Structure of dormant bacterial ribosomes reveals an elusive translation factor with evolutionary connections to eukaryotic protein synthesis factors

During starvation and stress, virtually all organisms arrest protein synthesis to conserve energy. Inactive ribosomes are converted into a dormant state, in which they are protected from damage by hibernation factor proteins. In bacteria, two major families of hibernation factors have been described, but the low conservation of these proteins and the huge diversity of species, habitats, and environmental stressors has confounded their discovery. In this study, using proteomics and cryo-EM, we identify a new dormancy factor from the psychrophilic bacterium Psychrobacter urativorans. By isolating ribosomes under cold-shock conditions, we observe a previously unknown protein bound to the ribosomal A site, protecting critical elements of both the decoding and peptidyl transferase centers. We show that this new factor, which we term Balon, is a homolog of the archaeo-eukaryotic translation factor aeRF1, providing a long-predicted evolutionary "missing link" between the eukaryotic and bacterial translation machinery. Our structures reveal that Balon is delivered to both vacant and actively translating ribosomes by EF-Tu, highlighting an unexpected and previously unknown role for this elongation factor in the bacterial stress response. We describe several unique structural motifs that allow Balon to bind ribosomes in an mRNA-independent manner, initiating a new mode of ribosome dormancy that can commence while ribosomes are still engaged in protein synthesis. Our bioinformatic analysis shows that putative Balon-encoding genes can be found within stress-response operons in nearly 20 % of all known bacterial species, including many human pathogens. Taken together, our work suggests that Balon/EF-Tu regulated ribosome dormancy is likely to be a ubiquitous stress-response mechanism throughout the bacterial kingdom. These findings call for a revision of our model of bacterial translation inferred from common model organisms and hold numerous implications for how we understand and study ribosome dormancy.