When a baby latches, the exchange isn’t just nutritional. Microscopic passengers in breast milk (bacteria adapted to digest milk sugars and support early development) transfer directly into the infant’s developing gut. This silent microbial hand-off shapes everything from how babies absorb nutrients to how their immune systems mature, yet scientists haven’t always known exactly which microbes make the journey or how persistently they settle in.
New research tracking 195 mother-infant pairs over six months provides the clearest picture yet of this biological dialogue. Published in Nature Communications, the study analyzed 507 metagenomic samples and found that breast milk carries a distinct microbial community dominated by Bifidobacterium longum. This single species drove most of the overlap between milk and infant gut microbiomes, and infants whose guts were dominated by it showed significantly more stable microbial communities over time.
The researchers identified 12 instances where identical bacterial strains appeared in both a mother’s milk and her baby’s stool. That’s the microbial equivalent of matching fingerprints. These shared strains included helpful commensals and so-called pathobionts like Klebsiella pneumoniae, microbes that live harmlessly in healthy people but can cause problems under certain conditions. Because all participants were healthy, their presence reflects normal diversity rather than disease risk.
The Mechanics of Microbial Seeding
Getting these results required overcoming significant technical challenges. Breast milk is notoriously difficult to study. It’s packed with fat and human cells but carrying relatively few bacteria. The team used shotgun metagenomic sequencing rather than older, lower-resolution methods, allowing them to identify bacteria not just by species but by exact strain.
“Metagenomic analysis is trickier and more complicated, but it really paid off because it allowed us to obtain information at the level of different bacterial strains,” Pamela Ferretti explains.
That precision revealed something unexpected: the transfer isn’t entirely one-way. Streptococcus salivarius and other mouth-associated bacteria turned up in milk samples, suggesting “retrograde flow” where tiny amounts of the baby’s oral bacteria travel back into the breast during feeding. It’s a vivid reminder of breastfeeding’s physical intimacy, where fluids and microscopic life move in both directions.
The study also uncovered antimicrobial resistance genes shared between mothers’ milk and their infants’ guts. While resistance genes can sound alarming, the authors emphasize this reflects normal microbial inheritance in healthy populations rather than antibiotic exposure or infection.
When Seeding Matters Most
The microbial overlap between milk and infant gut actually decreases as babies age. At one month, roughly 10 percent of bacterial taxa in infant stool also appeared in their mother’s milk. By six months, that figure dropped to 7 percent. The seeding process is most active in the very earliest weeks of life, when the gut is essentially empty real estate being colonized for the first time.
Infants born vaginally showed more persistent bacterial strains compared to those delivered via C-section, adding another dimension to how birth method and feeding practices work together. The dataset nearly doubles the amount of public metagenomic breast milk data available, opening possibilities for tracking whether these early microbial patterns predict later health outcomes like asthma or obesity risk.
For now, the findings reframe breast milk as something beyond nutrition and immunity. It’s a living system that delivers microbes already adapted to build a functioning gut community from the ground up. Those first bacterial passengers, transferred in silence during every feeding, lay groundwork that may echo through a lifetime of health.
Nature Communications: 10.1038/s41467-025-66497-y
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.