Orthogonal pair-directed codon reassignment as a tool for evaluating the factors affecting translation in E. coli

Abstract

Proteins are polymers of amino acids that are essential for life, central to cellular function, and have applications in fields ranging from materials science to biomedicine. Proteins in nature are composed of 20 amino acids with limited variability in size and chemical properties. Expanding the genetic code to contain non-canonical amino acids (ncAAs) that contain functionalities not contained in nature is a powerful strategy for probing and extending the properties of proteins. Current in vivo systems for expanding the genetic code have focused on using an engineered orthogonal aminoacyl-tRNA and aminoacyl tRNA-synthetase pairs (tRNA/aaRS) to direct incorporation of ncAAs at amber stop codons. In order to further expand the genetic code to 22 or more amino acids, additional codons must be targeted for reassignment to ncAAs. The genetic code is degenerate; 18 of the 20 canonical amino acids are encoded by more than one codon. Breaking the degeneracy of the genetic code by orthogonal pair directed sense codon reassignment is one pathway to genetic codes of 22 or more amino acids. However, orthogonal pair directed sense codon reassignment is hampered by a limited understanding of the relative importance of the factors that affect the translation of proteins. Here, we describe the repurposing of two commonly used orthogonal pairs from Methanocaldococcus jannaschii (M. jannaschiiI) and Methanosarcina barkeri (M. barkeri) to measure the in vivo reassignment efficiency of 30 different sense codons to tyrosine in E. coli with a simple fluorescence-based screen. The suite of sense codon reassignment efficiencies identified multiple promising codons for reassignment to ncAAs that have not been previously identified. Importantly, every sense codon was partially reassigned to tyrosine when either orthogonal tRNA/aaRS pair was used. Sense codons reassigned to tyrosine with high efficiency may be used directly for reassignment to ncAAs, and any sense codon with measurable reassignment to tyrosine may be improved through directed evolution. The sets of in vivo sense codon reassignment also revealed that E. coli are broadly tolerable to a large number of amino acid substitutions to tyrosine throughout the proteome. The codon reassignment efficiency measurements also enabled an analysis of the in vivo importance of local codon context effects, tRNA abundance, aminoacylation level, tRNA modifications, and codon-anticodon binding energy in determining translational fidelity. Quantitative sense codon reassignment efficiency measurements showed that the process of translation is highly balanced and both tRNA abundance and aminoacylation efficiency do not appear to be dominant factors in determining translational fidelity. Furthermore, quantitative measurements of amber stop codon reassignment efficiencies to tyrosine with the orthogonal M. jannaschii pair revealed that local codon context is an important factor for orthogonal pair directed amber stop codon reassignment.