(P27-023-24) Minimally-Invasive Measurement of Small Intestinal and Stool Microbiomes in Humans Reveals Distinct Community Structure and Composition: A Novel Tool for Dietary Studies

Objectives: The human gut microbiota differs along the gastrointestinal (GI) tract and may offer a target for dietary intervention. Yet, few studies have directly compared the microbiota of the small intestine (SI) to that found in stool using simple, minimally-invasive sampling methods. The objective of this study was to evaluate microbiota from the duodenum and jejunum and compare to stool.

Methods: Individuals (n=6, age 58.0 ± 5.3 years, 66% White, 66% male, BMI 31.8 ± 3.9 kg/m²) were enrolled in a randomized, multi-intervention, semi-controlled feeding trial (mini-MED). At baseline, participants provided stool samples and completed fasted sampling of the SI using the EnteroTracker^® device. Briefly, individuals swallowed a 2-cm capsule with a collection string deployed along the length of the upper GI and extending into the jejunum. An external tethering string was affixed to the cheek while the internal collection string obtained a liquid mucosal biopsy during a 3-hour dwell time before retrieval. String samples from sections corresponding to the duodenum and jejunum were determined using pH analysis, presence of bile, and distance measurements. String and stool samples were stored at -80 °C until analysis. Taxonomic classifications for microbiota from each sample were generated by 16S rRNA gene sequencing. Alpha and beta diversity metrics (e.g., composition, richness, diversity) were determined and differences between SI and stool assessed via independent t-tests.

Results: Sampling of the SI was feasible and well-tolerated by participants. Duodenal and jejunal microbiotas were more similar to each other within than between individuals. The most prevalent taxa in the SI were Firmicutes/Streptococcus, Fusobacteria/Fusobacterium, and Actinobacteria/Actinomyces. The most prevalent taxa in stool were Firmicutes/Lachnospiraceae, Bacteroidetes/Bacteroides, and Firmicutes/Blautia. Mean bacterial richness (Sobs) nominally differed between SI and stool (66.5 ± 13.2 vs 55.1 ± 10.8, p=0.07), but Shannon diversity (H) did not differ (3.2 ± 0.5 vs 3.2 ± 0.9, p=0.92).

Conclusions: In this pilot feasibility study, SI and stool microbiota differed in composition and richness, but not Shannon diversity. Additional analyses will be conducted to evaluate longitudinal changes in response to diet in the ongoing mini-MED trial.

Funding Sources: NIH/NIDDK; Beef Checkoff