An integrated multiscale computational framework deciphers SARS-CoV-2 resistance to sotrovimab

Area of Science:

• Virology
• Structural Biology
• Computational Chemistry

Background:

• Monoclonal antibody (mAb) treatments like sotrovimab target the SARS-CoV-2 spike (S) protein.
• Emergence of resistance mutations in the S protein challenges the efficacy of these therapies.
• Understanding mutation-driven resistance mechanisms is crucial for therapeutic development.

Purpose of the Study:

• To develop and validate an integrated computational framework for identifying SARS-CoV-2 S protein resistance mutations against sotrovimab.
• To elucidate the structural and dynamic features of resistance mutations.
• To inform the design of next-generation antibody therapeutics.

Main Methods:

• Integrated framework combining interface protein design, machine learning, QM/MM, and multiscale molecular dynamics (MD) simulations.
• Analysis of S protein-sotrovimab interactions and identification of critical residues.
• Machine learning for predicting resistance mutations based on structural, sequence, binding, and energetic features.
• QM/MM to assess the impact of mutations on complex stability and MD simulations for dynamic behavior analysis.
• Validation against clinical sequencing data and empirical evidence.

Main Results:

• Identified pivotal residues and plausible resistance mutations at the S protein-sotrovimab interface.
• Two key residues, E340 and Y508, were identified and their designs correlated with clinically observed resistance mutations.
• Machine learning models predicted novel S protein sequences with altered sotrovimab affinity, structurally validated by AlphaFold.
• Multiscale simulations captured conformational dynamics and stability impacts of mutations, including glycan interactions.

Conclusions:

• The integrated framework effectively identifies SARS-CoV-2 resistance mutations against sotrovimab, supported by clinical data.
• This multidimensional approach provides insights into resistance mechanisms and aids in designing effective antibody therapeutics.
• The methodology has translational relevance for developing antibodies against other viral systems.