Ovarian Cancer Cell Survival and Proliferation Is Mediated by the Warburg Effect.

The Warburg Effect was hypothesized in the 1930 and states, when normal cells differentiate into cancerous cells, this will result in a metabolic switch from oxidative phosphorylation (OXPHOS) to aerobic glycolysis. One difference between these two pathways is the rate of glucose consumption. Glycolysis will consume more glucose in a shorter amount of time than OXPHOS. Aerobic glycolysis provides the appropriate mechanism to increase the rate of energy production and permits tumor proliferation and survival. The Warburg effect is functioning in prostate and gastric cancers and leukemia, which all exhibit a dysregulated glucose metabolism. Ovarian cancer is the most lethal of gynecological malignancies and is referred to as a "silent killer." This cancer gradually progresses and is poorly diagnosed. Usually, there are concealed symptoms in the earlier stages of ovarian cancer but becomes detectable at stages III and IV; therefore, it is crucial to find early detections markers to improve therapeutic intervention. The objective of this study was to understand if a metabolic change occurs while ovarian cells are transitioning from an epithelial to mesenchymal phenotype; thus, this will give us insight if the Warburg effect is operative in ovarian cancer. In order to test this hypothesis, the cell's phenotype, lactic acid production and ATP production was inspected. The lactic acid and ATP levels were normalized to the cell's protein concentration. The human ovarian cancer cells lines used in the study: IOSE (donated by Nelly Auersperg), TOV112D (ATCC), SkOv3 (ATCC), and HEYC2 (donated by Jean Hurteau). In order to determine if an epithelial or mesenchymal phenotype was present in each cell line, immunocytochemistry was conducted for E-cadherin (CDH1) (BD Biosciences Pharmingen) expression. An upregulation of E-cadherin expression is proportional to an epithelial phenotype; thus, IOSE and TOV112D had high E-cadherin expression. In contrast, a downregulation of E-cadherin expression is related to a mesenchymal phenotype such as in SkOv3 and HEYC2, which had no E -cadherin expression. Secondly, lactic acid levels (l-lactate assay kit, Invitrogen) were proportionally related to a mesenchymal phenotype, such as the HEYC2 cells, and inversely related to an epithelial phenotype, such as the IOSE. Therefore, these results suggested a significant difference in the amount of lactic acid produced from the different cell lines. Next, energy production (luminescent cell viability assay kit, Promega) was examined in the different cell lines to asses ATP levels. Results were significantly different, illustrating high ATP production in TOV112D, then HEYC2 and finally the SkOv3. To determine if ATP is produced by OXPHOS, cells are treated with carbonyl cyanide m-chloroenyl hydrazone (CCCP), which will inhibit OXPHOS by dissipating the proton motive force in the mitochondria and lead to a decrease in mitochondrial membrane potential. 2-deoxy-D-glucose (2DG) is a glucose isomer that inhibits glycolysis. These treatment groups distinguish between ATP produced from glycolysis or OXPHOS. These observations will allow us to define a metabolic preference when human ovarian cells transition into a mesenchymal phenotype. If ovarian cancer is governed by a dysregulated glucose metabolism, then metabolite homeostasis is crucial for controlling the severity of the cancer and may provide a basis for early detection. Supported by: NIH/NCCAM AT004085; NCI CA133915.

(platform)