|
|
|
Back to 2015 Annual Meeting Program Disease-associated enterotypes and metabotypes in families with pediatric inflammatory bowel disease Jonathan P. Jacobs*1, Maomeng Tong2, Ian McHardy2, Maryam Goudarzi3, Paul Ruegger4, Dermot McGovern5, James Borneman4, Albert J. Fornace3, Marla Dubinsky6, Jonathan Braun2 1Division of Digestive Diseases, UCLA, Los Angeles, California, 2Department of Pathology, UCLA, Los Angeles, California, 3Georgetown University, Washington, District of Columbia, 4UC Riverside, Riverside, California, 5Cedars-Sinai Medical Center, Los Angeles, California, 6Mount Sinai Hospital, New York, New York Background: The intestinal microbiome and its metabolic products are believed to play a critical role in the pathogenesis of IBD. Existing studies of the IBD microbiome/metabolome have employed cross-sectional cohorts. We undertook a family-based study to identify microbial and metabolic features of IBD taking into account the shared environmental and genetic factors within families. Methods: Fecal samples (n=90) were collected from 21 pediatric IBD patients and their first-degree relatives for 16S sequencing (Illumina HiSeq) and mass spectrometry (Q-TOF Premier). The cohort included 36 individuals with IBD (21 probands, 6 siblings, 9 parents); of these, 26 had Crohn’s disease. Dirichlet multinomial models were used to group individuals into enterotypes and metabotypes based upon their microbial composition and metabolomic profile. Multivariate models incorporating gender, ethnicity, mode of delivery, anti-TNF therapy, and family were used to identify OTUs and spectral features associated with enterotype/metabotype and IBD. Spectral features were grouped into modules by WCGNA. Modules associated with metabotype were identified using set enrichment analysis. Results: Individuals in this cohort separated into 2 enterotypes and 3 metabotypes. Enterotype and metabotype were associated with disease state (p<10-6) and with one another. The association between enterotype and metabotype was preserved when healthy individuals were analyzed separately (p=2x10-4). Family had a statistically significant effect on metabotype in healthy individuals (p=0.008) and may have also influenced enterotype (p=0.08). The IBD-associated enterotype had increased Lachnospiraceae (Blautia, [Ruminococcus]) and decreased Ruminococcaceae (Faecalibacterium, Ruminococcus). The two enterotypes were not distinguished by Bacteroides abundance. Prevotella overgrowth was seen in only 10% of patients. 14 metabolite modules in the positive ionization mode and 4 metabolite modules in the negative ionization mode were associated with metabotype, representing 594 spectral features (212 with FDR<0.05). In contrast, only 38 spectral features were associated with IBD status in multivariate models at FDR<0.05. Validation of putative IDs was performed for two modules, revealing reduced abundance of glutamate, tryptophan, and tyrosine in the metabotype most strongly associated with IBD and reduced valine and phenylalanine in the second IBD-associated metabotype. Conclusion: Individuals in families with IBD can be divided into groups with shared fecal metabolomic profiles and microbial composition, suggesting common stable host-microbiome states. The existence of these states in healthy individuals argues that they do not simply represent the effects of intestinal inflammation. Healthy individuals who have an IBD-associated metabotype and/or enterotype may be at increased risk for developing IBD. Back to 2015 Annual Meeting Program |
||||||||
© 2024 Society for Surgery of the Alimentary Tract. All Rights Reserved. Read the Privacy Policy.