Infant Gut Microbiota Associated with Fine Motor Skills

Abstract

Background: During early life, dynamic gut colonization and brain development co-occur with potential cross-talk mechanisms affecting behaviour.

Methods: We used 16S rRNA gene sequencing to examine the associations between gut microbiota and neurodevelopmental outcomes assessed by the Bayley Scales of Infant Development III in 71 full-term healthy infants at 18 months of age. We hypothesized that children would differ in gut microbial diversity, enterotypes obtained by Dirichlet multinomial mixture analysis and specific taxa based on their behavioural characteristics.

Results: In children dichotomized by behavioural trait performance in above- and below-median groups, weighted Unifrac b-diversity exhibited significant differences in fine motor (FM) activity. Dirichlet multinomial mixture modelling identified two enterotypes strongly associated with FM outcomes. When controlling for maternal pre-gestational BMI and breastfeeding for up to 3 months, the examination of signature taxa in FM groups showed that Turicibacter and Parabacteroides were highly abundant in the below-median FM group, while Collinsella, Coprococcus, Enterococcus, Fusobacterium, Holdemanella, Propionibacterium, Roseburia, Veillonella, an unassigned genus within Veillonellaceae and, interestingly, probiotic Bifidobacterium and Lactobacillus were more abundant in the above-median FM group.

Conclusions: Our results suggest an association between enterotypes and specific genera with FM activity and may represent an opportunity for probiotic interventions relevant to treatment for motor disorders.

Keywords: fine motricity; gut–brain axis; microbiota; neurodevelopment; probiotics.

Figure 1

Gut microbial composition and structure in 71 full-term healthy infants at 18 months of age. (A) Phylogeny of infants’ gut microbiota at the genus level of 17 core genera detected in ≥98% of study subjects. Sample names are coloured according to enterotype: Firm in blue, Bact in red. Gut microbial composition and structure in 71 full-term healthy infants at 18 months of age. (B) Scatterplot from principal coordinate analysis using weighted Unifrac metrics in above-median (blue) and below-median (red) fine motor groups. Additional significant explanatory variables are represented by black arrows originating from the coordinate system origin. (C) Horizontal bars show the influence of anthropometric, maternal and nutritional variables (r²) on gut microbiota composition. * p < 0.05 for PERMANOVA test with 999 permutations.

Figure 2

Probabilistic modelling with Dirichlet multinomial mixtures of faecal samples revealed 2 enterotypes in 71 full-term healthy infants at 18 months of age. (A) The model was fit to k = 4 and rendered the lowest model fit value of Laplace approximation for four components which was taken as evidence for the presence of the two main community types. (B) Heatmap showing the relative abundance of the 30 most dominant genera per DMM enterotype. Abundance represents square root sequence reads. (C) Contribution of the 30 most dominant genera to overall community composition. (D) Contribution to model prediction of microbial core genera to each enterotype. (E) Contribution of core signature genera (mean relative abundances) to microbial profiles in each DMM enterotype.

Figure 3

Differential abundance of gut microbiota between infants with above- and below-median fine motricity groups. Volcano plot showing shrunken log₂ fold changes in mean abundance of genera versus −log₁₀ of FDR. Blue colours depict significance and label names denote the taxonomy of genera at phylum level: Firmicutes (black), Actinobacteria (red), Bacteroidetes (green), Fusobacteria (purple). Dashed lines denote thresholds of significance (FDR < 0.05).