Importance of single molecular determinants in the fidelity of expanded genetic codes Alicja K. Antonczaka,1, Zuzana Simovaa,1, Isaac T. Yonemoto2, Matthias Bochtlera,b,c, Anna Piaseckad, Honorata Czapińskaa,c, Andrea Brancalee, and Eric M. Tippmanna,3 aSchool of Chemistry and bSchool of Biosciences, Cardiff University, Cardiff CF10 3AT United Kingdom; dSchool of Medicine, Cardiff University, UHW Main Building, Heath Park, Cardiff CF14 4XN, United Kingdom; eWelsh School of Pharmacy, Cardiff University, Cardiff CF10 3AT United Kingdom; cThe International Institute of Molecular and Cell Biology, Trojdena 4, 02-109 Warsaw, Poland Edited* by Robert Huber, Max Planck Institute for Biochemistry, Planegg-Martinsried, Germany, and approved December 15, 2010 (received for review August 17, 2010) The site-selective encoding of noncanonical amino acids (NAAs) system’s inherent tendency to terminate translation in response is a powerful technique for the installation of novel chemical func- to the stop signal. As a result of this competitive recoding/sup- tional groups in proteins. This is often achieved by recoding a pression strategy, it is relatively easy to interpret results when stop codon and requires two additional components: an evolved a mutant AARS exhibits exquisite fidelity in charging its cognate – aminoacyl tRNA synthetase (AARS) and a cognate tRNA. Analysis NAA to tRNAstop (26 28). If the tRNAstop goes uncharged, the of the most successful AARSs reveals common characteristics. The stop codon inserted into the gene of interest reverts to its natural highest fidelity NAA systems derived from the Methanocaldococ- role as the molecular signal for termination of translation, and cus jannaschii tyrosyl AARS feature specific mutations to two resi- truncated products of translation ensue. Mutant AARSs may also dues reported to interact with the hydroxyl group of the substrate be capable of charging amino acids besides the desired NAA onto tyrosine. We demonstrate that the restoration of just one of these tRNAstop. In this situation, translation termination will not occur, determinants for amino acid specificity results in the loss of fidelity and low-resolution analysis techniques such as SDS-PAGE de- as the evolved AARSs become noticeably promiscuous. These re- signed to report solely on termination will not be able to discri- sults offer a partial explanation of a recently retracted strategy minate between proteins containing successfully installed NAAs for the synthesis of glycoproteins. Similarly, we reinvestigated a and unsuccessfully synthesized alloproteins with a substituting BIOCHEMISTRY tryptophanyl AARS reported to allow the site-selective incorpora- canonical amino acid. Selection schemes to minimize mischarging tion of 5-hydroxy tryptophan within mammalian cells. In multiple have been developed (29) and refined (30). experiments, the enzyme displayed elements of promiscuity de- In NAA incorporation systems, the amino acid most likely to spite its previous characterization as a high fidelity enzyme. Given be charged in place of the NAA is the amino acid that the evolved the many similarities of the TyrRSs and TrpRSs reevaluated here, or engineered AARS originally recognized. Translationally, the our findings can be largely combined, and in doing so they rein- ribosome exerts little if any editing power, and so for AARSs force the long-established central dogma regarding the molecular without editing domains, such as TyrRSs and TrpRSs, the fidelity basis by which these enzymes contribute to the fidelity of transla- of NAA incorporation with these AARSs will be directly related tion. Thus, our view is that the central claims of fidelity reported in to the relative activity that these synthetases have with their ori- several NAA systems remain unproven and unprecedented. ginal amino acid versus the NAA. We show that single residue elements in these AARS enzymes confer the enzyme’s specificity directed evolution ∣ protein engineering and, ultimately, the fidelity of incorporation. The applicability of this concept has been extended to TyrRS (31–33) and TrpRS (25) minoacyl tRNA synthetases (AARSs) are a central feature of derived systems whose reported fidelity is in question, or appears Aany effort to recode or reassign the genetic code (1). During at odds with other work in the field. translation AARSs enforce fidelity through a number of well- determined mechanisms (2–4). AARSs discriminate between Results and Discussion amino acids, but also all endogenous tRNAs, thereby conferring TyrRS. Fersht et al. identified two residues of the Bacillus stear- othermophilus Bst specificity. A cognate tRNA molecule is aminoacylated based on ( ) TyrRS, Tyr 34, and Asp 176, as hydrogen sequence identity elements, and a given amino acid comes to be bonding partners for the hydroxyl -OH group of substrate tyro- associated with the correct anticodon. Misacylation occurs, but sine (Fig. 1) (9, 34, 35). Their observations were based on