ATH-09 Metabolomics & multi-omics analysis of Crohn's disease. Issue 2 (June 2019)

• Record Type: Journal Article
• Authors: Frau, Alessandra; Hough, Rachael; Ijaz, Umer; Campbell, Barry; Kenny, John; Hall, Neil; Anson, Jim; Darby, Alistair; Probert, Chris
• Is Part Of: Gut. Volume 68:Issue 2(2019)
• Page Start: A68
• Page End: A68
• Publication Date: 2019-06
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2019-BSGAbstracts.131
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Background: Faecal metabolomic studies show that CD patients have specific volatile organic compounds (VOCs) profiles different from healthy subjects. Integration of metabolomics with bacterial and fungal data is now possible and allows the correlation of metabolites and the microorganisms that are likely involved in their production. We present metabolomics data from a cohort of CD patients and integrate this with microbiome data to assess which microorganisms are likely active in the disease. Methods: Stool samples from 43 donors (23 CD and 20 controls) were analysed. Briefly, gases from 450–500 mg faecal aliquots were analysed by headspace, solid phase micro extraction gas chromatography/mass spectrometry. Data were interpreted using AMDIS with NIST reference library. Statistical analysis was performed in Metaboanalyst. VOCs data were then integrated with bacterial 16S rRNA and fungal 18S rRNA data from the same cohort with DIABLO (MixOmics). This uses supervised analysis to highlight signature features and to identify correlated variables. Results: Analysis of VOCs showed that CD patients formed a separate cluster and several metabolites were increased in CD including VOCs related to fungi: 3, 7-dimethyl-1, 6-octadien-3-ol and octanal. Branched chained fatty acids were also increased in CD, along with esters, butanoic acid, nonanal and indole. Multi-omics integration analysis comparing CD and controls showed that branched-chain fatty acids (high in CD) were correlated with gut fermenters, mainly Firmicutes. A second comparison saw CD active (n= 11) vs controls (n=20). Correlation of signature variables showed that OTUs assigned to Saccharomycetales yeasts and a mould ( Aspergillus ) were correlated to fungal metabolites (heptanal and 3, 7-dimethylocta-1, 6-dien-3-ol), supporting a possible role of fungi in active CD. These fungi were also correlated to Clostridiales and Enterobacteriales. This last model gave interesting results, however, the unbalance in number of samples between the two categories contributed to give a balanced error rate (BER) of the model relatively high (≈45%), therefore these data are not definitive. Conclusion: This is the first study to integrate metabolomics, microbiome and mycobiome data in CD: its potential is evident. We were able to pinpoint which microorganisms are likely active in disease and understand which produce metabolites of interest. The metabolomics data are consistent with earlier literature. The correlation of fungi metabolites with fungal species is also very relevant. The high BER does not allow us to draw definite conclusions and further studies, with larger cohorts, are required. However, we can say that fungi are very likely to be active during relapse.