Prenylated proteins play key roles in several human diseases including cancer, atherosclerosis and Alzheimer’s disease. KRAS4b, which is frequently mutated in pancreatic, colon and lung cancers, is processed by farnesylation, proteolytic cleavage and carboxymethylation at the C-terminus. Plasma membrane localization of KRAS4b requires this processing as does KRAS4b-dependent RAF kinase activation. Previous attempts to produce modified KRAS have relied on protein engineering approaches or in vitro farnesylation of bacterially expressed KRAS protein. The proteins produced by these methods do not accurately replicate the mature KRAS protein found in mammalian cells and the protein yield is typically low. We describe a protocol that yields 5–10 mg/L highly purified, farnesylated, and methylated KRAS4b from insect cells. Farnesylated and methylated KRAS4b is fully active in hydrolyzing GTP, binds RAF-RBD on lipid Nanodiscs and interacts with the known farnesyl-binding protein PDEδ.

The RAS family of GTPases (HRAS, KRAS, NRAS) continue to present formidable challenges to the design of compounds that successfully and specifically target oncogenic RAS mutant populations in cancer. Although recent advances have led to the successful development of inhibitors specific to KRAS G12C (such as Sotorasib [Amgen] and Adagrasib [Mirati Therapeutics]), ongoing efforts seek to develop additional targeted inhibitors for oncogenic mutants of KRAS. Development of new KRAS targeted inhibitors requires the understanding of how a mutation impacts the accessibility of key residues within the protein, along with how each candidate compound interacts with populations of mutant and wild-type proteins. Proteomic analyses, typically comprising peptide-based methods to confirm compound localization or quantify binding kinetics, represent a key validation step for each phase of the inhibitor development process. However, these methods may not always provide a clear picture of how effectively the compound binds to the protein, how specific the compound is to the target residue, and how concentration or incubation time may impact these factors. To address this, we have developed a novel assay to evaluate target engagement in vitro employing top-down proteomics, which entails the analysis of intact protein molecules and their modified forms (proteoforms). By analyzing intact labeled or compound-treated recombinant KRAS4B proteins, we can directly visualize and precisely characterize each species within a sample. This method allows us to easily detect nonspecific binding, identify targeting of incorrect residues, and characterize unanticipated compound reaction products within a short LC-MS run, all facilitated by targeted MS2 fragmentation. Moreover, we have developed complementary methods for RAS proteoform quantitation, allowing for the evaluation of compound engagement to all proteoforms with greater accuracy than traditional peptide-based methods. We first performed a proof-of-concept experiment using a panel of KRAS4B G12D_C118S proteins with selected residues modified to cysteine to evaluate reactivity to maleimide-biotin over time. We then applied our optimized top-down assay to determine the engagement of targeted covalent inhibitors Sotorasib and Adagrasib with KRAS WT, G12C, or G13C in both the active (GppNHp) and inactive (GDPβS) states. Finally, we selected three compounds from a large-scale screen against a UCSF Small Molecule Discovery Center compound library and evaluated each for specificity to KRAS4B G13C over KRAS4B WT and KRAS4B G12C. Our results show the concentration dependence of target engagement and directly map compound binding location(s) without disrupting the protein’s primary structure. Moreover, our ability to target discrete compound additions for top-down MS2 fragmentation provides context and order of preferential labeling of KRAS4B. Our hope is that this methodology may help accelerate the identification of new successful targeted inhibitors for KRAS and other RAS isoforms by the greater RAS community. Citation Format: Robert A. D'Ippolito, Caroline J. DeHart, Dana Rabara, Maria Abreu Blanco, Emily Alberico, Nitya Ramakrishnan, Stephanie Widmeyer, Simon Messing, David Turner, Michelle Arkin, Anna Maciag, Andrew Stephen, Dominic Esposito, Frank McCormick, Dwight V. Nissley. A top-down proteomic assay to evaluate KRAS4B-compound engagement [abstract]. In: Proceedings of the AACR Special Conference: Targeting RAS; 2023 Mar 5-8; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Res 2023;21(5_Suppl):Abstract nr A018.