Host remodeling of the gut microbiome and metabolic changes during pregnancy

Abstract

Many of the immune and metabolic changes occurring during normal pregnancy also describe metabolic syndrome. Gut microbiota can cause symptoms of metabolic syndrome in nonpregnant hosts. Here, to explore their role in pregnancy, we characterized fecal bacteria of 91 pregnant women of varying prepregnancy BMIs and gestational diabetes status and their infants. Similarities between infant-mother microbiotas increased with children's age, and the infant microbiota was unaffected by mother's health status. Gut microbiota changed dramatically from first (T1) to third (T3) trimesters, with vast expansion of diversity between mothers, an overall increase in Proteobacteria and Actinobacteria, and reduced richness. T3 stool showed strongest signs of inflammation and energy loss; however, microbiome gene repertoires were constant between trimesters. When transferred to germ-free mice, T3 microbiota induced greater adiposity and insulin insensitivity compared to T1. Our findings indicate that host-microbial interactions that impact host metabolism can occur and may be beneficial in pregnancy.

Figure 1. 16S rRNA Gene Surveys Reveal Changes To Microbial Diversity During Pregnancy

(A–G) Microbial communities clustered using PCoA of the weighted UniFrac matrix. The percentage of variation explained by the principal coordinates is indicated on the axes. The same plots are shown for A–G, except one month postpartum samples are additionally included in panel A. Each point corresponds to a community colored by first trimester (T1), third trimester (T3), or 1 month postpartum (A); pre-pregnancy BMI (B); gestational diabetes (GDM; C); trimester and birth order of expected child (D); abundance gradient of Bacteroidetes (E); abundance gradient of Firmicutes (F); and abundance gradient of Proteobacteria (G). Arrows in (D) point to samples from women who received antibiotics in T1 (orange arrows) and T2 (not T3, grey arrows). (E–G) Gradients are colored from low abundance (blue) to high abundance (red). (H) Boxplots for community richness (α-diversity) for T1 and T3 samples. For both T1 and T3, data shown are Faith’s phylogenetic diversity (PD) for 100 iterations of 790 randomly selected sequences/sample. *** P<0.0001. See Figure S1.

Figure 2. Abundances Of Phyla And Enrichment of Bacterial Genera In T1 versus T3

(A) Mean relative abundances of the phyla present in samples for T1 (left, orange bar) and T3 (right, grey bar). Colors correspond to phyla (see legend). (B) Heatmap of OTU abundances found to discriminate between T1 and T3 by machine learning. Counts were standardized (Z-score, shown in legend) prior to unsupervised hierarchical clustering of samples (columns). The color bar indicates the origin of the samples (T1: orange, T3: grey). The taxonomic assignment of each OTU is indicated to the right of the rows (OTUs; note several OTUs may share the same taxonomic assignment). See Figure S2.

Figure 3. Microbial Diversity Of T1 Samples Is More Similar To Non-Pregnant HMP Controls Than T3 Samples

PCoA of the weighted UniFrac distances between T1 (orange), T3 (grey), normal healthy HMP male (black) and female (pink) controls. Each symbol represents a sample. The percent of variation explained by the PCs is indicated in parentheses on the axes. See Figure S3.

Figure 4. High Between-Individual Microbial Diversity In T3 Persists In The Women Postpartum And Is Observed In Their Neonates

(A) Mean weighted (±SEM) UniFrac distances between bacterial communities of women (sampled at T1, T3 and 1 month postpartum) and their children (1 month, 6 months and 4 years old). Different letters on bars indicate that means are significantly different at p≤0.05. (B) PCoA plot of weighted UniFrac distance matrix, percent variation explained by PCs in indicated on the axes. Each symbol represents a child’s microbiota, colored by age. See Figure S4.

Figure 5. Comparison Of Functional VariationAnd Taxonomic Abundances In T1 And T3 Microbiomes

(A) Relative abundances of genes categories across 20 microbiomes (10 women, each sampled at T1 and T3), based on MG-RAST functional categories (numbers in legend correlate to gene categories in the KEGG pathway database, See Supplemental Information). (B) Relative taxonomic abundances based on MG-RAST taxonomic classification of shotgun reads. Colors relate to taxa in legend. Numbers at the bottom of the figures refer to mother ID. See Figure 5S.

Figure 6. Transfer Of T3 Microbiota To Germ-Free Mice Causes Greater Metabolic Changes Than T1 Microbiota

Eleven to thirteen week old germ-free mice were intra-gastrically administered inoculum from T1 and T3 human donors (5 mothers, fecal samples were pooled by trimester) and monitored for 2 weeks. (A) Mouse cecal communities clustered based on PCoA of unweighted UniFrac matrix. Each sample corresponds to a mouse cecal microbiota harvested at 15 days and colored by trimester input. Variation explained by the principal coordinates is indicated on the axes. (B) Cecal cytokine levels in recipient mice. (C) Changes in adiposity (measured by DEXA) for mouse recipients of T1 (n=6) and T3 (n=5, one outlier removed) human gut microbiota. (D) Blood glucose levels in recipient mice during oral glucose tolerance testing. Data are means ± SEM. o denotes p<0.1; * denotes p≤0.05. See Figure S6.