Maternal gut microbiota Bifidobacterium promotes placental morphogenesis, nutrient transport and fetal growth in mice

Abstract

The gut microbiota plays a central role in regulating host metabolism. While substantial progress has been made in discerning how the microbiota influences host functions post birth and beyond, little is known about how key members of the maternal gut microbiota can influence feto-placental growth. Notably, in pregnant women, Bifidobacterium represents a key beneficial microbiota genus, with levels observed to increase across pregnancy. Here, using germ-free and specific-pathogen-free mice, we demonstrate that the bacterium Bifidobacterium breve UCC2003 modulates maternal body adaptations, placental structure and nutrient transporter capacity, with implications for fetal metabolism and growth. Maternal and placental metabolome were affected by maternal gut microbiota (i.e. acetate, formate and carnitine). Histological analysis of the placenta confirmed that Bifidobacterium modifies placental structure via changes in Igf2P0, Dlk1, Mapk1 and Mapk14 expression. Additionally, B. breve UCC2003, acting through Slc2a1 and Fatp1-4 transporters, was shown to restore fetal glycaemia and fetal growth in association with changes in the fetal hepatic transcriptome. Our work emphasizes the importance of the maternal gut microbiota on feto-placental development and sets a foundation for future research towards the use of probiotics during pregnancy.

Keywords: Bifidobacterium; Fetus; Metabolism; Microbiota; Pregnancy.

Fig. 1

Effects of the maternal gut microbiome and B. breve supplementation during pregnancy on fetal viability, growth and hepatic transcriptome. A Experimental design. B Number of viable fetuses per litter (One-way ANOVA with Tukey’s multiple comparison). C Fetal weight. D Circulating glucose concentrations in fetal blood. E Fetal organ weights. F–G RNA-Seq analysis of fetal liver samples obtained at GD18.5. F PCA plot and G volcano plots showing up and down-regulated differentially expressed genes (DEGs) in BIF group (compared to GF group). H Heat map of the 20 most up and down-regulated DEGs (BIF group). I, J Functional profiling (g:Profiler) on 602 DEGs. Key enriched GO terms and REACTOME pathways are shown in the figure. Fetal data are obtained on GD16.5 from: SPF (49 fetuses/6 dams), GF (33 fetuses/5 dams), BIF (34 fetuses/6 dams). Dots represent raw data for each variable assessed (individual values). However, the statistical analysis and the mean ± SEM reported in the graphs were obtained with a general linear mixed model taking into account viable litter size as a covariate and taking each fetus as a repeated measure followed by Tukey multiple comparisons test (further explanations can be found in the Materials and Methods, statistical analysis section). Identification of outliers was performed with the ROUT Method. RNA-seq was performed on fetal livers obtained at GD18.5 from a total of 3 GF and 4 BIF pregnant dams/litters. RNA-Seq data analysis is described in the material and methods section (NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001)

Fig. 2

Effects of maternal gut microbiome and B. breve supplementation during pregnancy on placental structure on day 16.5 of gestation. A Placenta weight. B Placental efficiency determined by dividing fetal by placental mass. C Placental regional analysis. Scale bar = 1 mm. D Representative staining of placental glycogen with PAS and glycogen abundance. Scale bar = 2.5 mm and 250 μm. E Representative image of lectin and cytokeratin staining for labyrinth zone structural quantification. Scale bar = 500 μm and 50 μm. F–I Stereological parameters determined in placental labyrinth zone. J Representative image of TUNEL staining for apoptosis quantification in labyrinth zone. Scale bar = 2.5 mm and 100 μm. All data were analyzed by a general linear mixed model, taking into account litter size as a covariate and taking each fetus as a repeated measure followed by Tukey multiple comparisons test. ROUT test was used for the identification of outlier values. Dots represent raw data (individual values). However, the statistical analysis and the mean ± SEM reported within the graphs were obtained with the general linear mixed model (further explanations can be found in the Materials and Methods statistical analysis section). Placental weight-efficiency was obtained from: SPF (49 fetuses/6 dams), GF (33 fetuses/5 dams), BIF (34 fetuses/6 dams). Laboratorial analysis was performed with: SPF (14–15 placentas from 6 dams), GF (10 placentas from 5 dams) and BIF (9–11 placentas from 6 dams). Only placentas collected on day 16.5 of gestation were analysed. One to three placentas per litter were randomly selected and used for assessment. Placentas were analysed blind to the experimental groups. (NS, not significant; *P < 0.05; ***P < 0.001). D decidua, Jz junctional zone, Lz labyrinth zone, TB trophoblasts, FC fetal capillaries, MBS maternal blood spaces

Fig. 3

Effects of maternal gut microbiome and B. breve supplementation during pregnancy on placental gene and protein levels on day 16.5 of gestation. A Gene expression levels in micro-dissected labyrinth zones. B Immunoblots and protein quantification by Western blot in micro-dissected labyrinth zones. C–E Gene expression levels in micro-dissected labyrinth zones for amino-acids, glucose and lipid transporters. Western blot data were analysed by one-way ANOVA. qPCR data were analyzed by a general linear mixed model, taking into account litter size as a covariate and taking each fetus as a repeated measure followed by Tukey multiple comparisons test. ROUT test was used for the identification of outlier values. Dots represent raw data (individual values). However, the statistical analysis and the mean ± SEM reported within the graphs (for qPCR data) were obtained with the general linear mixed model (further explanations can be found in the Materials and Methods, statistical analysis section). Gene expression analysis was performed with: SPF (13 placentas from 6 dams), GF (11 placentas from 5 dams) and BIF (14 placentas from 6 dams). Protein quantification was performed with: SPF (4 placentas from 4 dams), GF (5 placentas from 5 dams) and BIF (5 placentas from 5 dams). Only placentas collected on day 16.5 of gestation were analysed. For qPCR, one to three placentas per litter were assessed and selection of the samples was conducted at random. For protein expression analysis, 1 placenta per litter was selected (NS, not significant; *P < 0.05; **P < 0.01; ****P < 0.0001)

Fig. 4

Metabolomic profiling of placental labyrinth zone and fetal liver on day 16.5 of gestation. Data were analysed by Kruskal–Wallis test followed by multiple comparisons using the Benjamini & Hochberg false discovery rate method and Dunn's test. ROUT test was used for the identification of outlier values. Data presented as mean ± SEM. Number of litters analysed per group: SPF (8 placentas-livers from 4 dams), GF (8–7 placentas-livers from 4–5 dams), BIF (6–7 placentas-livers from 5 dams). Only tissues collected at GD16.5 were analysed. Selection of the samples was conducted at random (NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001)

Fig. 5

Summary illustration showing the most relevant results on how the maternal gut microbiota and B. breve affects mother, placenta and fetus during gestation. The effects of lacking maternal gut microbiota on maternal, placental and fetal phenotype are shown in red circles (SPF vs GF comparisons). Our results suggest that lacking maternal gut microbiota aside from inducing changes in the maternal digestive tract, pancreas and caecum metabolites, has important implications for the correct growth of the fetus and its placenta. The effects of B. breve administration compared to the SPF and GF groups are shown in blue and red arrows, respectively. Overall, B. breve induces changes in the maternal compartment that affect the structure, metabolome and function of the placenta in association with alterations in fetal metabolism, growth and hepatic transcriptome. SPF specific-pathogen-free mouse, GF germ-free mouse, BIF germ-free mouse treated with B. breve UCC2003, Lz labyrinth zone, MBS maternal blood spaces, FC fetal capillaries, SA surface area for exchange, BT barrier thickness, DEG differentially expressed genes