Comparison of different translation fixation conditions for ribosome profiling

Ribosome profiling is a powerful technique to measure translation activity and to identify translational pausing sites. The method is based on collecting and sequencing RNA fragments protected by ribosomes from endonuclease digestion. During isolation of these ribosome protected fragments, or RPFs, translation needs to be arrested, and this is generally accomplished by the addition of specific translation inhibitors, such as the antibiotic chloramphenicol in case of bacteria. However, it has been shown that the harvesting of cells and the use of certain translation inhibitors, including chloramphenicol, can cause a bias in ribosome pausing on the mRNA. In this study, we have benchmarked five different translation blocking agents against chloramphenicol, including tetracycline, methanol, the crosslinkers di-thiobis-succinimidyl-propionate (DSP) and formaldehyde, and a combination of the latter two. For this comprehensive analysis, we used an amylase overexpressing Bacillus subtilis strain as test model. The methods were evaluated based on RPF yields, ribosomal A-site prediction accuracy, codon bias, mean ribosome load, ribosome pausing sites, and sequence bias. All methods yielded largely comparable results, including the control method lacking any translation blocking agent. This indicates the robustness of the method, and suggests that rapidly cooling of cells already causes sufficient translation arrest. Of the different agents tested, tetracycline, DSP and the formaldehyde-DSP combination scored best on multiple parameters, and better than chloramphenicol. A curious observation was an apparent bias for guanosine residues in the 4th codon downstream of ribosome pausing sites. This was also observed for E. coli. Why this is the case, we do not know.