Structure-Guided Redesign of RBD-Specific Antibodies to Improve Binding Affinity to SARS-CoV-2 Sublineages

The continuous emergence of new SARS-CoV-2 variants carrying numerous mutations associated with greater transmissibility, infectivity, and evasion of neutralizing antibodies poses a significant challenge to limiting the spread of the disease. This study aimed to use a computational approach to redesign the Complementarity-Determining Regions (CDRs) of the CV30 antibody to increase antibody binding to the Receptor Binding Domain (RBD) of SARSCoV-2 sublineages.

RBD sequences from SARS-CoV-2 sublineages were analyzed in comparison with the Wuhan RBD, and 3D structures were generated. RBD-specific antibodies were selected based on the lowest Kd values or calculated free binding energy (ΔGbind[Rosetta]) using the MM-GBSA method. The CV30 mAb was selected for guided redesign to increase RBD binding affinity to BA.4, BA.5, BQ.1.1, and KP.2 SARS-CoV-2 variants.

Computational analysis indicated that the redesigned antibodies (CV30 BA.4/5, CV30 BA.5 BQ.1.1, and CV30 KP.2) exhibited higher binding affinity for their respective RBDs. The redesigned antibodies showed a more pronounced amino acid change in heavy chain CDR3 and in all light CDRs. Long CDR H3 was particularly found in both CV30 BA.4/5 and CV30 BA.5 BQ.1.1 antibodies.

Potent RBD-neutralizing antibodies with long CDRs H3 have also been reported in the literature, suggesting that long CDRs H3 may also be an alternative framework for the design of new antibodies. Furthermore, all novel antibodies showed increased interaction interface to the RBDs of BA.4, BA.5, BQ.1.1, and KP.2 sublineages.

The redesigned antibodies represent promising candidates for neutralizing emerging SARS-CoV-2 sublineages.