Multiscale Model for Quantitative Prediction of Insulin Aggregation Nucleation Kinetics

Area of Science:

• Biophysics
• Computational Biology
• Biochemistry

Background:

• Protein aggregation is implicated in diseases like Alzheimer's and diabetes.
• Understanding the nucleation phase is critical for controlling aggregation.
• Insulin aggregation serves as a model for studying general protein aggregation mechanisms.

Purpose of the Study:

• To determine accurate kinetic parameters for protein aggregation nucleation.
• To develop a predictive model for aggregation kinetics.
• To gain mechanistic insights into the aggregation process.

Main Methods:

• Combined single-molecule and two-molecule explicit-solvent simulations.
• Employed structural bioinformatics and transition complex theory.
• Utilized a population balance model with derived kinetic parameters.

Main Results:

• Developed a method to account for aggregation-prone species heterogeneity.
• Identified a dominant binding mode in the kinetic pathway, similar to native protein association.
• Accurately predicted aggregation nucleation time over two orders of magnitude.
• Demonstrated the transferability of physically determined parameters, unlike empirical sets.

Conclusions:

• The developed simulation-based approach accurately predicts protein aggregation nucleation kinetics.
• Physically determined kinetic parameters offer mechanistic insights into aggregation.
• This method provides a superior alternative to empirical parameterization for predicting ligand effects and aggregation behavior.