P698. Characterization of the tissue-associated microbiome in active and inactive IBD
A. Tyler1, J. Stempak1, D. Gevers2, C. Huttenhower3, M. Silverberg1, 1Mount Sinai Hospital, Toronto, Canada, 2The Broad Institute of MIT and Harvard, Boston, United States, 3Harvard University, School of Public Health, Boston, United States
Recent studies have suggested a role for microbes in the etiology of Crohn's disease (CD) and ulcerative colitis (UC). The aim of this study was to evaluate the composition of the tissue-associated microflora obtained from a minimally medicated, well-characterized cohort.
Patients with UC and CD (not on antibiotics, steroids, immunomodulators or biologics) and non-IBD healthy controls (HC) were recruited from Mount Sinai Hospital, Toronto. Demographic and clinical data were collected and sigmoid biopsies obtained from all patients during standard of care endoscopic evaluation. Terminal ileum samples were available on a subset of subjects. Endoscopic appearance at biopsy site was recorded and clinical activity assessed by CDAI and Mayo scores. Microbial DNA was extracted and the V4-V5 hypervariable region of the 16S rRNA gene was sequenced using the Illumina MiSeq platform. Paired end reads were stitched, quality trimmed, assembled into OTUs with 97% sequence identity and assigned to a taxonomy using QIIME. Raw counts were converted to relative abundance, and statistical comparisons were conducted using non-parametric statistics in R with false discovery rate (FDR) correction for multiple testing, and LEfSe.
In total 179 samples (115 sigmoid, 64 terminal ileum) from 115 individuals were included in this analysis (44 HC, 34 CD, 37 UC). 15 patients with UC and 11 with CD had endoscopically confirmed active colonic inflammation. 5 patients with CD had ileal inflammation. 108 genera were detected which met the inclusion criteria. Sigmoid samples from individuals with CD had decreased levels of genera previously associated with IBD compared to HC, including Ruminococcus, Roseburia and Clostridium (pFDR < 0.05). Smaller reductions in these organisms were also detected in UC. Faecalibacterium was decreased among individuals with CD compared to HC (pFDR < 0.05). Additional organisms showed evidence of association with IBD, however with marginal significance. Several genera also demonstrated inflammation-specific changes, including Bifidobacteria, which were detected in higher frequency in samples taken from inflamed tissue of CD and UC patients compared to those from uninflamed tissue. No associations with outcome were seen in the terminal ileum.
Our results demonstrate that specific microbes including members of the Firmicutes, are associated with IBD outcome and inflammatory status. These results are further evidence for a role of microbial ‘dysbiosis’ in IBD.