Nbs1-mediated DNA damage repair pathway regulates haematopoietic stem cell development and embryonic haematopoiesis

Abstract

Objectives: DNA damages pose threats to haematopoietic stem cells (HSC) maintenance and haematopoietic system homeostasis. Quiescent HSCs in adult mouse bone marrow are resistant to DNA damage, while human umbilical cord blood-derived proliferative HSCs are prone to cell death upon ionizing radiation. Murine embryonic HSCs proliferate in foetal livers and divide symmetrically to generate HSC pool. How murine embryonic HSCs respond to DNA damages is not well-defined.

Materials and methods: Mice models with DNA repair molecule Nbs1 or Nbs1/p53 specifically deleted in embryonic HSCs were generated. FACS analysis, in vitro and in vivo HSC differentiation assays, qPCR, immunofluorescence and Western blotting were used to delineate roles of Nbs1-p53 signaling in HSCs and haematopoietic progenitors.

Results: Nbs1 deficiency results in persistent DNA breaks in embryonic HSCs, compromises embryonic HSC development and finally results in mouse perinatal lethality. The persistent DNA breaks in Nbs1 deficient embryonic HSCs render cell cycle arrest, while driving a higher rate of cell death in haematopoietic progenitors. Although Nbs1 deficiency promotes Atm-Chk2-p53 axis activation in HSCs and their progenies, ablation of p53 in Nbs1 deficient HSCs accelerates embryonic lethality.

Conclusions: Our study discloses that DNA double-strand repair molecule Nbs1 is essential in embryonic HSC development and haematopoiesis. Persistent DNA damages result in distinct cell fate in HSCs and haematopoietic progenitors. Nbs1 null HSCs tend to be maintained through cell cycle arrest, while Nbs1 null haematopoietic progenitors commit cell death. The discrepancies are mediated possibly by different magnitude of p53 signaling.

Keywords: Cell fate; DNA damage response; Haematopoietic stem cells; Nbs1; p53.

Figure 1

Nbs1 deletion generates persistent DNA breaks in HSCs. A, Genomic instability, indicated by γ‐H2AX foci (Red), of freshly sorted HSCs from control and Nbs1‐HSCΔ E16.5 embryos. The frequency of cells with different amount of DNA damages is summarized on the lower panel. Data are generated from HSCs isolated from 2 controls and 2 Nbs1‐HSCΔ embryos. 189 control LSK HSCs and 246 Nbs1 deficient HSCs are used for the final quantification. ***, P < .001. Chi‐square test is used. B, Chromosome abnormalities in E15.5 foetal liver haematopoietic cells from 3 control and 4 Nbs1‐HSCΔ embryos. Red arrows mark the chromosome abnormalities in Nbs1‐HSCΔ haematopoietic cells. ‘N’ denotes the numbers of metaphases used for the quantification (right panel). Please note: ‘Other chromosome abnormalities’ include chromosome fusions and endo‐replication

Figure 2

Nbs1 deletion causes perinatal lethality and depletion of HSCs in mice. A, Kaplan‐Meier survival curve of control (Co) and Nbs1‐HSCΔ (Nbs1Δ) newborns. ‘N’ denotes the number of mice used for the quantification. B, Blood smear (left panels) from P1 control (Co) and Nbs1‐HSCΔ (Nbs1Δ) mice and HE staining of tibia (right panels) from P3 control (Co) and Nbs1‐HSCΔ (Nbs1Δ) mice. C, Representative FACS profiles of haematopoietic stem cells (LSK HSCs) and haematopoietic progenitors (LK progenitors) from control and Nbs1‐HSCΔ mice. Frequencies of LK progenitors and LSK HSCs (in gated bone marrow mononuclear cells) are shown. D, Nbs1 deficiency results in depletion of HSCs (left panel, LSK HSCs) and haematopoietic progenitors (right panel, LK progenitors) in p1 control (Co) and Nbs1‐HSCΔ (Nbs1Δ) mice. Note: N = 3 for each group; ‘P1’ denotes postnatal day 1. Note: *, P < .05; **, P < .01; ***, P < .001. Unpaired Student's t test is used

