P006 Host-genetics, dysbiosis, and clinical history explains fecal metabolic alterations in patients with Inflammatory Bowel Disease

• Published in: Journal of Crohn's and colitis Vol. 15; no. Supplement_1; p. S128
• Main Authors: Vich Vila, A, Hu, S, Andreu-Sánchez, S, Collij, V, Jansen, D, Ruigrok, R, Abu-Ali, G, Parkinson, J, Al Garawi, A, Schneider, J, Kurilshikov, A, Björk, J, Gacesa, R, Fu, J, Zhernakova, A, Weersma, R K
• Format: Journal Article
• Language: English
• Published: 27-05-2021
• ISSN: 1873-9946; 1876-4479
• DOI: 10.1093/ecco-jcc/jjab076.135

Abstract
Background
The pathogenesis of inflammatory bowel diseases is characterized by gastrointestinal dysbiosis. Small molecules -or metabolites- present in feces can improve our understanding of host and microbes’ interactions. By exploring the relationship between host genetics, gut microbiota, and fecal metabolites composition from 500 patients with IBD and 255 population controls, we aimed to identify potential new biomarkers for the disease and to reveal host-microbiota interactions while considering factors like life-style and clinical history.
Methods
In short, we used shotgun sequencing to profile the microbiota composition, and an untargeted metabolomics platform (Metabolon INC., USA) was used to identify the levels of 1684 fecal metabolites. Whole exome sequencing and GSA Illumina chip were used to characterize host’s genetics. Multi-omics penalized regressions were used to explore the relationship between metabolite levels and the use of 44 types of medication, 144 long-term dietary patterns, and multiple patient characteristics. Random forest models were used to determine the discriminative power of metabolites.
Results
The levels of 597 fecal metabolites were altered in Crohn’s disease and 209 in patients with ulcerative colitis compared to population controls. Disease heterogeneity was reflected in the fecal metabolite profile; for example, disease location had a large impact on fecal bile-acid profile. Interestingly, metabolomics data could discriminate IBD samples from non-IBD samples (Random Forest AUC= 0.81). Genetic polymorphisms in the NAT2 gene were associated with the levels of multiple coffee metabolites. In both cohorts, IBD and non-IBD individuals, the increase in the abundance of several pathobionts was related to changes in the tryptophan, fatty-acids, and bilirubin metabolism. For example, the increase in abundance of R.gnavus was correlated with higher tryptamine levels, and the presence of bacterial starch degradation enzymes with increased concentrations of butyrate (FDR<0.05). In fact, microbial composition and bacterial enzyme abundance could predict more than 40% of the variation in the levels of 80 metabolites.
Conclusion
Patients with IBD present a different fecal metabolite profile that can be used to distinguish IBD from non-IBD individuals. Metabolite patterns are strongly related to microbial composition, diet, and patient’s clinical characteristics. Overall, these results can help to understand the disease’s complexity considering host-microbiota interaction and point to new therapeutical directions.