Upgrading aminoacyl-tRNA synthetases for genetic code expansion

Abstract

Synthesis of proteins with non-canonical amino acids via genetic code expansion is at the forefront of synthetic biology. Progress in this field has enabled site-specific incorporation of over 200 chemically and structurally diverse amino acids into proteins in an increasing number of organisms. This has been facilitated by our ability to repurpose aminoacyl-tRNA synthetases to attach non-canonical amino acids to engineered tRNAs. Current efforts in the field focus on overcoming existing limitations to the simultaneous incorporation of multiple non-canonical amino acids or amino acids that differ from the l-α-amino acid structure (e.g. d-amino acid or β-amino acid). Here, we summarize the progress and challenges in developing more selective and efficient aminoacyl-tRNA synthetases for genetic code expansion.

Figure 1

Accurate protein synthesis and conventional methodology to alter aaRS specificity. (a) Each aaRS (XRS) must selectively attach a cognate AA (denoted as X) to a dedicated tRNA with the correct anticodon to form aminoacyl-tRNA (X-tRNA^X). Correct codon-anticodon matching in the ribosome ensures accurate incorporation of X at a defined position in a protein sequence. (b) Polyspecific aaRSs recognize more than one amino acid (depicted by YRS and ZRS) or more than one tRNA (ZRS). These errors in aminoacylation result in expression of a gene into proteins with varied primary structures. (c) General steps for directed evolution of aaRSs.

Figure 2

Advanced methodology for aaRS evolution. (a) Multiplex-automated genome engineering (MAGE) [18,19^••] uses a modified λ-Red allelic replacement system to introduce changes in a targeted chromosomal gene (e.g. orthogonal aaRS) by integrating mutagenic oligos. Short regions flanking the randomized sequence of the oligo direct β protein to introduce the primer into the replication fork. After two rounds of duplication, only one of the DNA copies from the depicted replication fork will incorporate mutagenic oligos, resulting in the production of no more than 25% of recombinant clones for every round of MAGE. The desired level of diversity is achieved by repeating MAGE cycles, increasing the size of the oligo library, and the number of targeted sites. (b) Phage-assisted continuous evolution (PACE) [21,22^••] employs a bacteriophage that lacks the gene for protein III (pIII), which is required for infectivity. A copy of the pIII gene, with at least one TAG codon, is encoded in a plasmid in E. coli. Active aaRS mutants aminoacylate the suppressor tRNA leading to pIII expression and propagation of bacteriophage. (c) Post-translational proofreading (PTP) [39^•] exploits the N-end rule for protein degradation in E. coli. Green fluorescent protein (GFP) is fused to ubiquitin (Ub), separated by a TAG stop codon. Expression of the full-length reporter is the result of aminoacylation of a suppressor tRNA by orthogonal aaRS variants. Ubiquitin cleavase protein 1 (UBP1) from S. cerevisiae cleaves Ub from the reporter, exposing the N-terminus residue of GFP. ClpS surveys the N-terminus of GFP and applies the N-end rule to facilitate degradation of GFP if an undesired AA is incorporated. In contrast, if the desired ncAA is installed at the N-terminus, GFP accumulates and cells with high fluorescence emission are sorted and collected. The system is adaptable as ClpS can be engineered to discriminate against specific ncAAs.