Abstract
Cancer is generally thought to be caused by expansion of a single mutant cell. However, analyses of human early lesions show that tumours can originate from several genetically distinct cell populations1–5. How neighbouring mutant clones interact to shape tumourigenesis, and which driver genes mediate these effects is largely unexplored. Here, we use an in vivo mosaic Drosophila epithelial model to systematically test interclonal interactions during early Ras-driven tumourigenesis. We screened 88 recurrent RAS co-mutated driver genes in human carcinomas for their ability to modify Ras-clone growth when disrupted in neighbouring epithelial clones. This uncovered two opposing classes of interactions: Interclonal cooperativity, where neighbouring mutant clones promote the overgrowth of Ras mutant tumours, and interclonal suppression, in which neighbours restrain Ras tumours, unexpectedly improving host survival. The strongest suppressive modifiers included canonical cell competition regulators, including Myc, archipelago (ago/FBXW7), and taiman (tai/NCOA1-3). In contrast, the strongest cooperative modifiers were disruptions of XNP/ATRX and SWI/SNF chromatin remodelling subunits (including Osa/ARID1A, Bap170/ARID2, Polybromo/PBRM1, among others), which in neighbouring cells induce a wound-like inflammatory program and drive an anabolic, pro-growth state in Ras tumours. Notably, the disruption of SWI/SNF components cell autonomously within Ras tumours confers no cooperativity. We show that interclonal cooperative Ras tumour growth requires reactive oxidative species and prostaglandin synthesis in SWI/SNF-disrupted clones. Together, this study provides a catalogue and emerging principles of cooperative and suppressive interclonal interactions among recurrent RAS co-mutated drivers, extending the rules of oncogenic cooperation beyond cell intrinsic co-mutation.
Competing Interest Statement
The authors have declared no competing interest.
