Background: Among all tissues, bone marrow (BM) is the most sensitive tissue to ionizing radiation (IR)-induced acute tissue damage (ATD) and chronic long-term residual damage (LT-RD). BM failure and a significant reduction in blood cells (pancytopenia) often occurs within days after exposure to IR due to the massive death of proliferative hematopoietic progenitor cells (HPCs). However, due to their quiescent cell cycle status and reduced fidelity of DNA repair feature, many hematopoietic stem cells (HSCs) cannot fully eliminate such damage and enter senescence; this results in LT-RD. Abnormal dysplastic hematopoiesis is the most common LT-RD in most victims of IR, followed by an increased risk of leukemia/lymphoma development. Thus IR exposure is an established cause of BM failure and leukemia. A significant increase in the production of inflammatory cytokines is induced by IR which contributes to the pathogenesis of both ATD and LT-RD. Such inflammatory cytokines induce the activation of Ripk3-Mlkl-mediated necroptotic signaling in HSCs. However, the role of Ripk3-Mlkl signaling in IR-induced damage has not studied.

Experimental procedures: The self-renewal capacity of HSCs among Ripk3-/-, Mlkl-/- and WT mice were examined and compared by serial transplantation assay. The phenotypes of ATD and LT-RD induced by different dosages of IR were compared among Ripk3-/-, Mlkl-/- and WT mice. The mechanism by which Ripk3 signaling prevents IR-induced leukemia development was studied.

Results: Ripk3-Mlkl signaling is not required for hematopoiesis during homeostatic condition. However, during serial transplantation, inactivation of such signaling prevents stress-induced loss of HSCs. Interestingly, Ripk3 signaling also induces an Mlkl-independent ROS-p38-p16-mediated senescence in HSCs. Thus Ripk3-/- HSCs showed better competitive hematopoietic ability compared to Mlkl-/- and WT HSCs during serial transplantation.

A sub-lethal dosage of IR (6Gy) induces Ripk3-dependent NF-κB activation and pro-survival gene expression in HSCs, which is necessary for the survival of damaged HSCs. After 6Gy IR, although DNA damage is repaired in most HSCs within 2 days, a proportion of HSCs in WT and Mlkl-/- mice fail to fully repair the damage and undergo p53-p21-dependent senescence. However such cells in Ripk3-/- mice die from apoptosis. Thus the remaining HSCs in Ripk3-/- mice should be functionally normal, while a proportion of the remaining HSCs in Mlkl-/- and WT mice remain damaged but senescent, all as demonstrated by competitive hematopoietic reconstitution assay.

Multiple low-doses of IR (1.75Gy once week × 4) induce HSC exhaustion in WT mice but not in Ripk3-/- and Mlkl-/- mice. Interestingly, almost all Ripk3-/- mice develop acute lymphoblastic leukemia within 200 days after such low dose IR, while 45% of WT and 60% of Mlkl-/- mice develop thymomas within 360 days (see Figure). Mechanistically, such low-dose IR stimulates chronic inflammatory cytokine production. Such cytokines induce Ripk3-Mlkl-mediated necroptosis in response to HSC exhaustion observed in WT mice. These cytokines also induce Ripk3-ROS-p38-p16-mediated senescence in response to impaired HSC functioning observed in both WT and Mlkl-/- mice. In Ripk3-/- mice, due to the lack of both necroptotic and senescent signaling, mutant HSCs accumulate and leukemia development is accelerated.

Conclusion: Ripk3 signaling plays distinct roles in HSCs in response to different doses of IR. High-dose IR induces Ripk3-dependent NF-κB/survival signaling, which is required for the survival of HSCs which fail to repair the damage. Thus temporal inhibition of Ripk3-NF-κB signaling might help to remove the damaged HSCs thus preventing the occurrence of LT-RD. However multiple low-doses of IR induces Ripk3 activation in HSCs which represses leukemia development by inducing both ROS-p38-p16-mediated senescence and Ripk3-Mlkl-mediated necroptosis. Induced activation of Mlkl-necroptosis might help to repress leukemia development by removing damaged HSCs.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution