Enhancement of protein thermostability by three consecutive mutations using loop-walking method and machine learning.

Area of Science:

• Biochemistry
• Protein Engineering
• Computational Biology

Background:

• Protein thermostability is crucial for industrial applications.
• Enzyme engineering often involves modifying protein structures to enhance stability.
• Directed evolution and computational methods are key strategies in protein engineering.

Purpose of the Study:

• To develop and apply a novel method, termed the "loop-walking method", to improve the thermostability of Burkholderia cepacia lipase.
• To identify specific loops and mutations that significantly impact protein thermal stability.
• To integrate machine learning with experimental screening for efficient identification of highly thermostable enzyme variants.

Main Methods:

• Random mutagenesis was performed on 12 loops of Burkholderia cepacia lipase, creating 12 mutant libraries.
• High-throughput screening identified L7 as a critical "hot-spot" loop affecting thermostability.
• Machine learning models were trained on experimental data to predict thermostability and guide the selection of promising mutants for further assay.

Main Results:

• The L7 library yielded a mutant (P233G/L234E/V235M) with improved thermostability.
• A computational model predicted thermostability potentials for thousands of potential mutants.
• A top-ranked mutant (P233D/L234P/V235S) exhibited remarkable thermostability, retaining 66% activity after heat treatment, significantly outperforming wild-type (5%) and the initial L7 mutant (35%).

Conclusions:

• The "loop-walking method" combined with machine learning is an effective strategy for enhancing protein thermostability.
• Specific loop regions, like L7, can be targeted for significant improvements in enzyme thermal resistance.
• The identified P233D/L234P/V235S mutant represents a substantial advancement in Burkholderia cepacia lipase thermostability for potential biotechnological applications.