Almost 180 million children across the globe are stunted, a severe, disabling consequence of malnutrition and repeated childhood infections that puts them at risk for cognitive impairment and disease. New studies now point to another player in stunting: the gut microbiome. The right combination of microbes, it seems, can tip the balance between stunting and healthy growth, even when calories are scarce—a tantalizing, if preliminary, clue to possible interventions.

The three studies, reported in this week’s issues of Science and Cell, “are a watershed moment in global health generally, and in nutrition specifically,” says William Petri, Jr., an infectious disease expert at the University of Virginia in Charlottesville. Petri was not involved in the current work, but he has spent years tracking the health of infants in Bangladesh. He and others have been long been frustrated by the inability of dietary supplements to reverse the negative effects of poor nutrition.

When Petri heard that gut bacteria might influence obesity, he reasoned that they might also influence a person’s response to hunger. So he and Tahmeed Ahmed from the International Centre for Diarrheal Disease Research in Bangladesh teamed up with Jeffrey Gordon, a microbiologist at Washington University in St. Louis in Missouri, to collect monthly fecal samples from healthy and malnourished Bangladeshi children under 2. Petri and Gordon found that as children matured, their community of gut bacteria normally shifted as well. But as they reported in 2014, stunted children didn’t have the appropriate bacterial community for their age, but an “immature” one more typical of a younger child.

Gordon’s team reports finding the same pattern in infants in Malawi, and they present evidence that these microbial communities influence growth. Working with mice bred to have no gut microbes of their own, Gordon’s graduate student Laura Blanton fed them a mash of the same food typically eaten by Malawian children. Germ-free mice given the “immature” microbiomes of children with symptoms of malnourishment grew poorly, whereas mice on the same diet given “mature” microbiomes of healthy children put on more muscle and developed seemingly denser bones.

François Leulier, a biologist from the École Normale Supérieure de Lyon in France and colleagues report a similar finding. His postdoc Martin Schwarzer showed that young, germ-free mice don’t put on as much muscle or grow bones as big as mice that have the normal complement of bacteria, even when they eat the same amount of food. The team also glimpsed a mechanism: They found that the microbes affect the animals’ own hormones.

In healthy animals, growth hormone stimulates an increase in a second hormone, insulinlike growth factor 1 (IGF-1), which in turn promotes tissue growth. Germ-free mice still have the same amount of growth hormone as other mice, but the activity of IGF-1 in their blood, liver, and muscles is lower, Leulier’s team found. It’s not clear how the microbes influence the hormones. But injecting germ-free mice with IGF-1 brought their growth on par with the other mice, as did giving them one particular strain of lactobacillus.

A breakdown in this microbiome-hormone connection may help explain stunting in undernourished children. But why do only some of them end up with an immature microbiome? In the Cell study, Gordon, graduate student Mark R. Charbonneau, and their colleagues show that breast-feeding may help the right microbes get established and set the baby’s growth on the right trajectory. Healthy mothers typically produce modified sugar molecules called sialylated human milk oligosaccharides. Babies don’t make use of these nutrients, but their gut microbes thrive on them, recent research has shown. The new work shows that mothers of children who show signs of severe malnutrition make less of this microbiome “food.”

When the researchers added oligosaccharides purified from whey to the Malawian diets of mice with microbes from a severely malnourished infant, the mice grew more muscle, bigger bones, and had quite “dramatic changes” in brain and liver metabolism, Gordon says. He speculates that by processing these sugars, the bacteria may in turn produce molecular building blocks to help the host’s body grow well. The group has also seen this beneficial effect in germ-free piglets, whose physiology is more humanlike than mice, they report.

To David Relman, a microbiologist at Stanford University in Palo Alto, California, the implications are “profound. … The nutritional status of children might be modifiable through manipulation of the gut microbiota.” Gordon and Leulier plan to test that possibility in people, particularly malnourished children, this year with dietary interventions.

A safe and effective probiotic “would be tremendous,” Petri says. Success in laboratory animals may not translate to humans, of course. And researchers need to be sure that any added microbes don’t cause unwanted effects, such as inflammatory disease or obesity, adds Eric Pamer, an infectious disease expert at the Memorial Sloan Kettering Cancer Center in New York City. But one thing is certain, he says: “The health-promoting impact of [these] microbiota is astonishing.”