coli ΔssrA buy INK1197 growth defect. This is surprising since in H. pylori, the SsrASTOP mutation is not essential for in vitro growth strongly suggesting that it is still effective in release of stalled ribosomes [10]. In a previous study [15], an equivalent mutation was introduced see more into E. coli SsrA,

however only phage propagation phenotype is reported and no mention was made of the growth rate of this mutant. The most straightforward interpretation of our data is that trans-translation by Hp-SsrASTOP in E. coli is not efficiently using the resume codon. Indeed, there are striking differences between Hp-SsrA and Ec-SsrA. In particular, the resume codon of Hp-SsrA is GUA encoding Valine and in E. coli, the resume codon GCA encodes Alanine (Figure 4) [5]. Replacement of the Ec-SsrA resume codon by GUA or GUC encoding Valine is functional in E. coli [22]. However, mass spectrometry analysis revealed that breakage of the peptide tag occurred frequently after certain residues like a Valine Sepantronium molecular weight encoded by GUA and that these SsrA-tag added to proteins are ineffective in growth competition with ΔssrA mutants [22]. Therefore, we hypothesize that the GUA resume codon of Hp-SsrA is a poor resume codon for trans-translation

in E. coli and that additional downstream sequence compensate for this deficiency. As a consequence, the introduction of two stops immediately after the resume codon as in the Hp-SsrASTOP mutant might render this compensation impossible and translation restart ineffective. These data emphasize the strict constraints on SsrA sequence to achieve

ribosome rescue in a given organism. The functionality of Hp-SsrA in E. coli was also examined using the phage λimm P22 propagation test. Several studies illustrated in Table 4 conclude that λimm P22 propagation in E. coli is mainly dependent on efficient ribosome rescue and that the inactivation of the tagging activity did not affect phage growth. It was also reported that the Farnesyltransferase threshold SsrA function required for plaque formation in E. coli is fairly low [23]. Thus, the absence of phage λimm P22 propagation in the E. coli ΔssrA expressing wild type Hp-SsrA (that complements growth defect) was unexpected (Table 3). In contrast to Hp-SsrA, wild-type SsrA from Neisseria gonorrhoeae (NG-SsrA) restores phage propagation in E. coli ΔssrA [20]. Interestingly, NG-SsrA mutant versions carrying mutations affecting either the ribosome rescue function (NG-SsrAUG) or the functionality of the tag sequence (SsrADD and SsrAOchre) were defective in complementing the phage propagation in E. coli ΔssrA. This suggests that under conditions of heterologous complementation of E. coli ΔssrA either with Hp-SsrA (this work) or with NG-SsrA [20], λimm P22 phage propagation requires trans-translation-dependent protein tagging in addition to ribosome rescue. The proposition of a secondary role of protein tagging in λimm P22 propagation in E.