Integrated fecal microbiome-metabolome signatures reflect stress and serotonin metabolism in irritable bowel syndrome

Abstract

To gain insight into the complex microbiome-gut-brain axis in irritable bowel syndrome (IBS), several modalities of biological and clinical data must be combined. We aimed to identify profiles of fecal microbiota and metabolites associated with IBS and to delineate specific phenotypes of IBS that represent potential pathophysiological mechanisms. Fecal metabolites were measured using proton nuclear magnetic resonance (¹H-NMR) spectroscopy and gut microbiome using shotgun metagenomic sequencing (MGS) in a combined dataset of 142 IBS patients and 120 healthy controls (HCs) with extensive clinical, biological and phenotype information. Data were analyzed using support vector classification and regression and kernel t-SNE. Microbiome and metabolome profiles could distinguish IBS and HC with an area-under-the-receiver-operator-curve of 77.3% and 79.5%, respectively, but this could be improved by combining microbiota and metabolites to 83.6%. No significant differences in predictive ability of the microbiome-metabolome data were observed between the three classical, stool pattern-based, IBS subtypes. However, unsupervised clustering showed distinct subsets of IBS patients based on fecal microbiome-metabolome data. These clusters could be related plasma levels of serotonin and its metabolite 5-hydroxyindoleacetate, effects of psychological stress on gastrointestinal (GI) symptoms, onset of IBS after stressful events, medical history of previous abdominal surgery, dietary caloric intake and IBS symptom duration. Furthermore, pathways in metabolic reaction networks were integrated with microbiota data, that reflect the host-microbiome interactions in IBS. The identified microbiome-metabolome signatures for IBS, associated with altered serotonin metabolism and unfavorable stress response related to GI symptoms, support the microbiota-gut-brain link in the pathogenesis of IBS.

Keywords: Irritable bowel syndrome; fecal metabolome; gut metabolome; gut microbiota; gut-brain; host-microbiome interaction; microbiome; serotonin; stress.

Figure 1.

Distinguishing HC and IBS using microbiome at different taxonomic ranks, and using fecal water metabolites.

Figure 2.

Multi-omics integration to predict differences between HC and IBS.

Figure 3a.

Unsupervised clustering using kernel t-SNE (kt-SNE)

Figure 3b.

Important features for SVR/SVC models of clinical markers significantly different between clusters.

Figure 4.

Metabolites involved in reactions mediated by enzymes found in one or more of the microbial families. (a) Enzymatic reactions from all microbial species in the KEGG database that belong to the families that were identified in our microbiome analysis were included in this figure. Two metabolites were considered associated with each other if a biochemical reaction entry in KEGG indicates that they are a main reactant pair and the enzyme involved is linked to a human or microbial gene. Left column shows microbial families and Homo sapiens for reference, the right column the metabolites and the middle column the enzymes (enzyme code) that are encoded by genes from microbiota (and Homo sapiens) that mediate reactions involving one or more metabolites on the right. Microbial families and metabolites in green are increased in HCs and purple in IBS patients. This representation is based on the full metabolic reaction network in Figure S5. (b-e) Four reactions mediated by microbial enzymes that involve two or more metabolites associated with HC (green) or IBS (purple). The numbers in green and purple indicate the number of microbial families that have a gene encoding this enzyme. For web-based access of the Figure 4a, please check on the link https://imperialcollegelondon.box.com/s/ka1f3c3g64t0t1p0p6g4hgys0friaghm. In which you can point at any dot in the graph and all the lines attached to that dot are made visible. This provides perfect insight into our data and any connections that are present between different datasets.