Abstract The RAS oncogenes are mutated in one third of human cancers, but therapies against Ras-driven cancers have been unsuccessful. Ras proteins themselves are not yet druggable, and the Ras pathway contains uncharacterized redundancies, feedback mechanisms, and tributaries that have stymied the development of other targeted therapies. Multiple-agent combination therapies hold some promise, but their development requires a more thorough understanding of Ras cell biology than we currently possess. We therefore pursued creation of a physical and genetic map of the Ras pathway in non-small-cell lung cancer (NSCLC). Using H, K-, and NRas and twelve other Ras pathway proteins as baits, we conducted tandem affinity purification experiments to create a high-confidence protein-protein interaction (PPI) network. We integrated our data with public PPI, genetic susceptibility, and patient data to assemble an interpretable interaction map encompassing 360 proteins and 1000 physical interactions among them. Guided by the topology and annotations of this PPI map, we constructed a library of 1000 sgRNAs covering 120 genes. These were screened for pairwise genetic interactions (GIs) in A549 and H23 NSCLC lines using a dual sgRNA vector system to discover over 250 synthetic lethal genetic interactions. Each suggests a strategy for combination therapy. The combined GI/PPI network also produced myriad mechanistic hypotheses. Pursuing these, we made new discoveries that demonstrate the power of the approach. First, we find that KRas binds the cell adhesion regulator Radil, and that Ras regulates cell morphology by modulating the Rap signaling pathway. Second, we identify the critical guanine nucleotide exchange factor and effector by which KRas up-regulates macropinocytosis. Third, we demonstrate that that the in vivo physical interactions between Ras, Raf, and RalGEF family proteins depend crucially on the specific paralogs involved. The equivalent in vitro interactions are not selective, demonstrating that many undiscovered factors direct the strong specificity observed in vivo. Distinct genetic interaction patterns between paralogs in these same families support the non-equivalence of their members. Fourth, we detect a new synthetic lethal genetic interaction between the GTPase chaperone Rap1GDS1 and the GTPase RhoA. The strength of interaction varies among NSCLC lines, but correlates with the KRas dependence of each. This and other interactions show that several other small GTPases work in concert with Ras signaling to regulate tumor progression. Together, these discoveries show that our multiomic approach furthers two critical goals: it produces testable hypotheses that point to new Ras cell biology and reveals combination susceptibilities for the development of new therapies against Ras-driven cancers. Citation Format: Marcus R. Kelly, Kyuho Han, Kaja Kostyrko, Edwin E. Jeng, Nancie Mooney, Alejandro Sweet-Cordero, Michael Bassik, Peter K. Jackson. Proteomic and genetic interaction mapping of the Ras pathway reveals new effectors and vulnerabilities [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 959.
Read full abstract