tetano
Editor, Senior Moderator
Heliyon
. 2022 Aug 8;8(8):e10101.
doi: 10.1016/j.heliyon.2022.e10101. eCollection 2022 Aug.
In-silico modelling studies of 5-benzyl-4-thiazolinone derivatives as influenza neuraminidase inhibitors via 2D-QSAR, 3D-QSAR, molecular docking, and ADMET predictions
Mustapha Abdullahi[SUP] 1 2 [/SUP], Adamu Uzairu[SUP] 1 [/SUP], Gideon Adamu Shallangwa[SUP] 1 [/SUP], Paul Andrew Mamza[SUP] 1 [/SUP], Muhammad Tukur Ibrahim[SUP] 1 [/SUP]
Affiliations
Abstract
Influenza virus disease is one of the most infectious diseases responsible for many human deaths, and the high mutability of the virus causes drug resistance effects in recent times. As such, it became necessary to explore more inhibitors that could avert future influenza pandemics. The present research utilized some in-silico modelling concepts such as 2D-QSAR, 3D-QSAR, molecular docking simulation, and ADMET predictions on some 5-benzyl-4-thiazolinone derivatives as influenza neuraminidase (NA) inhibitors. The 2D-QSAR modelling results revealed GFA-MLR ( R2train
=0.8414, Q[SUP]2[/SUP] = 0.7680) and GFA-ANN ( R2train =0.8754, Q[SUP]2[/SUP] = 0.8753) models with the most relevant descriptors (MATS3i, SpMax5_Bhe, minsOH and VE3_D) for predicting the inhibitory activities of the molecules which has passed the global criteria of accepting QSAR models. The results of the 3D-QSAR modelling results showed that CoMFA_ES ( R2train =0.9030, Q[SUP]2[/SUP] = 0.5390) and CoMSIA_EA ( R2train
=0.880, Q[SUP]2[/SUP] = 0.547) models are having good predicting ability among other developed models. The molecules were virtually screened via molecular docking simulation with the active site of NA protein receptor (pH1N1) which confirms their resilient potency when compared with zanamivir standard drug. Molecule 11 as the most potent molecule formed more H-bond interactions with the key residues such as TRP178, ARG152, ARG292, ARG371, and TYR406 that triggered the catalytic reactions for NA inhibition. Furthermore, six (6) molecules (9, 10, 11, 17, 22, and 31) with relatively high inhibitory activities and docking scores were identified as the possible leads for in-silico exploration of novel NA inhibitors. The drug-likeness and ADMET predictions of the lead molecules revealed non-violation of Lipinski's rule and good pharmacokinetic profiles respectively, which are important guidelines for rational drug design. Hence, the outcome of this study overlaid a solid foundation for the in-silico design and exploration of novel NA inhibitors with improved potency.
Keywords: Binding energy; Influenza; Modelling; Neuraminidase; Residual interaction.
. 2022 Aug 8;8(8):e10101.
doi: 10.1016/j.heliyon.2022.e10101. eCollection 2022 Aug.
In-silico modelling studies of 5-benzyl-4-thiazolinone derivatives as influenza neuraminidase inhibitors via 2D-QSAR, 3D-QSAR, molecular docking, and ADMET predictions
Mustapha Abdullahi[SUP] 1 2 [/SUP], Adamu Uzairu[SUP] 1 [/SUP], Gideon Adamu Shallangwa[SUP] 1 [/SUP], Paul Andrew Mamza[SUP] 1 [/SUP], Muhammad Tukur Ibrahim[SUP] 1 [/SUP]
Affiliations
- PMID: 36016519
- PMCID: PMC9396554
- DOI: 10.1016/j.heliyon.2022.e10101
Abstract
Influenza virus disease is one of the most infectious diseases responsible for many human deaths, and the high mutability of the virus causes drug resistance effects in recent times. As such, it became necessary to explore more inhibitors that could avert future influenza pandemics. The present research utilized some in-silico modelling concepts such as 2D-QSAR, 3D-QSAR, molecular docking simulation, and ADMET predictions on some 5-benzyl-4-thiazolinone derivatives as influenza neuraminidase (NA) inhibitors. The 2D-QSAR modelling results revealed GFA-MLR ( R2train
=0.8414, Q[SUP]2[/SUP] = 0.7680) and GFA-ANN ( R2train =0.8754, Q[SUP]2[/SUP] = 0.8753) models with the most relevant descriptors (MATS3i, SpMax5_Bhe, minsOH and VE3_D) for predicting the inhibitory activities of the molecules which has passed the global criteria of accepting QSAR models. The results of the 3D-QSAR modelling results showed that CoMFA_ES ( R2train =0.9030, Q[SUP]2[/SUP] = 0.5390) and CoMSIA_EA ( R2train
=0.880, Q[SUP]2[/SUP] = 0.547) models are having good predicting ability among other developed models. The molecules were virtually screened via molecular docking simulation with the active site of NA protein receptor (pH1N1) which confirms their resilient potency when compared with zanamivir standard drug. Molecule 11 as the most potent molecule formed more H-bond interactions with the key residues such as TRP178, ARG152, ARG292, ARG371, and TYR406 that triggered the catalytic reactions for NA inhibition. Furthermore, six (6) molecules (9, 10, 11, 17, 22, and 31) with relatively high inhibitory activities and docking scores were identified as the possible leads for in-silico exploration of novel NA inhibitors. The drug-likeness and ADMET predictions of the lead molecules revealed non-violation of Lipinski's rule and good pharmacokinetic profiles respectively, which are important guidelines for rational drug design. Hence, the outcome of this study overlaid a solid foundation for the in-silico design and exploration of novel NA inhibitors with improved potency.
Keywords: Binding energy; Influenza; Modelling; Neuraminidase; Residual interaction.