Deletion of Puma protects hematopoietic stem cells and confers long-term survival in response to high-dose gamma-irradiation

Abstract

Molecular paradigms underlying the death of hematopoietic stem cells (HSCs) induced by ionizing radiation are poorly defined. We have examined the role of Puma (p53 up-regulated mediator of apoptosis) in apoptosis of HSCs after radiation injury. In the absence of Puma, HSCs were highly resistant to gamma-radiation in a cell autonomous manner. As a result, Puma-null mice or the wild-type mice reconstituted with Puma-null bone marrow cells were strikingly able to survive for a long term after high-dose gamma-radiation that normally would pose 100% lethality on wild-type animals. Interestingly, there was no increase of malignancy in the exposed animals. Such profound beneficial effects of Puma deficiency were likely associated with better maintained quiescence and more efficient DNA repair in the stem cells. This study demonstrates that Puma is a unique mediator in radiation-induced death of HSCs. Puma may be a potential target for developing an effective treatment aimed to protect HSCs from lethal radiation.

Figure 1

Transcriptional expression of Puma in HSCs and HPCs. (A) Basal expression of Puma mRNA in different hematopoietic cell subsets under homeostatic conditions. The different subsets of hematopoietic cells were sorted for real-time RT-PCR analysis to quantify the levels of gene expression. LT-HSC (CD34⁻LKS); ST-HSC (CD34⁺LKS); CMP, common myeloid progenitor (CD34⁺FcγR^lowCD127⁻LKS⁻); MEP, megakaryocyte/erythroid progenitors (CD34⁻FcγR^lowCD127⁻LKS⁻); CLP, common lymphoid progenitor (CD127⁺L⁻K^lowS^low); GMP, granulocyte-macrophage progenitor (CD34⁺FcγR^hiCD127⁻LKS⁻). The values on y-axis indicate the fold change normalized to β-actin. (B) Induced expression of Puma mRNA in LT-HSCs after irradiation. Each bar shows the mean ± SD from 3 replicates. *P < .05.

Figure 2

Quantitative measurements of HSCs and HPCs with or without 4-Gy irradiation. (A) Quantitative analysis of the frequencies of LT-HSCs, ST-HSCs plus multipotent progenitors (CD34⁺LKS) and HPCs (LKS⁻) in the Puma^+/+ and Puma^−/− mice without irradiation by flow cytometry. The percentage of each cell population (top) was calculated by the acquired number of the each population divided by the acquired number of CD45⁺ nucleated cells. The absolute number per harvest (bottom) was calculated by percentage of the cells multiplied by the number of BMNCs. The Student t test was used for the statistical analysis between groups (n = 7). (B) Relative increase of LKS cells after TBI in the absence of Puma. The Puma^+/+ and Puma^−/− mice (n = 3 in each phenotype) were subjected to 4-Gy irradiation (IR). LKS cells were measured in the bone marrow of the mice 6 hours after radiation in comparison with the nonirradiated (NR) controls (n = 3 in each phenotype). Representative profiles in flow cytometry from NR (left) and IR (right) groups were shown. (C) The percentage of LKS cells in the bone marrow (left) and the absolute number per harvest (right) with and without radiation were summarized (*P < .05).

Figure 3

Kaplan-Meier analysis of overall survival in Puma^+/+, Puma^+/−, Puma^−/−, and p53^−/− mice challenged with 10 Gy of γ-irradiation. The plots of short-term analysis (42 days, top) and long-term analysis (more than 600 days, bottom) were shown, respectively. All the Puma^+/+ (n = 23, red) and p53^−/− (n = 21, green) mice died within 35 days and there was no significant difference between these 2 groups (P = .07). The Puma^−/− (n = 24, blue) and Puma^+/− (n = 32, purple) mice survived much longer. There are significant differences when Puma^−/− or Puma^+/− groups were compared with Puma^+/+ or p53^−/− groups (P < .01in all groups).

Figure 4

Radiation resistance of engrafted hematopoietic stem cells in the absence of Puma. (A) Radioprotection of HSCs by deletion of Puma in competitive transplantation model. The experimental design was shown in detail in supplemental Figure 1A. The hematopoietic contribution in blood by 100 LT-HSCs before and after 4-Gy irradiation was indicated by the ratios of CD45.2 to CD45.1/.2 cells (n = 4 in both Puma^+/+ and Puma^−/− groups). (B-C) Self-renewing capacities of Puma^−/− HSCs remained after serial competitive transplantation and 4-Gy radiation. The experimental design was shown in detail in supplemental Figure 1B. The engraftment levels of Puma^−/− HSCs relative to competitor cells in primary recipients (B) and secondary recipients (C) were shown, respectively. (D-E) Kaplan-Meier analysis of overall survival of the mice reconstituted with Puma^+/+ or Puma^−/− hematopoietic cells (supplemental Figure 1C) exposed to first round 9-Gy irradiation (D) and second round 9-Gy irradiation (E). P values are indicated in the graphs.

Figure 5

Decreased apoptosis and remained quiescence of HSCs in the absence of Puma. (A) Detection of apoptosis of LKS cells by the TUNEL assay. LKS cells sorted from Puma^+/+ and Puma^−/− mice were treated with 4-Gy irradiation. The percentage of cells undergoing apoptosis at 24 hours after irradiation was compared with those without irradiation. Total 250 cells were calculated in each group. Percentage of apoptotic cells in Puma^+/+ and Puma^−/− was compared using the Fisher exact test. **P < .01. (B) Flow cytometric analysis of cell-cycle change of LKS cells after radiation. G₀ versus G₁ fraction in cell cycle was measured at 20 hours after 4-Gy irradiation. Representative flow cytometric profiles from 3 experiments were shown (left). The average percentages of G₀ fraction in LKS cells from 3 experiments were summarized (right). *P < .05.

Figure 6

Confocal microscopy for the expression of γ-H2AX protein. (A-C) Phosphorylated γ-H2AX protein was detected with its antibody and then measured by confocal microscopy at different time points after 4-Gy of γ-irradiation (×40 magnification). (D) Fold change of fluorescence intensity. The intensity of Alexa 488–labeled γ-H2AX was measured by Adobe PhotoshopCS3 imaging process software (Adobe System Inc). The bars show the fold change of mean fluorescence intensity per cell (n = 50 cells picked randomly) compared with that of Puma^+/+ nonirradiated group. **P < .01.