Hijacking a bacterial membrane transporter for efficient genetic code expansion

The site-specific encoding of non-canonical amino acids (ncAAs) provides a powerful tool for expanding the functional repertoire of proteins. Its widespread use for basic research and biotechnological applications is, however, hampered by low efficiencies of current ncAA incorporation strategies. We uncover poor cellular ncAA uptake as a main obstacle to efficient genetic code expansion and overcome this bottleneck by hijacking a bacterial membrane transporter to actively import isopeptide-linked ncAAs within easily synthesizable tripeptide-scaffolds. Using this approach, we enable efficient encoding of eleven previously inaccessible ncAAs, decorating proteins with bioorthogonal and crosslinker moieties, posttranslational modifications, and functionalities for chemoenzymatic conjugation. To enhance scalability of protein production, we evolve the membrane transporter for preferential import of isopeptide-linked tripeptides, creating a novel Escherichia coli strain that facilitates single and multi-site ncAA incorporation with wild type efficiencies. Additionally, we adapt the tripeptide-scaffolds for co-transport of two different ncAAs, enabling their efficient dual incorporation. This work underscores the importance of optimizing ncAA-uptake for high-yielding production of modified proteins and will accelerate the development of generalizable transport systems, aiding incorporation of non-canonical building blocks to broaden the chemical space of proteins without the need to design for passive membrane permeability.