P831 The role of adherent-invasive E. coli in pediatric Crohn’s disease: cause or effect?

N. Faqerah, Michael. Logan, Richard. Russell, Konstantinos. Gerasimidis, Daniel. Walker

University of Glasgow, Bacteriology, Glasgow, UK

Background

The ability of adherent-invasive E. coli (AIEC), to adhere and invade intestinal epithelial cells and survive autophagy implicates them in the pathogenesis of Crohn’s disease (CD). Previous studies focused mostly on differences in the abundance of E. coli between patients with CD and healthy controls. Here, we studied changes in E. coli levels, E. coli strains and their ability to utilise carbon sources during treatment with exclusive enteral nutrition (EEN), at food reintroduction, and compared with healthy controls.

Methods

E. coli strains were isolated from culturing stool samples of twelve children with CD before and during induction treatment with EEN and following that, at food reintroduction. 10 healthy children acted as controls. 69 samples (CD, n = 59; healthy children, n = 10) were investigated in total. Absolute concentration of E. coli was measured with qPCR, changes in E. coli strains using colicin sensitivity spot tests and their ability to utilise carbon sources with Biolog phenotype microarrays.

Results

There was no significant change in the absolute levels of E. coli in patients with CD before or during EEN, and after food reintroduction. In comparison to healthy children, patients with CD had significantly higher levels of E. coli (p-value= 0.001). There were changes in E. coli at a strain-level in nine out of twelve CD patients as tested using colicin sensitivity tests before, during EEN, and after food reintroduction. Biolog microarrays showed a significant difference in E. coli community utilisation of 17 carbon sources in CD patients at different time points. Utilisation of polysaccharides, D-cellobiose, sucrose, and D-raffinose was increased during EEN compared with treatment initiation and after treatment cessation, at food reintroduction. In contrast, there was a significant reduction in utilisation for L-fucose and L-Rhamnose monosaccharides, sugar alcohols like D-glucose 6-PO4, and fatty acids such as propionic acid and acetoacetic acid by the end of EEN compared with treatment initiation and at reintroduction of habitual diet. In comparison to healthy controls, E. coli utilisation of different carbon sources in patients with CD was significantly different compared with CD patients during EEN only.

Conclusion

Children with CD experience changes in E. coli strains and in their ability to utilise certain carbon sources during treatment with EEN and compared with healthy controls. Analysis of strain-level metagenomic data will provide a comprehensive depiction of changes in E. coli population during EEN and compared with the healthy status.