Metabolic Cross Talk Between Foxo3 and mTOR Is Essential for Hematopoietic Stem Cell Function

Abstract 856

Unbalanced accumulation of reactive oxygen species (ROS) compromises self-renewal of hematopoietic stem cells (HSC). Nonetheless, survival of HSC in the hypoxic niche requires mitochondrial ROS-mediated activation of hypoxic response and significant metabolic adaptation. Hypoxia favors glycolysis to mitochondrial oxidative phosphorylation for energy production. In agreement with this, increased mitochondrial function results in defective maintenance of HSC as observed in Tsc1^−/− mice through overactivation of mammalian target of rapamycin (mTOR) signaling pathway. Mechanisms that regulate the strict coordination of mitochondrial function with upstream metabolic pathways that is required for HSC maintenance and balancing ROS remain largely unknown. Foxo3 Forkhead transcription factor is a strong candidate for coordinating metabolic pathways in HSC. Evidence from our laboratory and others has identified Foxo3 as a critical regulator of HSC quiescence and a key modulator of oxidative stress in HSC. To address whether Foxo3 has a more global metabolic control of HSC activity, we investigated the mitochondrial function in Foxo3 mutant HSC. To achieve this, we measured ATP content and oxygen consumption, two major defining mitochondrial parameters, using ATP Bioluminescence Assay and Oxygen Biosensor analysis respectively in freshly isolated Lin⁻Sca-1⁺cKit⁺(LSK) CD34⁻Flk2⁻ bone marrow cells isolated from wild type and Foxo3^−/− mice. We showed that loss of Foxo3 leads to significant mitochondrial defects in HSC as indicated by a strong decrease in both ATP content and oxygen consumption. In agreement with this, the glycolytic flux in Foxo3 mutant HSC, as analyzed by ¹³C lactate production using gas chromatography-mass spectrometry, was increased indicating a shift in the ATP production from mitochondria to the cytosolic glycolysis in Foxo3 mutant HSC. In addition, mitochondrial mass and membrane potential, that is generated during mitochondrial oxidative phosphorylation for energy production, were assessed by flow cytometry in WT and Foxo3^−/− LSK using mitotracker green and JC-1 probes respectively. Loss of Foxo3 increased both mitochondrial mass and membrane potential, likely reflecting a compensatory mechanism to the defective mitochondria in Foxo3^−/− HSC. These anomalies may partly contribute, in addition to the known defective ROS detoxification, to the increased ROS levels previously observed in Foxo3^−/− HSC. Importantly, the mitochondrial dysfunction was not due to the abnormal increase of ROS observed in Foxo3^−/− HSC since in vivo treatment of Foxo3^−/− mice with ROS scavenger N-Acetyl-Cysteine (NAC) for two weeks, did not revert the increased mitochondrial membrane potential. Altogether these results strongly suggest that Foxo3 is critical for the regulation of mitochondrial function in HSC. mTOR signaling controls major cellular metabolic processes and is critical for the regulation of mitochondrial function in HSC. We have previously found that mTOR signaling is amplified by a redox-mediated mechanism in Foxo3^−/− hematopoietic progenitors (Yalcin et al., 2010). Thus, we asked whether mTOR signaling is involved in the Foxo3 regulation of mitochondria function in HSC. Interestingly, in vivo treatment of Foxo3^−/− mice for two weeks with rapamycin, a specific inhibitor of mTOR complex 1 (mTORC1) activity, as measured by the decrease in phosphorylation of ribosomal protein S6 in HSC, mitigates the increased mitochondrial membrane potential and normalizes ROS levels in Foxo3^−/− HSC, suggesting that Foxo3 regulation of mitochondria is mediated by mTOR signaling in HSC. Notably, rapamycin treatment rescued partially Foxo3^−/− HSC pool and function as measured by the number and frequency of LSK in Foxo3^−/− mice, as well as their long-term repopulation ability measured by the capacity of CD48⁻CD150⁺LSK cells to repopulate the hematopoietic compartment in lethally irradiated recipient mice within 8 weeks. Taken together, our findings reveal a new function for Foxo3 in the control of mitochondria in HSC and support a model in which mitochondria is key to the maintenance of HSC. We propose that a Foxo3-mTOR signaling node partly controls mitochondrial function in HSC. These findings are likely to have an important impact on our understanding of the metabolic regulation of hematopoietic and leukemic stem cells and may be of therapeutic value.