• 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • While we previously demonstrated that TraG is inhibited by


    While we previously demonstrated that TraG is inhibited by specific transglycosylase inhibitors and that enzymatic activity is strongly reduced upon mutation of the potential catalytic core (Arends et al., 2013), the enzymatic mechanism of both domains was not assigned. Here, by mass spectrometry analyses of characteristic peptidoglycan digestion products generated by the TraG variants, we could show that TraG has both lytic transglycosylase and endopeptidase activity, with the CHAP domain alone displaying endopeptidase activity. We therefore suggest that the lytic transglycosylase activity could be associated with the proposed SLT domain of TraG. The absence of typical lytic transglycosylase cleavage products in the TraG SLT variant lacking the CHAP domain either suggests complete loss or dramatically reduced lytic transglycosylase activity of the SLT domain. This could for instance be due to misfolding or limited substrate accessibility of the isolated SLT domain of TraG and is in line with results obtained in complementation experiments with the truncated variant TraGΔCHAP. With a bacterial-2-hybrid approach we observed self-interaction of both TraG and TraM. TraM participation in the translocation channel has been postulated, since TraM shows structural similarity to A. tumefaciens VirB8 protein that contributes to formation of the inner membrane complex (Goessweiner-Mohr et al., 2013, Trokter et al., 2014). While insertion of the complex likely requires GDC-0032 what degrading enzymes, a role for TraG in complex assembly remained elusive. We could show a direct interaction between TraG and TraM and a similar localization pattern of both proteins in defined spots in the E. faecalis cell membrane, possibly representing the actual T4SS complex. To verify the bacterial-2-hybrid results, additional biochemical assays will be necessary. Hence, we propose a crucial role for TraG in the translocation channel assembly and hypothesize its involvement in the T4SS complex. For G+ bacteria participation of VirB1-like proteins, such as PrgK, in the T4SS complex has been already proposed (Bhatty et al., 2013, Laverde Gomez et al., 2014). In cells devoid of TraG, the distinct TraM foci at the cell membrane disappeared and a rather diffuse distribution was observable. Further analysis of TraM localization by proteinase K digestion of cell surface proteins revealed correct localization of TraM at the cell membrane in wild type but not in traG mutant cells. Interestingly, this mislocalization was also present in the traG knockout complemented with TraGΔTMH, whereas complementation with TraGΔCHAP revealed correct localization of TraM. Thus, proper localization of TraM depends on the transmembrane helix of TraG, while the enzymatic activity apparently is not required for TraM localization at the cell membrane, even though conjugative transfer does not occur in knockout cells with in trans expression of TraGΔCHAP. This indicates an essential role of TraG, likely as scaffolding factor in the localization process of TraM and possibly in the assembly of the translocation channel. Interestingly, TraM levels in general were significantly decreased in traG-knockout cells. This might be due to degradation of TraM, as a functional T4SS apparatus cannot be formed due to lack of local opening in the peptidoglycan caused by missing TraG activity. This observation coincides with the results of the mating assays, where traG deletion led to fully abolished plasmid transfer. Since traG proved to be an essential gene and its enzymatic activities could be assigned, inhibition of T4SS associated peptidoglycan degrading enzymes might be a promising strategy to inhibit conjugative transfer in G+pathogens.
    Competing interest
    Funding This work was supported by a grant from the Austrian Science Foundation (FWF project P27383) to W.K. Work in the E.G. lab was supported by DLR grant 50WB1166. S.B. lab was supported by the Austrian Science Foundation FWF (grant P27183-B24) and the Swedish Research Council Vetenskapsrådet (grant 2015-05468).