Codevelopment of gut microbial metabolism and visual neural circuitry over human infancy

Kevin S. Bonham, Emma T. Margolis, Guilherme Fahur Bottino, Ana Pina Sobrino, Fadheela Patel, Shelley McCann, Michal R. Zieff, Marlie Miles, Donna Herr, Lauren Davel, Cara Bosco, Curtis Huttenhower, Nicolò Pini, Daniel C. Alexander, Derek K. Jones, Steven Williams, Dima Amso, Melissa Gladstone, William P. Fifer, Kirsten A. Donald, Laurel J. Gabard‐Durnam, Vanja Klepac‐Ceraj, Layla Bradford, Simone Williams, Tembeka Mhlakwaphalwa, Bokang Methola, Khanyisa Nkubungu, Candice Knipe, Zamazimba Madi, Nwabisa MlandumBio 2025 · DOI

TL;DR

This study reveals a significant link between the developing gut microbiome and visual brain development in infants. Microbial genes involved in metabolizing neuroactive compounds at 4 months of age were associated with visual cortical development (measured by VEP) between 9 and 14 months, suggesting that early gut microbial activity plays a crucial role in shaping neural plasticity.

What Problem Does This Paper Address?

While the connection between the gut microbiome and the brain is well-established in adults, particularly in the context of neuropsychiatric disorders, how this relationship emerges during early human development remains largely unknown. Infancy is a critical period of rapid brain and sensory development, characterized by high neuroplasticity. The gut microbiome also undergoes significant changes during this time. This paper aims to bridge this knowledge gap by investigating the co-development of the gut microbiome's metabolic potential and visual neural circuitry during human infancy. Specifically, it explores whether early microbial activity can influence the maturation of the visual cortex.

Methods

The researchers employed a longitudinal study design, collecting data from 194 South African infants over the first 18 months of life.

  • Microbiome Analysis: Shotgun metagenomic sequencing was used on infant stool samples to identify microbial genes and their metabolic potential. This allowed for the analysis of functional profiles rather than just taxonomic identification.
  • Neurodevelopmental Assessment: Electroencephalography (EEG) was used to measure visual cortical development through Visual-Evoked Potentials (VEPs). VEPs provide insights into the maturation of the visual cortex, including amplitude (reflecting cortical circuit function) and latency (reflecting structural integrity and myelination).
  • Statistical Analysis:
    • Feature Set Enrichment Analysis (FSEA): This was the core analytical approach to link microbial gene sets (specifically those involved in metabolizing neuroactive compounds) with VEP features. FSEA identifies whether a group of genes (a "gene set") is significantly enriched or depleted in association with a particular outcome (VEP features).
  • Generalized Linear Models (GLMs): Used for initial analyses and to control for covariates like age and number of EEG trials.
  • Longitudinal Models: Employed to examine prospective associations, where early microbiome data was linked to later VEP measurements.

Datasets

  • Longitudinal Infant Cohort: Data from 194 infants in Gugulethu, Cape Town, South Africa, collected at up to three time points within the first 18 months of life. This cohort is part of the prospective longitudinal "Khula" Study.
  • Microbiome Data: Shotgun metagenomic sequencing data from infant stool samples.
  • Neurodevelopmental Data: EEG data, specifically Visual-Evoked Potentials (VEPs), collected from the infants.

Key Insights and Ideas

  • Early Microbiome as a "Developmental Sculptor": The most striking finding is that microbial gene sets from early infancy (around 4 months) showed a stronger and more numerous association with later visual development (9-14 months) than concurrent associations. This suggests a prospective influence, where early microbial activity shapes future neural development.
  • Temporal Specificity of Influence: Different microbial gene sets were associated with VEP features at different time points, indicating that the microbiome's influence is not static but evolves alongside infant development and microbial maturation. For example, GABA metabolism genes were more relevant for early VEP changes, while SCFA and menaquinone genes showed stronger associations with later VEP features.
  • Metabolites of Neuroplasticity: The study highlights specific classes of microbial genes that are crucial: those involved in the metabolism of neurotransmitters (GABA, glutamate), amino acids (tryptophan), and short-chain fatty acids (SCFAs like acetate, butyrate) known to be involved in myelination and neuroinflammation.
  • VEP Components as Sensitive Indicators: The P1 and N2 VEP components, which are known to undergo significant developmental changes in the first year of life and reflect underlying visual learning and plasticity, were consistently the most sensitive to microbial gene associations.

