Abstract 456: An ultrahigh throughput phenotypic screening platform to discover novel inhibitors of methyl transferases

Abstract Discovering drug-like molecules for hard-to-target proteins remains a significant challenge. Methyltransferases (DNA, RNA, and protein) are among such targets. At Zafrens, we have developed an ultra-high-throughput discovery platform that leverages large arrays of isolated nano-wells and single-bead-single-compound DNA-encoded library (DELs) to uncover target-specific inhibitory impacts on cells. Using a combination of next-generation sequencing and spatially indexed optical imaging of biomarkers and cell paint probes, we successfully linked transcriptomic signatures and phenotypic responses to a combinatorial chemical library of over 15, 000 compounds, achieving approximately fivefold hit coverage. Incorporation of known inhibitors of a methyltransferase into the chemical library revealed that both transcriptome-driven clustering and machine learning-driven linkage of compound SMILES with cell phenotype could successfully identify reproducible signatures that were aligned well with public databases, including expected methyltransferase-dependent genes such as MTF1. The robustness and consistency of our platform have allowed us to identify several novel compounds that mirrored the transcriptomic and phenotypic profiles of the control compounds. Subsequent biochemical assays confirmed that many of these were bona fide inhibitors of the target methyltransferase. Among these, ZF165 emerged as a promising candidate due to its high hit frequency across multiple screens and across two cell lines. ZF165 was then resynthesized and further validated by both immunofluorescence- and western blot analyses for its impact on biomarker responses. Bulk transcriptomic mRNA sequencing analysis confirmed common profiles between ZF165 and a control methyltransferase inhibitor. Furthermore, ZF165 demonstrated initial anti-tumor cell activity with a potency comparable to known pre-clinical compounds, further supporting its potential as a lead molecule. With ZF165 as the new series lead, we designed a secondary DEL library featuring novel warheads and improved physicochemical properties. The secondary library, ZEL29, was synthesized and screened, yielding new hits as additional (or) providing SAR, with enhanced potency and better drug-like properties relative to ZF165. These results support the utility of our platform in both the initial hit discovery and subsequent hit optimization toward lead compounds. Validation-design cycles for lead optimization are currently ongoing. By coupling this powerful screening platform with robust data analytics, we aim to revolutionize the discovery of novel therapeutics to address other challenging drug targets. Citation Format: Nathan G. Hedrick, Maina Ndungu, Ivy Nai-Jung Hung, Devin K. Porter, Elliot Imler, Gopi Chandran RaviChandran, Aurelien Lauerre, Chris M. Glinkerman, Martin Smrcina, Mark R. Hansen, Allan Nojadera, Alexandra Cole, Cindy Huynh, Michael Miller, Dmitri rozanov, Logan Van Meter, Nicholas Sm-Soon, Steven Brown, Yi Zhang, Vikas K. Goel. An ultrahigh throughput phenotypic screening platform to discover novel inhibitors of methyl transferases [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 456.