Errors can occur at any level during the replication and transcription of genetic information. Genetic mutations derived mainly from replication errors have been extensively studied. However, fundamental details of transcript errors, such as their rate, molecular spectrum, and functional effects, remain largely unknown. To globally identify transcript errors, we applied an adapted rolling-circle sequencing approach to Escherichia coli, Bacillus subtilis, Agrobacterium tumefaciens, and Mesoplasma florum, revealing transcript-error rates 3 to 4 orders of magnitude higher than the corresponding genetic mutation rates. The majority of detected errors would result in amino-acid changes, if translated. With errors identified from 9929 loci, the molecular spectrum and distribution of errors were uncovered in great detail. A G→A substitution bias was observed in M. florum, which apparently has an error-prone RNA polymerase. Surprisingly, an increased frequency of nonsense errors towards the 3' end of mRNAs was observed, suggesting a Nonsense-Mediated Decay-like quality-control mechanism in prokaryotes.
Keywords: B. subtilis; E. coli; RNA quality-control; base substitutions; chromosomes; gene expression; genetics; genomics; transcript errors; transcriptional fidelity.
Most cells contain molecules of DNA that carry instructions to make the proteins cells need to perform different tasks. When a cell requires a certain protein, the corresponding DNA sequence is first transcribed into molecules of ribonucleic acid (RNA) known as transcripts. These sequences of RNA are then read by the cell and translated into the desired protein sequence. Errors in copying DNA before a cell divides, can lead to genetic mutations that affect the ability of the cell to carry out certain roles, influencing the overall ‘fitness’ of the cell. Similar to genetic mutations, errors that arise when forming RNA transcripts may also alter the tasks a cell performs. However, it is difficult to find out what kinds of errors cells have in their transcripts and how often these mistakes occur. This is because current methods for sequencing RNA are prone to technical inaccuracies that interfere with the ability to detect true transcript errors. Now, Li and Lynch have adapted a method for high-throughput sequencing of RNA, which can accurately identify transcript errors in Escherichia coli and other species of bacteria. The experiments showed that errors in RNA molecules occurred more frequently than genetic mutations in the same sequence of DNA. Li and Lynch also found that the transcripts contained more nonsense errors – that is, mutations which prematurely stop transcripts from being translated, resulting in shorter proteins – at the end of the RNA molecule than at the beginning or middle. It is possible that transcripts with errors at the beginning or the middle are more efficiently eliminated than those at the end, suggesting that bacteria have a quality-control mechanism for removing transcripts with premature stop sequences. These findings suggest that at any one-time cells carry thousands of transcripts with inaccuracies in their sequence, which likely impact the tasks cells perform. The next step will be to investigate how these different transcript errors affect the fitness of cells.
© 2020, Li and Lynch.