Maximizing the native concentration and shelf life of protein: a multiobjective optimization to reduce aggregation

Area of Science:

• Biochemistry and Biophysics
• Protein Chemistry
• Computational Biology

Background:

• Protein aggregation reduces shelf life and therapeutic efficacy.
• Optimizing formulation parameters is crucial for protein stability.

Purpose of the Study:

• To maximize native protein concentration and shelf life of ASD.
• To determine optimal formulation conditions (pH, temperature, stabilizers) using computational methods.

Main Methods:

• Multiobjective optimization using Artificial Neural Network (ANN) and Genetic Algorithm (GA).
• Central Composite Design (CCD) for experimental data collection.
• Size exclusion chromatography and turbidity measurements for target variables.

Main Results:

• Achieved a nearly fourfold increase in shelf life (~196 h) at GA-optimized conditions.
• Maintained maximum relative area of the dimer peak (64.49%) at optimal conditions, indicating no aggregation.
• Identified storage temperature and glycerol concentration as key factors influencing aggregation.

Conclusions:

• Multiobjective optimization strategy effectively enhances protein shelf life and stability.
• ANN and GA provide a powerful tool for predicting and optimizing protein formulations.
• The methodology was also applied to optimize actinomycin D production.