035 - Maternal gut microbiota-derived EVs form a distinct taxonomic entity during pregnancy

Publication Number: 35.5819

Anna Kaisanlahti, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Jenni Turunen, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Jenni Hekkala, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Surbhi Mishra, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Sonja Karikka, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Sajeen Bahadur. Amatya, University of oulu, Oulu, Pohjois-Pohjanmaa, Finland; Niko Paalanne, Oulu university hospital, Oulu, Pohjois-Pohjanmaa, Finland; Seppo Vainio, Oulu University, Oulu, Pohjois-Pohjanmaa, Finland; Anatoliy Samoylenko, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Genevieve Bart, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Leo Lahti, University of Turku, Turku, Varsinais-Suomi, Finland; Justus Reunanen, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Tejesvi Mysore V., University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland; Terhi Ruuska-Loewald, University of Oulu, Oulu, Pohjois-Pohjanmaa, Finland

Background: Human gut microbiota is a complex ecosystem essential for human health. Recent research underscores the importance of maternal gut microbiota in placenta formation and fetal development. The underlying mechanisms, however, remain poorly understood. Extracellular vesicles (EVs) are nanoparticles secreted by both gram-negative and gram-positive bacteria. EVs carry various biomolecules and cross biological barriers. Previously, we have shown that amniotic fluid contains bacterial EVs, which shows that maternal microbiota communicates with the fetus during pregnancy through EVs (Kaisanlahti et al., Microbiome, 2023).

Objective: Here, we set out investigate the taxonomy of EVs secreted by maternal gut microbiota during pregnancy and compared that to the taxonomy of maternal gut microbiota.

Design/Methods: Fecal samples were collected from pregnant women (n = 23) undergoing elective cesarean section delivery after a term pregnancy at Oulu University Hospital, Oulu, Finland. DNA and EVs were isolated from the fecal samples, followed by EV characterization and RNA isolation from EVs. Gut microbiota composition and taxonomy of gut microbiota-derived EVs was analyzed using 16S rRNA gene sequencing. Machine learning approach was used to investigate differences in the taxonomy of maternal gut microbiota and gut microbiota-derived EVs.

Results: The taxonomic composition of the gut microbiota-derived EVs differs from that of gut microbiota. Paired samples of maternal gut microbiota and gut microbiota-derived EVs showed distinct clustering when visualizing beta diversity with PCoA (figure 2A). Machine learning model was able to accurately distinguish between gut microbiota and microbiota-derived EV samples from pregnant women based on their taxonomic composition (AUC = 1.00, figure 2B). Using ANCOM-BC analysis, we identified 100 genera as differentially abundant between gut microbiota and EVs, with 67 genera enriched in EVs. The genera most important for classification in the machine learning analysis, including Porphyromonas, Lawsonella, Staphylococcus, Gemella, Alloprevotella, Peptoniphilus, Neisseria and Finegoldia, were enriched in EVs in ANCOM-BC analysis (figure 2C).

Conclusion(s): These findings indicate that gut microbiota-derived EVs form a distinct taxonomic entity from gut microbiota during pregnancy. This suggests that some bacterial strains actively secrete EVs whereas some bacteria may not interact with fetus via EVs. Our data further suggest that EVs from maternal gut microbiota, and potentially from other sites of maternal microbiota, could be a source of a distinct “EV-biome” in fetal environment.

Figure 1.Maternal gut microbiota-derived EVs in the fetal environment are likely to play an important role in the development of the fetus.
Artwork by Sonja Karikka.

Figure 2. Maternal gut microbiota-derived EVs form a distinct entity from maternal gut microbiota composition during pregnancy.
Analysis of 16S rRNA gene sequencing data from paired maternal gut microbiota samples (n = 23) and gut microbiota-derived EVs (n = 22) revealed differences in their taxonomy. Gut microbiota and EVs showed distinct clustering when visualizing beta-diversity with PcoA (A). A machine learning model was able to accurately differentiate between gut microbiota and gut microbiota-derived EV samples of pregnant women (B). The most important genera for classification of gut microbiota and gut microbiota-derived EVs of pregnant women were enriched in EVs (C).