Are Your Gut Bacteria Vegetarian?

This spring, scientists announced that each person seems to have a signature set of gut bacteria, like a blood type for the microbiome. Their human subjects fell into three separate "enterotypes," each one representing a distinct microbial ecosystem. The enterotypes didn't correlate with subjects' age, gender, or nationality. But a new study has found something that does predict what enterotype you'll host: eating a plant- or animal-based diet.

Researchers surveyed 98 subjects about their diets, both recently and in the long term. They also took stool samples from their subjects. The scientists gave this portion of their study the very belabored acronym COMBO, for "Cross-sectional study Of diet and stool MicroBiOme composition."(It's not clear whether COMBO is meant to reference the combination of survey data and fecal data, or the resemblance a cheese-filled pretzel tube has to a cross-section of the colon.)

The researchers used these diet surveys to create profiles of subjects' nutrient intake. They also sequenced the DNA in subjects' stool samples to create profiles of their bacterial communities.

What they found was that subjects could be clustered into two bacterial ecosystems. One group's guts were dominated by bacteria in the genus Bacteroides; the others were dominated by Prevotella bacteria. (These match up with two of the three enterotypes described in the earlier study. In the current study, there was some evidence for the third enterotype the other authors had described--characterized by Ruminococcus bacteria--but in most statistical analyses this enterotype was merged with Bacteroides.)

Comparing enterotype data to dietary data, they saw a strong correlation between a person's bacterial type and his or her preferred nutrient sources. The Bacteroides enterotype showed up in people who consumed more animal proteins and fats. The Prevotella type, on the other hand, went along with higher consumption of carbohydrates and fiber--in other words, plants.

The second portion of the study was called CAFE, short for Controlled Feeding Experiment. You might notice that there's no A in the experiment's actual title--but that's fitting, given that the experiment was basically the opposite of a café. A patron in a café usually expects to choose food from a menu, whereas the 10 subjects in the CAFE experiment had to stay in the research center for 10 days and eat exactly what was put in front of them, no more or less. (This sounds more like a CAFO, or Concentrated Animal Feeding Operation, also called a factory farm.) Half the subjects were put on a diet that was high-fat/low-fiber, and the other half ate a low-fat/high-fiber diet. All subjects ate identical meals; the only difference between the two diets was in the proportions of foods on the plate.

Also unlike in a café, subjects had to provide daily stool samples to the researchers. All 10 subjects started out with the Bacteroides (high protein and fat) enterotype. Over the course of their 10-day diet, none of the high-fiber subjects switched over to the opposite enterotype. Even though their bacterial types seemed to be based on their long-term diets, a 10-day change in diet wasn't enough to seriously affect subjects' bacterial ecosystems.

The different diet did affect the members within each ecosystem, though. Within 24 hours of starting their new diet, subjects' bacterial populations shifted noticeably. So while a temporary change in diet may not permanently alter your enterotype, it could significantly change the balance of different microbes within that enterotype.

In their dietary surveys, the researchers found a few other factors that were tied to the balance of microbes in subjects' guts, but not to a particular enterotype. BMI was one unsurprising example. Others included red wine and aspartame consumption--suggesting either that aspartame kills off gut bacteria, or that something has actually evolved to live on it.

The earlier study, which used a smaller sample size, found no link between enterotype and nationality. But a larger sample size might be able to capture connections between gut bacteria and nationality that reflect different countries' dietary traditions. The current paper references a study that compared European children to children in the African nation of Burkina Faso. While European children tended to have the Bacteroides (animal nutrients) enterotype, the African children were more likely to have the plant-related Prevotella enterotype, reflecting their high-carbohydrate diet.

Increasingly, connections are cropping up between human health--both mental and physical--and the microbiome. We know very little about our relationship with our gut bacteria, but untangling more connections like this one will help us understand how to stay on good terms with them. In the future, keeping ourselves healthy may mean keeping our internal ecosystems well-fed.

Wu, G., Chen, J., Hoffmann, C., Bittinger, K., Chen, Y., Keilbaugh, S., Bewtra, M., Knights, D., Walters, W., Knight, R., Sinha, R., Gilroy, E., Gupta, K., Baldassano, R., Nessel, L., Li, H., Bushman, F., & Lewis, J. (2011). Linking Long-Term Dietary Patterns with Gut Microbial Enterotypes Science DOI: 10.1126/science.1208344