Differential Regulation of Autophagy by Therapeutic and Physiologic Glucose Restriction under Normoxia and Hypoxia

Advances in our understanding of tumor biology achieved in the past decade have convincingly revealed the coupling of oncogenic transformation with increased glucose utilization, also known as aerobic glycolysis or the “Warburg effect”, that was observed more than 80 years ago. Although the altered glucose metabolism renders tumor cells selective growth and survival advantages, it also creates a therapeutic window to target cancer using extrinsic sugar analogs such as 2-deoxyglucose (2-DG). On the other hand, certain malignant cell populations inevitably face glucose starvation (GS) intrinsically arising as a solid tumor grows and the nutrient demand exceeds supply. In order to overcome the deleterious effects of both therapeutic (2-DG) and physiologic (GS) glucose restriction, tumor cells mount an evolutionarily conserved intracellular bulk degradation process called autophagy, among others, for adaptation and survival under these metabolic stresses. Therefore, delineating the precise role and regulation of autophagy in response to therapeutic as well as physiologic glucose restriction is critical not only for a better understanding of the biology of cancer but improving the treatment outcomes. Acting as a glucose analog, 2-DG blocks glycolysis and thereby reduces cellular ATP levels. Due to its unique structure, 2-DG also mimics mannose and thus interferes with N-linked glycosylation leading to the induction of endoplasmic reticulum (ER) stress. Thus, in Chapter 1 of this dissertation we set out to determine which pathway 2-DG interferes with is responsible for activating autophagy. We show that 2-DG activates autophagy predominantly through the induction of ER stress rather than ATP reduction. Furthermore, we find that 2-DG-induced ER stress stimulates autophagy through Ca2+-CaMKKβ signaling-dependent AMPK activation. In contrast, we show that although GS leads to ER stress which contributes to autophagy activation, it does so by a different mechanism. In addition to ER stress, GS also stimulates autophagy through lowering ATP and activating the canonical LKB1-AMPK energy sensing pathway as well as through increasing reactive oxygen species (ROS) resulting in the activation of ERK. While under normoxic conditions 2-DG only results in moderate ATP reduction, under hypoxic conditions this sugar analog elicits severe ATP depletion due to the lack of ATP production from the mitochondria. Accordingly, in Chapter 2 we investigate how 2-DG regulates autophagy under hypoxia. In a model of chemical hypoxia where mitochondrial ATP synthase is inhibited, 2-DG suppresses rather than induces autophagy. Similar observations are observed in a genetic model of hypoxia where cells are devoid of mitochondrial DNA. Moreover, in a more physiologic environmental model of hypoxia, both 2-DG and GS reduce autophagy in cells cultured under low O2 tensions. This blockage is accompanied by the disruption of the PI3K III-Beclin1 complex for autophagy initiation, conjugation of Atg12 to Atg5 for autophagosome expansion, as well as inhibition of the functional lysosomal compartment for autophagic degradation. It has been reported that autophagy responses to various cellular stresses to play either a pro- or anti-death role. Therefore, in Chapter 3 we study the functional role of autophagy activation in modulating 2-DG’s cytotoxic effects. Our results demonstrate that autophagy protects cancer cells against 2-DG-elicited cell death and apoptosis, apparently through relieving 2-DG-induced ER stress. Mechanistically, our data support a model where therapeutic and physiologic glucose restriction differentially activate autophagy under normoxia, while similarly inhibit this process under hypoxia. Functionally, our work shows that 2-DG-induced autophagy ameliorates ER stress and counteracts 2-DG cytotoxicity. Overall, this study delineates the molecular mechanisms and functional roles of autophagy regulation by glucose restriction under different environmental conditions, and therefore may provide useful information for improving 2-DG’s anti-tumor efficacy as well as for a better understanding of the influence of GS on tumor pathophysiology