The L-shape form of tRNA is maintained by tertiary interactions occurring in the core. Base changes in this domain can cause structural defects and impair tRNA activity. Here, we report on a method to safely engineer structural variations in this domain utilizing the noncanonical scaffold of tRNAPyl. First, we constructed a naïve hybrid between archaeal tRNAPyl and tRNATyr, which consisted of the acceptor and T stems of tRNATyr and the other parts of tRNAPyl. This hybrid tRNA efficiently translated the UAG codon to 3-iodotyrosine in Escherichia coli cells, when paired with a variant of the archaeal tyrosyl-tRNA synthetase. The amber suppression efficiency was slightly lower than that of the "bench-mark" archaeal tRNATyr suppressor assuming the canonical structure. After a series of modifications to this hybrid tRNA, we obtained two artificial types of tRNATyr: ZtRNA had an augmented D (auD) helix in a noncanonical form and the D and T loops bound by the standard tertiary base pairs, and YtRNA had a canonical auD helix and non-standard interloop interactions. It was then suggested that the ZtRNA scaffold could also support the glycylation and glutaminylation of tRNA. The synthetic diversity of tRNA would help create new tRNA⁻aminoacyl-tRNA synthetase pairs for reprogramming the genetic code.
Keywords: amber suppression; genetic code expansion; pyrrolysine tRNA; tRNA secondary structure; tertiary base pairs.