Selection of a ‘minimal’ glutaminyl‐tRNA synthetase and the evolution of class I synthetases.

The evolution of the aminoacyl‐tRNA synthetases is intriguing in light of their elaborate relationship with tRNAs and their significance in the decoding process. Based on sequence motifs and structure determination, these enzymes have been assigned to two classes. The crystal structure of Escherichia coli glutaminyl‐tRNA synthetase (GlnRS), a class I enzyme, complexed to tRNA(Gln) and ATP has been described. It is shown here that a ‘minimal’ GlnRS, i.e. a GlnRS from which domains interacting with the acceptor‐end and the anticodon of the tRNA have been deleted, has enzymatic activity and can charge a tRNA(Tyr)‐derived amber suppressor (supF) with glutamine. The catalytic core of GlnRS, which is structurally conserved in other class I synthetases, is therefore sufficient for the aminoacylation of tRNA substrates. Some of these truncated enzymes have lost their ability to discriminate against non‐cognate tRNAs, implying a more specific role of the acceptor‐end‐binding domain in the recognition of tRNAs. Our results indicate that the catalytic and substrate recognition properties are carried by distinct domains of GlnRS, and support the notion that class I aminoacyl‐tRNA synthetases evolved from a common ancestor, jointly with tRNAs and the genetic code, by the addition of non‐catalytic domains conferring new recognition specificities.