Interface‐Guided Computational Protein Design Reveals Bebtelovimab‐Resistance Mutations in SARS‐CoV‐2 RBD: Correlation with Global Viral Genomes and Bebtelovimab‐Escape Mutations

Abstract

AbstractThe emergence of novel mutations in the SARS‐CoV‐2 spike protein challenges monoclonal antibody (mAb) effectiveness. Comprehending resistance mutations and pinpointing vulnerable spike protein residues is vital for enhanced antibody design. To address this issue, we employed an interface‐guided computational protein design (CPD) approach to decode bebtelovimab‐resistance mutations and uncover susceptible residues within the receptor‐binding domain (RBD). Utilizing structural‐modeling and high‐throughput techniques, we mapped the bebtelovimab‐RBD interface, identifying critical resistance mutations through analysis of binding energetics and residue interactions. Our design protocol integrated stability predictions, side‐chain conformational sampling, and binding affinity calculations to prioritize substitutions that restore antibody recognition and neutralization. Previously unexplored susceptible RBD residues were also discovered, offering new therapeutic avenues. Comparative analysis with COVID‐19 patient data validated the predicted resistance mutations (69 %–100 % correlation, based on different MinProp cut‐offs). Precision and recall values, calculated by comparing our predictions with experimentally reported bebtelovimab‐escape mutants, demonstrated the performance and accuracy of our predictions. Investigation of intermolecular interactions highlighted the importance of van der Waals forces, hydrogen bond energy, and electrostatic contributions in bebtelovimab‐RBD binding affinity. This computational design empowers the decoding of resistance mutations and the development of next‐generation antibodies against viral variants, strengthening our response to SARS‐CoV‐2 and related coronaviruses.