Researchers say they have identified how infants thrive on what were previously believed to be indigestible components of milk from humans and cows—a development that could point the way to helping save some seriously ill children. The research began with a puzzling question: Why is about a third of the content of breast milk completely indigestible by newborn babies? It turns out that indigestible components called oligosaccharides—complex molecules of sugars and proteins—are food for bacteria that live in the human gut and are crucial to early development. Breast milk and these microbes have coevolved in a symbiotic relationship over 400 million years of mammalian and human evolution, says University of California, Davis, biochemist David Mills, who led some of the new research.

Bifidobacterium longum subspecies infantis is usually the first bacterium to colonize a newborn’s gut. It produces lactic acid that inhibits growth of pathogenic organisms, and it serves as the foundation for the development of the adult gut’s complex microbial ecosystem. B. infantis recedes in number and importance as the infant is weaned and exposed to a greater variety of microbes—but if it is not present early on, then other bacteria, perhaps pathogenic ones, will fill the ecological niche.

Scientists hope recent findings about this interplay may lead to a better treatment for necrotizing enterocolitis (NEC), an often deadly condition associated with a lack of protective microbes in a premature or newborn infant’s gut. It can be caused by insufficient exposure to the helpful organisms, pathogens crowding them out or a shortage of proper “food” for them—and in these beneficial microbes’ absence, pathogens run rampant and kill intestinal cells. NEC occurs around the world and mortality is about 40 percent, frequently from sepsis; survivors often require extensive intestinal surgery.

Mills and others believe supplements of B. infantis and oligosaccharides might protect children—both those at risk for NEC or in its early stages—and early findings presented at conferences have been encouraging. Mills co-founded and serves as a consultant to Evolve Biosystems, a company working on commercial application of the researchers’ findings.

Earlier research had found that B. infantis has an enzyme called EndoBI-1 “that sits on the outside of the cell and just lops off the whole oligosaccharide from the protein” of the milk molecule, Mills explains. His most recent paper, published last week in Applied and Environmental Microbiology, clearly establishes that it is the sugar component rather than the protein that allows B. infantis to flourish. This occurs with both breast milk and cow’s milk, although the sugars are more plentiful in human milk.

The cleaving process that separates the sugar molecule may also make the remaining protein portion more digestible for an infant, and Mills is investigating that hypothesis. He says it makes sense from an evolutionary perspective that most useful parts of a food would eventually be utilized by the body.

Washington University in Saint Louis systems biologist Jeffrey Gordon has discovered much about how food and microbes interact within the gut ecosystem in health and disease. In a paper published in February he described how he collected gut microbiota from a starving Malawi child, transplanted it into a mouse model—with mice raised in a sterile environment and lacking a typical gut microbiota—and then challenged that system with various interventions. Gordon hoped reactions to those challenges might tell him what the transplanted mouse ecosystem was lacking, and therefore why the child was malnourished. One of the interventions was “bovine milk oligosaccharides [which] dramatically improved weight gain in the mice,” says Mills, who collaborated on that study. The result suggested the child had a deficiency that could be ameliorated by supplementing those missing sugars.

But that wasn’t the whole story; B. infantis did not survive transplantation to the mice, so other bacteria, including a nonpathogenic strain of Escherichia coli, feasted on the sugars. Mills sees this as a warning that there may not be a single silver bullet solution—sometimes more than one component may be needed in order to restore balance to a gut ecosystem. He worries that if the milk sugars had been given to an animal or child carrying a less-benign strain of E. coli, it might have made the things worse.

Understanding the function of sugars in breast milk is important but any broad application of those findings on a commercial scale would likely involve cows. “Cows’ milk glycoproteins are not the same as human glycoproteins,” says Bernd Stahl, director of human milk research at Nutricia Research in the Netherlands. Stahl, who did not participate in Mills’s or Gordon’s research, finds their work interesting because it shows that beneficial bacteria can harvest and use complex carbohydrates—and can do so across species.

Milk has evolved to fit the needs of different species’ offspring and the respective paces of their physiological development, so content differs from mammal to mammal, Stahl explains. “A rabbit needs only six days to double its birth weight but a human being takes 180 days,” he says. That is possible because protein content of rabbit milk is 14 percent, cow’s milk 3 to 4 percent and human milk only 1 percent.

Most infant formula products are based on cows’ milk and adapted to the specific needs of human babies; initially they used higher protein content but that has been reduced in contemporary products because of fears of obesity. Stahl believes the basic research findings of Mills’s team “will add to ways that cows’ milk–derived products can mimic the complex sugars found in human milk—still the ‘gold standard’—to support beneficial bacteria.”