Microbiome-producing SCFAs are associated with preterm birth via trophoblast function modulation

Abstract

Although preterm birth (PTB) is one of the major causes of perinatal mortality and neonatal morbidity, little is known about its complex etiology. An abnormal cervicovaginal microbiome during pregnancy is associated with an increased risk of PTB. The cervicovaginal microbiota and its active metabolites, such as short-chain fatty acids (SCFAs), might be effectively used to predict and diagnose PTB. However, the roles of these proteins and the underlying mechanisms involved remain elusive. We conducted 16S rRNA gene sequencing and used a targeted metabolomics approach to study cervicovaginal swabs obtained from 51 singleton pregnancies and 52 twin pregnancies in the second trimester. Next, functional in vitro experiments were performed to investigate the roles and mechanisms of SCFAs in placental trophoblast cells (HTR8/SVneo cells). Significant cervicovaginal microbiome dysbiosis, characterized by a substantial reduction in the abundance of lactobacilli and overgrowth of anaerobes, was revealed in the second trimester and was strongly associated with subsequent PTB (P = 0.036). Among the paired samples (n = 103), acetic acid was significantly greater in the preterm group than in the term group (P = 0.047). Data obtained from integrated gas chromatography‒mass spectrometry and 16S RNA studies revealed metabolites that were distinctly associated with particular microbial communities. Gardnerella vaginalis was the species most positively associated with acetic acid content. In addition, we identified a marker set consisting of the pregnancy type, acetic acid concentration, and community state type to accurately diagnose PTB. Acetate was associated with increased interleukin (IL)-8 and IL-6 levels and extravillous trophoblast cell migration and invasion through the activation of the extracellular signal-regulated kinase 1/2 signaling pathway in HTR8/SVneo cells. Cervicovaginal microbiota dysbiosis is an important etiological factor of PTB. The cervicovaginal microbiota and its active metabolites can be efficiently used to predict and diagnose PTB. Our findings enrich the microbiota-placenta axis theory and contribute to the development of microecological products for pregnancy.

Importance: Preterm birth (PTB) is a leading cause of infant mortality and long-term health issues, affecting millions of families worldwide. Despite its prevalence, the exact causes of PTB remain unclear. Our study reveals that certain bacteria and their metabolic byproducts in the cervicovaginal environment, specifically short-chain fatty acids (SCFAs), are linked to the risk of preterm birth. By analyzing samples from pregnant women, we found that an imbalance in the vaginal microbiota and increased levels of SCFAs are associated with changes in cells that can lead to early labor. This research provides new insights into how the microbiome influences pregnancy outcomes and highlights potential biomarkers for predicting preterm birth. Understanding these microbial influences could lead to innovative strategies for early diagnosis and prevention, ultimately improving maternal and infant health.

Keywords: 16S rRNA; SCFAs; cervicovaginal microbiota; pregnancy; preterm birth.

Fig 1

Heatmap of the relative abundance of microbial taxa and four main SCFAs identified in the cervicovaginal microbiota of 103 pregnant women. (A) Schematic representation of the study design. (B) Ward linkage hierarchical clustering of the Jensen–Shannon metric identified five community state types (CST I, II, III, IV, V). The upper color bar shows the three groups of community state types, while the lower color bar shows each sample’s pregnancy outcome and type (PTB, preterm birth; TB, term birth; TP, twin pregnancy; SP, singleton pregnancy). Colored bars represent the relative abundance of microbial taxa in each replicate sample. The color intensity of the heatmap increases with the taxa’s relative abundance from low (blue) to high (red).

Fig 2

Relationships between CVF microbiota communities and pregnancy outcomes. (A) The α diversity of the microbiome was determined by the Simpson index and Shannon index values. *P < 0.05 and **P < 0.01. (B) β diversity of the microbiome was determined using PCoA analysis based on unweighted (left) and weighted (right) UniFrac distances. The P value corresponding to the PERMANOVA test was less than 0.05. The microbial taxa of the genes and their relative abundance in each group at the genus (C) and species (D) levels. (E) Stacked bar charts showing that the distribution of CSTs differed between TB and PTB patients. P for trend <0.05 for linear-by-linear association. PTB, preterm birth; CST, community state type; CST I, Lactobacillus crispatus-dominant; CST II, Lactobacillus gasseri-dominant; CST III, Lactobacillus iners-dominant; CST IV, Lactobacillus spp.-depleted; CST V, Lactobacillus jensenii-dominant.