Key Results

  • Prospective Associations Dominate: Microbial gene sets from early visits (e.g., 4 months) were significantly associated with VEP features measured later (e.g., 9-14 months) more frequently than concurrent associations. For instance, 13 out of 17 tested gene sets from visit 1 were associated with visit 2 VEP features.
  • Specific Gene Sets Show Strong Links:
    • GABA and Glutamate Synthesis/Degradation: Gene sets related to these neurotransmitters showed associations with all VEP features, particularly at later visits.
  • Tryptophan Metabolism: Pathways involved in tryptophan metabolism (including quinolinic acid) were also strongly linked to VEP development.
  • Short-Chain Fatty Acids (SCFAs): Acetate synthesis was broadly associated with VEP features. Butyrate and propionate metabolism genes showed significant associations with VEP latencies, particularly in the second half of the study period, aligning with the role of SCFAs in myelination.
  • Menaquinone (Vitamin K2): Gene sets for menaquinone synthesis also showed persistent associations with VEP features, particularly at later time points.
  • Age-Related Changes in VEP: As infants aged, VEP amplitudes became more negative (N1) or smaller (P1, N2), and latencies shortened, reflecting expected maturation of the visual cortex.

Limitations and Future Work

  • Metabolite Concentration vs. Gene Abundance: The study analyzed microbial gene abundance, not the actual concentration of metabolites. This limits direct conclusions about metabolite levels in the gut, blood, or brain and their precise impact. The relationship between gene presence and actual metabolic activity can be complex.
  • Causality: While the prospective associations strongly suggest a causal role for the microbiome in brain development, the study design does not definitively prove causality. The authors acknowledge this limitation.
  • Future Directions:
    • Mechanistic Studies: The authors propose using animal models (germ-free or defined-microbiome animals) to test hypotheses generated by this study. This could involve assessing VEPs in these models and supplementing them with specific molecules or microbial species.
  • Direct Molecule Measurement: Future research could focus on measuring metabolite concentrations in biological samples (e.g., plasma) to better understand systemic exposure.
  • Intervention Targets: The findings highlight specific microbial pathways (e.g., GABA, tryptophan metabolism) as potential targets for future interventions aimed at supporting healthy infant brain development.

Summary updated

Abstract

Infancy is a time of elevated neuroplasticity supporting rapid brain and sensory development. The gut microbiome, also undergoing extensive developmental changes in early life, may influence brain development through the metabolism of neuroactive compounds. Here, we leverage longitudinal data from 194 South African infants across the first 18 months of life to show that microbial genes encoding enzymes that metabolize molecules playing a key role in modulating early neuroplasticity are associated with visual cortical neurodevelopment, measured by the Visual-Evoked Potential (VEP). Neuroactive compounds included neurotransmitters GABA and glutamate, the amino acid tryptophan, and short-chain fatty acids involved in myelination, including acetate and butyrate. Microbial gene sets around 4 months of age were strongly associated with the VEP from around 9-14 months of age and showed more associations than concurrently measured gene sets, suggesting that microbial metabolism in early life may affect subsequent neural plasticity and development.IMPORTANCEOver the past decade, extensive research has revealed strong links between the gut microbiome and the brain, at least in adults or those with neuropsychiatric disorders. This study explores how these associations emerge in early development using a longitudinal sample of 194 infants with repeated microbiome metabolism and electroencephalography (EEG) measures during the critical early period of visual cortex neuroplasticity. We examined microbial genes encoding enzymes for neuroactive compounds (e.g., GABA, glutamate, tryptophan, and short-chain fatty acids) and their association with the visual-evoked potential (VEP). Genes from 4-month stool samples strongly correlated with VEP features between 9 and 14 months, suggesting that early microbial metabolism influences later visual neurodevelopment. These prospective associations were more numerous than the concurrent ones. Our findings suggest that early gut microbiome metabolic potential plays a crucial role in shaping neural plasticity and visual neurodevelopment.
BibTeX
@inproceedings{bonham2025codevelopment,
  author = {Kevin S. Bonham and Emma T. Margolis and Guilherme Fahur Bottino and Ana Pina Sobrino and Fadheela Patel and Shelley McCann and Michal R. Zieff and Marlie Miles and Donna Herr and Lauren Davel and Cara Bosco and Curtis Huttenhower and Nicolò Pini and Daniel C. Alexander and Derek K. Jones and Steven Williams and Dima Amso and Melissa Gladstone and William P. Fifer and Kirsten A. Donald and Laurel J. Gabard‐Durnam and Vanja Klepac‐Ceraj and Layla Bradford and Simone Williams and Lauren Davel and Tembeka Mhlakwaphalwa and Bokang Methola and Khanyisa Nkubungu and Candice Knipe and Zamazimba Madi and Nwabisa Mlandu and Nwabisa Mlandu},
  title = {{Codevelopment of gut microbial metabolism and visual neural circuitry over human infancy}},
  booktitle = {mBio},
  year = {2025},
  doi = {10.1128/mbio.00835-25},
}
Figure