Ebola virus disease is a deadly pathogenic disease with a fatality rate of 25–90 % as recorded in previous outbreaks. The Ebola Virus glycoprotein (EBOV-GP) plays a crucial role in the entry of viruses into human cells, making it an interesting target for therapeutic discovery. Therefore, inhibiting this protein can directly limit the virus replication and disease progression at an early stage of infection. The present study focuses on the design of novel potent EBOV-GP inhibitors using multiple computational techniques. In this context, two QSAR models were built from a set of 86 amodiaquine derivatives as anti-EBOV-GP using Monte Carlo and genetic algorithm multiple linear regression methods. Both models confirmed their predictive performance with satisfactory statistical parameters of the validation (R2 = 0.9129; Q2 = 0.8848 for the CORAL model and R2 = 0.8848; Q2 = 0.8148 for the GA-MLR model). From the outputs of the CORAL model, the structural fragments responsible for increasing and decreasing the inhibition activity were extracted and interpreted. This molecular information was used to design 26 new potentially safe and active candidate drugs. Molecular docking and dynamics simulations have affirmed the efficacy of the designed compounds. Specifically, compounds D2 (pIC50_coral = 7.12; pIC50_GA-MLR = 7.07), D3 (pIC50_coral = 7.83; pIC50_GA-MLR = 7.10), and D5 (pIC50_coral = 7.26; pIC50_GA-MLR = 7.55) displayed notable predicted inhibitory activity, according to both models. These compounds also exhibited conformational and structural stability, as well as a favorable binding profile. Furthermore, these potential drug candidates were found to be non-toxicity and have acceptable pharmacological properties.

Drug design of new anti-EBOV inhibitors: QSAR, homology modeling, molecular docking and molecular dynamics studies