We examined the effect of miaA, a mutation conferring a deficiency in 2-methylthio-N6-isopentenyladenosine in tRNA, on patterns of peptidyl-tRNA accumulation in Escherichia coli strains deficient in peptidyl-tRNA hydrolase activity. A specific reduction in peptidyl-tRNA accumulation was seen for tRNAs which normally contain the 2-methylthio-N6-isopentenyladenosine modification. These results provide new evidence in support of the ribosome editor model, which links peptidyl-tRNA release to mistranslation events.The isolation of a temperature-sensitive mutation in an Escherichia coli gene coding for peptidyl-tRNA hydrolase (pth) was reported by Atherly and Menninger in 1972 (1). This enzyme specifically cleaves peptides from cytoplasmic peptidyl-tRNA. The pth mutation is conditionally lethal (16). This implies that peptidyl-tRNA hydrolase is an essential enzyme in E. coli. Yet, in the classical description of protein synthesis, there is no step where cytoplasmic peptidyl-tRNA is generated. In addition, the release of a nonfunctional, half-finished peptide from the ribosome seems extremely wasteful.These considerations led Menninger (11) to propose that the release of peptidyl-tRNA is associated with a proofreading mechanism, the so-called ribosome editor. In this model, the premature release of peptidyl-tRNA is the result of an active editing mechanism which recognizes and specifically releases noncognate tRNA molecules that have escaped proofreading in the acceptor (A) site (23). Recent evidence suggests that peptidyl-tRNA release is associated with the translocation step (15).The ribosome editor hypothesis leads to a very precise prediction. The vast majority of, if not all, tRNA molecules released as peptidyl-tRNA should be noncognate; i.e., they should have been released in the process of mistranslating a codon. However, testing this precise prediction directly has proven difficult (2, 3).Here we describe a new approach to test the ribosome editor hypothesis. We previously studied the effect of miaA, a mutation which causes a deficiency in 2-methylthio-N6-isopentenyladenosine in tRNA, on the readthrough and suppression of nonsense mutations in E. coli (21). This modification in 2-methylthio-N6-isopentenyladenosine, hereafter called isopentenyladenosine, occurs 3' adjacent to the anticodon in all tRNAs that read codons beginning with uracil (17). Those rely ,more heavily on tRNA modifications such as the isopentenyladenosine modification for stabilization in the A site of a programmed ribosome.The miaA mutation can be used to tag genetically a subpopulation of tRNAs. Based on our previous results, introduction of the miaA allele should specifically reduce the ability of miaA tRNAs to misread other codons but leave all other aspects of protein synthesis more or less unaltered. Here we report how the miaA mutation influenced patterns of peptidyl-tRNA release. Our approach had the advantage that the ribosome and therefore the ribosome editor remained unaltered throughout these experiments. modification was also ...