Hypervariable loop profiling decodes sequence determinants of antibody stability

Area of Science:

• Biophysics
• Immunology
• Protein Engineering

Background:

• Antibody folding and aggregation present significant hurdles in developing therapeutic and reagent antibodies.
• The diversity of complementarity-determining regions (CDRs), essential for antigen binding, often compromises antibody folding stability.
• Understanding the relationship between CDR sequences and antibody folding is limited due to sequence diversity and data scarcity.

Purpose of the Study:

• To develop a high-throughput method for quantifying the folding fitness of diverse CDR sequences.
• To identify sequence-level rules governing CDR folding propensity.
• To engineer more stable antibody scaffolds using folding insights.

Main Methods:

• Development of a high-throughput 'deep loop profiling' technique.
• Training machine learning models on large-scale CDR folding data.
• Utilizing sequence-based predictions to guide protein engineering.

Main Results:

• Quantification of folding fitness across millions of CDRs.
• Identification of CDR1 and CDR2 as critical determinants of antibody folding.
• Successful rescue of unstable nanobodies, including a SARS-CoV-2 binder and a GPCR-targeting intrabody.
• Creation of next-generation synthetic antibody libraries with enhanced biophysical properties.

Conclusions:

• Deep loop profiling offers a scalable framework for understanding antibody folding.
• Sequence-based rules can predict and improve antibody folding competence.
• This approach facilitates the engineering of stable, functional antibody-based therapeutics and reagents.