The Fecal Mycobiome in Chronic Pancreatitis Is Characterized by an Increase in Candida species and Nakaseomyces

Abstract

Introduction: The exocrine pancreas is an important determinant of the intestinal microbiome composition and stability. Although chronic pancreatitis (CP) is known to severely affect the bacterial community, its impact on the intestinal mycobiome is currently unknown.

Methods: A total of 93 patients with clinical and imaging evidence of CP were prospectively recruited and compared with 2 equally sized matched control cohorts. One control group was matched for age, sex, body mass index, and smoking (Con-1), and the other additionally for exocrine pancreatic function (stool elastase) and diabetes (Con-2). Fecal samples were collected from all 279 individuals to determine the fecal mycobiome by internal transcribed spacer 2 sequencing.

Results: In CP patients, fungal reads were increased (3.7-fold and 2.0-fold) as compared with Con-1 and Con-2. In comparison with Con-1, CP patients demonstrated higher total abundance of Candida (4.5-fold, q = 0.009) and higher mean relative abundance (11.4% vs 1.0%, q = 0.014) and presence (25.8% vs 9.7%, q = 0.025) of Nakaseomyces . In contrast to Con-2, CP patients showed higher Candida total abundance (1.9-fold, P = 0.016) which was, however, not significant after correction for multiple testing ( q = 0.056).

Discussion: Not only the microbiome but also the mycobiome in CP patients is characterized by distinct changes, with higher abundances of Candida or Nakaseomyces . Exocrine pancreatic dysfunction in CP patients likely contributes to this observation. This may result in increased rates of fungal infections, chronic inflammation, and could be contributing to the development of pancreatic cancer.

Keywords: acute pancreatitis; fungeome; microbiome; microbiota; pancreatic cancer.

Figure 1.

Fecal mycobiome comparison between chronic pancreatitis cases and controls. (a) Stacked bar plots show the mean absolute (left) and relative composition (right) of the fecal mycobiome of CP cases as well as Con-1 and Con-2. (b) PCoA depicting the mycobiome community structure of CP cases (blue dots) as compared with Con-1 (left, magenta dots) and Con-2 (right, yellow dots). The centroids of both groups are displayed as diamonds, and the respective samples are surrounded by a 95% data ellipse. The difference in the mycobiome composition is larger between CP cases and Con-1 than between CP cases and Con-2. R² and P values were derived from fitting the CP phenotype on the ordination, and its significance assessed by 1,000 permutations. Con-1, control group 1; Con-2, control group 2; CP, chronic pancreatitis; PCoA, principal coordinate analysis.

Figure 2.

Mycobiome changes in chronic pancreatitis. Figures show the log-fold mean abundance or presence changes comparing CP cases as compared with Con-1 and Con-2. (a) Depicts abundance changes based on absolute data whereas (b) is calculated based on relative data. (c) Shows the changes based on presence/absence data. * depicts significant results. Con-1, control group 1; Con-2, control group 2; CP, chronic pancreatitis.

Figure 3.

Candida species distribution. Stacked bar plot shows the composition of the Candida fraction within the intestinal mycobiome of CP cases and Con-1 and Con-2. In CP patients, the proportion of Candida tropicalis was increased as compared with Con-1. Con-1, control group 1; Con-2, control group 2; CP, chronic pancreatitis.

Figure 4.

Mycobiome richness and diversity comparison. Box plots show the distribution of (a) species richness (N0), (b) Shannon diversity (H), and (c) Inverse Simpson index (N2) between CP cases and Con-1 and Con-2. The 2-sided Mann-Whitney test was applied for assessment of statistical significance. No significant differences were found. Con-1, control group 1; Con-2, control group 2; CP, chronic pancreatitis.