crystal- for the AARSs where this is problematic, for example an IleRS lographic evidence (36), and their observations can be applied Methanocaldococcus jannaschii Mj (5) or LeuRS (6), additional editing domains result in the hydro- to the highly homologous ( ) Mj lysis of the misacylated amino acid. However, TyrRSs and TrpRSs, TyrRS (37). In the majority of TyrRSs evolved to charge Mj the focus of this work, are not known to have additional editing NAAs, these two residues ( numbering Tyr32 and Asp158) mechanisms despite several decades of intensive molecular and have been mutated to more hydrophobic residues as a require- biochemical investigation (7–9). TyrRSs and TrpRSs, evolved or engineered, are responsible for Author contributions: A.K.A., Z.S., I.T.Y., M.B., A.B., and E.M.T. designed research; A.K.A., a significant proportion of all noncanonical amino acids (NAAs) Z.S., A.P., H.C., A.B., and E.M.T. performed research; A.K.A., Z.S., I.T.Y., M.B., A.P., H.C., A.B., now routinely incorporated into proteins (10–14). These two and E.M.T. analyzed data; and I.T.Y. and E.M.T. wrote the paper. AARSs probably share a common ancient origin given the simi- The authors declare no conflict of interest. larities in protein sequence and overall structure (15–18). Thus, *This Direct Submission article had a prearranged editor. we believe that concepts about the determinants of specificity of Data deposition: The crystallography, atomic coordinates, and structure factors have been one AARS are often applicable to the other and vice versa. deposited in the Protein Data Bank, www.pdb.org (PDB ID codes 3PRH and 3FHJ-1). With some exceptions (19–21), the site-selective incorporation 1A.K.A. and Z.S. contributed equally to this work. of NAAs is commonly achieved by recoding a single stop codon. 2Present address: 3086 West Fox Run Way, San Diego, CA 92111. This process has been reported using mutant TyrRSs or TrpRSs in 3To whom correspondence should be addressed. E-mail: [email protected]. – a range of host organisms (22 25). In these systems, suppression This article contains supporting information online at www.pnas.org/lookup/suppl/ of a stop codon is in constant competition with the translation doi:10.1073/pnas.1012276108/-/DCSupplemental. www.pnas.org/cgi/doi/10.1073/pnas.1012276108 PNAS Early Edition ∣ 1of6 Downloaded by guest on September 30, 2021 We contend that it is more likely that the 18432.3 Da signal is due to competing incorporation of tyrosine, which is consistent with other mass spectra but also the importance of this particular TyrRS’s Asp158 binding the Tyr -OH group (vide infra). The importance of Tyr32 as a “gatekeeper” residue of TyrRS fidelity, especially in NAA systems, was investigated by reverting three evolved TyrRSs to code for Tyr at this position. The three TyrRSs chosen corresponded to the enzymes reported for p-iodoPhe (45), p-acetylPhe (28), and p-boronoPhe (46). The evolved enzymes were reported with either a Leu or Ser in place of Tyr at position 32. Each revertant and cognate tRNACUA were expressed along with a UAG mutant for T4 lysozyme in the presence (þ) and absence (−) of each NAA. The results suggest high fidelity tRNACUA charging for the original enzymes with p-iodoPhe and p-boronoPhe, and marginal fidelity for charging with p-acetylPhe (Fig. 2). The results also show that the three TyrRS revertants would appear to charge tRNACUA in the ab- sence of NAA. The presence of full-length protein in the absence Fig. 1. Molecular interactions between hydrogen-bonding side chains of NAA (Fig. 2) is consistent with WT enzyme activity where from Bst tyrosyl tRNA synthetase with bound Tyr-AMP. Bst residues Tyr34 tRNACUA is charged with a canonical amino acid, presumably and Asp174 map onto Mj Tyr32 and Asp158, respectively. (Reproduced with tyrosine. It was also observed that the revertants displayed similar permission from ref. 9 (Copyright 1985, Nature Publishing Group.) fidelity to the AARS reported to have high fidelity for GlcNAc- Ser. Again, this is consistent with the fact that the TyrRS variant reported to have high fidelity for GlcNAc-Ser possessed the same ment for achieving high fidelity incorporation of the NAA versus Tyr32 residue found in the WT enzyme (39). native tyrosine (Table S1). Söll has suggested that an exogenous AARS produced in Whereas most Mj TyrRS variants have changed Tyr32 to a Escherichia coli requires as little as 1% of the activity of the WT smaller, more hydrophobic residue, the TyrRS evolved to charge enzyme to complement an auxotrophic phenotype (15). Our 3-aminoTyrosine (3-aminoTyr) featured Gln, Glu, Lys, or Arg at results suggest that the activity of the GlcNAc-SerRS and the re- this position (38). This can be rationalized from the overall struc- vertants are well above this threshold. Other insight into why the tural similarity between 3-amino Tyr and Tyr and their relative GlcNAc-SerRS and revertant TyrRSs would charge tRNACUA ability to engage in salt bridging interactions.