Background: The intestinal microbiota is implicated in the pathogenesis of several immune-mediated disorders including inflammatory bowel diseases and has subsequently been the target of different therapeutic interventions. We designed the Food influence on the Intestinal microbioTa (FIT) trial to study the effects of diet on intestinal microbiota changes and inflammation in healthy individuals (part 1) and patients with ulcerative colitis (part 2). We here report the results of the first part of the study.
Methods: The FIT diet consists of a semi-vegetarian diet, high in fibre (>30g/day), low in saturated fat and sulphites and exclusion of added sugar, processed foods, carrageenan, and polysorbate-80. Following informed consent, 29 volunteers followed the diet for 1 month and were followed up for 6 months. Faecal calprotectin was measured on fresh faecal samples (Bühlmann ELISA). Dietary compliance was followed with food frequency questionnaires and 3-day food records.
16S rDNA paired-end sequencing targeting the V4 hypervariable region was performed using Illumina MiSeq sequencer. Sequencing depth was downsized to 10000 reads/sample. The RDP classifier was used for taxonomic annotation. Statistical analyses were performed with R.
Results: A significant weight loss was observed after 4 weeks following the FIT diet (t-test, p<0.0001, mean −2.3 kg, SD −1.5). Strikingly, faecal calprotectin – although within normal ranges in all but 1 individual – significantly decreased after dietary intervention (Wilcoxon test, p=0.0008) and microbial richness significantly increased (OTU observed richness, Wilcoxon test, p=0.004). There was an inverse correlation between the microbial richness at baseline and the magnitude of increase in richness following the diet (Spearman's ρ=−0.51, p=0.0113). At genus level, Roseburia decreased after the diet, although after multiple testing correction, this was no longer significant. At enterotype level, 27% of individuals which were Bacteroides at baseline shifted towards the Ruminococcus enterotype, 11% of Ruminococcus shifted towards Bacteroides and no shifts were observed in the Prevotella enterotype.
Conclusions: The FIT diet significantly increased intestinal microbial richness in healthy individuals, especially in individuals with low-richness at baseline. The Bacteroides enterotype, frequently associated with dysbiosis, was less resilient to dietary changes. Furthermore, a significant decrease in faecal calprotectin was seen after the diet suggesting additional anti-inflammatory metabolic effects beyond microbial richness and composition. A proof-of-concept study using the FIT diet is currently ongoing in patients with quiescent ulcerative colitis but recent flare, to see if the diet could prevent relapse.