Abstract 2982: Utility of S. cerevisiae genetic interactions in the mechanistic validation and therapeutic potential of highly conserved targets for drug discovery

Abstract

Abstract Use of chemo-genetic interaction mediated synthetic lethality represents a strategy for mechanistic validation of targets and gene-based patient stratification for clinical development. Given that core pathways supporting cancer development and progression (cell cycle regulation, genome integrity, metabolism) are highly conserved, orthogonal model organisms represent powerful systems in which to identify synthetic lethal pairs for highly conserved human drug targets. BPM42522 is an enzyme in the ubiquitin proteasome system whose anti-cancer potential has been validated through genetic and pharmacologic modulation. The S. cerevisiae homolog of BPM42522 (yBPM42522) is a critical cell cycle-regulatory gene with 273 genetic interactions affecting fitness (Costanzo 2016). Negative genetic interactors regulate cellular processes validated for therapeutic intervention in oncology such as DNA replication and repair, protein turnover, and mitosis. Thus, mapping yBPM42522 negative genetic interactors to their human homologs and curating tumors in which they are altered may aid in identification of clinical contexts more susceptible to pharmacologic inhibition of BPM42522. To this end, conservation analysis of genetic interactors of yBPM42522 was performed, and coding mutations and deletion events were then characterized using The Cancer Genome Atlas (TCGA) tumor data. The most frequently altered genetic interactors across human tumors were reviewed for the direction and strength of the genetic interaction between their yeast homologs and yBPM42522, and mutations were curated by their likelihood to result in a loss of function. Those with strong negative genetic interactions with yBPM42522 and frequent mutation or deletion in its human homolog were prioritized. The resulting candidates included tumor suppressors (FBXW7, NF1), cell cycle regulators (CCNB1, CCNB3, BUB1B), and the NIMA kinase and GTPase-activating protein families (NEK3, NEK4, RASA1, RASAL2). TCGA database was used to determine if alterations in these candidate genes were prevalent in specific tumor types and whether they co-occurred with alterations in established cancer driver genes. Several candidate genetic interactors identified were frequently mutated or deleted in specific tumor types including uterine carcinosarcoma, uterine corpus endometrial carcinoma, ovarian serous cystadenocarcinoma, and mesothelioma. Moreover, these alterations were mutually exclusive with 96 reported cancer driver genes. Validation of the candidate genetic interactors for synthetic lethality with BPM42522 and their role in specific cancer types highlights the approach for rapid identification of synthetic lethality and its potential use to stratify patient populations most likely to benefit from therapeutic agents targeting highly conserved drug targets. Citation Format: Anne R. Diers, Kris Richardson, Leonardo O. Rodrigues, Rangaprasad Sarangarajan, Niven R. Narain, Jennifer A. Benanti, Stephane Gesta. Utility of S. cerevisiae genetic interactions in the mechanistic validation and therapeutic potential of highly conserved targets for drug discovery [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2982.