Using watermelons the uncover metabolic adaptations of cancer persister cells

Cancer persister cells, a subpopulation of tumor cells that transiently resist therapy in the absence of a resistance-mediating genetic alteration, contribute to disease recurrence. However, little is known about the non-genetic mechanisms underlying the ability of persister cells to resist therapy and why only a small fraction can re-enter the cell cycle under constant drug treatment. To address this question, we developed the watermelon library, a high-complexity expressed barcode library that enables simultaneous tracing of the lineage as well as the transcriptional and proliferative states of cells. We combined time-lapse imaging with single-cell RNA sequencing and metabolomics to show that cycling persister cancer cells do not acquire a facilitating resistance mutation but rather shift to a new metabolic state. This drug-induced metabolic reprograming is associated with high oxidative stress and a shift to fatty acid oxidation both in preclinical models and in tumor samples. Moreover, blocking this metabolic shift eliminates persister cells. By uncovering metabolic adaptions of this rare yet clinically relevant population, we expose new vulnerabilities that can be targeted to delay or even prevent disease recurrence.