Reconstruction of a Context-Specific Metabolic Model of Epithelial to Mesenchymal Transition

The epithelial-mesenchymal transition (EMT) is a fundamental developmental process, where epithelial cells lose their epithelial phenotype to gain a mesenchymal phenotype. Alterations in metabolism during EMT may play a causal role in transformation [1]. Studying the metabolic changes that occur during EMT is therefore of prime importance in order to understand what triggers this transformation. This can be done by the use of context-specific metabolic network models. We used transcriptomic and metabolomic data to define two phenotypes of breast epithelial cells; epithelial and mesenchymal.  In the first step, metabolomic data was integrated with a human genome scale global metabolic network (RECON1) for defining model uptake and secretion rates [2]. Metabolomics data was in the form of changes in metabolites concentration in media over time. These measurements were converted to uptake/secretion fluxes which were then integrated with the model, by using optimization to generate a feasible model while minimizing the deviation of the rates from the experimentally measured values. In the second step, transcriptomic data was used to define the set of reactions to be included and the FASTCORE algorithm was used to generate a consistent network [3] common to both conditions. The reason for generating a common model is the observation that EMT is a reversible process and is sustained by transient molecular changes in the cell rather than permanent genetic changes [1]. Random sampling of the flux solution space was then performed in order to get a flux distribution of the common model. Based on this flux distribution and differences in expression in transcriptomic data, two models were generated: epithelial phenotype and mesenchymal phenotype. Thereafter both the models were sampled to obtain flux differences between the two metabolic phenotypes. This method helped us study the differences in the two phenotypes associated with EMT. We further plan to use the models to predict the targets for preventing EMT and lethal for one/both of the phenotypes.

References:

1. Li, Linna, and Wenliang Li. "Epithelial–mesenchymal transition in human cancer: Comprehensive reprogramming of metabolism, epigenetics, and differentiation." Pharmacology & therapeutics 150 (2015): 33-46.
2. Duarte, Natalie C., et al. "Global reconstruction of the human metabolic network based on genomic and bibliomic data." Proceedings of the National Academy of Sciences 104.6 (2007): 1777-1782.
3. Vlassis, Nikos, Maria Pires Pacheco, and Thomas Sauter. "Fast reconstruction of compact context-specific metabolic network models." PLoS computational biology 10.1 (2014).