Tumorigenicity of hypoxic respiring cancer cells revealed by a hypoxia–cell cycle dual reporter

Abstract

Although aerobic glycolysis provides an advantage in the hypoxic tumor microenvironment, some cancer cells can also respire via oxidative phosphorylation. These respiring ("non-Warburg") cells were previously thought not to play a key role in tumorigenesis and thus fell from favor in the literature. We sought to determine whether subpopulations of hypoxic cancer cells have different metabolic phenotypes and gene-expression profiles that could influence tumorigenicity and therapeutic response, and we therefore developed a dual fluorescent protein reporter, HypoxCR, that detects hypoxic [hypoxia-inducible factor (HIF) active] and/ or cycling cells. Using HEK293T cells as a model, we identified four distinct hypoxic cell populations by flow cytometry. The non-HIF/ noncycling cell population expressed a unique set of genes involved in mitochondrial function. Relative to the other subpopulations, these hypoxic "non-Warburg" cells had highest oxygen consumption rates and mitochondrial capacity consistent with increased mitochondrial respiration. We found that these respiring cells were unexpectedly tumorigenic, suggesting that continued respiration under limiting oxygen conditions may be required for tumorigenicity.antiangiogenesis | metabolism | mitochondria C hanges in cancer-cell metabolism have been linked to genetic alterations of oncogenes, tumor suppressors, and metabolic enzymes (1-3). The hypoxic tumor microenvironment further modifies metabolism through activation of hypoxia-inducible factors (HIFs). The HIFs enhance tumorigenesis by stimulating glycolysis, cell motility, and angiogenesis (4, 5). Thus, hypoxia portends poor prognosis in common cancers, such as gastric, lung, ovarian, pancreatic, prostate, and renal carcinomas (5).Although Otto Warburg observed respiration in certain cancer types, his obsession with aerobic glycolysis as a cause of cancer promulgated the prevailing misconception that cancers only exhibit the Warburg effect exclusive of respiration (6). Because the hypoxic tumor microenvironment activates HIFs and diminishes respiration, whether hypoxia enhances tumorigenicity at the expense of respiration is not fully understood (7). We found recently that oxidative and glycolytic metabolism coexist in hypoxic B lymphocytes, such that the shunting of glucose to lactate away from the tricarboxylic acid cycle (TCA) cycle by hypoxia is compensated through glutamine oxidation in the TCA cycle (8). These metabolic aberrations suggest the existence of hypoxic respiring (herein termed non-Warburg) cells capable of continued oxidative metabolism under hypoxic conditions. Further, it is believed that cancer cells within the tumor microenvironment are either aerobic or hypoxic because of oxygen gradients coming from nearby imperfect blood vessels. An intriguing commensal metabolic relationship between hypoxic and aerobic cells has been documented, whereby hypoxic cells produce lactate that is converted to pyruvate for respiration by aerobic cancer cells located nearby the blood vessel (9). We hypothesi...

Antiproliferative effects of mitochondria-targeted cationic antioxidants and analogs: Role of mitochondrial bioenergetics and energy-sensing mechanism

“…However, it is clear that tumor heterogeneity should be taken into consideration for proper interpretation. Recent studies indicate metabolic heterogeneity among hypoxic cancer cells, and that hypoxic tumor cells contain a sub-population of non-Warburg respiring cells [43]. Previously, we have shown that Mito-CP and other mitochondria-targeted antioxidants inhibit Nox2 expression and activity in mice and inhibit nitration and carbonylation of proteins induced by NADPH oxidases (Nox) and inducible NOS [44,45].…”

Combination therapy with BPTES nanoparticles and metformin targets the metabolic heterogeneity of pancreatic cancer

“…Transduced cells were selected with 1 μg/mL puromycin and observed by fluorescence microscopy to detect the cycling cell subpopulation (mCherry + ). Cells were then exposed to 1% O 2 overnight for hypoxia assessment (GFP + ) after 0, 2, 4, 8, and 24 h of reoxygenation (21). To generate orthotopic tumors, 5 × 10 6 cells from each HypoxCR-pancreatic cancer cell line in 50 μL PBS and Matrigel in a 1:1 ratio were injected into the pancreas.…”

“…Although the efficacy of BPTES-NPs in mice was found to be similar to CB-839, BPTES-NPs were better-tolerated, showing no liver toxicity, in contrast to CB-839-treated mice. In vivo imaging using the HypoxCR reporter (21) revealed that BPTES-NPs targeted the replicating tumor cell subpopulation without affecting hypoxic cells. Using NMR and liquid chromatography-mass spectrometry (LC-MS), we found that the residual tumor cells after BPTES-NP treatment were reliant on glycolysis and glycogen synthesis.…”

“…4A), which simultaneously detects hypoxic cells (hypoxia-inducible factor-dependent GFP expression) and cycling cells (G2/M/S phase-dependent mCherry expression) in vivo. This reporter has been a valuable tool for identifying and quantifying specific tumor cell subpopulations that are sensitive to drug effects (21). We transduced patient-derived PDAC cells with HypoxCR and injected these cells into mouse pancreas.…”

Mitochondrial functions in stem cells

“…HIF1-a-induced expression of pyruvate dehydrogenase kinases (PDKs) decreases oxidative metabolism of glucose, while glycolytic targets of HIF1-a stimulate lactate generation. However, non-HIF-1a-expressing cells can maintain higher respiration rates under hypoxic conditions [88], of potential relevance to naïve ES cells, which exhibit low HIF-1a expression and a HIF-inducible shift to EpiSC phenotype [21 ].…”