Discussion
Emma Newport, Lydia Ashton, Michelle Potter, Tim Claridge1, James McCullagh1, Karl Morten and Alessandro Valli2
Nuffield Department of Obstetrics and Gynaecology, University of Oxford, John Radcliffe Hospital
1 Chemistry Research Laboratory, University of Oxford, 12 Mansfield Road, Oxford. 2 TDI, Nuffield Department of Medicine, NDM Research Building, University of Oxford, Old Road Campus, Headington
Otto Warburg’s experiments in the 1950’s indicated that cancer cells in the presence of oxygen produced most of their energy from glycolysis rather than through mitochondrial oxidative phosphorylation (OXPHOS). This became known as the Warburg effect. From 2005 clinical studies began to emerge suggesting that drugs proposed to impact on mitochondrial energy metabolism could significantly reduce the incidence of cancer. For example, diabetics taking the drug metformin developed fewer cancers than diabetics not taking metformin (1-2). Recent work using nuclear magnetic resonance (C13 NMR) to study glucose oxidation in primary tumours in patients and animal models identified significant amounts of mitochondria glucose oxidation taking place in cancer cells (3-4).
Our studies using probes (Luxcel Biosciences) to measure extracellular oxygen and media acidification in a range of cancer cell lines identified a range of bio-energetic phenotypes. In these assays media acidification is used as an indicator of lactate production (glycolysis) while extracellular oxygen consumption indicates levels of OXPHOS. On high glucose (25mM) culture media cancer cells showed a variety of phenotypes, some showed the Warburg effect, others had high rates of OXPHOS while other lines showed high levels of both. As plasma glucose levels in patients range from 4-6mM with considerably lower glucose levels at the centre of tumour we investigated the effect of more physiological glucose levels of cancer cell bio-energetics. Interestingly when glucose levels were reduced to 1mM all cancer cell lines tested switched to use OXPHOS to generate energy.
Further studies were carried out using mass spectrometry to track the fate of C13 labelled glucose and glutamine under high (10-25mM), low (1mM) and trace glucose conditions. Surprisingly, fully labelled C13 glucose label experiments under high glucose (10mM) conditions showed similar levels of glycolytic intermediates irrespective of whether high or low media acidification was observed. All cell lines showed significant levels of C13 labelled TCA cycle intermediates. Studies with C13 glutamate highlighted some interesting metabolic pathways. The expected TCA intermediates (malate and succinate) were observed but the most abundant intermediate was citrate which is likely produced, at least in some part, by glutamine carboxylation. Cell lines with the highest rates of mitochondrial respiration tended to show the highest levels of TCA intermediates. Surprisingly on both high (25mM) and trace glucose media a significant amount of C13 glutamine label ended up in the glycolytic pathway. On trace glucose media evidence was found for C13 glutamine breakdown products ending up in glycogen.
These studies highlight the significant effect cell culture conditions have on cancer cell metabolism and bio-energetics. The effectiveness of drugs targeting cancer cell metabolism are likely to be influenced by the cancer cell environment and culture conditions. Using a metabolomics approach in combination with higher throughput bio-energetic profiling will allow potential therapeutic metabolic targets to be selected on a cell line /patient specific basis.
References (1) Evans, J.M. et al (2005) BMJ 330, 1304-1305. (2) Decensi, A. et al (2010) Cancer Pev. Res, 3, 1451-1461. (3)Marin-Valencia (2012) Cell Metabolism, 15, 827-837. (4) Mahler (2012) NMR Biomed 25, 1234-1244.
