Stress-Mediated Activation of Dormant Hematopoietic Stem Cells In Vivo

Abstract SCI-41

Maintenance of the blood system is dependent on dormant hematopoietic stem cells (HSCs), which are characterized by pluripotency and lifelong self-renewal capacity. In order to both maintain a supply of mature blood cells and not exhaust HSCs throughout the lifespan of the organism, most adult HSCs remain deeply quiescent during homeostasis, and only a limited number are cycling at any given time. The balance between self-renewal and differentiation of HSCs is controlled by external factors such as chemokines and cytokines, as well as by interactions of HSCs with their niche environment. The transcriptome of dormant CD34-CD150+CD48-LSK- HSCs significantly differs from that of active HSCs with the same phenotype, while the latter are highly similar to MPP1 progenitors which express CD34. One of the genes differentially expressed is the cylindromatosis (CYLD) gene, which encodes a negative regulator of the NF-κB signaling pathway. HSCs failing to express functional CYLD show various defects associated with a disturbed balance between dormant and active HSCs, suggesting a role for NF-κB signaling in establishing dormancy in HSCs. In addition, our studies have recently shown that the cytokine interferon-α (IFNα) very efficiently activates dormant HSCs in vivo. Within hours after treatment of mice with IFNα, HSCs exit G₀ and enter an active cell cycle. In general, IFNα is produced in response to viral infections by cells of the immune system, and plays an important role in the antiviral host defense. We now questioned whether endogenous IFNα is also produced in response to other types of bone marrow stress and whether this affects the proliferation rate of HSCs. To monitor IFNα production in the bone marrow in vivo, we have generated MxCre ROSA-R26-EYFP mice and found that treatment with both the chemotherapeutic agent 5-FU as well as the endotoxin LPS leads to the production of IFNα in the vicinity of HSCs and progenitors. In addition, LPS treatment in vivo induced a strong increase in HSC cycling. Surprisingly, since mice lacking the IFNα receptor (Ifnar^−/−) still respond to LPS, this effect is independent of IFNAR signaling. Strikingly, LPS-induced HSC activation correlated with increased expression of Sca-1, similar to what occurs upon IFNα treatment. Moreover, as for IFNα, the upregulation of SCA-1 is required for LPS-induced proliferation, since Sca-1^−/− mice fail to respond to LPS stimulation. In summary, these data suggest that not only virus-inducible IFNα, but also infections by gram-negative-bacteria-produced LPS induce cycling of progenitors and otherwise dormant HSCs.