Metabolic Diversity in Human Non-Small Cell Lung Cancer Cells




Chen, Pei-Hsuan

Journal Title

Journal ISSN

Volume Title


Content Notes


Cancer cells display oncogene-driven rewiring of metabolism to produce energy and macromolecules for growth. Inhibition of growth-promoting metabolic pathways may prove to be a useful therapeutic strategy in cancer. However, neither the full breadth of cancer cell metabolic diversity, nor the complement of mechanisms by which tumor mutations elicit metabolic reprogramming, are known. We set out to characterize cell-autonomous metabolic heterogeneity in non-small cell lung cancer (NSCLC) and to use orthogonal high-content data sets to understand the mechanisms by which metabolic phenotypes are established in lung cancer. A major goal is to understand whether these metabolic phenotypes predict therapeutic liabilities to novel metabolic inhibitors, targeted therapies, or conventional chemotherapeutic agents. We used a highly annotated panel of more than 80 NSCLC cell lines to develop the most comprehensive database of cancer cell metabolism to date. These cell lines were analyzed for a set of ~100 metabolic parameters derived from nutrient utilization, nutrient addiction, and isotope labeling patterns following culture with 13C-glucose or 13C-glutamine. Orthogonal data sets included analysis of the genome, epigenome, transcriptome and proteome, as well as sensitivity to over 40 chemotherapeutic agents. Several cell lines were also subjected to high-throughput chemical compound and genome-wide siRNA screens.
NSCLC cell lines display a surprising degree of cell-autonomous metabolic heterogeneity in culture. Many canonical hallmarks of cancer cell metabolism, including the Warburg effect, were observed to span at least a 10-fold range among cell lines grown under identical conditions. Affinity propagation clustering using metabolic features alone produced families that were largely distinct from clusters based solely on gene expression. Nevertheless, databases of metabolic features and orthogonal data sets could be cross-queried to identify robust, novel relationships connecting metabolic preferences to oncogenotypes, transcriptomic phenotypes and therapeutic responses.
Focused metabolic assays can produce a highly informative view of the metabolic phenotyping among large panels of cell lines. NSCLC cell metabolism is highly heterogeneous in every parameter so far assessed. Functional metabolic families describe an unparalleled view of the connections between genetics, drug sensitivity and cell-autonomous metabolism in NSCLC.

General Notes

Table of Contents


Related URI