Improved porcine model for Shiga toxin-producing Escherichia coli infection by deprivation of colostrum feeding in newborn piglets.

Porcine edema disease (ED) is a toxemia caused by enteric infection with Shiga toxin 2e (Stx2e)-producing Escherichia coli (STEC). ED occurs most frequently during the weaning period and is manifested as emaciation associated with high mortality. In our experimental infection with a specific STEC strain, we failed to cause the suppression of weight gain in piglets, which is a typical symptom of ED, in two consecutive experiments. Therefore, we examined the effects of deprivation of colostrum on the sensitivity of newborn piglets to STEC infection. Neonatal pigs were categorized into two groups: one fed artificial milk instead of colostrum in the first 24 h after birth and then returned to the care of their mother, the other breastfed by a surrogate mother until weaning. The oral challenge with 1011  colony-forming units of virulent STEC strain on days 25, 26 and 27 caused suppression of weight gain and other ED symptoms in both groups, suggesting that colostrum deprivation from piglets was effective in enhancing susceptibility to STEC. Two successive STEC infection experiments using colostrum-deprived piglets reproduced this result, leading us to conclude that this improved ED piglet model is more sensitive to STEC infection than the previously established models.

Establishment and characterization of cultured epithelial cells lacking expression of ZO-1.

In well polarized epithelial cells, closely related ZO-1 and ZO-2 are thought to function as scaffold proteins at tight junctions (TJs). In epithelial cells at the initial phase of polarization, these proteins are recruited to cadherin-based spotlike adherens junctions (AJs). As a first step to clarify the function of ZO-1, we successfully generated mouse epithelial cell clones lacking ZO-1 expression (ZO-1-/- cells) by homologous recombination. Unexpectedly, in confluent cultures, ZO-1-/- cells were highly polarized with well organized AJs/TJs, which were indistinguishable from those in ZO-1+/+ cells by electron microscopy. In good agreement, by immunofluorescence microscopy, most TJ proteins including claudins and occludin appeared to be normally concentrated at TJs of ZO-1-/- cells with the exception that a ZO-1 deficiency significantly up- or down-regulated the recruitment of ZO-2 and cingulin, another TJ scaffold protein, respectively, to TJs. When the polarization of ZO-1-/- cells was initiated by a Ca2+ switch, the initial AJ formation did not appear to be affected; however, the subsequent TJ formation (recruitment of claudins/occludin to junctions and barrier establishment) was markedly retarded. This retardation as well as the disappearance of cingulin were rescued completely by exogenous ZO-1 but not by ZO-2 expression. Quantitative evaluation of ZO-1/ZO-2 expression levels led to the conclusion that ZO-1 and ZO-2 would function redundantly to some extent in junction formation/epithelial polarization but that they are not functionally identical. Finally, we discussed advantageous aspects of the gene knock-out system with cultured epithelial cells in epithelial cell biology.