Using an ex vivo human organ culture model, we showed that the invasive but not the non invasive strain of S. flexneri induced significant desquamation of the IEB which was significantly reduced following infection with SepA deficient S. flexneri strains. In addition, S. flexneri also induced rapid neuronal morphological alterations suggestive of cell death. These alterations were associated with a significant increase in the proportion of VIP-IR neurons. Morphological changes induced by S. flexneri but not the increase in VIP-IR were blocked by the NMDA receptor antagonist MK-801. In addition, the absence of data describing early effects of shigellosis in humans is mainly due to the lack or scarse material from patients infected with S. flexneri prior to the development of their disease and symptoms. In this context, our model has the advantage to allow the characterization of early events following S. flexneri infection in the human colon upon both the IEB and the ENS. It also presents the advantage over reductionist cellular models to integrate various components of the mucosa such as IEC, submucosal neurons and resident inflammatory cells. One major finding of the study was the presence of a massive desquamation of surface epithelium induced by S. flexneri, as early as three hours post-infection. These results are reminiscent of observations obtained in vivo, both in animal models and in patients with shigellosis. In particular, human intestinal segments grafted into SCID mice infected with S. flexneri showed focal desquamation of the epithelial lining without ulcerations, similar to our C188-9 observations. Interestingly, the lesions of the mucosa were not reduced in polynuclear deficient mice. This is consistent with our observations since no recruitment of neutrophils can occur in the colonic explants used in our experiments. In addition, damages of the mucosa occurred in area devoid of M cells suggesting that, in our model, S. flexneri does not require entry through Peyer’s patches. A striking finding was the rapidity of the lesions induced by S. flexneri which occurred within 3 h post-infection. This could be due to the high inoculum of bacteria used and the absence of peristaltism and secretory response in our ex vivo model. In addition, a basal inflammation in the explant, probably due to hypoxic conditions generated by the culture and absence of vascularisation, might also have potentiated the mucosal alterations induced by S. flexneri. Interestingly, no bacteria was identified within intestinal epithelial cells, although they were present in some immune cells, but mainly directly beneath the shedding epithelial layer. This observation is also consistent with ND-630 in vivo animal studies showing few bacteria in the epithelium but localized mainly in the lamina propria and submucosa. Our study shows that growth-induced solid stress can affect cell phenotype, and suggests that there may be a ‘‘dynamic equilibrium’’ of proliferation and apoptosis that maintains tumor size in the plateau phase, as proposed by Holmgren et al. If only a small amount of IL-4 or 13 cross the placenta, then the level phosphorylation may be very low, and not detectable by Western Blot.