Gastrointestinal immune-mediated diseases: what role for the microbiota and the virome

The microbiota has been already proven to modulate inflammation in the gastrointestinal tract and airways by stimulating the host epithelial cells to produce pro-inflammatory cytokines. Eosinophilic esophagitis, EoE, is a Th-2 antigen-mediated chronic inflammatory disease which is triggered by food and airborne allergens. The chronic inflammation reshapes the oesophageal epithelium, leading to difficulties in swallowing. To date, the diagnosis requires an endoscopy with the collection of at least 6 biopsies along the oesophageal tract, followed by the counting of eosinophils and the exclusion of other morbidities. The aim of this project is to better characterise the microbiome in saliva and stool of EoE patients compared with healthy controls, potentially leading to the identification of biomarkers for non-invasive diagnosis. Saliva and stool samples were collected through specific self-collection kits together with clinical and lifestyle metadata from subjects with a confirmed EoE diagnosis and from matched healthy controls, HCs. Shotgun metagenomics and metatranscriptomics were performed for saliva and stool samples. After raw reads preprocessing, the taxonomic and bacteriophages profiles, and the functional profiles were obtained with MetaPhlAn 4.1.1 and HUMAnN 3.9 respectively. Among the most significant results, EoE diagnosis significantly impacted on the salivary microbiota beta diversity, while alpha diversity directly correlated with the endoscopic severity score of the disease. Peptidiphaga genus was more prevalent in EoE compared to HCs and Capnocytophaga bilenii was enriched in active EoE compared to inactive EoE. Stool microbiota did not show significant differences in beta nor in alpha diversity according to the disease diagnosis or activity. However, an uncharacterised taxon, SGB3991, was enriched in active EoE compared to HCs. Interestingly, an uncharacterised bacteriophage, M857, was enriched in HCs compared to patients with active disease in salivary samples, while another uncharacterised bacteriophage, M406 showed a correlation with the disease activity score in stool samples. Although limited by the sample size, this study identified some interesting microbiome signatures in both saliva and stool samples. Further studies are needed to confirm and better elucidate these findings.