Adequate prediction for inhibitor affinity of Aβ₄₀ protofibril using the linear interaction energy method

Area of Science:

• Computational chemistry
• Neuroscience
• Drug discovery

Background:

• Alzheimer's disease (AD) is characterized by amyloid-beta (Aβ) oligomers and fibrils.
• Developing effective inhibitors for Aβ aggregation is crucial for AD treatment.
• Accurate and efficient methods for predicting ligand binding affinity to Aβ are needed.

Purpose of the Study:

• To evaluate the Linear Interaction Energy (LIE) approach for predicting binding free energies between ligands and Aβ peptides.
• To compare the accuracy and computational cost of LIE with Free Energy Perturbation (FEP) and Molecular Mechanic/Poisson-Boltzmann Surface Area (MM/PBSA) methods.
• To identify the dominant interaction types contributing to ligand-Aβ binding.

Main Methods:

• Calculated binding free energies for 30 ligands interacting with Aβ_11-40 peptides using the LIE approach.
• Validated LIE results against experimental data (R = 0.79).
• Compared LIE with FEP (R = 0.72) and MM/PBSA (R = 0.27) methods for accuracy and computational efficiency.

Main Results:

• The LIE method demonstrated good correlation with experimental binding affinity data.
• Van der Waals interactions were found to be more significant than electrostatic interactions in ligand-Aβ complexes.
• LIE proved to be substantially less time-consuming than FEP and MM/PBSA.

Conclusions:

• The LIE approach provides an accurate and computationally efficient method for predicting ligand binding affinity to Aβ peptides.
• LIE facilitates rapid screening of potential drug candidates for Alzheimer's disease.
• This method can accelerate the design and development of novel Aβ-targeting ligands.