Characterization of Bacterial and Fungal Microbiome in Children with Hirschsprung Disease with and without a History of Enterocolitis: A Multicenter Study

Abstract

Development of potentially life-threatening enterocolitis is the most frequent complication in children with Hirschsprung disease (HSCR), even after definitive corrective surgery. Intestinal microbiota likely contribute to the etiology of enterocolitis, so the aim of this study was to compare the fecal bacterial and fungal communities of children who developed Hirschsprung-associated enterocolitis (HAEC) with HSCR patients who had never had enterocolitis. Eighteen Hirschsprung patients who had completed definitive surgery were enrolled: 9 had a history of HAEC and 9 did not. Fecal DNA was isolated and 16S and ITS-1 regions sequenced using Next Generation Sequencing and data analysis for species identification. The HAEC group bacterial composition showed a modest reduction in Firmicutes and Verrucomicrobia with increased Bacteroidetes and Proteobacteria compared with the HSCR group. In contrast, the fecal fungi composition of the HAEC group showed marked reduction in diversity with increased Candida sp., and reduced Malassezia and Saccharomyces sp. compared with the HSCR group. The most striking finding within the HAEC group is that the Candida genus segregated into "high burden" patients with 97.8% C. albicans and 2.2% C. tropicalis compared with "low burden" patients 26.8% C. albicans and 73% C. tropicalis. Interestingly even the low burden HAEC group had altered Candida community structure with just two species compared to more diverse Candida populations in the HSCR patients. This is the first study to identify Candida sp. as potentially playing a role in HAEC either as expanded commensal species as a consequence of enterocolitis (or treatment), or possibly as pathobioants contributing to the pathogenesis of HAEC. These findings suggest a dysbiosis in the gut microbial ecosystem of HAEC patients, such that there may be dominance of fungi and bacteria predisposing patients to development of HAEC.

Conflict of interest statement

Competing Interests: The authors have read the journal's policy and have the following competing interests: PKF is a consultant for Karl Storz Endoscopy - America. The remaining authors declare that they have no conflicts of interest. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials.

Figures

Fig 1. Bacterial phyla in HSCR and HAEC patients.

A,16S rRNA gene sequence of fecal bacteria of nine HSCR patients and nine HAEC patients. The pie charts show average relative abundance of five major phyla and six subdominant phyla (summarized as “All others”). B, histograms demonstrating the phyla level bacterial composition of individual subjects with HSCR and HAEC. Individual subject numbers are labeled on the X axis and expressed as relative OTU abundance per each subject. Colors were assigned for each of the 11 detected phyla with the scheme at the right side.

Fig 2. Fungal genera of HAEC patients have increased Candida sp. abundance compared with HSCR patients.

A, Genera level distribution of fungi in nine HSCR patients and eight HAEC patients expressed as OTU abundance of 18S ITS-1 sequences (upper panel). Candida species composition in HSCR and HAEC patients (lower panel). B, Histograms demonstrating the fungal genera composition of individual subjects with HSCR and HAEC. Individual subject numbers are labeled on the X axis and expressed as relative OTUs abundance per each subject. Seventy-four different genera were identified by ITS-1 sequencing. The histogram shows 13 most abundant genera, unclassified genera, and 61 infrequent genera being summarized as “All others”.

Fig 3. Quantitative PCR of Candida albicans.

The quantitation of C. albicans by quantitative PCR on total fecal DNA from HSCR and HAEC patients.

Fig 4. Candida albicans and tropicalis OTU abundance by phenotype.

A, The OTU abundance of Candida in feces of HSCR and HAEC patients. Three out of eight HAEC patients showed very elevated Candida OTU’s. B, relative distribution of C. albicans to C. tropicalis in the “high burden” patients compared with the “low burden” patients. Comparisons between each group using t-test are noted. There was no significant difference between the HSCR and HAEC-Candida low burden groups.