Microbial Community Dynamics in Mother's Milk and Infant's Mouth and Gut in Moderately Preterm Infants

Abstract

Mother's own milk represents the optimal source for preterm infant nutrition, as it promotes immune defenses and gastrointestinal function, protects against necrotizing enterocolitis, improves long-term clinical outcome and is hypothesized to drive gut microbiota assembly. Preterm infants at birth usually do not receive their mother's milk directly from the breast, because active suckling and coordination between suckling, swallowing and breathing do not develop until 32-34 weeks gestational age, but actual breastfeeding is usually possible as they grow older. Here, we enrolled moderately preterm infants (gestational age 32-34 weeks) to longitudinally characterize mothers' milk and infants' gut and oral microbiomes, up to more than 200 days after birth, through 16S rRNA sequencing. This peculiar population offers the chance to disentangle the differential contribution of human milk feeding per se vs. actual breastfeeding in the development of infant microbiomes, that have both been acknowledged as crucial contributors to short and long-term infant health status. In this cohort, the milk microbiome composition seemed to change following the infant's latching to the mother's breast, shifting toward a more diverse microbial community dominated by typical oral microbes, i.e., Streptococcus and Rothia. Even if all infants in the present study were fed human milk, features typical of healthy, full term, exclusively breastfed infants, i.e., high percentages of Bifidobacterium and low abundances of Pseudomonas in fecal and oral samples, respectively, were detected in samples taken after actual breastfeeding started. These findings underline the importance of encouraging not only human milk feeding, but also an early start of actual breastfeeding in preterm infants, since the infant's latching to the mother's breast might constitute an independent factor helping the health-promoting assembly of the infant gut microbiome.

Keywords: breastfeeding; infant gut microbiota; infant oral microbiota; latching; microbiota assembly; milk microbiota; moderately preterm infants.

FIGURE 1

Microbiota structure of the infant’s gut and mouth and mother’s milk in moderately preterm infants, in relation to postnatal age. Principal Coordinates Analyses (PCoA) based on Bray-Curtis distances between genus-level relative abundance profiles of fecal (A), oral (B), and mother’s milk (C) samples. Samples are depicted as dots filled in with different shades of gray, from light (earlier samples) to dark (later samples), according to the sampling time (days from birth). A legend for the gray scale is provided. First and second coordination axes are reported in each plot; percentages of variation in the datasets explained by each axis are reported. Postnatal age (PNA, expressed in days from birth at which each sample was taken) as a quantitative environmental variable is depicted as a blue arrow. The biplot of the average bacterial coordinates weighted by the corresponding bacterial relative abundance per sample was superimposed on the PCoA plot for abundant (average relative abundance > 1%) bacterial genera that most significantly contributed to the ordination space (envfit, P < 0.001) (red arrows). Some of the genus names are abbreviated as follows: Staph., Staphylococcus; Bif., Bifidobacterium; Pseud., Pseudomonas; Janth., Janthinobacterium; Strept., Streptococcus; Enter., Enterococcus; Sedim., Sediminibacterium; Acin., Acinetobacter. See Supplementary Figures S1–S3 for details about sampling times for each infant.

FIGURE 2

Microbiota structure of the infant’s gut and mouth and mother’s milk in moderately preterm infants, in relation to latching. PCoA based on Bray-Curtis distances between genus-level relative abundance profiles of fecal (A), oral (B), and mother’s milk (C) samples. Samples taken before and after latching are depicted as empty dots and filled triangles, respectively. First and second coordination axes, with percentages of plotted variation, are reported as in Figure 1.

FIGURE 3

Features of the three Milk Community Types (MCTs). (A) MCTs were defined by Ward clustering on Euclidean distances calculated between the genus-level milk microbiota profiles. Samples assigned to MCT1, MCT2, and MCT3 are depicted in pink, gold and blue, respectively. (B–I) Relative abundances of bacterial genera for which a significant difference among MCTs was found (Kruskal-Wallis test, FDR corrected P < 0.05) are shown as boxplots. (J) Levels of alpha diversity calculated as Shannon index are depicted as boxplots for each MCT. Lines and stars below each plot indicate significant differences for pairwise comparisons (post-hoc Tukey and Kramer test). (K) Dynamics of the milk microbiome through each MCT for each enrolled lactating mother. Symbols represent sampling times for each subject. Triangles indicate samples taken after latching and start of actual breastfeeding.

FIGURE 4

Features of the moderately preterm infants’ fecal and oral bacterial ecosystems associated with the three MCTs, in comparison to healthy, exclusively breastfed, full term infants. Boxplots of the relative abundances of OTUs assigned to the genera Bifidobacterium and Pseudomonas in fecal (A) and oral (B) samples, respectively, taken from the enrolled moderately preterm infants at time points in which the corresponding milk samples harbored a MCT1 (pink), MCT2 (gold) or MCT3 (blue). In gray, the data obtained in samples taken from healthy, full term, actively and exclusively breastfed 21-day-old infants (Biagi et al., 2017; sequence reads re-analyzed together with those obtained in the present study). Lines and stars below each plot indicate significant differences for pairwise comparisons (post-hoc Tukey and Kramer test, after Kruskal-Wallis test, P < 0.05).