Binding symmetry and surface flexibility mediate antibody self-association.

Area of Science:

• Biochemistry
• Structural Biology
• Biopharmaceutical Development

Background:

• Solution stability of engineered biotherapeutics like monoclonal antibodies is crucial for manufacturability, efficacy, and safety.
• Understanding protein self-association mechanisms is key to developing predictive tools for drug development.

Purpose of the Study:

• To investigate the atomic-level mechanisms of self-association in natively folded antibody fragments (Fabs).
• To identify factors contributing to differing aggregation propensities in antibody variants.
• To evaluate the accuracy of current in silico tools in predicting antibody aggregation.

Main Methods:

• Investigated affinity-matured full-size antibodies and their single amino-acid variants.
• Performed biophysical testing on antigen-binding fragments (Fabs).
• Determined crystal structures (PDB: 6MXR, 6MXS, 6MY4, 6MY5).
• Utilized hydrogen-exchange mass spectrometry for structural analysis.
• Tested existing in silico aggregation prediction tools.

Main Results:

• Antibody variants with single amino acid changes exhibited significantly different aggregation behaviors.
• Fabs from aggregating antibodies showed reversible self-association in the low-micromolar range.
• Fab self-association occurred through complementarity-determining regions in a symmetric manner.
• Point mutations induced local conformational changes, creating complementary interfaces for dimerization.
• In silico tools demonstrated low reliability in predicting observed aggregation propensities.

Conclusions:

• Single amino acid substitutions can drastically alter antibody aggregation propensity through specific interface formation.
• Current in silico prediction methods are insufficient for native antibody aggregation.
• Further development of predictive algorithms should incorporate intermolecular docking, flexibility, and packing interactions.