Hypoxic stabilization of mRNA is HIF-independent but requires mtROS

Abstract

Background: Tissue ischemia can arise in response to numerous physiologic and pathologic conditions. The cellular response to decreased perfusion, most notably a decrease in glucose and oxygen, is important for cellular survival. In response to oxygen deprivation or hypoxia, one of the key response elements is hypoxia inducible factor (HIF) and a key protein induced by hypoxia is vascular endothelial growth factor (VEGF). Under hypoxia, we and others have reported an increase in the half-life of VEGF and other hypoxia related mRNAs including MYC and CYR61; however, the mediator of this response has yet to be identified. For this study, we sought to determine if HIF-mediated transcriptional activity is involved in the mRNA stabilization induced by hypoxia. Methods: HEK293T or C6 cells were cultured in either normoxic or hypoxic (1% oxygen) conditions in the presence of 1 g/L glucose for all experiments. Pharmacological treatments were used to mimic hypoxia (desferroxamine, dimethyloxaloglutamate, CoCl2), inhibit mitochondrial respiration (rotenone, myxothiazol), scavenge reactive oxygen species (ROS; ebselen), or generate mitochondrial ROS (antimycin A). siRNAs were used to knock down components of the HIF transcriptional apparatus. mRNA half-life was determined via actinomycin D decay and real time PCR and western blotting was used to determine mRNA and protein levels respectively. Results: Treatment of HEK293T or C6 cells with hypoxic mimetics, desferroxamine, dimethyloxaloglutamate, or CoCl2 showed similar induction of HIF compared to hypoxia treatment, however, in contrast to hypoxia, the mimetics caused no significant increase in VEGF, MYC or CYR61 mRNA half-life. Knockdown of HIF-alpha or ARNT via siRNA also had no effect on hypoxic mRNA stabilization. Interestingly, treatment of HEK293T cells with the mitochondrial inhibitors rotenone and myxothiazol, or the glutathione peroxidase mimetic ebselen did prevent the hypoxic stabilization of VEGF, MYC, and CYR61, suggesting a role for mtROS in the process. Additionally, treatment with antimycin A, which has been shown to generate mtROS, was able to drive the normoxic stabilization of these mRNAs. Conclusion: Overall these data suggest that hypoxic mRNA stabilization is independent of HIF transcriptional activity but requires mtROS.