Home/Issues/Issue 06/Study

Prospective birth cohort with longitudinal metagenomic sequencing Publication date verified at source

Maternal influences on infant gut microbiome and health

AI narration, generated on first listen
Journal
Nature
Authors
Sinha T
Institution
University of Groningen and University Medical Center Groningen
Published
12 August 2026
Source
PMID 42587158 · DOI 10.1038/s41586-026-10922-9
Design
Prospective birth cohort with longitudinal shotgun metagenomic sequencing of mother and infant stool, plus breast milk and vaginal microbiome sequencing, integrated with 474 clinical and exposure variables.
Sample
Mother and infant pairs, Lifelines NEXT cohort, Netherlands

What strain sharing is

Tracking bacteria at a finer level than species, so a specific lineage in the infant can be matched to the same lineage in the mother. It is measured here because only strain level matching can show that a microbe was actually passed on, rather than picked up separately from the environment.

Drawn from the paper's approach as described.

Why they ran it

The authors note that the establishment of the infant gut microbiome is critical for later health, yet how it is shaped by maternal and early life factors remains unclear.

That is why they sampled repeatedly from early pregnancy through the first year, and sequenced the breast milk and vaginal microbiomes alongside the gut, rather than sampling once and inferring the rest.

Drawn from the paper's introduction, which states the open question rather than a formal hypothesis.

The Lifelines NEXT birth cohort followed 714 mother and infant pairs with repeat stool sampling from twelve weeks of pregnancy through one year postpartum, producing 4,526 longitudinal samples. Those were integrated with 474 clinical and exposure variables and with ultra deep sequencing of breast milk and vaginal microbiomes, so the three candidate routes of transmission could be compared against each other rather than assumed.

The maternal gut microbiome itself changed only subtly across pregnancy and postpartum, shifted by diet, infections and pre pregnancy smoking. It was also the main source of the infant's gut strains. Transmission from the vaginal and breast milk microbiomes happened only occasionally. Strain sharing was time dependent, and the more abundant a species was in the mother's gut, the more likely it was to be passed on. The maternal gut microbiome also predicted infant eczema.

Mode of delivery and feeding mode were still the primary shapers of the infant gut microbiome and its functional profiles. But of 585 vaginally delivered infants, 155 were born at home, and home delivery was only moderately associated with infant gut composition, on a par with birth parameters like duration of pushing and ruptured membranes.

The numbers

undefinedundefined
undefinedundefined
undefinedundefined
undefinedundefined
undefinedundefined

Why this might happen

Shown in this study The study measured this step directly.

Because the team matched bacteria at strain level rather than species level, they could show which lineages were actually passed on rather than acquired separately. On that measure the maternal gut is a major reservoir for infant gut strains, and the vaginal and breast milk microbiomes contribute only occasionally.

Sharing is time dependent, and the more abundant a species is in the mother's gut, the more likely it is to be transmitted, which is the dose relationship you would expect if the maternal gut were the source.

Drawn from the paper's results, where strain level matching was the measurement.

What this does not show

  • This does not show that changing a mother's microbiome would change her infant's health. The cohort tracks natural transmission. No intervention altered anyone's microbiome, so the study cannot say what would follow if one did.
  • The eczema result is prediction, not cause. The maternal gut microbiome predicted infant eczema in this cohort. The authors do not demonstrate a causal pathway from one to the other.
  • It does not test vaginal seeding as a practice. What it measures is how much colonisation naturally comes from the vaginal route, which turns out to be occasional. Whether deliberately applying vaginal fluid to a caesarean infant does anything is a different question and a different study.
  • It is one country and one year. A Dutch cohort, followed to twelve months. Transmission patterns and the health outcomes that follow may differ elsewhere and later.

Where this leaves us

It is now more likely that the mother's own gut, rather than the birth canal, supplies most of what colonises an infant's gut. That refines an origin story in which vaginal transmission has carried most of the emphasis, while confirming that delivery mode still matters through some route this study does not isolate.

Mother and infant pairs in a Dutch birth cohort, followed from twelve weeks of pregnancy to twelve months postpartum.

An intervention that deliberately alters the maternal gut microbiome and then follows infant colonisation and health outcomes, which would move this from a map of transmission to a test of it.

Caveats worth holding

  • Observational cohort. Transmission is tracked rather than manipulated, so causal claims are not available from this design.
  • Two authors are affiliated with Nestle Research and Nestle Nutrition.
  • Single country cohort, and follow up stops at one year.
  • The eczema finding is a predictive association reported alongside a very large number of measured variables.

Newsletter

Each issue by email, when the newsletter launches. Leaving your address puts you on the list, nothing is sent yet.