Highly tumor-selective pan RAS inactivators for targeting oncogenic RAS-driven cancers

Abstract Cancer driver mutations in the RAS-RAF-MEK-ERK pathway occur in 46% of all human cancers. This has inspired the successful development of many small molecule inhibitors of BRAF and MEK. However, development of RAS small molecule inhibitors has not been as successful, in large part due to the lack of a sufficiently large and deep hydrophobic pocket in RAS proteins to allow for small molecule binding. Thus, oncogenic RAS proteins have long been considered as “undruggable” until the recent discovery and FDA- approval of the allele-specific covalent inhibitors Sotorasib and Adagrasib for treating non-small cell lung cancer (NSCLC) with the KRASG12C mutation. However, treated patients inevitably developed drug resistance, mostly due to other secondary KRAS mutations (G12D, G12R, G12V, G13D, or Q61H) that are not targeted by these inhibitors. KRASG12D allele-specific inhibitors are under development at preclinical stages, but similar drug resistance mechanisms are expected to limit these allele-specific inhibitors. Notably, in this regard, many bacterial pathogens have evolved potent protein toxins to disrupt the RAS-RAF-MEK-ERK pathway. Fortunately, these naturally designed potent toxins, such as anthrax toxin and DUF5, can be structurally modified to achieve high tumor specificity. DUF5 is a toxin effector domain (amino acids Glu3581-Gln4089) of MARTX (multifunctional-autoprocessing repeats-in-toxin) from Vibrio vulnificus. DUF5 was recently identified to be a potent and specific endopeptidase that cleaves and inactivates all oncogenic RAS proteins (KRAS, NRAS, and HRAS). To address the above unmet medical needs, in this application, we will develop a potent and highly tumor-selective RAS inactivator by engineering the anthrax toxin protein delivery system and DUF5. This tumor- selective RAS inactivator can specifically bind to the major anthrax toxin receptor CMG2 on tumor cells, and strictly relies on the simultaneous presence of two distinct tumor-associated proteases, MMPs and urokinase, to gain entry specifically into tumor cells where it proteolytically inactivate RAS signaling, achieving potent selective tumor targeting. Our preliminary data show that our highly tumor-selective pan RAS inactivator displays potent cytotoxicity to cancer cells whose survival depends on oncogenic RAS signaling. In this application, we will comprehensively characterize and evaluate this novel agent's anti-tumor activity using a wide range of tumor models. We propose therefore to develop a tumor-selective, safe, and highly potent pan-RAS inactivator that has high clinical utility for targeting tumors with various oncogenic RAS mutations.