Activity-Based DNA-Encoded Library Technology

Project Summary The vast majority of the human proteome is considered “undruggable.” Undruggable proteins may be difficult to express, lack surface binding clefts, do not have corresponding activity assays, or some combination thereof. This concept is symptomatic of a major liability of contemporary drug discovery, which requires significant

investment to generate and scale up protein expression or cell culture and engineering an activity assay for every new target. It may be possible to bypass these bottlenecks by directly targeting translation intermediates, or “ribosome nascent chains” (RNCs), with small molecules that selectively inhibit protein synthesis by

interacting with an RNC and stalling translation. RNCs represent a vast source of new drug targets that do not follow the rules of druggability, but high-throughput screens for RNC-targeting “Selective Terminators of Protein Synthesis” (SToPS) have been roundly unsuccessful due to the limited scope of structures in standard

compound screening decks. During the previously funded project, our instrumentation and systems engineering laboratory developed solid-phase DNA-encoded library (DEL) synthesis methods and microfluidic DEL screening technology that collectively enabled unprecedented activity-based screens on these large

collections of novel chemical matter. We demonstrated that this platform can efficiently search DELs of drug-like small molecules to identify novel bioactive molecules for several clinically relevant drug targets. The proposed MIRA program will leverage our activity-based DEL screening capabilities to establish a SToPS

discovery platform through two parallel technology development initiatives. The first is a synthetic biology-driven microfluidic droplet-scale in vitro translation-based approach to identifying small molecule SToPS of a specific target RNC. The second is a polymer/tissue culture engineering approach that will explore

cellular assays of translation stalling, the screening format that identified the original examples of SToPS targeting the hypercholesterolemia-associated protein, PCSK9. Both approaches will benefit from DEL-based chemical diversity, which can be designed to explore chemical space known to elicit ribosome binding and

selective translation stalling. Cellular DEL screening technology will ensure that screening hits are cell active, and more broadly will deliver a long-sought screening modality to the field of drug discovery. Following proof-of-concept SToPS screens, we will develop computational workflows that mine publicly available

ribosome profiling data sets to predict candidate stall sites for SToPS screening, tackling CCR5 (anti-HIV) and the bacterial signal sequence as examples of undruggable targets. We envision a completely plug-and-play chemical probe discovery strategy for translating human genome sequence directly into SToPS as chemical

probes, thereby fulfilling the original vision of the Human Genome Project and eliminating “undruggable” from the drug discovery lexicon.