Researchers sequenced 4,526 samples from 714 mother-infant pairs and found that a mother’s gut bacteria colonize her baby’s gut, not her vaginal or breast milk bacteria. And her gut predicts whether the baby will get eczema.
The clinical advice given to expectant parents about gut health has been organized around two primary mechanisms for decades. Vaginal birth is preferred over cesarean section partly because passage through the birth canal exposes the newborn to the mother’s vaginal microbiome, seeding the infant gut with its first bacterial colonizers. Breastfeeding is encouraged partly because human milk contains live bacteria and prebiotic sugars that support the growth of beneficial gut microbes in the infant. Both of these pathways are real and documented. The question is how much they matter relative to another maternal source that received considerably less attention.
A new study published in Nature, conducted by researchers at the University of Groningen and collaborating institutions, provides the most comprehensive answer yet to where babies actually get their gut bacteria. The study enrolled 714 mother-infant pairs from the Dutch Lifelines NEXT birth cohort, following each pair from 12 weeks of pregnancy through the first year of the baby’s life. Over that period, the research team collected and sequenced 4,526 fecal samples, along with vaginal microbiome samples and breast milk samples from the mothers. They also gathered data on 474 clinical, dietary, and environmental variables from each participant.
The scale and longitudinal design of this dataset allowed the researchers to ask a question that smaller, cross-sectional studies had not been able to answer reliably: when you find a specific bacterial strain in a baby’s gut, where did it come from?
The answer was the maternal gut, not the vaginal microbiome and not breast milk.
“The maternal gut microbiome is a major reservoir for infant gut strains, with only occasional transmission from vaginal and breast milk microbiomes,” the researchers concluded.
What the strain-level analysis revealed
Most previous research on the infant microbiome worked at the level of bacterial species, asking whether a species present in the mother was also present in the baby. This approach misses a critical distinction: the same species can have hundreds of different strains, some of which are shared and some of which are not. Finding the same species in mother and baby does not prove transmission. Finding the identical strain in both does.
The Groningen team worked at the strain level, using high-resolution metagenomic sequencing that can distinguish between strains of the same species with far greater precision than species-level approaches. This allowed them to ask not just whether a species was shared but whether the specific genetic lineage of a bacterium had passed from mother to child.
At this resolution, the maternal gut signal was dominant. Strains that appeared in the infant gut in the weeks after birth matched strains found in the mother’s gut microbiome during pregnancy. The match rate was substantially higher than the match rate for vaginal or breast milk strains.
The mechanism follows from abundance. The more of a particular bacterial species a mother carried in her gut, the more likely that exact strain was to be found in her baby’s gut in the weeks that followed. Transmission was not random across bacterial species. It was proportional to how well-established a bacterium was in the maternal gut ecosystem.
This finding has immediate implications for how the field thinks about interventions to support infant gut health. If the maternal gut is the primary source of infant microbial colonization, then the microbiome interventions most likely to benefit babies are the ones targeted at the maternal gut during pregnancy, not the vaginal or breast milk microbiomes that have received more research attention in recent years.
What the birth canal and breast milk do contribute
The study was not designed to argue that vaginal birth and breastfeeding do not matter. Both remained significant predictors of infant gut microbiome composition in the analysis, and the researchers are direct about the importance of each.
Delivery mode and feeding mode were the strongest single predictors of how the infant gut microbiome developed overall, a finding consistent with the existing literature. Babies born vaginally received an initial seeding of maternal vaginal bacteria that influenced their gut composition in the first days and weeks of life. Babies who were breastfed showed different microbiome trajectories than formula-fed babies, reflecting both the bacteria in breast milk and the prebiotic substrates that human milk oligosaccharides provide to specific beneficial bacteria.
What changed in this study is the understanding of where the persistent, shared bacterial strains in a baby’s gut come from over the first year of life. The initial colonization at birth may involve vaginal and skin bacteria, but the strains that take up residence and persist through the first year more reliably trace back to the maternal gut than to any other maternal microbial compartment.
Home birth was included in the analysis because 155 of the 585 vaginally delivered babies in the study were born at home, a practice more common in the Netherlands than in most other countries. Home delivery was only moderately associated with infant gut microbiome composition, similar to other birth parameters such as duration of labor and membrane status at delivery, providing a natural experiment that the researchers used to further parse what delivery-related factors matter most.
The eczema prediction finding
The second major finding in the study is the one with the most immediate clinical implications for individual families. The composition of a mother’s gut microbiome during pregnancy predicted whether her baby would develop eczema in the first year of life.
Eczema, or atopic dermatitis, is the most common inflammatory skin condition in infancy, affecting roughly 20% of children in high-income countries. Its causes are multifactorial, involving genetic predisposition, skin barrier function, immune development, and environmental exposure. The microbiome’s role in eczema has been studied for over a decade, with evidence that the infant gut microbiome influences immune programming in ways that affect atopic disease risk.
