Modelling the Gut Fungal-Community in TIM-2 with a Microbiota from Healthy Individuals

Abstract

Most research on the human microbiome focuses on the bacterial component, and this has led to a lack of information about the fungal component (mycobiota) and how this can influence human health, e.g., by modulation through the diet. The validated, dynamic computer-controlled model of the colon (TIM-2) is an in vitro model to study the microbiome and how this is influenced by interventions such as diet. In this study, it was used to the study the gut fungal-community. This was done in combination with next-generation sequencing of the ITS2 region for fungi and 16S rRNA for bacteria. Different dietary interventions (control diet (SIEM), high-carbohydrate, high-protein, glucose as a carbon source) were performed, to see if diet could shape the mycobiome. The mycobiome was investigated after the adaptation period, and throughout the intervention period which lasted 72 h, and samples were taken every 24 h. The fungal community showed low diversity and a greater variability when compared to bacteria. The mycobiome was affected most in the first hours of the adaptation period. Taxonomic classification showed that at the phylum-level Ascomycota and Basidiomycota dominated, while Agaricus, Aspergillus, Candida, Penicillum, Malassezia, Saccharomyces, Aureobasidium, Mycosphaerella, Mucor and Clavispora were the most abundant genera. During the intervention period, it was shown that the change of diet could influence the diversity. Clustering of samples for different time points was analyzed using Bray–Curtis dissimilarities. Samples of t0 clustered together, and samples of all other time points clustered together. The Bray–Curtis-dissimilarity analysis also showed that for the different dietary interventions, samples treated with glucose clustered together and were different from the other groups (p < 0.05, PERMANOVA). Taxonomic classification showed that the genera Alternaria, Thanatephorus, Candida and Dekkera differentially changed for the various diet groups (p < 0.05, Kruskal–Wallis). These results show that the mycobiota could be modelled in TIM-2; however, the low diversity and high variability make studying fungal, as compared to bacterial, communities, much more challenging. Future research should focus on the optimization of the stability of the fungal community to increase the strength of the results.