Preterm Birth Is Correlated With Increased Oral Originated Microbiome in the Gut

Abstract

Background: Preterm birth is one of the leading causes of perinatal morbidity and mortality. Gut microbiome dysbiosis is closely related to adverse pregnancy outcomes. However, the role of the gut microbiome in the pathogenesis of preterm birth remains poorly studied.

Method: We collected fecal samples from 41 women (cases presenting with threatened preterm labor =19, 11 of which delivered preterm; gestational age-matched no-labor controls, all of which delivered at term = 22) were recruited for the study. We performed 16S rRNA amplicon sequencing to compare the composition of the gut microbiome in threatened preterm labor cases and controls and among women who delivered preterm and at term. By annotating taxonomic biomarkers with the Human Oral Microbiome Database, we observed an increased abundance of potential oral-to-gut bacteria in preterm patients.

Results: Patients with preterm birth showed a distinct gut microbiome dysbiosis compared with those who delivered at term. Opportunistic pathogens, particularly Porphyromonas, Streptococcus, Fusobacterium, and Veillonella, were enriched, whereas Coprococcus and Gemmiger were markedly depleted in the preterm group. Most of the enriched bacteria were annotated oral bacteria using the Human Oral Microbiome Database. These potential oral-to-gut bacteria were correlated with clinical parameters that reflected maternal and fetal status.

Conclusions: This study suggests that patients who deliver preterm demonstrate altered gut microbiome that may contain higher common oral bacteria.

Keywords: bacteria translocation; gut microbiome; maternal gut microbiome; preterm birth; preterm birth subtypes.

Figure 1

The diversity and composition of the gut microbiome. (A) Observed ASVs of all groups. Observed ASVs between the healthy group and the preterm group (Wilcoxon rank-sum test), along with the preterm symptoms delivered group (Sym PTB), late-week preterm group (Late PTB) and early week preterm group (Early PTB). (Subgroups are compared with healthy group using Wilcoxon rank-sum test and adjusted by the Benjamini and Hochberg method). (B, C) Bray–Curtis distances PCoA of all groups. PCoA of Bray–Curtis distances for the bacterial community structure of the gut microbiome between the healthy group and the preterm group (B), the healthy group and the Sym PTB group and the late PTB group and the early PTB group. (C) The eigenvalues of axe PC1 and PC2 were 0.29 (17.781%) and 0.46 (11.029%), respectively. The eigenvalues of axe PC1 and PC2 were 0.60 (17.781%) and 0.97 (11.029%), respectively. PERMANOVA was employed. (D) Preterm subgroup distances to healthy group. The distance of the preterm birth subgroups to the healthy group, based on Bray–Curtis distances. (E) Relative abundance of all groups. Comparison of the relative abundance of the dominant phylum in the healthy group, the preterm group, and the preterm birth subgroups.

Figure 2

The association of the gut microbiome and host parameters. (A–C) Clinical features’ projection on all samples. The projection of continuous parameters in the healthy group and the preterm group samples, based on the Bray–Curtis distance metric, gestational weeks (A), Apgar score at 5 min (B) and Apgar score at 1 min (C), respectively. (D–J) Clinical features projection on preterm birth subgroups. The projection of continuous parameters in the preterm subgroup samples, based on the Bray–Curtis distance metric, BMI at delivery (D), CRP (E), age (F); Apgar score at 1 min (G) gestational weeks (H); Apgar score at 5 min (I) and neonatal weight (J), respectively.

Figure 3

Identification of the gut microbial biomarkers between the healthy and preterm groups. (A) LEfSe analyses of healthy and preterm groups. Linear discriminant analysis effect size identified the genus between the healthy and preterm groups. Preterm-enriched taxa are indicated with a positive LDA score, and taxa enriched in healthy controls have a negative score. Only taxa meeting an LDA significant threshold of >3 are shown. (B) Total relative abundance of the common oral bacteria. The relative abundance of the common oral bacteria in the healthy group, the preterm group, and the preterm birth subgroups (healthy group and preterm group are compared using Wilcoxon rank-sum test; subgroups are compared with healthy group using Wilcoxon rank-sum test and adjusted by the Benjamini and Hochberg method). (C) Genus comparison between healthy and preterm group. Bacteria identified by the Wilcoxon rank-sum test with p < 0.05 are shown in the heatmap. The bar on the top indicates the group information of each sample. Red genus represents the common oral bacteria that overlapped with LEfSe results. Yellow genera represent the common oral bacteria but not overlapped with LEfSe results. *p < 0.05; ***p < 0.001.

Figure 4

Correlations of parameters and abundance of oral-to-gut bacteria. Coefficient of correlation between clinical features and total potential oral bacteria. The relationship between the relative abundance of oral and non-oral bacteria and host parameters (gestational weeks; neonatal weight; Apgar scores in 1, 5, and 10 min; BMI at delivery).