Search for: All records

Therapeutic antibody development requires selection and engineering of molecules with high affinity and other drug-like biophysical properties. Co-optimization of multiple antibody properties remains a difficult and time-consuming process that impedes drug development. Here we evaluate the use of machine learning to simplify antibody co-optimization for a clinical-stage antibody (emibetuzumab) that displays high levels of both on-target (antigen) and off-target (non-specific) binding. We mutate sites in the antibody complementarity-determining regions, sort the antibody libraries for high and low levels of affinity and non-specific binding, and deep sequence the enriched libraries. Interestingly, machine learning models trained on datasets with binary labels enable predictions of continuous metrics that are strongly correlated with antibody affinity and non-specific binding. These models illustrate strong tradeoffs between these two properties, as increases in affinity along the co-optimal (Pareto) frontier require progressive reductions in specificity. Notably, models trained with deep learning features enable prediction of novel antibody mutations that co-optimize affinity and specificity beyond what is possible for the original antibody library. These findings demonstrate the power of machine learning models to greatly expand the exploration of novel antibody sequence space and accelerate the development of highly potent, drug-like antibodies.

Monoclonal antibodies that target SARS-CoV-2 with high affinity are valuable for a wide range of biomedical applications involving novel coronavirus disease (COVID-19) diagnosis, treatment, and prophylactic intervention. Strategies for the rapid and reliable isolation of these antibodies, especially potent neutralizing antibodies, are critical toward improved COVID-19 response and informed future response to emergent infectious diseases. In this study, single B cell screening was used to interrogate antibody repertoires of immunized mice and isolate antigen-specific IgG1⁺memory B cells. Using these methods, high-affinity, potent neutralizing antibodies were identified that target the receptor-binding domain of SARS-CoV-2. Further engineering of the identified molecules to increase valency resulted in enhanced neutralizing activity. Mechanistic investigation revealed that these antibodies compete with ACE2 for binding to the receptor-binding domain of SARS-CoV-2. These antibodies may warrant further development for urgent COVID-19 applications. Overall, these results highlight the potential of single B cell screening for the rapid and reliable identification of high-affinity, potent neutralizing antibodies for infectious disease applications.

The COVID‐19 pandemic continues to be a severe threat to human health, especially due to current and emerging SARS‐CoV‐2 variants with potential to escape humoral immunity developed after vaccination or infection. The development of broadly neutralizing antibodies that engage evolutionarily conserved epitopes on coronavirus spike proteins represents a promising strategy to improve therapy and prophylaxis against SARS‐CoV‐2 and variants thereof. Herein, a facile multivalent engineering approach is employed to achieve large synergistic improvements in the neutralizing activity of a SARS‐CoV‐2 cross‐reactive nanobody (VHH‐72) initially generated against SARS‐CoV. This synergy is epitope specific and is not observed for a second high‐affinity nanobody against a non‐conserved epitope in the receptor‐binding domain. Importantly, a hexavalent VHH‐72 nanobody retains binding to spike proteins from multiple highly transmissible SARS‐CoV‐2 variants (B.1.1.7 and B.1.351) and potently neutralizes them. Multivalent VHH‐72 nanobodies also display drug‐like biophysical properties, including high stability, high solubility, and low levels of non‐specific binding. The unique neutralizing and biophysical properties of VHH‐72 multivalent nanobodies make them attractive as therapeutics against SARS‐CoV‐2 variants.