Fingerprint-Based Machine Learning Approach to Identify Potent and Selective 5-HT2BR Ligands.

Krzysztof Rataj,José Brea,Andrzej Bojarski,María Loza,György Keserű,Ádám Kelemen

doi:10.3390/molecules23051137

Abstract

The identification of subtype-selective GPCR (G-protein coupled receptor) ligands is a challenging task. In this study, we developed a computational protocol to find compounds with 5-HT2BR versus 5-HT1BR selectivity. Our approach employs the hierarchical combination of machine learning methods, docking, and multiple scoring methods. First, we applied machine learning tools to filter a large database of druglike compounds by the new Neighbouring Substructures Fingerprint (NSFP). This two-dimensional fingerprint contains information on the connectivity of the substructural features of a compound. Preselected subsets of the database were then subjected to docking calculations. The main indicators of compounds’ selectivity were their different interactions with the secondary binding pockets of both target proteins, while binding modes within the orthosteric binding pocket were preserved. The combined methodology of ligand-based and structure-based methods was validated prospectively, resulting in the identification of hits with nanomolar affinity and ten-fold to ten thousand-fold selectivities.

Highlights

There is an increasing need of efficacious CNS drugs with reduced off-target activity that often connected to notable subtype selectivity
We built a Neighbouring Substructures Fingerprint (NSFP) fingerprint-based machine learning model using in vitro activity data available for human 5-HT1B R and 5-HT2B R receptors in ChEMBL [34]
Since our design concept was based on the role of the secondary binding pocket in selectivity, only compounds with 22 or more heavy atoms were considered, as they are more likely to bind to both the orthosteric binding pocket (OBP) and SBP of

Summary

Introduction

There is an increasing need of efficacious CNS (central nervous system) drugs with reduced off-target activity that often connected to notable subtype selectivity. A significant number of drugs was introduced to the market targeting different 5-HT subtypes ranging from 5-HT1–7 R [2]. The continuously growing number of available relevant class A GPCR (G-protein coupled receptor) X-ray structures (bovine rhodopsin [3,4], β2 AR [5], 5-HT1B R [6], 5-HT2B R [7,8,9], D3 R [10], M2 R [11], etc.) revealed certain important structural motifs of molecular recognition and ligand binding, including determinants of selectivity across the certain subtypes. GPCRs consist of seven transmembrane helices (and the additional intramembrane helix 8) connected in a bunch through

Methods

Results

Conclusion