Associations between the Gut Microbiota, Urinary Metabolites, and Diet in Women during the Third Trimester of Pregnancy

Abstract

Background: Pregnancy causes many metabolic and physiologic changes. However, associations between gut microbiota, dietary intake, and urinary metabolites are poorly characterized in pregnant women.

Objectives: The research objective was to identify dietary and microbial associations with urinary metabolites during pregnancy to elucidate potential biomarkers and microbial targets to improve maternal-fetal health. This is a secondary outcome of the study.

Methods: Pregnant women (n = 27) in the Pregnancy EAting and POstpartum Diapers pilot study provided dietary intake information in addition to fecal and urine samples at 36 wk gestation. The gut microbiota was characterized following fecal DNA extraction and 16S rRNA gene sequencing. Urinary metabolites were identified using liquid chromatography high-resolution mass spectrometry.

Results: Urinary glycocholate was consistently and negatively correlated with α-carotene intake. There were 9 significant correlations between microbial taxa and urinary metabolites and 13 significant correlations between microbial taxa and dietary intake. On average, Bacteroides were the most abundant taxon in the participants' gut microbiotas. Notably, the gut microbiotas of some pregnant women were not dominated by this taxon. Bacteroides-dominant women consumed more protein, fat, and sodium, and their gut microbiotas had lower alpha diversity than those of nondominant participants.

Conclusions: Several urinary metabolites and microbial taxa were associated with maternal diet and gastrointestinal community composition during the third trimester of pregnancy. Future work should determine the mechanisms underlying the associations identified herein.

Keywords: bacteria; diet; gastrointestinal; liquid chromatography; metabolites; microbiota; pregnancy; stool; urine.

FIGURE 1

Participant recruitment and enrollment flowchart.

FIGURE 2

Urinary glycocholate is negatively correlated (Spearman; Q < 0.1) to (A) proximal and (B) habitual α-carotene intake.

FIGURE 3

Correlation plots of taxa significantly associated with (A) urinary metabolites and (B) dietary constituents. ∗Q<0.1; ∗∗Q<0.05. GABA, γ-aminobutyric acid.

FIGURE 4

Bacteroides-dominant individuals (HighB) consume more total fat, sodium, and proteinaceous foods (P < 0.05) than Bacteroidesnondominant individuals (LowB). In this study, proteinaceous foods include total meat, poultry, organ meat, cured meat, seafood, eggs, soy, and nuts and seeds, but exclude legumes.