Age and Gender Affect the Composition of Fungal Population of the Human Gastrointestinal Tract

Abstract

The fungal component of the human gut microbiota has been neglected for long time due to the low relative abundance of fungi with respect to bacteria, and only recently few reports have explored its composition and dynamics in health or disease. The application of metagenomics methods to the full understanding of fungal communities is currently limited by the under representation of fungal DNA with respect to the bacterial one, as well as by the limited ability to discriminate passengers from colonizers. Here, we investigated the gut mycobiota of a cohort of healthy subjects in order to reduce the gap of knowledge concerning fungal intestinal communities in the healthy status further screening for phenotypical traits that could reflect fungi adaptation to the host. We studied the fecal fungal populations of 111 healthy subjects by means of cultivation on fungal selective media and by amplicon-based ITS1 metagenomics analysis on a subset of 57 individuals. We then characterized the isolated fungi for their tolerance to gastrointestinal (GI) tract-like challenges and their susceptibility to antifungals. A total of 34 different fungal species were isolated showing several phenotypic characteristics associated with intestinal environment such as tolerance to body temperature (37°C), to acidic and oxidative stress, and to bile salts exposure. We found a high frequency of azoles resistance in fungal isolates, with potential and significant clinical impact. Analyses of fungal communities revealed that the human gut mycobiota differs in function of individuals' life stage in a gender-related fashion. The combination of metagenomics and fungal cultivation allowed an in-depth understanding of the fungal intestinal community structure associated to the healthy status and the commensalism-related traits of isolated fungi. We further discussed comparatively the results of sequencing and cultivation to critically evaluate the application of metagenomics-based approaches to fungal gut populations.

Keywords: antifungal resistance; commensal fungi; fungal metagenomics; fungi-host interactions; human gut mycobiota.

Figure 1

Gender-related and age-related differences in the gut mycobiota of 111 healthy volunteers. Histogram of the mean of (A) abundances and (B) the richness ± standard error of fungal isolates in female and male subjects; box-plot representation of the (C) abundance and (D) richness of fungal isolates in different age groups i.e., infants (0–2 years old), children (3–10 years old), adolescents (11–17 years old) and adults (≥18 years old). ^**p < 0.005, ^*p < 0.05, Wilcoxon rank-sum test.

Figure 2

Box-plot representation of fungal alpha-diversity measures using the number of observed OTUs between (A) genders or (B) age groups and measures of fungal beta-diversity by PCoA of the between samples distances measured using (C) the unweighted UniFrac distance and (D) the Bray-Curtis dissimilarity. ^*p < 0.05, Wilcoxon rank-sum test.

Figure 3

Stacked bar-plot representation of the relative abundances at the genus level of the fecal mycobiota of healthy subjects from metagenomics analysis distributed according to individuals' life stage and gender.

Figure 4

Box-plot representation of the comparison of fungal isolates growth ability at 37°C (control condition) vs. different stressful conditions mimicking the gastrointestinal tract challenges. ^**p < 0.0005, ^***p < 0.0001, Wilcoxon rank-sum test.

Figure 5

Box-plot representation of fungal isolates able (or not) to produce hyphae or pseudohyphae in relationship with (A) their ability to be invasive on YPD solid medium, (B) their resistance to itraconazole. ^*p < 0.05, ^**p < 0.005, ^***p < 0.001, Wilcoxon rank-sum test.

Figure 6

Spearman's r correlation analysis between antifungals susceptibility and growth ability of the tested fungal isolates under different stress conditions. Solid squares represent the degree of correlation among the variables taken into account. Crossed squares indicate non-significant correlations; significant results with p < 0.05.