Stringent regulation of redox status is critical to the control of hematopoietic stem cell (HSC) quiescence and to the maintenance of HSC pool. However mechanisms by which oxidative stress controls HSC quiescence versus cycling remain unknown. Foxo3 transcription factor is required for the regulation of HSC quiescence and for the maintenance of hematopoietic and leukemic stem cell pool. Redox regulation is key to the Foxo3 control of HSC pool. ROS accumulation in Foxo3 null HSC mediates in vivo activation of p53, and increased p21 expression leading to an arrest in the G2/M phase of cell cycle associated with loss of quiescence. We hypothesized that ROS may regulate HSC quiescence versus cycling via control of DNA damage repair program. To address this question, we examined whether Foxo3 is involved in DNA damage response of HSC. We first evaluated by immunostaining phosphorylation of histone H2AX variant (γH2AX), a hallmark sensor of DNA strand break, in LSK (Lin−Sca-1+c-Kit+) cells freshly isolated from Foxo3−/− bone marrow. We found the number of cells with nuclear γH2AX foci significantly increased in Foxo3−/− LSK cells (n=100; >5 foci/nuclei) in comparison with wild type (WT)-LSK. We subsequently confirmed and quantified these data by flow cytometry analysis of γH2AX. Together these analyses showed that loss of Foxo3 leads to increased γH2AX levels in LSK cells at the steady state. We next evaluated the presence of DNA breaks, by submitting Foxo3−/− versus WT LSK FACS-sorted cells to single-cell gel electrophoresis (Comet Assay). These investigations confirmed that LSK cells from Foxo3−/− mice accumulate DNA breaks at the steady state, as the percentage of comet shape cells (4 fold) and comet length (3 fold) were all increased in Foxo3 mutant LSK. We then asked whether the increased ROS accumulation had any direct role in damaging DNA in Foxo3−/− LSK. Using a fluorescent probe specific for the most common oxidative DNA damage lesion, the 8-hydroxyguanine base (8-OxoG), we further showed that Foxo3−/− LSK cells exhibit oxidative DNA damage. To further investigate the potential function of ROS in the control of HSC DNA damage response, we treated Foxo3−/− and WT mice for 14 days with the ROS scavenger N-acetyl-cysteine (NAC; 100 mg/Kg/day) in vivo. NAC treatment reduced by four fold γH2AX in Foxo3−/− LSK cells to levels similar to that in WT-LSK cells. Similarly, comet assay analysis of FACS-sorted LSK cells from NAC-treated WT and Foxo3−/− mice showed a two fold reduction of DNA breaks. These results suggest that increase in ROS damage DNA and triggers DNA damage response in Foxo3−/− LSK cells at the steady state. Additionally, expression of a number of genes involved in DNA damage repair including Xrcc5 (Ku80) and Xrcc6 (Ku70) was highly downregulated in both long-term-HSC (LT-HSC, LSK-CD150+CD48−) and LSK populations as evidenced by Q-RT-PCR on the Fluidigm™ microfluidics array technology. Together these results strongly suggest that Foxo3-mediated redox regulation is required for protection of DNA from accumulating damage at the steady state in HSC. We further investigated whether ROS-mediated activation of p53 in Foxo3 null HSCs limits the extent of accumulation of DNA damage in HSC. To address this question we crossed p53+/−Foxo3+/− double heterozygous animals to generate p53-Foxo3 double knockout mice. Loss of p53 in Foxo3−/− mice led to significant rise in lymphocyte counts and decrease in neutrophil counts in comparison with Foxo3−/−, indicating a potential shift in lineage determination from HSC. To our surprise, loss of one allele of p53 in Foxo3-null mice significantly reduced gH2AX staining and DNA breaks, as analyzed respectively by flow cytometry and comet assay of sorted LSK cells. While the rescue of DNA damage in Foxo3−/− HSCs as result of loss of p53 was unexpected it is not clear whether it is related to the impact on the fate of HSC. The clarification of these questions in future studies will be important for understanding mechanisms that control the emergence of leukemic stem cells. Together these studies suggest that Foxo3 guards DNA from damage in HSC at the steady state. In addition they indicate an important function for ROS modulation in the in vivo regulation of DNA damage response in HSC. Altogether understanding mechanisms that control ROS modulation of DNA damage response are likely to advance our understanding of the regulation of normal hematopoietic and leukemic stem cell quiescence.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.