Discovery and SAR Study of Boronic Acid-Based Selective PDE3B Inhibitors from a Novel DNA-Encoded Library.

Human genetic evidence shows that PDE3B is associated with metabolic and dyslipidemia phenotypes. A number of PDE3 family selective inhibitors have been approved by the FDA for various indications; however, given the undesirable proarrhythmic effects in the heart, selectivity for PDE3B inhibition over closely related family members (such as PDE3A; 48% identity) is a critical consideration for development of PDE3B therapeutics. Selectivity for PDE3B over PDE3A may be achieved in a variety of ways, including properties intrinsic to the compound or tissue-selective targeting. The high (>95%) active site homology between PDE3A and B represents a massive obstacle for obtaining selectivity at the active site; however, utilization of libraries with high molecular diversity in high throughput screens may uncover selective chemical matter. Herein, we employed a DNA-encoded library screen to identify PDE3B-selective inhibitors and identified potent and selective boronic acid compounds bound at the active site.

Building Block-Based Binding Predictions for DNA-Encoded Libraries.

DNA-encoded libraries (DELs) provide the means to make and screen millions of diverse compounds against a target of interest in a single experiment. However, despite producing large volumes of binding data at a relatively low cost, the DEL selection process is susceptible to noise, necessitating computational follow-up to increase signal-to-noise ratios. In this work, we present a set of informatics tools to employ data from prior DEL screen(s) to gain information about which building blocks are most likely to be productive when designing new DELs for the same target. We demonstrate that similar building blocks have similar probabilities of forming compounds that bind. We then build a model from the inference that the combined behavior of individual building blocks is predictive of whether an overall compound binds. We illustrate our approach on a set of three-cycle OpenDEL libraries screened against soluble epoxide hydrolase (sEH) and report performance of more than an order of magnitude greater than random guessing on a holdout set, demonstrating that our model can serve as a baseline for comparison against other machine learning models on DEL data. Lastly, we provide a discussion on how we believe this informatics workflow could be applied to benefit researchers in their specific DEL campaigns.