Development of novel ALOX15 inhibitors combining dual machine learning filtering and fragment substitution optimisation approaches, molecular docking and dynamic simulation methods

Abstract

The oxidation of unsaturated lipids, facilitated by the enzyme Arachidonic acid 15-lipoxygenase (ALOX15), is an essential element in the development of ferroptosis. This study combined a dual-score exclusion strategy with high-throughput virtual screening, naive Bayesian and recursive partitioning machine learning models, the already established ALOX15 inhibitor i472, and a docking-based fragment substitution optimisation approach to identify potential ALOX15 inhibitors, ultimately leading to the discovery of three FDA-approved drugs that demonstrate optimal inhibitory potential against ALOX15. Through fragment substitution-based optimisation, seven new inhibitor structures have been developed. To evaluate their practicality, ADMET predictions and molecular dynamics simulations were performed. In conclusion, the compounds found in this study provide a novel approach to combat conditions related to ferroptosis-related injury by inhibiting ALOX15.

Keywords:

• Ferroptosis
• ALOX15
• virtual screening
• machine learning
• molecular dynamic simulation

Acknowledgements

We thank the Public Service Platform of South China Sea for R&D Marine Biomedicine Resources for support.

Disclosure statement

The authors declare no competing interests.

Data availability

The data that support this study are available from the corresponding author upon reasonable.

Additional information

Funding

This research was funded by the Science and Technology Special Project of Zhanjiang (2022A01034); The Guangdong Provincial Department of Education Research Project (2022KTSCX); The Science and technology program of Guangdong Province (2023A1515010850); The Key Discipline Construction Project of Guangdong Medical University (4SG22004G, 2022KTSCX043); The National Natural Science Foundation of China (82370564)