Accurate prediction of the structurally diverse complementarity determining region heavy chain 3 (CDR-H3) loop structure remains a primary and long-standing challenge for antibody modeling. Here, we present the H3-OPT toolkit for predicting the 3D structures of monoclonal antibodies and nanobodies. H3-OPT combines the strengths of AlphaFold2 with a pre-trained protein language model and provides a 2.24 Å average RMSDCα between predicted and experimentally determined CDR-H3 loops, thus outperforming other current computational methods in our non-redundant high-quality dataset. The model was validated by experimentally solving three structures of anti-VEGF nanobodies predicted by H3-OPT. We examined the potential applications of H3-OPT through analyzing antibody surface properties and antibody-antigen interactions. This structural prediction tool can be used to optimize antibody-antigen binding and engineer therapeutic antibodies with biophysical properties for specialized drug administration route.

Therapeutic antibody discovery using synthetic diversity has been proved productive, especially for target proteins not suitable for traditional animal immunization-based antibody discovery approaches. Recently, many lines of evidences suggest that the quality of synthetic diversity design limits the development success of synthetic antibody hits. The aim of our study is to understand the quality limitation and to properly address the challenges with a better design. Using VH3-23 as a model framework, we observed and quantitatively mapped CDR-H3 loop length-dependent usage of human IGHJ4 and IGHJ6 germline genes in the natural human immune repertoire. Skewed usage of DH2-JH6 and DH3-JH6 rearrangements was quantitatively determined in a CDR-H3 length-dependent manner in natural human antibodies with long CDR-H3 loops. Structural modeling suggests choices of JH help to stabilize antibody CDR-H3 loop and JH only partially contributes to the paratope. Our observations shed light on the design of next-generation synthetic diversity with improved probability of success.

Covalent target modulation with small molecules has been emerging as a promising strategy for drug discovery. However, covalent inhibitory antibody remains unexplored due to the lack of efficient strategies to engineer antibody with desired bioactivity. Herein, we developed an intracellular selection method to generate covalent inhibitory antibody against human rhinovirus 14 (HRV14) 3C protease through unnatural amino acid mutagenesis along the heavy chain complementarity-determining region 3 (CDR-H3). A library of antibody mutants was thus constructed and screened in vivo through co-expression with the target protease. Using this screening strategy, six covalent antibodies with proximity-enabled bioactivity were identified, which were shown to covalently target HRV14-3C protease with high inhibitory potency and exquisite selectivity. Compared to structure-based rational design, this library-based screening method provides a simple and efficient way for the discovery and engineering of covalent antibody for enzyme inhibition.