We previously showed that glucose deprivation can activate POX to maintain autophagy and survival. Since nutrient deprivation is usually accompanied by hypoxia, we tested the effects of hypoxia on POX expression. A variety of cultured cells subjected to hypoxia showed an increase in POX expression either monitored at the level of mRNA by real time PCR or by a luciferase assay for POX promoter activity. POX mRNA and POX protein by Western blot increased as a function of hypoxia (5%, 0.5%, 0.05% oxygen) and duration of hypoxia. Interestingly, the increase in POX is not mediated by HIF-1alpha. Instead, the POX response was mediated by AMPK. Consistent with the dependence on AMPK, we found that ATP levels which were decreased with hypoxia, decreased further with POX knockdown. Importantly, the decrease in cell proliferation due to hypoxia was accentuated by knockdown of POX with siRNA. Although POX was responsible, in large part, for the increase in ROS with hypoxia, it did not induce apoptosis as measured by PARP cleave. Instead, hypoxia induced autophagy and this autophagy was decreased by the knockdown of POX by siRNA. Although the mechanism for POX induction was identified in tissue culture, we desired confirmation of the effects of hypoxia in xenograft tumors in vivo. In collaboration with Kristine Glunde?s laboratory at Johns Hopkins, we tested POX expression in tumors in vivo. This model uses cells expressing green fluorescent protein (GFP) under control of a hypoxia response element (HRE) controlled promoter. We first showed that these cells exposed to hypoxia or chemical hypoxia (CoCl2) expressed POX as monitored by qPCR. These cells were then injected into nude mice and when the tumors had attained a size with inadequate vascularization, they were removed, sectioned and stained for GFP and for POX. With the help of Dr. Miriam Anver and Donna Butcher of the histotechnology lab, a method was developed with a chicken anti-GFP antibody and a rabbit anti-POX antibody. Differential fluorescence of the secondary antibodies distinguished the expression of GFP and POX, respectively. Using a mathematical model, colocalization was established. Thus, the coupling between hypoxia and POX expression was confirmed in vivo. These in vitro and in vivo studies led to the conclusion that with hypoxia, AMPK activates PRODH/POX to produce ROS which initiates autophagy. On the other hand, with low glucose in the presence or absence of hypoxia, AMPK activates PRODH/POX to generate ATP for cell survival. After neovascularization with an adequate blood supply when tumor cells are proliferating, this state is maintained by the downregulation of PRODH/POX by the increased levels of miR-23b* which is increased by c-MYC. We are currently identifying modulatory proteins for this regulatory axis. An interesting finding is that FOXO proteins are involved in the POX-mediated, ROS generated signaling. This mechanism is being actively explored.
Showing the most recent 10 out of 15 publications
