A Structure Guided QSAR: A Rapid and Accurate technique to predict IC50: A Case Study

Area of Science:

• Biochemistry
• Computational Chemistry
• Medicinal Chemistry

Background:

• Leukotriene A4 hydrolase (LTA4H) is a key enzyme in inflammatory pathways.
• Understanding enzyme active site interactions is crucial for drug discovery.
• The role of structured water molecules in enzyme-ligand binding is often overlooked.

Purpose of the Study:

• To investigate the influence of structured water molecules in the LTA4H active site on docking experiments.
• To develop and validate quantitative structure-activity relationship (QSAR) models for LTA4H inhibitors.
• To identify novel LTA4H inhibitors using in-silico methods.

Main Methods:

• Docking experiments were performed with and without considering bound water molecules.
• Two structure-guided, bi-parametric linear regression QSAR models were constructed using Glide and FlexX docking scores.
• Models were validated using test and validated sets, and subsequently used to predict activities of new compounds.

Main Results:

• QSAR models incorporating bound waters showed improved predictive power (Q(2) = 0.73) compared to those without (Q(2) = 0.67).
• In-silico screening of 409 compounds yielded 39 promising candidates.
• Correlation between predicted and experimental IC50 values for synthesized compounds showed a high R(2) of 0.87.

Conclusions:

• Structured water molecules play a significant role in LTA4H ligand binding and should be included in computational studies.
• The developed QSAR models are effective for predicting LTA4H inhibitory activity.
• This study successfully identified novel, potent LTA4H inhibitors through an integrated in-silico and experimental approach.