Lig3-dependent rescue of mouse viability and DNA double-strand break repair by catalytically inactive Lig4

Abstract

Recent studies have revealed a structural role for DNA ligase 4 (Lig4) in the maintenance of a repair complex during non-homologous end joining (NHEJ) of DNA double-strand breaks. In cultured cell lines, catalytically inactive Lig4 can partially alleviate the severe DNA repair phenotypes observed in cells lacking Lig4. To study the structural role of Lig4 in vivo, a mouse strain harboring a point mutation to Lig4's catalytic site was generated. In contrast to the ablation of Lig4, catalytically inactive Lig4 mice are born alive. These mice display marked growth retardation and have clear deficits in lymphocyte development. We considered that the milder phenotype results from inactive Lig4 help to recruit another ligase to the repair complex. We next generated a mouse strain deficient for nuclear Lig3. Nuclear Lig3-deficient mice are moderately smaller and have elevated incidences of cerebral ventricle dilation but otherwise appear normal. Strikingly, in experiments crossing these two strains, mice lacking nuclear Lig3 and expressing inactive Lig4 were not obtained. Timed mating revealed that fetuses harboring both mutations underwent resorption, establishing an embryonic lethal genetic interaction. These data suggest that Lig3 is recruited to NHEJ complexes to facilitate end joining in the presence (but not activity) of Lig4.

Graphical Abstract

Figure 1.

Generation and characterization of mice with inactive Lig4. (A) K273S mutation of Lig4 inactivate its enzymatic activity in vitro for ligating a linear DNA fragment into dimers and multimers. M, mouse. H, human. X4, XRCC4. L4, Lig4. (B) K273S (codon AAG to AGT) mutation was introduced by CRISPR gene editing and homology-directed repair with a single-strand template in fertilized mouse eggs. The guide RNA and PAM sequences are indicated by blue and red lines, respectively. (C) Genotyping of mutant allele. PCR amplicon and restriction sites are shown. +, wild type allele. S, Spe I. H, Hind III. (D) Mating between Lig4^+/K273S mice yields offspring of all three possible genotypes. *P < 0.05, calculated based on Chi-squared test with Yates continuity correction. (E) Both male and female Lig4^K273S/K273S mice are growth retarded as compared to their littermates. WT and Lig4^+/K273S mice have no difference in body weights. Error bars represent standard deviations. ***P< 0.001, based on two-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test.

Figure 2.

Thymus and thymocyte defects in Lig4^K273S/K273S mice. (A) The Lig4^K273S/K273S thymus is tiny and most times barely visible as compared to the Lig4^+/K273S thymus. (B) CD4⁺CD8⁺, CD4⁺ and CD8⁺ cells are 100∼1000× fold less in the Lig4^K273S/K273S thymus as compared to the Lig4^+/K273S thymus. (C) Flow cytometry analysis of thymic CD4⁺CD8⁺, CD4⁺ and CD8⁺ cells in the Lig4^K273S/K273S thymus at 3 days and 8 weeks after birth. **P< 0.005, based on unpaired t-tests.

Figure 3.

Bone marrow B-cell defect in Lig4^K273S/K273S mice. (A) Viable cells isolated from femur bones of Lig4^+/K273S and Lig4^K273S/K273S mice. (B) The greatly diminished B lineage cells (B220⁺) in the Lig4^K273S/K273S bone marrow. (C) The lack of naïve B cells (IgM⁺) in the Lig4^K273S/K273S bone marrow. ***P < 0.0005, based on unpaired t-tests.

Figure 4.

Splenocytes in Lig4^K273S/K273S spleen. (A) The Lig4^K273S/K273S spleen is much smaller than that of the Lig4^+/K273S spleen. (B) The Lig4^K273S/K273S spleen contains much less T cells (CD3+) and no B cells (CD19+). (C) Lack of mature B cell (B220⁺IgM⁺) in the Lig4^K273S/K273S spleen. (D) Helper (CD4⁺) and cytotoxic (CD8⁺) T cells are greatly reduced in the Lig4^K273S/K273S spleen. *P < 0.05; **P < 0.005, based on unpaired t-tests.

Figure 5.

Generation and characterization of mice lack of nuclear Lig3. (A) Isoforms of Lig3 peptides resulted from alternative translation initiation target to mitochondrion and nucleus, depending on the presence or absence of a MLS at the N-terminus. (B) CRISPR gene editing and HDR in fertilized mouse eggs to generate mutation at the second initiation codon that disable the translation of the nuclear Lig3. The initiation codon for nuclear Lig3 is indicated. The guide RNA and PAM sequences are indicated by blue and red lines, respectively. (C) Genotyping of mutant allele by PCR followed by restriction digestion. +, wild type allele. −, mutant allele that disables nuclear Lig3 expression. (D) Western blot analysis of Lig3 expression in various tissues in Lig3^+/m and Lig3^m/m mice. Stain-free gel images are shown as loading controls. (E) Mating between Lig3^+/m mice results in offspring at near Mendelian ratios. ns, P > 0.05. (F) Mating between Lig3^+/m and Lig3^m/m mice results in offspring at the expected ratios. ns, P > 0.05, calculated based on Chi-squared test with Yates continuity correction. (G) Body weights of Lig3^m/m mice as compared to their littermates. Error bars represent standard deviations. *P< 0.05, based on two-way ANOVA with Tukey’s multiple comparisons test.

Figure 6.

Lig3 is required for the viability of Lig4^K273S/K273S mice. (A) Mating between Lig3^m/mLig4^+/K273S mice fails to generate Lig3^m/mLig4^K273S/K273S offspring. ****P < 0.001, calculated based on Chi-squared test with Yates continuity correction. (B) Embryos isolated from pregnant Lig3^m/mLig4^+/K273S females at day 15.5 after mating with Lig3^m/mLig4^+/K273S male mice. ns, P > 0.05, calculated based on Chi-squared test with Yates continuity correction.

Figure 7.

DNA repair in MEFs. (A) Western blot analysis of Lig4 expression in MEFs of various genotypes. Asterisk indicates a non-specific band that is cross-reactive to the Lig4 antibody. (B) Growth curves of MEFs of various genotypes. (C) MTT assays of zeocin sensitivity of MEFs of various genotypes. (D) Cell fractionation and western blot analysis of Lig3 in MEFs of the indicated genotypes. (E) Growth curves of MEFs of the indicated genotypes. (F) MTT assays of camptothecin sensitivity of MEFs of the indicated genotypes. (G) MTT assays of zeocin sensitivity of Lig4^-/- and Lig3^m/mLig4^K273S/K273S MEFs transduced with lentivirus with a mouse nuclear Lig3 complementary DNA (cDNA). Error bars indicate standard deviations from at least three independent experiments. *P < 0.05; **P< 0.005; ns, P > 0.05, based on unpaired t-tests.