Computationally designed sensors for endogenous Ras activity reveal signaling effectors within oncogenic granules

Abstract Genetically encoded biosensors have accelerated biological discovery, however many important targets such as active Ras (Ras-GTP) are difficult to sense as strategies to match a sensor’s sensitivity to the physiological range of target are lacking. Here, we use computational protein design to generate and optimize intracellular sensors of Ras activity ( LOCKR -based S ensor for Ras activity: Ras-LOCKR-S) and proximity labelers of the signaling environment of Ras ( LOCKR -based, Ras activity-dependent P roximity L abeler: Ras-LOCKR-PL). We demonstrate that our tools can measure endogenous Ras activity and environment at subcellular resolution. We illustrate the application of these tools by using them to identify Ras effectors, notably Src-Associated in Mitosis 68 kDa protein (SAM68), enriched in oncogenic EML4-Alk granules. Localizing these sensors to these granules revealed that SAM68 enhances Ras activity specifically at the granules, and SAM68 inhibition sensitizes EML4-Alk-driven cancer cells to existing drug therapies, suggesting a possible therapeutic strategy.