metaphase spreads and summarized. b Copy number profiling in five lymphoma samples by whole-genome sequencing analysis. Log2-transformed copy ratios are plotted to show gain/loss in genomic regions. c Summary of genomic instabilities identified in five sequenced lymphoma samples. Genomic instabilities include single nucleotide polymorphisms (SNP), small insertions and deletions (Indel), and large structural variations (SV). d Mutational status in genes known to be correlated with thymic lymphoma onset. e Distributions of different types of structural variants in each lymphoma sample. Bar graph plotted the number of intra-chromosomal changes (deletion and inversion) and inter-chromosomal changes (translocation). f Circos plots of lymphoma samples showing locations of structural variations.

Fig. 6. BRCA1-53BP1 DKO cells have increased microhomology-mediated end joining.

a An example of structural variation with microhomology identified in sequencing analysis. Each row represents one aligned read that supports this translocation contig. Only sequences near the breakpoint joint are shown. The two genomic regions are colored differently in red and blue. The 3-bp microhomology sequence in this breakpoint is marked. b Analysis of the microhomology length distribution in structural variation breakpoints. Two-tailed t-test was applied to calculate the statistical significance between ratios in different length group. c Schematic representation of the MMEJ reporter for analyzing efficiency of DSB repair by MMEJ. d Western blotting analyses of BRCA1 and 53BP1 protein in WT and 53BP1 KO U2OS MMEJ-GFP cells after control or BRCA1 knockdown. α-tubulin was used as loading control. e Summary of MMEJ efficiency in WT and 53BP1 KO U2OS MMEJ-GFP cells after control or BRCA1 knockdown. Quantification of POLQ (f), LIG3 (g), and RAD52 (h) mRNA levels in MEFs with indicated genotypes. Error bars represent SEM from three independent experiments. *p < 0.05; **p < 0.01; ns not significant.

Fig. 7. BRCA1-53BP1 DKO cells have defective in G2/M cell cycle checkpoint.

a Representative flow cytometry analyses of mitotic cells defined as 4 N cells positive for histone H3 phosphorylation (S10) in MEFs with indicated genotypes 1 h after 5 Gy IR exposure. Error bars represent SEM from three independent experiments. *p < 0.05; **p < 0.01. b Western blotting analyses of CHK2, KAP1 phosphorylation (S824), KAP1, CHK1 phosphorylation (S345), and CHK1 in MEFs with indicated genotypes 1 h after 10 Gy IR exposure. α-tubulin was used as loading control.

Fig. 8. Working model: 53BP1 loss partially rescue embryonic lethality of BRCA1 total knockout mice without restoring HR.

BRCA1 is critical for DSB repair by homologous recombination (HR). In WT mice, BRCA1 promotes DNA end resection by removing 53BP1 from DSB ends and facilitates HR repair (left panel). 53BP1 fails to be removed in BRCA1 total KO mice (Brca1^−/−) and occupies DSB ends. DNA end resection is inhibited, which results in HR deficiency and embryonic lethality of BRCA1 total KO mice (middle panel). Loss of 53BP1 lifts the barrier for DNA end resection in BRCA1 total KO mice. HR is not restored but microhomology-mediated end joining (MMEJ) is increased. Brca1^−/−;Trp53bp1^−/− mice are partially viable, but they have severe genomic instability and develop thymic lymphomas (right panel).