Nature has developed quality-control mechanisms to ensure accurate protein synthesis. In a report in Molecular Cell, Hayes and Sauer now show that when a bacterial ribosome pauses during polypeptide synthesis, the messenger RNA that is being translated is cleaved at the codon that occupies the ribosomal A site. They propose that this quality-control mechanism — which they refer to as A-site mRNA cleavage — together with another translational quality-control system (the tmRNA system), provides a way of reducing translational errors and therefore the production of harmful polypeptides.

Ribosomal pausing can occur at rare codons or stop codons because the cognate transfer RNA or protein-release factor, which is needed for translation termination, is scarce. The authors used a version of the Escherichia coli ybeL gene, which encodes a protein that ends with two proline residues (YbeL-PP), to study its effects — the carboxy-terminal proline residue of YbeL-PP causes inefficient translation termination, which results in ribosome pausing. Northern-blot analysis of cellular RNA showed a truncated ybeL-PP mRNA that was close in size to a control ybeL transcript, which ends after the second base of the stop codon. More precise mapping confirmed that ybeL-PP mRNA was cleaved at, or near, the stop codon.

Interestingly, Hayes and Sauer found that tmRNA (a specialized RNA that ensures the tagging of polypeptides for degradation) is not essential for A-site mRNA cleavage. Yet, when tmRNA is present, it ensures that the polypeptide at the pausing ribosome is tagged for degradation — thereby linking A-site mRNA cleavage to the tmRNA quality-control system.

By introducing an amber stop codon in the ybeL-PP gene, which prevents efficient translation of the ybeL-PP transcript, the level of truncated mRNA was reduced significantly. So, translation is clearly important for A-site mRNA cleavage.

By growing cells in the presence of a proline analogue instead of proline, both ribosomal pausing and polypeptide tagging were reduced, which correlated with a reduction in the level of truncated transcript. The same effect on A-site mRNA cleavage was shown when protein release factor 1 (RF1), which is known to reduce ribosomal pausing, was overproduced. This led the authors to conclude that both translation and ribosome pausing are required for A-site mRNA cleavage. Indeed, when Hayes and Sauer did the reverse experiment, using conditions that increased ribosomal pausing, the level of truncated mRNA increased.

Next, the authors tested the E. coli RelE toxin, which is known to be a mediator of A-site mRNA cleavage, and a range of other bacterial toxins for their possible involvement in ybeL-PP mRNA cleavage in vivo. None of the toxin-deleted strains had any effect on A-site mRNA cleavage, as shown by northern blots and by assaying for polypeptide tagging. The ribosomal alarmone (p)ppGpp (a guanine nucleotide that is synthesized when ribosomes stall) also does not seem to have a role in either the cleavage or tagging process.

As is often the case in science, new findings give rise to further questions — the nuclease that is responsible for A-site mRNA cleavage is still elusive, and the authors speculate that the ribosome itself might mediate this cleavage reaction. In addition, whether a similar mechanism exists in other bacteria, archaebacteria or eukaryotes is, for the moment, anyone's guess.