Computer-guided design of novel nitrogen-based heterocyclic sphingosine-1-phosphate (S1P) activators as osteoanabolic agents.

Autor: Tangporncharoen R; Department of Clinical Microscopy, Faculty of Medical Technology, Mahidol University, Bangkok 10700, Thailand., Phanus-Umporn C; Department of Community Medical Technology, Faculty of Medical Technology, Mahidol University, Bangkok 10700, Thailand., Prachayasittikul S; Center for Research Innovation and Biomedical Informatics, Faculty of Medical Technology, Mahidol University, Bangkok 10700, Thailand., Nantasenamat C; Streamlit Inc., San Francisco, CA 94121, USA., Prachayasittikul V; Center for Research Innovation and Biomedical Informatics, Faculty of Medical Technology, Mahidol University, Bangkok 10700, Thailand., Supokawej A; Department of Clinical Microscopy, Faculty of Medical Technology, Mahidol University, Bangkok 10700, Thailand.
Jazyk: angličtina
Zdroj: EXCLI journal [EXCLI J] 2024 May 27; Vol. 23, pp. 818-832. Date of Electronic Publication: 2024 May 27 (Print Publication: 2024).
DOI: 10.17179/excli2024-7214
Abstrakt: Osteoanabolic agents, or drugs that promote bone formation, have gained considerable attention for osteoporosis management due to their curative and preventive potentials. Sphingosine-1-phosphate receptor 2 (S1PR2) is an attractive drug target, in which its activation leads to osteogenesis-promoting effect. Nitrogen-containing heterocyclic scaffolds (i.e., quinoxaline and indole) are promising pharmacophores possessing diverse bioactivities and were reported as S1PR2 activators. Quantitative structure-activity relationship (QSAR) modeling is a computational approach well-known as a fundamental tool for facilitating successful drug development. This study demonstrates the discovery of new S1PR2 activators using computational-driven rational design. Herein, an original dataset of nitrogen-containing S1PR2 activators was collected from ChEMBL database. The retrieved dataset was separated into two datasets according to their core scaffolds (i.e., quinoxaline and indole). QSAR modeling was performed using multiple linear regression (MLR) algorithm to successfully obtain two models with good predictive performance. The constructed models also revealed key properties playing essential roles for potent S1PR2 activation, such as Van der Waals volume (R2v+ and E3v), mass (MATS5m and Km), electronegativity (H3e), and number of 5-membered rings (nR05). Subsequently, the constructed models were further employed to guide rational design and predict S1PR2 activating effects of an additional set of 752 structurally modified compounds. Most of the modified compounds were predicted to have higher potency than their parents, and a set of promising potent newly designed compounds was highlighted. Additionally, drug-likeness prediction was performed to reveal that most of the newly designed compounds are druggable compounds with possibility for further development.
Competing Interests: There are no conflicts to declare.
(Copyright © 2024 Tangporncharoen et al.)
Databáze: MEDLINE
Externí odkaz: