Abstract
Cancer cells are often home to profound metabolic specialization, such as the Warburg effect wherein cancer cells exhibit high glucose uptake with conversion to lactate in the presence of oxygen. To understand the mechanisms underlying the Warburg effect in hepatocellular carcinoma and to investigate whether and how both the essential nutrient concentrations and the cells’ gene expression, impact on this metabolic alteration, we integrated computational genome-wide metabolic modeling with an experimental study on cell lines. We converted transcriptome sequencing (RNA-seq) data from two hepatocellular carcinoma lines HuH7 and PLC/PRF/5 to flux bounds in the most comprehensive human metabolic network model to date, Recon3D. A new method was developed to enable changes in medium concentrations to impact Flux Balance Analyses (FBA) of genome wide metabolic maps. By integrating the two methods, the cells’ metabolic behavior could be studied due to applying a new Nature Nurture Scaling (NNS) factor, which assesses the relative importance of gene expression (‘Nature’) and medium composition (‘Nurture’). To examine whether the medium concentrations of glucose and glutamine affect the rate of aerobic glycolysis and growth rate in vitro, we further cultured the same two hepatocellular carcinoma lines on media that contained different amounts of glucose and glutamine.
Our integrated approach where expression level (nature) and the environment (nurture) work together, led us to conclude that proliferation, glucose consumption, and lactate production are associated with the presence of glucose, but do not necessarily increase with its concentration when the latter exceeds the physiological concentration. There was no such association with the presence of glutamine. The observed dependencies on glucose concentration could hereby be understood in terms of a balance between Nature and Nurture, as could be the lack of the cells’ response to glutamine.
Author summary Metabolism is directly involved in many human diseases including cancer, and indirectly in virtually all, because disease causes metabolic changes that can accompany or even affect etiologies and be read as biomarkers. The best-known metabolic abnormality in cancer cells is an increased glycolysis followed by lactic acid production even in the presence of oxygen and fully functioning mitochondria, a process known as the Warburg Effect. So-called genome-scale metabolic reconstructions integrate all known metabolic reactions occurring in an organism into a single map. Together with a mathematical method for simulating the optimal balance of metabolic fluxes, this holds promise for studying the mechanisms by which networks control life. We combined this genome-scale metabolic modeling approach with in vitro experiments to investigate whether the behavior of cancer cells is determined by their nutrition or/and the expression of their genes. We hereto developed a new method that integrates the genome (nature) and the environment (nurture) and identified the influence of cell-nutrition changes on the Warburg effect in hepatocellular carcinoma.
Highlights 1. Cell nutrition and gene expression can now be dealt with simultaneously by FBA through a newly developed Nature Nurture Scaling (NNS) factor.
2. The physiological concentration of glucose is sufficient for the production of lactate by Huh7 and PLC.
3. Huh7 and PLC growth rates are independent of glutamine in media containing glucose.
4. On glutamine alone Huh7 and PLC growth rates are much lower than on glucose alone.