Chemical screening studies have identified drug sensitivities across hundreds of cancer cell lines but most putative therapeutics fail to translate. Discovery and development of drug candidates in models that more accurately reflect nutrient availability in human biofluids may help in addressing this major challenge. Here we performed high-throughput screens in conventional versus Human Plasma-Like Medium (HPLM). Sets of conditional anticancer compounds span phases of clinical development and include non-oncology drugs. Among these, we characterize a unique dual- mechanism of action for brivudine, an agent otherwise approved for antiviral treatment. Using an integrative approach, we find that brivudine affects two independent targets in folate metabolism. We also traced conditional phenotypes for several drugs to the availability of nucleotide salvage pathway substrates and verified others for compounds that seemingly elicit off-target anticancer effects. Our findings establish generalizable strategies for exploiting conditional lethality in HPLM to reveal therapeutic candidates and mechanisms of action.
Forward genetic screens across hundreds of human cancer cell lines have been used to identify genetic dependencies of proliferating human cells and how these vary by genotype and cell lineage. However, while it is appreciated that environmental factors influence cell physiology, there has been little investigation into how nutrient availability impacts gene essentiality. Further, most such screens have been performed in vitro using culture media that poorly reflect the metabolic composition of human blood. Previously, we developed Human Plasma‐Like Medium (HPLM), a physiologic medium designed to more closely reflect the biochemical conditions in human blood. To determine how medium composition affects gene essentiality, we recently performed CRISPR‐based screens of K562 leukemia cells in HPLM versus RPMI, the medium historically used to culture human blood cells. Among the strongest scoring HPLM‐essential genes is Nicotinamide Adenine Dinucleotide Kinase (NADK), which encodes one of two human NAD kinases (NADK, cytosolic; NADK2, mitochondrial) that can catalyze the phosphorylation of NAD(H)to NADP(H). While most human cells co‐express both isozymes, only NADK was identified as HPLM‐essential, suggesting that endogenous NADK2 expression is not sufficient to complement NADK knockout in HPLM‐cultured cells. Interestingly, NADK has been annotated as an essential gene in less than 1% of over 1000 genome‐wide CRISPR‐based screens from the Broad Institute Cancer Dependency Map project. To begin to investigate the conditional CRISPR phenotype for NADK, we first engineered NADK‐knockout cells and then found that they showed a 40% growth defect versus control cells in HPLM relative to RPMI using a short‐term growth assay. Next, to test the hypothesis that cytosolic NAD kinase activity is required for the conditionally essential role of NADK, we performed similar short‐term growth assays with NADK‐knockout cells expressing cDNAs for wild‐type and cytosol‐restricted NADK2, as well as kinase‐dead NADK. These experiments revealed that neither kinase‐dead NADK nor NADK2 overexpression could rescue the growth defect of NADK‐knockout cells, but that cytosol‐restricted NADK2 could. Collectively, these results indicate that the conditionally essential role of NADK is linked to its cytosolic NAD kinase activity. Currently, we are applying systematic approaches in media engineering and metabolomics to identify the medium component(s) that contribute to the NADK gene‐nutrient interaction, and to gain insights into the non‐redundant, context‐dependent role of NADK. Lastly, through further conditional essentiality profiling, we identified NADK2as DMEM‐essential versus HPLM. Consistent with two recently published studies, we traced this effect to the differential availability of proline between HPLM and DMEM. Collectively, our results demonstrate the human NADKs fulfill distinct cell‐essential roles and form distinct gene‐nutrient interactions that vary with their subcellular localization.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.