Development of the gut microbiota in healthy twins during the first 2 years of life and associations with body mass index z-score: Results from the Wuhan twin birth cohort study

Hong Mei, Shaoping Yang, Anna Peng, Ruizhen Li, Feiyan Xiang, Hao Zheng, Yafei Tan, Ya Zhang, Aifen Zhou, Jianduan Zhang, Han XiaoFrontiers in Microbiology 2022 · DOI

Abstract

The first three years of life are a critical period for gut microbiota maturation, influencing long-term metabolic and immune health. While well-studied in singletons, the microbial development of twins—who often face unique early-life stressors—remains less understood. This study analyzed 344 stool samples from 204 twin families in the Wuhan Twin Birth Cohort at 6, 12, and 24 months of age. Using group-based trajectory models, researchers identified distinct developmental paths for various microbial genera. Notably, the trajectory of the genus Parabacteroides was significantly associated with body mass index z-score (BMI_Z) curves (coefficient = 0.75, p = 0.04), suggesting that early-life microbial growth patterns are linked to physical development in twins.

Introduction

The assembly of the gut microbiome during infancy is a dynamic process that sets the stage for adult health. Disruptions during this "critical window" are linked to various metabolic disorders. Twins represent a unique population for studying these dynamics due to higher incidences of preterm birth, low birth weight, and C-section deliveries. This research aims to characterize the maturation of the twin gut microbiome from 6 to 24 months and determine if specific microbial developmental trajectories correlate with growth outcomes, specifically the BMI z-score (BMI_Z).

Methods and Data

Data were drawn from the Wuhan Twin Birth Cohort (WTBC), involving 204 twin families. Stool samples (n=344) were collected at 6, 12, and 24 months. Microbial composition was determined via 16S rRNA gene sequencing (V3–V4 region).

To move beyond static "snapshots," the researchers employed Group-Based Trajectory Models (GBTM) to identify Distinct Trajectory Curves (DTCs) for 61 genera that showed significant abundance changes over time. The association between these microbial DTCs and BMI_Z trajectories was then analyzed using Generalized Estimation Equation (GEE) models, which account for the inherent clustering in twin data and adjust for confounders such as birth weight, delivery mode, and feeding patterns.

Results: Microbial Maturation and Diversity

The study observed a clear successional shift in the twin gut microbiota. At 6 months, the community was largely dominated by Bifidobacterium, typical of milk-fed infants. By 24 months, as the infants transitioned to diverse solid foods, Bacteroides became the dominant genus, reflecting a shift toward an adult-like state.

Figure 1: Community of the gut microbiota among the three age groups (6, 12, and 24 months). (A) Venn plot of shared OTUs. (B, C) Relative abundance of top phyla and genera. (D, E) Sankey diagrams showing phyla and genera transitions.

Both alpha diversity (richness and evenness) and beta diversity (compositional differences) increased significantly with age (p < 0.05). The predominant phyla identified across the cohort were Proteobacteria, Actinobacteriota, Firmicutes, Bacteroidota, and Verrucomicrobiota.

The Parabacteroides and BMI Connection

A key finding of this research is the link between microbial "growth curves" and physical growth. Out of 61 genera analyzed, five showed trajectories correlated with BMI growth. However, after rigorous adjustment for confounding factors in the GEE model, only Parabacteroides remained a significant predictor. Twins following a "high" abundance trajectory for Parabacteroides exhibited higher BMI z-score trajectories throughout the first two years of life. This suggests that the developmental path of specific microbes may be more clinically relevant than their absolute abundance at any single time point.

Discussion and Limitations

This study underscores the importance of longitudinal trajectory modeling in understanding the infant microbiome. The identification of Parabacteroides as a potential biomarker for growth patterns in twins provides a new avenue for metabolic research.

However, several limitations exist. The study lacked stool samples from the first 6 months of life, missing the initial colonization phase. Additionally, while feeding patterns were tracked, detailed data on specific solid food introduction—which significantly impacts Parabacteroides levels—was unavailable. Finally, as the cohort was limited to Wuhan, China, further multi-ethnic and multi-geographic studies are required to confirm the generalizability of the Parabacteroides-BMI association.

Summary updated

Abstract

The gut microbiota undergoes rapid and vital changes to microbial community structure and the microbial-immune crosstalk during the first 3 years of life, which is thought to be involved in the pathobiology of later-life disease. Compared to single-born children, little is known about the gut microbiota of twins in early childhood. Based on the Wuhan Twin Birth Cohort study, 344 stool samples from 204 twin families were analyzed to investigate the difference in gut microbiota composition at 6, 12, and 24 months of age. Furthermore, this study evaluated the association between gut microbiota development curves and body mass index z-score (BMI_Z) curves at 6, 12, and 24 months of age. The predominant microbiota phyla identified in twins were <i>Proteobacteria, Actinobacteriota, Firmicutes, Bacteroidota</i>, and <i>Verrucomicrobiota</i>. The richness and diversity of gut microbiota increased from 6 to 24 months old (alpha diversity with <i>p</i> < 0.05). Beta diversity revealed 61 gut microbiota genera that were significantly different in relative abundance among the three age groups. Among the 61 gut microbiota genera, 30 distinct trajectory curves (DTCs) were generated by group-based trajectory models after log2 transformation of their relative abundance. Subsequently, Spearman correlation analysis revealed that only five gut microbiota DTC were correlated with the BMI_Z DTC. Therefore, we further examined the association between the five gut microbiota genera DTC and BMI_Z DTC using generalized estimation equation models. The results revealed a significant association between the DTC groups of <i>Parabacteroides</i> and that of BMI_Z (coefficient = 0.75, <i>p</i> = 0.04). The results of this study validated the hypothesis that the richness and diversity of gut microbiota developed with age in twins. Moreover, participants with a higher DTC of log2-transformed <i>Parabacteroides</i> had a higher BMI_Z DTC during the first 2 years of life. Further studies are needed to confirm the association between <i>Parabacteroides</i> and BMI_Z in other populations.
BibTeX
@inproceedings{mei2022development,
  author = {Hong Mei and Shaoping Yang and Anna Peng and Ruizhen Li and Feiyan Xiang and Hao Zheng and Yafei Tan and Ya Zhang and Aifen Zhou and Jianduan Zhang and Han Xiao},
  title = {{Development of the gut microbiota in healthy twins during the first 2 years of life and associations with body mass index z-score: Results from the Wuhan twin birth cohort study}},
  booktitle = {Frontiers in Microbiology},
  year = {2022},
  doi = {10.3389/fmicb.2022.891679},
}
Figure