Search

Protein synthesis in cells has been thoroughly investigated and characterized over the past 60 years. However, some fundamental issues remain unresolved, including the reasons for genetic code redundancy and codon bias. In this study, we changed the kinetics of the Eschrichia coli transcription and translation processes by mutating the promoter and ribosome binding domains and by using genetic code expansion. The results expose a counterintuitive phenomenon, whereby an increase in the initiation rates of transcription and translation lead to a decrease in protein expression. This effect can be rescued by introducing slow translating codons into the beginning of the gene, by shortening gene length or by reducing initiation rates. On the basis of the results, we developed a biophysical model, which suggests that the density of co-transcriptional-translation plays a role in bacterial protein synthesis. These findings indicate how cells use codon bias to tune translation speed and protein synthesis.

The overall process of protein biosynthesis across all domains of life is similar; however, detailed insights reveal a range of differences in the proteins involved. For decades, the process of protein translation in archaea has been considered to be closer to eukaryotes than to bacteria. In archaea, however, several homologues of eukaryotic proteins involved in translation initiation have not yet been identified; one of them being the initiation factor eIF2B consisting of five subunits (α, β, γ, δ and ε). Three open reading frames (PH0440, PH0702 and PH0208) in Pyrococcus horikoshii have been proposed to encode for the α-, β- and δ-subunits of aIF2B, respectively. The crystal structure of PH0440 shows similarity toward the α-subunit of eIF2B. However, the capability of PH0702 and PH0208 to function as the β- and δ-subunits of eIF2B, respectively, remains uncertain. In this study, we have taken up the task of annotating PH0702 and PH0208 using bioinformatics methods. The phylogenetic analysis of protein sequences belonging to IF2B-like family along with PH0702 and PH0208 revealed that PH0702 belonged to methylthioribose-1-phosphate isomerase (MTNA) group of proteins, whereas, PH0208 was found to be clustered in the group of ribose-1,5-bisphosphate isomerase (R15PI) proteins. A careful analysis of protein sequences and structures available for eIF2B, MTNA and R15PI confirms that PH0702 and PH0208 contain residues essential for the enzymatic activity of MTNA and R15PI, respectively. Additionally, the protein PH0208 comprises of the residues required for the dimer formation which is essential for the biological activity of R15PI. This prompted us to examine all eIF2B-like proteins from archaea and to annotate their function. The results reveal that majority of these proteins are homologues of the α-subunit of eIF2B, even though they lack the residues essential for their functional activity. A better understanding of the mechanism of GTP exchange during translation initiation in archaea is henceforth required., (Copyright © 2015 Elsevier B.V. All rights reserved.)

Deinococcus deserti is a desiccation- and radiation-tolerant desert bacterium. Differential RNA sequencing (RNA-seq) was performed to explore the specificities of its transcriptome. Strikingly, for 1,174 (60%) mRNAs, the transcription start site was found exactly at (916 cases, 47%) or very close to the translation initiation codon AUG or GUG. Such proportion of leaderless mRNAs, which may resemble ancestral mRNAs, is unprecedented for a bacterial species. Proteomics showed that leaderless mRNAs are efficiently translated in D. deserti. Interestingly, we also found 173 additional transcripts with a 5'-AUG or 5'-GUG that would make them competent for ribosome binding and translation into novel small polypeptides. Fourteen of these are predicted to be leader peptides involved in transcription attenuation. Another 30 correlated with new gene predictions and/or showed conservation with annotated and nonannotated genes in other Deinococcus species, and five of these novel polypeptides were indeed detected by mass spectrometry. The data also allowed reannotation of the start codon position of 257 genes, including several DNA repair genes. Moreover, several novel highly radiation-induced genes were found, and their potential roles are discussed. On the basis of our RNA-seq and proteogenomics data, we propose that translation of many of the novel leaderless transcripts, which may have resulted from single-nucleotide changes and maintained by selective pressure, provides a new explanation for the generation of a cellular pool of small peptides important for protection of proteins against oxidation and thus for radiation/desiccation tolerance and adaptation to harsh environmental conditions.