Hypoxia and the Mitochondrial Unfolded Protein Response

Thesis (Ph.D.)--University of Rochester. School of Medicine & Dentistry. Dept. of Pathology, 2017.Mitochondria are central to cellular metabolism, generating ATP through oxidative phosphorylation. Additionally, mitochondria are signaling organelles that regulate stress-induced cell death and adaptation. Altered production of small molecule metabolites and/or changes in the localization and distribution of mitochondrial proteins can contribute to signaling cascades that impact mitochondrial and cellular fitness. In particular, perturbations in mitochondrial protein homeostasis trigger a compartment-specific mitochondrial unfolded protein response (UPRmt), which protects against a broad range of stressors including reactive oxygen species (ROS), respiratory chain dysfunction, mitochondrial DNA damage and pathologic bacteria. Furthermore, mitochondrial stress in one tissue can activate the UPRmt in distal tissues suggesting mitochondrial health is monitored remotely for prophylactic adaptive purposes. In the genetic model organism Caenorhabditis elegans, canonical UPRmt signaling requires the transcription factor Activating Transcription Factor associated with Stress 1 (ATFS-1), which contains dual mitochondrial and nuclear localization motifs. Healthy mitochondria import and degrade ATFS-1; whereas dysfunctional mitochondria fail to import ATFS-1 resulting in its nuclear accumulation and binding to promoters of UPRmt target genes. Here, we show that either genetic or pharmacologic stresses that induce UPRmt protect against anoxia-reperfusion (A/R) injury in C. elegans. ATFS-1 is essential for the protective effects of UPRmt activation, and gain-of-function (gf) atfs-1 mutants that exhibit constitutive UPRmt activation are protected against A/R injury. To investigate cell non-autonomous UPRmt signaling we developed a novel genetic system to restrict atfs-1(gf) expression to a single tissue, and using this model we find that atfs-1(gf) alone triggers an autonomous UPRmt in the tissue of expression, but does not activate UPRmt in other tissues. Likewise, cells with atfs-1(gf) showed cell autonomous protection from A/R, but protection is not extended to remote tissues. Our findings contrast with de facto mitochondrial stress, which initiates both local and remote protection. This suggests ATFS-1's centrality is limited to cell autonomous responses to stress and that an alternate pathway underlies remote responses. Preliminary work is also described testing the hypothesis that metabolites accumulated under mitochondrial stress can function as signaling molecules. Specifically, genetic reagents were developed to assess the potential signaling roles of L-2-hydroxyglutarate (L-2-HG), which is produced by lactate dehydrogenase (LDH) under hypoxia, and its enantiomer D-2-HG, which is produced by cancer associated mutations in isocitrate dehydrogenases (IDH1 or IDH2). Both L- and D-2-HG can inhibit epigenetic regulatory enzymes (e.g. HIF prolyl-hydroxylases and JmjC histone demethylases) which utilize α-ketoglutarate as a substrate, providing a potential link between hypoxia and epigenetic modifications that influence UPRmt gene induction. These reagents will be useful to determine the impact of modulating 2-HGs levels under stress, and to investigate their potential signaling roles in the UPRmt. Overall, the results presented herein provide new insights to mitochondrial nuclear-communication and signaling during A/R stress. These insights will help to direct future work in mito-centric and hypoxia-relevant mammalian disease fields including cancer, stem cell biology, and ischemia-reperfusion injuries such as heart attack and stroke, where the UPRmt may contribute to cell survival