Gut Feeling: Microbiome at Work

No one would dispute that what we eat, and how we eat, are important to our health. But in recent years, it has become increasingly clear that it's not enough to think simply in terms of a binary system — our bodies, and the food we put in them. The fact is, our bodies, or more particularly, our digestive tracts, are, it turns out, host to many other creatures: microorganisms that are in us, but not, strictly speaking, part of us. This community of microorganisms — the microbiome — has a profound impact on our health. Bacteria are the big players here — the mix of bacteria in our digestive systems influences the way we break down and process the foods we eat; it facilitates our absorption of nutrients and helps us obtain the vitamins that we need.


Because the bacteria inhabiting our digestive tracts, or guts, are so critical to our everyday life, it's not surprising that scientists have been delving into how all these bacteria function. There is a lot to know: our bodies (made up of roughly 10 trillion cells) house about 100 trillion microbes — so we are still in the early stages of teasing apart the processes at work.
A new paper has made an important breakthrough, bringing a little order to the chaos. Manimozhiyan Arumugam and Peer Bork, both of the European Molecular Biology Laboratory in Heidelberg, Germany, and Jeroen Raes at the University of Brussels, along with 82 colleagues from Japan, China, Brazil and a number of European countries, recently published a study in Nature that finds that there are apparently three distinct types of microbiomes in the human gut. Given the multiplicity of both microbes and humans, this is a striking result.

Looking for Hay in a Haystack

The research team was not so much searching for a needle in a haystack as trying to take every piece of hay in the stack, categorize it and sort it; instead of hay, though, the scientists were working with fragments of DNA. The group took fecal samples from 22 people (selected for diversity from Denmark, France, Spain and Italy) and extracted DNA fragments from the samples. After sequencing the DNA fragments, they found that the vast majority of the fragments belonged to bacteria (92.76%), with a few viruses (5.8%), archaea (0.8%), other eukaryotes (0.5%), and human contaminants (0.14%) thrown in.
Arumugam, Bork and colleagues focused on the bacterial DNA fragments. The study also looked at bacterial DNA fragments extracted from samples collected from 11 other people in a previous study (bringing the total sample size to 33 individuals). Then they compared the DNA fragments against genome databases, matching their DNA sequences to those of known species. They met with remarkable success; on average they were able to identify the phylum of 80% of the DNA fragments in a sample, and the genus of 52.8% of the fragments. (As a refresher, scientific taxonomy goes — from general to specific — in the following order: kingdom, phylum, class, order, family, genus, species; getting to the genus level is pretty impressive for something as diverse as bacteria). Once these identifications had been made, the fragments were sorted into their phylogenetic groups. (The phylogenetic group of a sample would be the group it belongs to, based on evolutionary development and history.)

Three's No Crowd

The results of this sorting were unexpected and remarkable. While everyone may, in fact, be as individual as a snowflake, it turns out that our gut microbiomes are not; only 3 types of gut microbiomes, or enterotypes, were found in the 33 people studied. Within each enterotype, members of one genus of bacteria vastly outnumbered members of all other genera. The three different enterotypes were named for the dominant genus of bacteria in that enterotype: Bacteroides, Prevotella, and Ruminococcus. The division into three enterotypes was based not just on the dominant genus; individuals who had the same dominant genus of bacteria in their gut also shared a characteristic makeup for the remaining bacteria — that is, the genera present in smaller amounts, and those that were missing, were specific to the enterotype. This finding implies that the enterotypes depend on the composition of the entire bacterial community in the digestive tract, not just the presence of one genus.


Using a small sample size (and 33 is fairly small) can confound data analysis; it can overlook certain features or give unwarranted importance to others. So although their results seemed to be conclusive, the researchers conducted additional analysis to try to ensure that the bacterial genera pattern they found was not an artifact of their relatively small sample size.
They performed the same analysis as above on data from two previously published gut microbiome studies (involving a total of 139 subjects from Denmark and America), and came up with the same results. Arumugam, Bork, and their collaborators then analyzed their original data (the first 33 individuals) from another perspective. They divided the bacterial DNA fragments by orthologous genes, genes that are shared between genera, because they were inherited from a common ancestor. Investigating orthologous genes allows the researchers to explore the possible functions performed by the bacteria, not just their relatedness (which the DNA sequencing analysis above does), because the genes possessed by a bacteria dictate what it is able to do. The distribution of orthologous gene groups was, with a few exceptions, the same as that of bacteria genera, further supporting the study's conclusion that human gut bacteria are separated into discrete groups (both functionally and phylogenetically) within a larger community and not just mixed together in a random assortment. This analytical approach also had a surprise in store for the researchers; although most of the more common functions performed by gut bacteria were found to be carried out by the dominant bacteria genera, as might be expected, some were performed by minor genera. This finding strengthens the idea that to really understand what is happening in the human gut, all of the bacteria and their contribution to the microbiome must be explored.

Only Skin Deep

The orthologous gene groups correlated with several different characteristics of the study group. Ten gene groups were found to vary with nationality, and 12 varied with age. One orthologous gene group and five other biomarkers correlated with gender, and three biomarkers correlated with the BMI (body mass index) of the experimental subjects.
Notwithstanding the correlation of some orthologous gene groups, there was no overall connection between enterotypes and common individual traits, such as nationality, age, gender and body mass index (BMI). The one exception to this observation was that the Japanese study participants had a higher likelihood of having the Bacteroides enterotype than either of the other two.


While the research has made an important contribution to the study of the human gut microbiome, it's too soon to know where further study of enterotypes will lead. Clearly, it can be hoped that the knowledge might ultimately permit more personalized diets and drug treatments, based on an individual's digestive microbial ecology. It is even conceivable that an improved understanding of enterotypes may even assist in the diagnosis of certain diseases.

Discussion Questions

What other steps could researchers take to try to nail down the claim that there are just three basic enterotypes? Can you think of other factors, besides those mentioned above (nationality, age, gender, etc.), that could be looked at for possible correlations with enterotypes? How might one investigate whether a person's enterotype remains unchanged throughout life or is subject to variation?