Fig 3

Correlation analysis of taxa and metabolites. (A) The acetic acid concentration in the preterm group was significantly greater than that in the term group (n = 103, P = 0.0152). (B) SP-PTB vs. SP-TB, P = 0.0341; SP-TB vs. TP-TB, P = 0.0208. (C) Heatmap showing Spearman’s correlation between metabolites and microbiota composition at the genus level. Asterisks indicate a statistically significant correlation at the level of (*: P value < 0.05; **: P value < 0.01; ***: P value < 0.001), and the colors denote positive (red) and negative (blue) correlation values. (D) Network plot highlighting the highly correlated metabolites and microbiota. The figure shows the network diagram based on Spearman correlation analysis to calculate the correlation between species and metabolite data, and the relationship pairs with P values < 0.05 were selected. The red line indicates a positive correlation, and the green line indicates a negative correlation. The thickness of the line indicates the correlation coefficient. (E) Receiver operating characteristic (ROC) curve areas and AUC for predicting spontaneous preterm birth in patients in the composite models. PT, pregnancy type; A, acetic acid.

Fig 4

Expression and distribution of FFAR2 in placental tissue. (A) H&E staining and immunolocalization of HLA-G, CK7, and FFAR2 in human placental villi from pregnant women (bar, 250 µm/50 µm). (B) HE staining and immunolocalization of HLA-G, CK7, and FFAR2 in full-term human placentas from normal pregnancies. (Bar, 250 µm/50 µm). The brown color indicates positive staining for HLA-G, CK7, and FFAR2. CCT, cell column trophoblast; STB, syncytiotrophoblast; CTB, cytotrophoblast. (C) Immunofluorescence detection of FFAR2 (green) and HLA-G (red) in the placenta of first trimester. Nuclei were counterstained with DAPI (blue). Merge, merge of FFAR2, HLA-G, and DAPI. (D) Western blotting of FFAR2 in first-trimester human villi and full-term placentas from normal pregnancies. Representative Western blot images are shown for FFAR2 and GAPDH. ***P < 0.001 indicates that the difference between the two groups was statistically significant.

Fig 5

Effects of acetate on HTR-8/SVneo cells. (A) Growth analysis of HTR-8/SVneo cells was performed after treatment with acetate for 16 h. EdU (pink) staining of HTR-8/SVneo cells treated with various concentrations of acetate (10 mM, 100 mM). Nuclei were counterstained with DAPI (blue). Scale bars, 200 mm. (B) Acetate induces inflammatory cytokine secretion in HTR8/SVneo cells. Migration (C) and invasion (D) were analyzed in HTR8/SVneo cells with or without acetate treatment. Migration (C) and invasion (D) were analyzed in HTR8/SVneo cells inhibited by treatment with GLPG0974, a FFAR2 inhibitor. Four fields of view were examined, and the number of cells in the field of view was calculated. Four fields of view were examined, and the number of cells in the field of view was calculated. The cell numbers are presented as the means ± SEM of 3 independent experiments performed in triplicate. (one-way ANOVA, *P < 0.05, **P < 0.01, ***P < 0.001).

Fig 6

The ERK signaling pathway is implicated in migration and invasion through the induction of acetate. (A) The ERK inhibitor (PD98059/PD0325901) inhibited HTR8/SVneo cell migration after pretreatment for 1 h. (B) Four fields of view were examined, and the number of cells in the field of view was calculated. The cell numbers are presented as the means ± SEM of 3 independent experiments performed in triplicate. (one-way ANOVA, *P < 0.05). (C) WB analysis of total and phospho-ERK1/2 levels in trophoblast cell lines. (D) Cartoon plot showing the mechanism by which acetate promotes the migration of HTR8/SVneo cells through the ERK pathway.