DOP52 The faecal bacterial and fungal microbiome of newly-diagnosed, treatment naïve children with Crohn’s disease and the modifying effects of exclusive enteral nutrition and re-introduction of habitual diet

Abstract

Abstract Background The fungal community (mycobiome) and immunological responses to specific fungi have been implicated in Crohn’s disease (CD) pathogenesis and its phenotype. There is limited data on profiling the mycobiome in paediatric CD and how it changes during treatment Methods Fungal (ITS2) and bacterial (16S rRNA) microbiome analysis in faecal samples (n=124) from n=34 CD children before, during (30 and 60 days) and after exclusive enteral nutrition (EEN) at food reintroduction (within 60 days of EEN completion), and from 31 healthy controls were profiled. IgA and IgG anti-Saccharomyces cerevisiae antibodies (ASCA) analysis by enzyme-linked immunosorbent assay, in blood, was also carried out for CD patients prior to EEN initiation. Results The global mycobiome composition (β diversity) was different between CD and healthy children (R2=0.050; p=0.007) and presented lower Chao1 richness (mean 24.3 vs 33.0; p=0.005) for the CD cases; 13 fungal operational taxonomic units (OTUs) differed between the two groups. Similar signals were observed for the bacterial microbiome but the difference in global community structure (β diversity) was more profound than the fungal community (R2=0.094; p<0.001). Treatment with EEN induced drastic changes in mycobiome composition (β diversity), which were more profound and more variable than the effects observed for the bacterial community. Of the OTUs with abundance significantly lower in CD than healthy controls prior to EEN, one, belonging to S. cerevisiae, significantly decreased during EEN (compared with baseline; p<0.001 and p=0.086 at 30 and 60 days on EEN). Patients whose faecal calprotectin was low (<250 mg/kg) during food reintroduction had a lower mycobiome richness than those whose FC was raised (>250 mg/kg). 29 fungal OTUs were significantly different between these two groups in samples taken during food reintroduction, following EEN. Patients who were ASCA positive had a higher abundance of three OTUs of S. cerevisiae. In co-occurrence network analysis, Simplicillium chinense, presented significant more interactions with members of the bacterial community, particularly OTUs which changed during EEN. After participants had returned to their regular diets these microbial effects reversed and community structure moved towards that observed prior to EEN. Conclusion The mycobiome of children with CD present features of dysbiosis similar to the bacterial microbiome and improvement in disease condition during EEN is strongly associated with changes in its composition.