Effects of FODMAPs and gluten on irritable bowel syndrome- from self-reported symptoms to molecular profiling

Sammanfattning: Irritable bowel syndrome (IBS) is a complex disorder of gut-brain interactions. The diagnosis of IBS is based on subjective reporting of abdominal pain and altered bowel habits in the absence of any clinical alterations of the gut or other pathological conditions. Dietary regimens for symptom management include a low fermentable oligosaccharides, disaccharides, monosaccharides, and polyols (FODMAPs) diet and a gluten-free diet. However, scientific evidence supporting these dietary recommendations for managing IBS symptoms is weak: trials have been non-blinded and underpowered. While mechanistic understanding and objective markers of response remain scarce. Therefore, the aim of this thesis was to conduct a large double-blind study to investigate the effect of FODMAPs and gluten on symptomatic and molecular data including 16S rRNA analysis of the gut microbiota and metabolomics analyses, both at a group and subgroup (differential response) level. The resulting data served also to assess the accuracy of the Bristol Stool Form Scale (BSFS) used in IBS subtype diagnosis, and thus overcome the lack of objective evaluation of IBS symptoms. Trial data revealed that gluten caused no symptoms and FODMAPs triggered only modest symptoms of IBS, albeit with large inter-individual differences. Subjective reporting according to the BSFS conformed only modestly with stool water content in IBS, warranting caution towards IBS subtyping. FODMAPs increased saccharolytic microbial genera, phenolic-derived metabolites and 3-indolepropionate, but decreased bile acids. The genera Agathobacter, Anaerostipes, Fusicatenibacter, and Bifidobacterium correlated with increased plasma concentrations of phenolic-derived metabolites and 3-indolepropionate, i.e, metabolites related to decreased risk of incident type 2 diabetes and inflammation. Indeed, among FODMAP-related metabolites, only weak correlations to IBS symptoms were detected, as in the case of 3-indolepropionate to abdominal pain and interference with quality of life, warranting further investigation. Gluten displayed a modest effect on metabolites involved in lipid metabolism, including carnitine derivates, an acyl-CoA derivate, a medium-chain fatty acid, and an unknown lipid, but with no interpretable link to health. No molecular markers of a differential response were found, despite a comprehensive exploration with multiple analytical approaches. This could be explained by the absence of baseline variables, such as other omics layers or psychological factors, that could have determined the difference. In summary, the results indicate that gluten does not cause IBS symptoms. Moreover, the minor effect of FODMAPs on IBS symptoms must be weighed against their potential beneficial health effects. While the complexity of IBS likely explains the absence of molecular evidence for differential responses, such data analytical approach has potential where clear benefits of dietary interventions exist. Finally, the use of BSFS should include training for self-assessment, as a tool for subtyping IBS.