(596g) Invited Talk: Developing Improved Biosynthesis Pathways for Non-Canonical Amino Acids

Non-canonical canonical amino acids have proven valuable tools for augmenting protein function and offer an expansive array of protein chemistries beyond those found in the twenty standard amino acids. While hundreds of new chemistries have now been introduced into proteins, two non-canonical amino acids of particular interest to protein engineers are the 21^st amino acid, selenocysteine (Sec, U), and 3,4-dihydroxyphenylalanine or L-DOPA. Importantly, while both offer novel bioorthogonal chemistry, they are also naturally occurring, providing researchers with multiple model systems to study and serve as inspiration for new catalysts and materials.

Selenocysteine is a rare amino acid with an unusual mosaic distribution across the proteome. The amino acid is analogous to cysteine (Cys) but replaces the thiol with a selenol moiety allowing it to share similar but intensified chemical properties, including a high affinity for metals, strong nucleophilicity, and reversible covalent bond formation. These properties are of broad interest to the protein engineering community and have great potential for the development of new biologics stabilized with diselenide bonds and industrial biocatalysts. However, unlike canonical amino acids, selenocysteine lacks a dedicated aminoacyl-tRNA synthetase and is instead biosynthesized directly on its tRNA by directly modifying a pre-charged serine residue prior to its incorporation into the nascent polypeptide chain. In addition, it is co-translationally incorporated in response to a stop codon, a process that is often extremely inefficient. Collectively, these limitations have largely restricted studies of recombinant selenoproteins to catalytically impaired variants containing Sec-to-Cys substitutions and prevented investigation of selenocysteine in novel contexts. This also presents an unfortunate ‘Chicken and Egg’ situation, where one of the key enzymes required for selenocysteine biosynthesis (selenophosphate synthetase, SelD) is frequently itself a selenoprotein, containing an essential selenocysteine residue in most bacteria with large selenoproteomes.

In contrast to both canonical amino acids and selenocysteine, L-DOPA is natively introduced into proteins as a post-translational modification of tyrosine residues. The amino acid is formed by direct oxidation of the tyrosine phenol group to form a catechol moiety, a reaction performed by several different enzymes and with varying degrees of selectivity for the free or incorporated amino acid. In Nature, the reactive catechol moiety frequently forms covalent crosslinks with nearby nucleophiles, bonds which play an important role in tuning the properties of many proteinaceous biomaterials. Several groups have developed orthogonal translation machinery capable of site-specific L-DOPA incorporation, thereby enabling new protein engineering applications which leverage the unique catalytic and metal coordinating properties of the catechol group.

Efforts to better understand the biosynthesis pathways for these two amino acids are a high priority; direct conversion of serine into selenocysteine is unfavourable and proceeds very slowly in comparison to the initial charging of the tRNA with serine. This can result in erroneous incorporation of serine into proteins, especially at high rates of translation and turn-over of the selenocysteinyl-tRNA pool. Similarly, selectivity of the engineered translational machinery for L-DOPA incorporation is not perfect, often resulting in misincorporation of tyrosine. The current lack of highly efficient biosynthesis pathways for these amino acids is a key barrier to large scale production of Sec- and DOPA-containing proteins. Here we present several efforts to identify, characterize, and improve enzymes involved in the biosynthesis of selenocysteine and L-DOPA, and discuss the development of novel assays and reporter systems to facilitate high-throughput engineering.