Automated orthogonal tRNA generation

Abstract

The ability to generate orthogonal, active tRNAs—central to genetic code expansion and reprogramming—is still fundamentally limited. In this study, we developed Chi-T, a method for the de novo generation of orthogonal tRNAs. Chi-T segments millions of isoacceptor tRNA sequences into parts and then assembles chimeric tRNAs from these parts. Chi-T fixes the parts, containing identity elements, and combinatorially varies all other parts to generate chimeric sequences. Chi-T also filters the variable parts and chimeric sequences to minimize host identity elements. We show here that experimentally characterized orthogonal tRNAs are more likely to have predicted minimum free energy cloverleaf structures, and Chi-T filters for sequences with a predicted cloverleaf structure. We report RS-ID for the identification of synthetases that may acylate the tRNAs generated by Chi-T. We computationally identified new orthogonal tRNAs and engineered an orthogonal pair generated by Chi-T/RS-ID to direct non-canonical amino acid incorporation, in response to both amber codons and sense codons, with an efficiency similar to benchmark genetic code expansion systems. Genetic code expansion and reprogramming require orthogonal tRNAs. Methods have now been developed for the automated generation of chimeric orthogonal tRNAs and discovery of their cognate synthetases. These approaches have been used to discover new orthogonal pairs for efficient non-canonical amino acid incorporation.