Diet strongly influences intestinal bacteria populations

By Daniel Strain

Web edition: May 19, 2011

On the Serengeti plain, gut bacteria reveal who’s been eating grass and who’s been sampling the gazelle.

What a mammal eats seems to determine the microbial species that live in its intestines and what jobs those microbes take on, an international team of researchers reports May 20 in Science. The nutrients inside meat and veggies predictably influence the different communities of friendly bacteria that inhabit both vegetarian and meat-eating species, the group says.

“This kind of work is foundational,” says Martin Blaser, a microbiologist at New York University who was not involved in this study. “It helps to define the macro rules that govern these things.”

In the study, researchers picked microbes from the feces of 33 mammal species and took a close look at the genes those bugs carried. Of 1,500 pieces of genes that seemed to code for enzymes, 495 were associated with microbes belonging to either herbivores, like the hapless gazelle, or carnivores, such as the lion. Herbivore and carnivore microbes may be different, but they also seem to have different jobs that are determined by diet, says coauthor Jeffrey Gordon, a systems biologist at Washington University in St. Louis.

Carnivores’ microbes carry genes seemingly suited to breaking down the abundant amino acids in the diets of their hosts. Herbivores’ bacteria, on the other hand, appear to be better at building new amino acids from scratch. The patterns seem to make sense given the lion’s protein-rich and the gazelle’s protein-poor diet, suggests study coauthor Brian Muegge, also at Washington University. For herbivores, “Those bugs need to make the amino acids they need for life,” he says.

Proteins and other nutrients also hold sway over the gut bacteria within any given species, Gordon adds. In a separate study appearing online May 19 in Science, he and his colleagues took 10 species of intestinal bacteria from humans and put them into mice. By precisely tweaking the ratios of nutrients present in the mice’s diet, the team could predict 60 percent of the changes to the friendly microbes that followed.

Such microbial sculpting may help provide tools for improving human health, Gordon says. “At a time when nutrition represents a global health problem,” he says, “how do we make appropriate and informed and more helpful recommendations for what we should eat?”

Still, it’s not yet clear what communities of bacteria, down to different strains of the same species, will do best for specific clinical outcomes, says David Relman, an infectious-disease researcher at Stanford University. Though general rules like those outlined in the study may help scientists as they dig deeper, he says, there are “all sorts of details we’re glossing over by looking at poop samples.”