Identification of promising anti-EBOV inhibitors: de novo drug design, molecular docking and molecular dynamics studies

Eslam A R Mohamed; Sayed F Abdelwahab; Ahmad M Alqaisi; Amaal Mohammed Salih Nasr; Heba Ali Hassan

doi:10.1098/rsos.220369

Identification of promising anti-EBOV inhibitors: de novo drug design, molecular docking and molecular dynamics studies

R Soc Open Sci. 2022 Sep 28;9(9):220369. doi: 10.1098/rsos.220369. eCollection 2022 Sep.

Authors

Eslam A R Mohamed¹, Sayed F Abdelwahab², Ahmad M Alqaisi³, Amaal Mohammed Salih Nasr⁴, Heba Ali Hassan⁵

Affiliations

¹ Department of Chemistry, Faculty of Science, Minia University, Minia 61511, Egypt.
² Department of Pharmaceutics and Industrial Pharmacy, College of Pharmacy, Taif University, PO Box 11099, Taif 21944, Saudi Arabia.
³ Chemistry Department, University of Jordan, Amman 11942, Jordan.
⁴ Department of Chemistry, Faculty of Science, Universiti Putra Malaysia, 43400 Serdang, Malaysia.
⁵ Department of Pharmacognosy, Faculty of Pharmacy, Sohag University, Sohag 82524, Egypt.

Abstract

The Ebola virus (EBOV) outbreak was recorded as the largest in history and caused many fatalities. As seen in previous studies, drug repurposing and database filtration were the two major pathways to searching for potent compounds against EBOV. In this study, a deep learning (DL) approach via the LigDream tool was employed to obtain novel and effective anti-EBOV inhibitors. Based on the galidesivir (BCX4430) chemical structure, 100 compounds were collected and inspected using various in silico approaches. Results from the molecular docking study indicated that mol1_069 and mol1_092 were the best two potent compounds with a docking score of -7.1 kcal mol^-1 and -7.0 kcal mol^-1, respectively. Molecular dynamics simulations, in addition to binding energy calculations, were conducted over 100 ns. Both compounds exhibited lower binding energies than BCX4430. Furthermore, compared with BCX4430 (%Absorption = 60.6%), mol1_069 and mol1_092 scored higher values of % Absorption equal to 68.1% and 63.7%, respectively. The current data point to the importance of using DL in the drug design process instead of conventional methods such as drug repurposing or database filtration. In conclusion, mol1_069 and mol1_092 are promising anti-EBOV drug candidates that require further in vitro and in vivo investigations.

Keywords: Ebola virus; LigDream; deep learning; molecular docking; molecular dynamics.

Associated data

figshare/10.6084/m9.figshare.c.6197378