pseudotuberculosis exoproteome. Non-classically secreted proteins Intriguingly, a much higher proportion (29.0%) of the exoproteome of the 1002 strain of C. pseudotuberculosis was composed by proteins predicted by SurfG+ as not having an extracytoplasmic location, when compared to only 4.5% in the exoproteome

of the strain C231 (Figure 2). The possibility of these proteins being non-classically secreted has been evaluated using the SecretomeP algorithm Proteasome inhibition assay [29]. We have also reviewed the literature for evidence of other bacterial exoproteomes that could support the extracellular localization found for these proteins in our study. High SecP scores (above 0.5) could be predicted for 5 of the 19 proteins in the exoproteome of the 1002 strain considered by SurfG+ as having a cytoplasmic location (additional files 2 and 3); this could be an indicative that they are actually being secreted by non-classical mechanisms this website [29]. Nonetheless, 2 of these 5 proteins ([GenBank:ADL09626] and [GenBank:ADL20555]) were also detected in the exoproteome of the C231 strain, in which they were predicted by SurfG+ as possessing an extracytoplasmic location (additional file 2). A comparative analysis of the sequences encoding these proteins

in the genomes of the two C. pseudotuberculosis strains showed that the disparate results were generated due to the existence of nonsense mutations in the genome sequence of the 1002 strain, which impaired the identification of signal peptides for the two proteins at the time of SurfG+ analysis (data not shown). We believe that it is unlikely that these differences represent true polymorphisms, as the proteins were identified in the extracellular

proteome, indicating the real existence of exportation signals. This indeed demonstrates the obvious vulnerability of the prediction tools to the proper annotation of the bacterial genomes. On the other hand, the assignment of high SecP scores to these two proteins, even though they are not believed to be secreted by non-classical mechanisms, would be totally expected, as the SecretomeP is a predictor much based on a VX-689 price neural network trained to identify general features of extracellular proteins; this means the prediction tool will attribute SecP scores higher than 0.5 to most of the secreted proteins, regardless the route of export [29]. We have found reports in the literature that strongly support the extracellular localization observed for 8 of the 14 remaining proteins considered as non-secretory by SurfG+ and SecretomeP in the exoproteome of the 1002 strain, and without any detectable signal peptide (additional files 2 and 3, Figure 2).