Modular Nanobody Conjugates with Controlled Topology Using Genetically Encoded Non-canonical Amino Acids

Bispecific antibody-based derivatives are traditionally generated by fusing short, flexible peptides to two variable heavy and light chain pairs that recognize distinct antigens. However, this method limits domain joining to the amino (N) or carboxyl (C) termini, thereby restricting our ability to study the role of domain orientation in key properties of bispecific molecules, such as affinity and specificity. Here, we present an adaptable, plug-and-play application of genetic code expansion technology for the rapid, modular creation of bispecific nanobody conjugates from non-canonical amino acid (ncAA)- integrated nanobody domains, offering precise control over domain topology. We implement this strategy using a single-plasmid genetic code expansion system and computation-guided ncAA site selection to preserve nanobody expression and target binding. We demonstrate the effective incorporation and crosslinking of azide- and tetrazine-modified lysine and phenylalanine, respectively, at four engineered positions within an anti-PD-L1 nanobody and two positions within an anti-CTLA-4 nanobody in a panel of immune cell- and cancer-cell-binding nanobodies. Using this approach, we demonstrate the modular synthesis of a library of bispecific nanobodies, comprising four PD-L1xCD16 immune synapse engagers and eight PD-L1xCTLA-4 immune checkpoint bispecifics. This work enables a new dimension of control over protein crosslinking for rapidly constructing immune cell engagers for dual immune checkpoint blockade therapy and to bridge the synapse between immune cells and cancer cells.