Hypoxia inducible factor (HIF)-1 is a master transcriptional regulator for adaptation of cells to hypoxia. In addition to hypoxic responses, HIF-1 also plays an important role in the development of hematopoietic stem cells. Genetic deletion of β subunit of HIF-1 causes impairment of hematopoiesis. Culture of hematopoietic stem cells under hypoxic condition induces elevation of HIF-1α , another subunit of HIF-1, and subsequently enhances the growth of these cells. In our previous work we found that thrombopoietin (TPO), an important and non-redundant cytokine required for normal stem cell development, induces HIF-1α elevation in the TPO-dependent human leukemic cell line UT-7/TPO and in Sca-1+/c-kit+/Gr-1- cells (Kirito, K. et.al. Blood 2005). Under normoxic conditions HIF-1α is hydroxylated on proline residues by prolyl hydroxylase (PHD), which leads to its recognition by the von Hippel-Lindau tumor suppressor protein (pVHL), leading to degradation of HIF-1α . Hypoxia inhibits PHD function, blocking ubiquitination of HIF-1α , stabilizing the protein. We found that TPO controls stability of HIF-1α even under normoxic conditions. However, the mechanism by which TPO controls the stability of the protein remains unclear. Recently, several groups have reported that mitochondrial ROS play crucial roles in stabilization of HIF-1α in response to hypoxia. Disruption of mitochondrial function, either by interfering RNA against complex III of the mitochondrial electron transport chain or genetic elimination of cytochrome c, completely abolished the hypoxia-induced HIF-1α response. Based on these findings we hypothesized that ROS might be involved in TPO-induced HIF-1α elevation. To examine our hypothesis, we first tested whether TPO induced ROS production in UT-7/TPO cells using 2′, 7′-dichlorofluorescein diacetate, a redox sensitive fluorescence dye, and found that the hormone clearly induced ROS production in these cells. Next, we analyzed whether TPO-induced ROS generation is required for accumulation of HIF-1α . Pre-treatment of UT-7/TPO cells with the ROS scavenger catalase completely blocked HIF-1α elevation after TPO treatment. Furthermore, diphenylene iodinium (DPI), an inhibitor for ROS generating flavoenzymes including mitochondrial respiratory complexes, also inhibited the effects of TPO on HIF-1α levels. These results indicate that TPO induced HIF-1α activation is mediated by ROS production. To study the molecular pathway(s) by which TPO affects ROS, we tested the effects of ROS blockade on several known TPO-responsive signaling molecules; neither DPI nor catalase affected the activation of JAK2, STAT5, p38-MAPK or p42/p44-ERK induced by TPO, although AKT activation was blocked. Moreover, LY294002, an inhibitor of PI3-kinase and its activation of AKT also blocked of the HIF-1α response to TPO. Finally, inhibition of mitochondrial function in UT-7/TPO cells with rotenone or oligomycin also inhibited TPO-dependent accumulation of HIF-1α without affecting Jak2 activation. In conclusion, we found that TPO regulates HIF-1α levels through activation of ROS generation within mitochondrial respiratory complexes. We speculate that TPO mimics hypoxia by induction of ROS generation at mitochondria and subsequent elevation of HIF-1α , and regulates important genes for metabolisms and survival of hematopoietic stem cells.

Author notes

Corresponding author

Sign in via your Institution