KRASQ61H preferentially signals through MAPK in a RAF dimer-dependent manner in non-small cell lung cancer.

Assembly of RAS molecules into complexes at the cell membrane is critical for RAS signaling. We previously showed that oncogenic KRAS codon 61 mutations increase its affinity for RAF, raising the possibility that KRAS(Q61H), the most common KRAS mutation at codon 61, upregulates RAS signaling through mechanisms at the level of RAS assemblies. We show here that KRAS(Q61H) exhibits preferential binding to RAF relative to PI3K in cells, leading to enhanced MAPK signaling in in vitro models and human NSCLC tumors. X-ray crystallography of KRAS(Q61H):GTP revealed that a hyperdynamic switch 2 allows for a more stable interaction with switch 1, suggesting that enhanced RAF activity arises from a combination of absent intrinsic GTP hydrolysis activity and increased affinity for RAF. Disruption of KRASQ61H assemblies by the RAS oligomer-disrupting D154Q mutation impaired RAF dimerization and altered MAPK signaling but had little effect on PI3K signaling. However, KRAS(Q61H) oligomers but not KRAS(G12D) oligomers were disrupted by RAF mutations that disrupt RAF-RAF interactions. KRAS(Q61H) cells show enhanced sensitivity to RAF and MEK inhibitors individually, whereas combined treatment elicited synergistic growth inhibition. Furthermore, KRAS(Q61H) tumors in mice exhibited high vulnerability to MEK inhibitor, consistent with cooperativity between KRAS(Q61H) andRAF oligomerization and dependence onMAPKsignaling. These findings support the notion that KRAS(Q61H) and functionally similar mutations may serve as predictive biomarkers for targeted therapies against the MAPK pathway. Significance: These findings show that oncogenic KRAS(Q61H) forms a cooperative RAS-RAF ternary complex, which renders RAS-driven tumors vulnerable to MEKi and RAFi, thus establishing a framework for evaluating RAS biomarker-driven targeted therapies.