No abstract

No abstract

Systemic hyporesponsiveness occurs following oral administration of antigen (oral tolerance) and involves the uptake and processing of antigen by the gut‐associated lymphoid tissue (GALT), which includes Peyer's patches (PP) lamina propria lymphocytes and mesenteric lymph nodes (MLN). Animals with targeted mutations of genes in the tumor necrosis factor (TNF) family have differential defects in the development of peripheral lymphoid organs including PP and MLN, and provide a unique opportunity to investigate the role of GALT structures in the induction of oral tolerance. Oral tolerance could not be induced in TNF/lymphotoxin (LT) α–/– mice, which are devoid of both PP and MLN, although these animals could be tolerized by intraperitoneal administration of antigen, demonstrating the requirement for GALT for oral tolerance induction. LTβ–/– mice and LTα/LTβ+/– animals do not have PP but could be orally tolerized, as measured by IFN‐γ production and delayed‐type hypersensitivity responses by administration of both low or high doses of ovalbumin. To further investigate the requirement for PP, we tested the progeny of LTβ‐receptor‐IgG‐fusion‐protein (LTβRigG)‐treated mice, which do not form PP but have an otherwise intact immune system. Although these animals had decreased fecal IgA production, they could be orally tolerized. Our results demonstrate that PP are not an absolute requirement for the induction of either high‐ or low‐dose oral tolerance, although oral tolerance could not be induced in animals devoid of both PP and MLN.

Immunization of C57BL / 6 mice with myelin oligodendrocyte glycoprotein (MOG) peptide (p) 35 – 55 induces chronic experimental autoimmune encephalomyelitis (EAE). The role of γ δ T cells in the regulation of EAE is unclear. We investigated γ δ T cells in C57BL / 6 wild‐type mice and C57BL / mice with a disrupted TCRδ chain gene (δ– / – mice) using MOG p35 – 55. We found significantly less disease in δ– / – mice immunized with MOG / complete Freund's adjuvant (mean maximal EAE score 4.3 ± 0.8 in wild‐type vs. 2.3 ± 0.5 in δ– / – mice). Transfer of wild‐type spleen cells restored the ability of δ– / – mice to develop equally severe EAE as wild‐type mice. In addition to IFN‐γ, IL‐2, IL‐5 and IL‐10 was decreased in δ– / – mice. Decreased immune responses were also seen in δ– / – animals immunized with OVA peptide or protein and in concanavalin A‐stimulated splenocytes from δ– / – mice. Enriched dendritic cells from δ– / – mice secreted significantly less TNF‐α in response to lipopolysaccharide stimulation. Furthermore, when EAE was induced by adoptive transfer of an anti‐MOG p35 – 55 α β T cell line, there was a striking reduction of disease incidence (0 %) and severity in δ– / – as compared to wild‐type mice (83 % incidence). δ– / – mice showed no cellular infiltration in the spinal cord whereas wild‐type animals had infiltration of macrophages, B cells, α β‐ and γ δ T cells. In adoptive transfer EAE, there was reduced IL‐2 and IFN‐γ secretion in δ– / – mice. These results demonstrate an impaired immune response in the δ– / – mouse that is associated with a defect in developing both actively induced and adoptively transferred EAE.

We have previously demonstrated the loss of oral tolerance (OT) in lymphotoxin § −/− (LT § −/−) and TNF § /lymphotoxin § deficient (TNF § /LT § −/−) mice which have defective Peyer's patches (PP) and lymph node (LN) development. We have now studied OT in BALB/c mice with differential defects of the gut-associated lymphoid tissue (GALT) caused by inhibition of LT g R signaling during fetal development. Treatment of pregnant mice with LT g R-IgG (LT g RIgG) and TNFR I(55)-IgG (TNFR55IgG) abrogates the formation of PP (LT g RIgG) or of PP and mesenteric LN (MLN) (LT g RIgG/TNFRIgG) without genetically deleting the respective cytokine pathways. OT was readily induced in mice without PP but retaining MLN (PP null/LN+). In contrast, OT could not be induced in mice lacking both MLN and PP (PP null/MLN null) as shown by the inability of these mice to suppress IFN-+ secretion or DTH reactions. We next assessed OT in 129 × B6 LT § −/− mice with and without MLN. Timed treatment of pregnant LT § −/− mice with an agonist anti-LT g R mAb induces formation of MLN but not of PP in LT § −/− mice. LN+ LT § −/− mice developed OT while LN LT § −/− mice were resistant to OT induction. Taken collectively, the data show that in the presence of MLN PP are not required for OT induction and that the presence of MLN is sufficient for OT induction in the LT § −/− model.

Inflammatory bowel disease is associated with immune activation in Peyer's patches and mucosal lymph nodes. The role of these organs in dextran sodium sulfate (DSS)-induced colitis was investigated. We used mice lacking Peyer's patches and/or lymph nodes because of lymphotoxin-alpha gene deficiency or treatment in utero with lymphotoxin-beta-receptor IgG and tumor necrosis factor-receptor-I (55)-IgG fusion proteins. Mice lacking Peyer's patches and lymph nodes because of lymphotoxin-alpha deficiency or in utero fusion protein treatment developed more severe colitis than control mice as indicated by more severe intestinal shrinking, longer colonic ulcers, and higher histological disease scores. Oral DSS triggered the formation of colonic submucosal lymphoid patches in these mice and caused an increase in the number of submucosal lymphoid patches in mice treated in utero with the fusion proteins. Mice lacking Peyer's patches only showed more submucosal lymphoid patches whereas intestinal length and histological disease score were similar to control mice. In conclusion, more severe DSS-induced colitis correlates with the loss of the mesenteric lymph nodes. However, neither the absence of Peyer's patches nor the presence of colonic lymphoid patches were correlated with increased disease severity.

No abstract

Bacterial flagellin has recently been identified as a ligand for Toll-like receptor 5 (TLR5). Human sites known to specifically express TLR5 include macrophages and gastric and intestinal epithelium. Because infection of intestinal epithelial cells with Salmonella leads to an active transport of flagellin to the subepithelial compartment in proximity to microvessels, we hypothesized that human intestinal endothelial cells functionally express TLR5, thus enabling an active inflammatory response upon binding of translocated flagellin. Endothelial expression of TLR5 in human macro- and microvascular endothelial cells was examined by RT-PCR, immunoblot analysis, and immunofluorescence. Endothelial expression of TLR5 in vivo was verified by immunohistochemistry. Endothelial modulation of ICAM-1 expression was quantitated using flow cytometry, and leukocyte transmigration in vitro was assessed by an endothelial transmigration assay. Epithelial-endothelial cellular interactions upon infection with viable Salmonella were investigated using a coculture system in vitro. We found that Salmonella-infected intestinal epithelial cells induce endothelial ICAM-1 expression in cocultured human endothelial cells. Both macro- (HUVEC) and microvascular endothelial cells derived from human skin (human dermal microvascular endothelial cell 1) and human colon (human intestinal microvascular endothelial cells) were found to express high constitutive amounts of TLR5 mRNA and protein. These findings were paralleled by strong immunoreactivity for TLR5 of normal human colonic microvessels in vivo. Furthermore, incubation of human dermal microvascular endothelial cells with flagellin from clinical isolates of Escherichia and Salmonella strains led to a marked up-regulation of ICAM-1, as well as to an enhanced leukocyte transendothelial cell migration. These results suggest that endothelially expressed TLR5 might play a previously unrecognized role in the innate immune response toward bacterial Ags.