What this study adds is a demonstration that the predictive signal is visible not just in the baby’s own microbiome but in the mother’s gut microbiome before the baby is born. The composition and specific bacterial abundances in the maternal gut during pregnancy could predict which babies would go on to develop eczema in their first year, even before the infant’s own microbiome had been established.
The mechanism through which this prediction operates is not fully understood. One possibility is that maternal gut bacteria transmitted to the infant gut in the first weeks of life set up the microbial conditions that then shape immune development in the direction of either tolerance or atopic reactivity. Another is that the same factors that produce a particular maternal gut microbiome composition also influence the baby’s immune programming through other pathways, including in utero signals or factors present in breast milk.
“We find that the maternal gut microbiome is a predictor of infant eczema,” the researchers reported. They were explicit that the mechanisms underlying this association remain unclear and require further study.
How the mother’s gut changes during pregnancy and postpartum
One of the study’s more surprising secondary findings concerned the stability of the maternal gut microbiome across pregnancy and the postpartum period.
Many researchers had expected to find that the maternal gut microbiome shifts substantially during pregnancy, as the dramatic hormonal, immunological, and metabolic changes of gestation alter the internal environment in which gut bacteria live. The study found that the changes were actually quite subtle. The maternal gut microbiome underwent only minor compositional changes from 12 weeks of pregnancy through the first year postpartum.
The factors that did produce detectable changes in the maternal gut microbiome over this period were diet, infections experienced during pregnancy, and pre-pregnancy smoking history. Hormonal changes alone did not produce the dramatic microbial shifts that some researchers had anticipated.
This stability is actually reassuring from the perspective of infant microbiome establishment. If the maternal gut microbiome shifted dramatically and unpredictably during pregnancy, the strains available to transmit to the baby would be highly variable and difficult to influence through intervention. The relative stability of the maternal gut across pregnancy means that interventions targeting the maternal gut microbiome could be implemented before or early in pregnancy and would reasonably be expected to persist through the period when transmission to the infant is most active.
What the findings mean for clinical practice
The study’s most direct implication for clinical practice is that the maternal gut microbiome deserves more attention as a target for interventions aimed at supporting infant gut health and reducing atopic disease risk.
Current probiotic research in pregnancy has focused primarily on strains chosen for their known beneficial effects in infants or their presumed ability to influence immune programming. This study provides a framework for thinking about which maternal microbiome states are associated with better infant outcomes, not just which specific bacterial strains might be supplemented.
The finding that higher maternal gut abundance of a species increases the probability of transmission to the infant suggests that building a diverse, abundant maternal gut microbiome during pregnancy, through diet, avoiding unnecessary antibiotic use, and other microbiome-supporting practices, may be one of the most effective ways to support the infant gut microbiome establishment that takes place in the weeks after birth.
For the eczema finding, the clinical implication is still speculative at this stage. The researchers do not have a specific intervention to recommend based on this finding. What they have demonstrated is that the predictive information is there, in the maternal gut during pregnancy, before any intervention would typically be offered. Whether targeting specific maternal gut bacteria could reduce infant eczema risk requires randomized controlled trials that have not yet been conducted.
What the study cannot establish
The study was conducted in a Dutch birth cohort that is predominantly white, relatively affluent, and living in a high-income country with specific dietary patterns, healthcare practices, and environmental exposures. The specific bacterial strains transmitted between Dutch mothers and their infants may differ from those in other populations, and the predictive relationship between maternal gut composition and infant eczema may not replicate in populations with different genetic backgrounds, dietary patterns, or microbiome compositions.
The study also cannot establish the causal mechanisms through which maternal gut bacteria colonize the infant gut. Transmission at the strain level was demonstrated by sequence matching, but the physical route, whether through fecal-oral exposure during delivery, through breast milk in trace amounts not detected by the study’s methods, or through other perinatal contact, remains to be fully characterized.
The eczema prediction finding is observational. Demonstrating that maternal gut composition predicts infant eczema is not the same as demonstrating that the maternal gut microbiome causes infant eczema, or that modifying the maternal gut microbiome would prevent it.
What the study establishes, at a scale and resolution no previous study had achieved, is that the conventional understanding of how babies acquire their microbiome was missing the most important part of the picture. The vaginal microbiome and breast milk matter. But the maternal gut is where the infant’s microbial life primarily begins.
The study, “Maternal influences on infant gut microbiome and health”, was authored by Trishla Sinha, Lianmin Chen, Ranko Gacesa, and colleagues at the University of Groningen and collaborating institutions in the Netherlands, and published in Nature in 2026.
Source: University of Groningen / Lifelines NEXT Cohort. DOI: 10.1038/s41586-026-10922-9
