The problem of genetic code misreading during protein synthesis

Saccharomyces cerevisiae has been an important model for determining the frequency of translational misreading events, those in which a tRNA pairs incorrectly to the mRNA and inserts an amino acid not specified by the codon in the mRNA. Misreading errors have been quantified in vivo using reporter protein systems or mass spectrometry with both approaches converging on a simple model for most misreading. The available data show that misreading tRNAs must form stereotypical base mismatches that correspond to those that can mimic Watson–Crick base pairs when formed in the ribosomal A site. Errors involving other mismatches occur significantly less frequently. This work debunks the idea of an average misreading frequency of 5 × 10−4 per codon that extends across the genetic code. Instead, errors come in two distinct classes—high frequency and low frequency events—with most errors being of the low frequency type. A comparison of misreading errors in S. cerevisiae and Escherichia coli suggests the existence of a mechanism that reduces misreading frequency in yeast; this mechanism may operate in eukaryotes generally.