Surprisingly, MK0683 BLG production was not increased. These results partly confirmed what we published recently with LL-FnBPA+BLG in vitro and in vivo[32]. Oral administration in mice of LL-FnBPA+BLG or LL-FnBPA+GFP elicited a GFP or BLG production in enterocytes. As with LL-mInlA+ the BLG production was not increased with LL-FnBPA+. However the number of mice producing BLG was significantly higher after oral administration of LL-FnBPA+BLG compared to non invasive LL-BLG. Considering these results

it seems that LL-FnBPA+strain is a better DNA delivery vehicle than LL-mInLA+. As no significant advantages were observed by using LL-mInlA+BLG compared to LL-BLG, we hypothesize that interactions of recombinant mInlA with their receptors were impeded in mouse intestinal epithelium. This lack of invasion in vivo was also observed by another group working with E. coli strain expressing invasin from Yersinia pseudotuberculosis as an oral vaccine for

cancer immunotherapy. They showed that invasive E. coli was unable to enter gut epithelial cells due to a basolateral localization of the receptor, B1-integrin [34]. They demonstrated that invasive E. coli expressing Y. pseudotuberculosis invasin were selectively uptaken from the intestinal lumen into Peyer’s patches MX69 mw using an ex vivo model. Similarly, E-cadherin, the mInlA receptor, is also expressed on the basolateral membrane of IECs which are strongly linked to each other in the gut making E-cadherin less available. It has been shown recently that L. monocytogenes could enter the epithelial membrane through extruding epithelial cells at the top of the villi but mainly

through goblet cells which are located deeper in the crypt [35]. It is thus possible that LL-mInlA+BLG strain is not able to reach its receptor deeply buried in the crypt. The pathway whereby bacteria could penetrate gut epithelial monolayers could be through Microfold (M) cells in Peyer’s patches. These cells are able to take up particles/bacteria from the lumen [36]. Nevertheless, we cannot exclude any possibility that lactococci can also interact with other cells from the epithelial membrane such as dendritic cells. Some subset of dendritic cells is now well selleck chemical known to produce dendrites, able to reach the lumen in order to sample its content [37]. The other hypothesis is that the plasmid would be released in the lumen by lysed lactococci and then captured by the enterocytes. It has been shown that lactococci do not persist in the gut and are very sensitive to its physico-chemical condition [38]. Most likely, plasmid transfer in vivo is a combination of both mechanisms, bacteria and released plasmid captures. Considering these data, the use of lactobacilli which persist longer in the gut than lactococci could be a better HDAC inhibitors cancer option for DNA delivery. Conclusions Mutated Internalin A protein was successfully expressed at the surface of L. lactis NZ9000, as demonstrated by FACS analysis.