Figure 3

Defective foetal liver haematopoiesis after Nbs1 deletion. A, Cellularity of foetal livers from control and Nbs1‐HSCΔ embryos at different developmental stages (n ≥ 3 for each group). B, Frequencies of T cells (CD3⁺), B cell (B220⁺), red blood cells (Ter119⁺), granulocytes and macrophages (Mac1⁺) in E15.5 control (Co) and Nbs1‐HSCΔ (Nbs1Δ) foetal livers (N = 3 for each genotype). C, Colony formation assay of control and Nbs1‐HSCΔ foetal liver haematopoietic cells. Please note that Nbs1‐HSCΔ foetal liver haematopoietic cells fail to form any haematopoietic colony in vitro. D, Kaplan‐Meier survival curve of lethally irradiated Ly5.1 mice transplanted with control or Nbs1‐HSCΔ foetal liver cells. Mock reconstitution (+PBS) is used to determine the lethal dose. Please note that Nbs1‐HSCΔ foetal liver haematopoietic cells could not reconstitute the haematopoietic system in lethally irradiated Ly5.1 mice. Note: *, P < .05; **, P < .01; ***, P < .001. Unpaired Student's t test is used

Figure 4

Cell fates of foetal liver HSCs and LK progenitors upon Nbs1 deletion. A, Representative FACS profile of HSCs and LK progenitors from E15.5 embryonic foetal livers of control (Co) and Nbs1‐HSCΔ (Nbs1Δ) mouse embryos. Frequencies of LK progenitors and LSK HSCs (in gated foetal liver cells) are shown. B, Absolute number of LK progenitors in foetal livers at different stage (N ≥ 3 for each group). C, Absolute number of HSCs in foetal livers at different stage (N ≥ 3 for each group). D, Apoptosis (Annexin V⁺) of LK progenitors and HSCs from E15.5 control and Nbs1‐HSCΔ foetal livers. ‘N’ denotes the number of embryos analysed. E, Ki67^‐ cells in HSCs and LK progenitors at E16.5. HSCs and LK progenitors sorted from 2 controls and 2 mutant embryos are used for the analysis. N denotes the number of cells used for the quantification (chi‐square test is used. ***, P < .001; ns, not significant). Note: *, P < .05; **, P < .01; ***, P < .001. Unpaired Student's t test is used

Figure 5

Activated DNA damage response in Nbs1‐HSCΔ foetal livers. A, Western blotting analysis on p53 and Chk2 phosphorylation in E15.5 control (Co) and Nbs1‐HSCΔ (Nbs1Δ) foetal liver samples. β‐Actin is used as the loading control. B, qRT‐PCR analysis of p53 and its downstream target genes (p21, Noxa, Puma and Bax) in E17.5 foetal livers from control (Co, N = 5) and Nbs1‐HSCΔ (Nbs1Δ, N = 3) mice embryos. The expression of individual gene in control samples was defined as ‘1’. β‐Actin expression level was used as the internal control. Note: *, P < .05; **, P < .01. Unpaired Student's t test is used

Figure 6

p53 signaling determines cell fates of Nbs1 deficient foetal liver HSC and LK progenitors. A, The expression of Nbs1, p53 and Bax in control (Co) and Nbs1‐HSCΔ (Nbs1Δ) HSCs and LK progenitors. The expression of these genes in control samples was defined as ‘1’. β‐Actin expression level was used as the internal control. N = 3 for each genotype. B, p21 expression in control (Co) and Nbs1 deficient (Nbs1Δ) HSCs and LK progenitors. The expression of p21 in control samples was defined as ‘1’. β‐Actin expression level was used as the internal control. N = 3 for each genotype. C, Relative expression of p21, p53, Bax in HSCs and LK progenitors from control mouse embryos. The expression of these genes in LK progenitors was defined as ‘1’. β ‐Actin expression level was used as the internal control. N = 3 for each group. Note: *, P < .05; **, P < .01; ***, P < .001. Unpaired Student's t test is used