Since Shiga-toxigenic Escherichia coli (STEC) strains are not considered to be enteroinvasive, the mechanism(s) by which Shiga toxin (Stx) gains access to the circulation and to target tissues expressing its target receptor Gb3 is crucial to the disease process. There is increasing evidence that by facilitating translocation of Stx across the intestinal epithelium and by transporting bound toxin to remote sites such as the renal endothelium, polymorphonuclear leucocytes (PMNs) play a key role in the pathogenesis of serious STEC disease. Plasma levels of PMN-attracting CXC chemokines such as IL-8 also appear to correlate in humans with the severity of disease. Thus, the capacity of STEC strains to elicit CXC chemokine responses in intestinal epithelial cells may be a crucial step in pathogenesis. In order to determine which STEC factor(s) are responsible for the induction of CXC chemokine responses by intestinal epithelial (HCT-8) cells, a real-time reverse transcription PCR assay was developed to quantitatively measure relative expression of chemokine mRNA for IL-8, ENA-78, GCP-2, MGSA, MIP-2α and MIP-2β. Similarly, a commercially available sandwich ELISA was used to measure levels of IL-8 protein secreted by HCT-8 cells in response to infection with STEC. When HCT-8 cells were infected with the wellcharacterised locus of enterocyte effacement (LEE)-negative O113:H21 strain 98NK2 or the LEE-positive STEC strain EDL933, there were significant differences in the levels of chemokine mRNA and IL-8 protein expression. In particular, the LEE-negative strain 98NK2 induced significantly higher and earlier levels of chemokine mRNAs, including IL-8, MIP-2α and MIP-2β at 1 and 4 h, and ENA-78 at 4 h. However, EDL933 elicited no significant upregulation of any of the chemokine mRNAs at 1 h, and only modest increases in IL-8, MIP-2α and MIP-2β by 4 h, post-infection. These results were confirmed by IL-8 ELISA which showed that 98NK2 elicited significant levels of IL-8 protein by 2 h post-infection, and remained high until 4 h post-infection. In comparison, EDL933 did not elicit significant IL-8 induction over that of control cells, even at 4 h post-infection. When a range of STEC isolates from clinical samples were tested for their capacity to induce chemokine production in HCT-8 cells, highly significant differences were observed between the strains. Infection of HCT-8 cells with a range of LEE-negative STEC strains isolated from patients with severe STEC disease resulted in significantly higher and earlier upregulation of IL-8 and MIP-2α mRNA than that elicited by several LEE-positive STEC strains. Similarly, levels of IL-8 protein in LEE-negative STEC-infected HCT-8 culture supernatants were significantly higher than in LEE-positive STEC-infected culture supernatants. Only one LEE-positive strain, an O26 strain 95ZG1, was capable of inducing chemokine responses comparable to that induced by infection with the LEE-negative STEC strains. These results were also shown not to be attributable to differences in the adherence, initial doses or growth of the strains during the assay, or to a loss of viability of the HCT-8 cells. These results, therefore, suggest that there may be interesting differences in the ability of STEC strains to induce chemokine production in intestinal epithelial cells. The factor(s) that contribute to chemokine induction by epithelial cells in response to STEC were then examined. The difference in responses could not be attributed to the expression or non-expression of LEE genes, the presence or absence of an STEC megaplasmid or to differences in O serogroup. Although purified Stx1 and Stx2 were able to induce IL-8 and MIP-2α mRNA, and IL-8 protein, the levels of chemokine induction in response to wild-type STEC did not correlate with the type or amount of Stx produced by these strains in vitro. Similarly, deletion of the single stx2 gene from 98NK2 had no significant effect on chemokine induction compared to wild-type 98NK2-infected HCT-8 cells. Interestingly, several of the LEE-negative STEC strains eliciting the strongest chemokine responses belonged to flagellar serotype H21. Incubation of HCT-8 cells with purified H21 flagella elicited IL-8 and MIP-2α mRNA responses similar to those seen in the presence of the most potent LEE-negative STEC strains. Deletion of the fliC gene largely abolished the capacity of 98NK2 to elicit IL-8 and MIP-2α mRNA and IL-8 protein responses in HCT-8 cells. Similarly, deletion of both stx2 and fliC from 98NK2 elicited a response similar to that observed with deletion of fliC alone. Flagella were then purified from the high chemokine-inducing STEC strains 95HE4 (O91:H7) and 95ZG1 (O26:H11). Purified H7 and H11 flagella were similarly able to induce both IL-8 and MIP-2α mRNA, and IL-8 protein, in HCT-8 cells at levels similar to their respective wild-type strains. Deletion of fliC from two other STEC strains, 97MW1 (O113:H21) and 86-24 (O157:H7), confirmed that flagellin was responsible for the majority of chemokine responses in these wild-type strains. However, an inability of EDL933 to induce these responses was unexpected and later found to be due to a lack of expression of H7 flagella by this strain. Purified H21 FliC (His6-FliC) alone was able to induce chemokine production (including IL-8, MIP-2α and MIP-2β at 1 and 4 h, and ENA-78 at 4 h) by HCT-8 cells at similar levels to that observed for 98NK2. Taken together, these data suggest that although Stx is capable of inducing CXC chemokine responses, the elevated responses observed in cells infected with certain STEC strains are largely attributable to the production of flagellin. Purified His6-H21 flagellin was also able to induce p38 MAPK activation in vitro and IL-8 and MIP-2α mRNA were superinduced in the presence of both Stx2 and H21 flagellin. Blockade of the p38 pathway with SB203580 resulted in a down-regulation of IL-8 protein levels (by up to 61%) in response to H21 flagellin, but not IL-8 mRNA, suggesting that this inhibition may occur post-transcriptionally. Blocking the ERK and JNK pathways similarly decreased IL-8 secretion in response to H21 flagellin, suggesting that all three MAPK pathways are involved in this response. Indeed, concurrent inhibition of all three pathways resulted in virtually complete inhibition of IL-8 protein production (98%). Transfected HeLa and MDCK cells stably expressing TLR5 activated p38 in the presence of purified H21 flagellin, whereas dominant-negative (DN) TLR5-expressing cells did not, supporting previous studies that show that flagellin acts via TLR5. These data suggest that TLR5 and the p38, ERK and JNK MAPK pathways all play an important role in the response of intestinal epithelial cells to H21 flagellin from STEC, and that the combined effects of Stx and flagellin on host intestinal epithelial cells may result in an augmented inflammatory response. A role for flagellin in virulence was then investigated. BALB/c mice were orally inoculated with wild-type 98NK2 or 98NK2ΔfliC. Of the 16 mice challenged with the wildtype strain 98NK2, 9 (56%) died during the experiment (median survival time 7.6 days). However, only 3 of 16 mice (19%) challenged with 98NK2ΔfliC died (median survival time > 14 days). The difference in survival rate was statistically significant. No significant differences in the level of intestinal colonisation of 98NK2 or 98NK2ΔfliC were observed. Thus, flagellin directly contributes to the virulence of STEC in streptomycin-treated mice. Since the streptomycin-treated mouse is a model for systemic Stx-mediated pathology, these results suggest that the pro-inflammatory effects of flagellin play an important role in the pathogenesis of Stx-mediated STEC disease in vivo. / Thesis (Ph.D.)--School of Molecular and Biomedical Science, 2004.
| Identifer | oai:union.ndltd.org:ADTP/263855 |
| Date | January 2004 |
| Creators | Rogers, Trisha Jayne |
| Source Sets | Australiasian Digital Theses Program |
| Language | en_US |
| Detected Language | English |
Page generated in 0.0036 seconds