Spatial and Temporal Analysis of the Stomach and Small-Intestinal Microbiota in Fasted Healthy Humans

Abstract

Although the microbiota in the proximal gastrointestinal (GI) tract have been implicated in health and disease, much about these microbes remains understudied compared to those in the distal GI tract. This study characterized the microbiota across multiple proximal GI sites over time in healthy individuals. As part of a study of the pharmacokinetics of oral mesalamine administration, healthy, fasted volunteers (n = 8; 10 observation periods total) were orally intubated with a four-lumen catheter with multiple aspiration ports. Samples were taken from stomach, duodenal, and multiple jejunal sites, sampling hourly (≤7 h) to measure mesalamine (administered at t = 0), pH, and 16S rRNA gene-based composition. We observed a predominance of Firmicutes across proximal GI sites, with significant variation compared to stool. The microbiota was more similar within individuals over time than between subjects, with the fecal microbiota being unique from that of the small intestine. The stomach and duodenal microbiota displayed highest intraindividual variability compared to jejunal sites, which were more stable across time. We observed significant correlations in the duodenal microbial composition with changes in pH; linear mixed models identified positive correlations with multiple Streptococcus operational taxonomic units (OTUs) and negative correlations with multiple Prevotella and Pasteurellaceae OTUs. Few OTUs correlated with mesalamine concentration. The stomach and duodenal microbiota exhibited greater compositional dynamics than the jejunum. Short-term fluctuations in the duodenal microbiota were correlated with pH. Given the unique characteristics and dynamics of the proximal GI tract microbiota, it is important to consider these local environments in health and disease states.IMPORTANCE The gut microbiota are linked to a variety of gastrointestinal diseases, including inflammatory bowel disease. Despite this importance, microbiota dynamics in the upper gastrointestinal tract are understudied. Our article seeks to understand what factors impact microbiota dynamics in the healthy human upper gut. We found that the upper gastrointestinal tract contains consistently prevalent bacterial OTUs that dominate the overall community. Microbiota variability is highest in the stomach and duodenum and correlates with pH.

Keywords: mesalamine; microbiota; pH; small intestine; stomach.

FIG 1

Bacterial community relative abundance and diversity in the upper GI tract. (A) The mean relative abundance of genera at each GI site (sample number [n] is indicated). (B) Box plots of the inverse Simpson index measuring community diversity across the GI tract (shown as median with first and third interquartile ranges). Statistical analysis was performed with the Kruskal-Wallis test (not significant).

FIG 2

Dissimilarity of the proximal GI tract within and across individuals. (A and B) Heat map of the Yue and Clayton dissimilarity index (θ_YC), comparing different proximal GI sites and stool across individuals (interindividual pairwise comparisons) (A) and within individuals (intraindividual pairwise comparisons) (B). (C and D) Interindividual (C) and intraindividual (D) dissimilarities in the stomach, duodenum, and jejunum (sites combined). Statistical analysis was performed with the Kruskal-Wallis test. We plot each sample at a given site rather than site averages, since this allows us to capture potential extreme states that those communities might adopt over time. Statistical analyses were performed with Dunn’s test for multiple comparisons with a Benjamini-Hochberg P value adjustment: **, P < 0.001; ***, P < 0.0001.

FIG 3

Fluctuations in prevalent OTUs observed within an individual across the proximal GI tract. (A) Box plots of the percentage of OTUs detected in a given sample out of all OTUs detected (all OTUs possible for that individual) at a subject site. (B) Box plots of the percentage of OTUs that were consistently detected at a subject site out of the total OTUs detected in a given sample. (C) The percentage of relative abundance explained by prevalent OTUs at a subject site in a given sample. Statistical analyses were performed with the Kruskal-Wallis test.

FIG 4

Longitudinal compositional dynamics, mesalamine levels, and pH in the duodenum. Shown are stream plots of genus-level composition over time in the duodenum of six individuals (percentage as indicated on the left y axis, with genera color-coded according to the color key at the bottom). White lines indicate pH measurements (black y-axis labels on the right), and red lines indicate mesalamine concentration (red y-axis labels on the right).

FIG 5

Relative abundance of significant OTUs versus pH. Shown is the log relative abundance [log₁₀(RA)] as a function of pH of OTUs found to be significantly correlated with pH using linear mixed models (all samples with measurable pH). Lines represent linear fit per OTU. OTUs classified as Firmicutes (A), Bacteroidetes (B), Proteobacteria (C), and Actinobacteria (D) are depicted. The genus-level OTU classification is defined in the color code key in each panel.