Properties Governing Native State Entanglements and Relationships to Protein Function.

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• Non-covalent lasso entanglements are prevalent structural motifs in globular proteins.
• Misfolding of these entanglements is associated with various biological consequences.

Purpose of the Study:

• To characterize the structural and physicochemical properties of protein lasso entanglements.
• To investigate sequence biases, functional site correlations, and universal features of these motifs across different species (E. coli, S. cerevisiae, H. sapiens).

Main Methods:

• Analysis of structural and sequence properties of lasso entanglement components.
• Correlation analysis between crossing residues and functional sites.
• Application of machine learning models to identify predictive features.

Main Results:

• Crossing residues are significantly more likely to be beta-strands and are often surrounded by hydrophobic sequences (Val, Ile, Phe).
• Enrichment of crossing residues at enzyme active sites, small molecule binding sites, and metal binding sites (H. sapiens).
• RNA-binding residues are enriched in entanglement components, while protein-binding sites show depletion.
• Machine learning models achieved AUROC scores of 0.8 in predicting entanglements across species.

Conclusions:

• Native lasso entanglements play a direct role in specific protein functions.
• Strong secondary structure and sequence preferences are identified within native entanglements.
• These findings provide insights into the functional significance and structural determinants of protein lasso entanglements.