Modelling Fusobacterium lifestyles transitions by integrating transcriptomics and growth data

Bacteria living inside the tumoral micro-environment play a crucial role in the development of cancer and its progression. Enrichment of Fusobacterium nucleatum in colorectal cancer (CRC) tissue has been acknowledged as a major driver of its proliferation and mortality. Representatives of the F. nucleatum species exhibit a remarkable variability, being linked to a growing list of diseases. In this process, cellular metabolism plays a key role, allowing bacterial cells to efficiently cope with an ever-changing environment. To date, however, a mechanistic understanding of its relationship(s) with virulence and/or cancer-associated phenotypes is missing. In this work we characterize the basal physiology of this bacterium by reconstructing an experimentally validated genome-scale metabolic model (GEM) to simulate the major phenotypical features of F. nucleatum in different nutritional conditions. Further, we used gene expression data obtained from in vitro models to contextualize this metabolic reconstruction and simulate relevant phenotypes such as its interaction with human cells. Our analyses revealed that adhesion triggers a metabolic rewiring, with suppression of branched-chain amino acid catabolism and increased uptake of specific nutrients (e.g., methionine and serine), while invasion leads to a partial reactivation of central carbon and nitrogen pathways. Moreover, we identified shifts in short-chain fatty acid production and redox balance that may contribute to bacterial persistence and modulation of the tumor microenvironment.