The microbiota of the mother at birth and its influence on the emerging infant oral microbiota from birth to 1 year of age: a cohort study

Abstract

Background: The acquisition of microbial communities and the influence of delivery mode on the oral microbiota of the newborn infant remains poorly characterised. Methods: A cohort of pregnant women were enrolled in the study (n = 84). All infants were born full term, by Spontaneous vaginal delivery (SVD) or by Caesarean section (CS). At delivery a saliva sample along with a vaginal/skin sample from the mother. Saliva samples were the taken from the infant within one week of birth, and at week 4, week 8, 6 months and 1 year of age. We used high-throughput sequencing of V4-V5 region 16S rRNA amplicons to compare the microbiota of all samples. Results: The vaginal microbiota had a lower alpha diversity than the skin microbiota of the mother, while the infant oral microbiota diversity remained relatively stable from birth to 8 weeks of age. The oral microbiota of the neonate differed by birth modality up to 1 week of age (p < 0.05), but birth modality did not have any influence on the infant oral microbiota beyond this age. Conclusions: We conclude thatbirth mode does not have an effect on the infant oral microbiota beyond 4 weeks of age, and the oral microbiota of infants continues to develop until 1 year of age.

Keywords: Microbiota; infant; mother; newborn; oral cavity; saliva; skin; transmission; vagina.

Figure 1.

Principle coordinates analyses (PCoAs) on Spearman distance matrices of samples from mother (saliva, skin and vagina) and from infant oral cavity at 5 time points (weeks 1, 4, 8, 6 months and 1 year). (a) Bray–Curtis dissimilarity, (b) UniFrac weighted, and (c) unweighted UniFrac between the 8 groups (mother saliva/oral cavity, skin and vaginal and infant oral cavity at weeks 1, 4, 8, 6 months and 1 year). (a) Plot of principle coordinates using Bray–Curtis dissimilarity. Points are coloured according to group and ellipses describe the distribution of points for each group. Percentage variation explained: PCA 1 (18.5%) and PCA 2 (9.2%). (b) Plot of principle coordinates using weighted UniFrac distance. Points are coloured according to group and ellipses describe the distribution of points for each group. Percentage variation explained: PCA 1 (39.2%) and PCA 2 (11%). (c) Plot of principle coordinates using unweighted UniFrac distance. Points are coloured according to group and ellipses describe the distribution of points for each group. Percentage variation explained: PCA 1 (18.4%) and PCA 2 (6.7%).

Figure 2.

Alpha-diversity measurement of the influence of mode of delivery (SVD vs. CS) on the infant oral microbiota at various time points (week 1, week 4, 6 months and 1 year). Boxplot of the chao1 diversity, observed species, phylogenetic diversity and Shannon diversity in the two groups (SVD and CS) at (a) week 1, (b) week 4, (c) 6 months and (d) 1 year. Outliers are represented by black points.

Figure 3.

PCoA (principle coordinates) plots, each demonstrating the effect of birth mode (SVD vs. CS) on the oral microbiota composition of the infant at different ages (weeks 1, 4, 8, 6 months and 1 year). Significance was calculated using permutational multivariate analysis of variance (PerMANOVA) (Table 2).

Figure 4.

Alpha-diversity comparisons of the eight subject groups [(1) mother saliva, (2) Skin (mother) and (3) vagina (mother) and (4–8) infant oral weeks 1, 4, 8, 6 months and 1 year]. (a) Boxplot of chao1 diversity in the three groups, (b) boxplot of observed species in the eight groups, (c) boxplot of PD in the eight groups and (d) boxplot of Shannon diversity in the eight groups. Outliers are represented by black points.

Figure 5.

Hierarchical clustering of microbiota data at genus level. Abundances are colour-coded according to the colour key on the top left with grey representing a value of zero. Euclidean distance and complete linkage were used to cluster the rows and columns of the heatmap. The colour bar on side of the heatmap corresponds to sample type. Each genus with a mean ≥0.5% across all samples was included. All taxa present at less than 1% in all groups are excluded from the heatmap.

Figure 6.

Hierarchical clustering of microbiota data at species level.

Figure 7.

Hierarchical clustering of data between maternal saliva microbiota and infant oral microbiota at bacterial genus level. Abundances are colour-coded according to the colour key on the top left with red representing a value of zero. Euclidean distance and complete linkage were used to cluster the rows and columns of the heatmap. The colour bar on side of the heatmap corresponds to sample type (green = mother, yellow = infant). All taxa present at less than 1% in all three groups are excluded from the heatmap.

Figure 8.

Broad and fine detail compositional differences at genus, phylum and species level. (a) Microbiota composition at phylum level. Percentages for each taxon represent the median abundance values for the sample types. (b) Bar plot of percentage abundance at the genus level. Percentages for each taxon represent the median values for the groups. (c) Bar plot of percentage abundance at species level. Percentages for each taxon represent the median values for the groups.

Figure 9.

Comparison of the microbiota composition of infants born by different birth modes (SVD and CS) across five time points from 1 week to 1 year of age. (a) Microbiota composition at the phylum level. Percentages for each taxon represent the median abundance values for the sample types. (b) Bar plot of percentage abundance at the genus level. Percentages for each taxon represent the median values for the groups. The graph includes data for genera found at >1% average in the total population. Genera found at <1% were grouped as ‘other’.

Figure 10.

Beta-diversity PCoA (principle coordinates) plots (Bray–Curtis) illustrating the influence of breastfeeding duration on the oral microbiota of SVD and CS infants. (a) Naturally delivered infants (SVD). (b) CS infants. In blue are infants that were breastfed for less than 4 months. In red are infants that were breastfed exclusively for longer than 4 months. (c) Beta-diversity PCoA (principle coordinates) plots (Bray–Curtis) illustrating influence of breastfeeding duration on the oral microbiota of combined (CS and SVD) infants.