Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

P715 Assessment of community structure and predictive functional profiling of the mucosa-associated microbiome implicates alterations in benzoate metabolism in 'de novo' IBD after pouch-surgery and in treatment-naïve pediatric IBD

J. Connors1, R. Hansen2, K.A. Dunn3, M.G. Langille4, R.K. Russell2, A.R. Otley1, B. MacIntyre1, E. El-Omar5, J.P. Bielawski3, G.L. Hold5, J. Van Limbergen*1, 6

1IWK Health Centre - Dalhousie University, Pediatric Gastroenterology & Nutrition, Halifax, Canada, 2Yorkhill Hospital, Pediatric Gastroenterology & Nutrition, Glasgow, United Kingdom, 3Dalhousie University, Biology, Halifax, Canada, 4Dalhousie University, Pharmacology, Halifax, Canada, 5University of Aberdeen, Gastrointestinal Research Group, Division of Applied Medicine, Aberdeen, United Kingdom, 6Dalhousie University, Microbiology & Immunology, Halifax, Canada


Novel analytical frameworks to assess ecology and predictive function of the microbiome are now available. Post-surgical recurrence of IBD and treatment-naïve pediatric IBD are key populations to elucidate microbiome changes associated with the development of IBD. Our aims were to assess the community structure of the mucosa-associated microbiota of 'de novo' intestinal inflammation post-pouch surgery as well as treatment-naïve pediatric IBD and pediatric controls and to perform functional prediction of altered metabolic pathways.


Publicly-available 16S rRNA sequences from a cohort (n=71) of ileal pouch-anal anastomosis (IPAA) biopsies (Tyler AD et al. PLoS One 2013) were assessed for microbial features associated with post-surgical recurrence of disease/pouchitis vs non-inflammatory outcomes. For community structure analysis, we used a Bayesian analytical framework (BioMiCo - Shafiei et al. PLoS Comput Biolog 2014). Predictive functional profiling was performed using PICRUSt (Langille et al. Nat Biotechnol 2014). This approach was also applied to a pediatric IBD cohort: mucosal biopsies taken from 11 Crohn's disease and 10 ulcerative colitis patients at diagnosis were compared with 12 normal colon pediatric controls using 16S rRNA pyrosequencing (BISCUIT study by Hansen et al. Am J Gastro 2012).


Community structure mapping revealed strong overlap between pediatric IBD and controls: >75% of all pediatric samples had a posterior probability > 85% of being labeled as IBD. In contrast, 28/37 (75%) non-inflamed samples from the adult IPAA cohort, had a posterior probability >85% to correctly differentiate non-inflammatory outcomes (no pouchitis in UC/Familial Adenomatous Polyposis) from inflammatory (pouchitis/complicated pouch). PICRUSt analysis of the IPAA cohort revealed significant alterations in several metabolic pathways (p<0.05): xenobiotics degradation and metabolism (including benzoate) were upregulated in inflamed pouches. PICRUSt analysis of the pediatric cohort recapitulated the observation that benzoate degradation was increased in the colonic mucosal microbiome of treatment-naïve pediatric IBD samples compared with pediatric non-inflamed controls (p=0.02).


Unlike the community structure divergences observed in post-pouch inflammatory outcomes, there was strong overlap between treatment-naïve pediatric-onset IBD colonic mucosal microbiome and pediatric controls. However, predictive functional profiling of the mucosal microbiome of both the post-pouch and pediatric IBD cohorts showed increased benzoate degradation relative to controls. Further investigations of the role of benzoate metabolism in IBD pathogenesis are warranted.