Milk- and solid-feeding practices and daycare attendance are associated with differences in bacterial diversity, predominant communities, and metabolic and immune function of the infant gut microbiome

Abstract

The development of the infant intestinal microbiome in response to dietary and other exposures may shape long-term metabolic and immune function. We examined differences in the community structure and function of the intestinal microbiome between four feeding groups, exclusively breastfed infants before introduction of solid foods (EBF), non-exclusively breastfed infants before introduction of solid foods (non-EBF), EBF infants after introduction of solid foods (EBF+S), and non-EBF infants after introduction of solid foods (non-EBF+S), and tested whether out-of-home daycare attendance was associated with differences in relative abundance of gut bacteria. Bacterial 16S rRNA amplicon sequencing was performed on 49 stool samples collected longitudinally from a cohort of 9 infants (5 male, 4 female). PICRUSt metabolic inference analysis was used to identify metabolic impacts of feeding practices on the infant gut microbiome. Sequencing data identified significant differences across groups defined by feeding and daycare attendance. Non-EBF and daycare-attending infants had higher diversity and species richness than EBF and non-daycare attending infants. The gut microbiome of EBF infants showed increased proportions of Bifidobacterium and lower abundance of Bacteroidetes and Clostridiales than non-EBF infants. PICRUSt analysis indicated that introduction of solid foods had a marginal impact on the microbiome of EBF infants (24 enzymes overrepresented in EBF+S infants). In contrast, over 200 bacterial gene categories were overrepresented in non-EBF+S compared to non-EBF infants including several bacterial methyl-accepting chemotaxis proteins (MCP) involved in signal transduction. The identified differences between EBF and non-EBF infants suggest that breast milk may provide the gut microbiome with a greater plasticity (despite having a lower phylogenetic diversity) that eases the transition into solid foods.

Keywords: breastfeeding; daycare; feeding transitions; infant gut microbiome; metagenomics.

Figure 1

(A) A comparison of species richness S and (B) phylogenetic diversity PD between feeding groups. Non-EBF infants showed significantly higher S and PD values than EBF infants, and both values were impacted by the introduction of solid foods (^*p < 0.05). (C–F) Comparison of unweighted UniFrac PCoA plots with repeated resampling (jackknifing) of microbiota from different feeding groups. ANOSIM R and P values are indicated in each figure.

Figure 2

A comparison of species richness S (A) and phylogenetic diversity PD (B) between infants attending daycare and infants cared for at home (^*p < 0.05). (C) PCoA analysis of samples with repeated resampling (jackknifing) according to daycare attendance regardless of feeding group. Blue symbols represent samples from infants attending daycare. Red symbols represent infants not attending daycare. ANOSIM R and P values are indicated in the figure.

Figure 3

Relative abundance of (A) phyla and (B) genera in the different feeding groups. The mid panel shows a diagram of the taxa represented in the samples. For clarity only genera represented over 0.1% were included. The total representation per sample was >99.9%.

Figure 4

Relative abundance of (A) phyla and (B) genera by feeding group and attendance to daycare (D+). The mid panel shows a diagram of the taxa represented in the samples. For clarity only genera represented over 0.1% were included. The total representation per sample was >99.9%.

Figure 5

The infant core microbiome by feeding group. The percent of samples in which each taxa was identified is shown on the top.

Figure 6

PICRUSt (Langille et al., 2013) predicted summary of COG categories from 16S amplicon pyrosequencing. Results are depicted by feeding group (A, Exclusively breastfed; B, Non-exclusively breastfed; C, Exclusively breastfed plus solids; and D, Non-exclusively breastfed plus solids) and show an over representation of pathways involved in environmental information processing, specifically membrane transport, and an under representation of pathways involved in the metabolism of terpenoids and polyketides in EBF infants.

Figure 7

Relative abundances of KEGG pathways involved in energy metabolism, and nitrogen and methane within energy metabolism in the different feeding groups (^*p < 0.05, §p < 0.1).