ABSTRACT
Live probiotic bacteria are effective in reducing gut permeability and inflammation. We have previously shown that probiotics release peptide bioactive factors that modulate epithelial resistance in vitro. The objectives of this study were to determine the impact of factors released from Bifidobacteria infantis on intestinal epithelial cell permeability and tight junction proteins and to assess whether these factors retain their bioactivity when administered to IL-10-deficient mice. B. infantis conditioned medium (BiCM) was applied to T84 human epithelial cells in the presence and absence of TNF-alpha and IFN-gamma. Transepithelial resistance (TER), tight junction proteins [claudins 1, 2, 3, and 4, zonula occludens (ZO)-1, and occludin] and MAP kinase activity (p38 and ERK) were examined. Acute effects of BiCM on intestinal permeability were assessed in colons from IL-10-deficient mice in Ussing chambers. A separate group of IL-1-deficient mice was treated with BiCM for 4 wk and then assessed for intestinal histological injury, cytokine levels, epithelial permeability, and immune response to bacterial antigens. In T84 cells, BiCM increased TER, decreased claudin-2, and increased ZO-1 and occludin expression. This was associated with enhanced levels of phospho-ERK and decreased levels of phospho-p38. BiCM prevented TNF-alpha- and IFN-gamma-induced drops in TER and rearrangement of tight junction proteins. Inhibition of ERK prevented the BiCM-induced increase in TER and attenuated the protection from TNF-alpha and IFN-gamma. Oral BiCM administration acutely reduced colonic permeability in mice whereas long-term BiCM treatment in IL-10-deficient mice attenuated inflammation, normalized colonic permeability, and decreased colonic and splenic IFN-gamma secretion. In conclusion, peptide bioactive factors from B. infantis retain their biological activity in vivo and are effective in normalizing gut permeability and improving disease in an animal model of colitis. The effects of BiCM are mediated in part by changes in MAP kinases and tight junction proteins.
Subject(s)
Bifidobacterium/metabolism , Culture Media, Conditioned/pharmacology , Animals , Bifidobacterium/chemistry , Cell Line, Tumor , Culture Media, Conditioned/chemistry , Gene Deletion , Gene Expression Regulation , Humans , Interleukin-10/genetics , Interleukin-10/metabolism , Intestines/drug effects , Intestines/microbiology , Membrane Proteins/genetics , Membrane Proteins/metabolism , Mice , Mitogen-Activated Protein Kinase Kinases/metabolism , Phosphorylation , ProbioticsABSTRACT
BACKGROUND: In the IL-10 gene-deficient mouse model, development of intestinal inflammation is associated with a defect in epithelial barrier integrity that is thought to allow sufficient passage of bacteria or bacterial antigens to initiate a mucosal immune response. Microbial monoassociation experiments into axenic animals have shown that some, but not all, endogenous bacteria will initiate an intestinal inflammatory response. For instance, Bacteroides vulgatus does not initiate intestinal inflammation in axenic IL-10 gene-deficient mice. We investigated whether B. vulgatus requires concomitant disruption of the intestinal epithelial barrier integrity in order to initiate an inflammatory response. METHODS: We first identified a dose of the indomethacin that would cause a primary disruption of the epithelial barrier without causing intestinal inflammation. IL-10 axenic mice were then administered this dose of indomethacin in their drinking water for 7 days and concomitantly monoassociated, by oral gavage, with B. vulgatus. RESULTS: Indomethacin treatment (2 microg/g/d) for 7 days resulted in disruption of epithelial barrier integrity, but it caused neither a systemic inflammatory response nor a mucosal inflammatory response in the colon or cecum. Monoassociation with B. vulgatus alone did not lead to a mucosal inflammatory response, despite a measurable systemic response. In contrast, administration of indomethacin plus B. vulgatus-monoassociation resulted in a marked intestinal inflammatory response in colon and cecum. CONCLUSIONS: Our data show that, in a genetically predisposed animal model, the nondisease-causing endogenous bacteria, B. vulgatus, is able to cause an intestinal inflammatory response provided that disruption of the intestinal epithelial barrier has occurred.
