Deep learning guided optimization of human antibody against SARS-CoV-2 variants with broad neutralization

Abstract

SignificanceSARS-CoV-2 continues to evolve through emerging variants, more frequently observed with higher transmissibility. Despite the wide application of vaccines and antibodies, the selection pressure on the Spike protein may lead to further evolution of variants that include mutations that can evade immune response. To catch up with the virus's evolution, we introduced a deep learning approach to redesign the complementarity-determining regions (CDRs) to target multiple virus variants and obtained an antibody that broadly neutralizes SARS-CoV-2 variants.

Keywords: SARS-CoV-2 variants; broadly neutralizing antibodies; computational biology; deep learning; geometric neural networks.

Fig. 1.

Deep learning guided antibody optimization platform. (A) Overview of the pipeline. It demonstrates the computational/experimental feedback loop to refine antibody design. (B) Geometric deep learning model. The WT complex and the mutated complex structures are encoded using a shared geometric attention network. The effect of mutation measured by ΔΔG is then predicted by a network that compares features of the two complexes. (C) P36-5D2 antibody optimization. Given the complex structure, we first simulate different variants and then evaluate potential CDR mutations that will improve binding by predicted ΔΔG values. Mutants with top ΔΔG scores are examined in laboratory experiments, and those with neutralizing potency are combined for the next round of optimization. (D) Optimization improves neutralization ability against SARS-CoV-2 and Delta variant. (E) The log fold changes of IC₅₀ relative to the original antibody.

Fig. 2.

Evaluation of the neutralization level of optimized antibodies. (A) IC₅₀ values of optimized antibodies against pseudotyped SARS-CoV-2 variants. Mutations of each optimized antibody and mutations of each variant on RBD are indicated. Results were calculated from three independent experiments. (B) The log fold changes of IC₅₀ relative to the original antibody against SARS-CoV-2 and Delta variants are calculated. (C) Neutralization curves of optimized against pseudotyped SARS-CoV-2 carrying Omicron mutations N440K or G446S.

Fig. 3.

Predicted structure of important mutations on optimized antibodies. (A) Predicted structure of interactions between original antibody carrying T31 (red) or optimized antibody carrying W31 (red) with related residue R452/L452 (blue) on Delta/WT RBD (cyan); heavy chain is labeled in purple, and light chain is labeled in pink. (B) Predicted structure of interactions between original antibody carrying N57 (red) or optimized antibody carrying L57 (red) with Y449 (blue) and R452/L452 (blue) on Delta/WT RBD. (C) Predicted structure of interactions between original antibody carrying R103 (red) or optimized antibody carrying M103 (red) with related residues R346 (blue) and R452/L452 (blue) on Delta/WT RBD. (D) Predicted structure of interactions between original antibody carrying L104 (red) or optimized antibody carrying F104 (red) with R452/L452 (blue) on Delta/WT RBD.