Search

Serotonin (5-hydroxytryptamine, 5-HT) acts via 13 different receptors in humans. Of these receptor subtypes, all but 5-HT 1e R have confirmed roles in native tissue and are validated drug targets. Despite 5-HT 1e R's therapeutic potential and plausible druggability, the mechanisms of its activation remain elusive. To illuminate 5-HT 1e R's pharmacology in relation to the highly homologous 5-HT 1F R, we screened a library of aminergic receptor ligands at both receptors and observe 5-HT1e/1FR agonism by multicyclic drugs described as pan-antagonists at 5-HT receptors. Potent agonism by tetracyclic antidepressants mianserin, setiptiline, and mirtazapine suggests a mechanism for their clinically observed anti-migraine properties. Using cryoEM and mutagenesis studies, we uncover and characterize unique agonist-like binding poses of mianserin and setiptiline at 5-HT 1e R distinct from similar drug scaffolds in inactive-state 5-HTR structures. Together with computational studies, our data suggest that these binding poses alongside receptor-specific allosteric coupling in 5-HT 1e R and 5-HT 1F R contribute to the agonist activity of these antidepressants., Competing Interests: Competing interests: The authors declare that they have no competing interests.

The opioid epidemic has cast a shadow over public health, necessitating immediate action to address its devastating consequences. To effectively combat this crisis, it is crucial to discover better opioid drugs with reduced addiction potential. Artificial intelligence-based and other machine learning tools, particularly deep learning models, have garnered significant attention in recent years for their potential to advance drug discovery. However, utilizing these tools poses challenges, especially when training samples are insufficient to achieve adequate prediction performance. In this study, we investigate the effectiveness of transfer learning using combined ligand-based and structure-based molecular descriptors from the entire opioid receptor (OR) subfamily in building robust deep learning models for enhanced bioactivity prediction of opioid ligands at each individual OR subtype. Our studies hold the potential to greatly advance opioid research by enabling the rapid identification of novel chemical probes with specific bioactivities, which can aid in the study of receptor function and contribute to the future development of improved opioid therapeutics.

Likely effective pharmacological interventions for the treatment of opioid addiction include attempts to attenuate brain reward deficits during periods of abstinence. Pharmacological blockade of the κ-opioid receptor (KOR) has been shown to abolish brain reward deficits in rodents during withdrawal, as well as to reduce the escalation of opioid use in rats with extended access to opioids. Although KOR antagonists represent promising candidates for the treatment of opioid addiction, very few potent selective KOR antagonists are known to date and most of them exhibit significant safety concerns. Here, we used a generative deep learning framework for the de novo design of chemotypes with putative KOR antagonistic activity. Molecules generated by models trained with this framework were prioritized for chemical synthesis based on their predicted optimal interactions with the receptor. Our models and proposed training protocol were experimentally validated by binding and functional assays.