De Novo Design of & kappa;-Opioid Receptor Antagonists Using a Generative Deep-Learning Framework

Likely effective pharmacological interventions for thetreatmentof opioid addiction include attempts to attenuate brain reward deficitsduring periods of abstinence. Pharmacological blockade of the & kappa;-opioidreceptor (KOR) has been shown to abolish brain reward deficits inrodents during withdrawal, as well as to reduce the escalation ofopioid use in rats with extended access to opioids. Although KOR antagonistsrepresent promising candidates for the treatment of opioid addiction,very few potent selective KOR antagonists are known to date and mostof them exhibit significant safety concerns. Here, we used a generativedeep-learning framework for the de novo design ofchemotypes with putative KOR antagonistic activity. Molecules generatedby models trained with this framework were prioritized for chemicalsynthesis based on their predicted optimal interactions with the receptor.Our models and proposed training protocol were experimentally validatedby binding and functional assays.