COLD SPRING HARBOR, N.Y. — Microbes aren’t completely the boss of their human hosts. People’s genes may have a say in which microbes come to live in and on the human body, a new study suggests.
Recent research has shown that the mix of microbes living in and on the human body is associated with some diseases. But exactly what determines which microbes settle a particular human host has been a mystery. Diet and geography are partially responsible, but the part human genetics plays in determining the microbial mix on the body has been unclear.
“We know there is a genetic component,” says Ran Blekhman, a geneticist at Cornell University. “We’re just not sure how big it is.”
To find out, Blekhman and his colleagues turned to data collected by the Human Microbiome Project, an effort to genetically catalog the microbes living in and on the human body. Though the project looks for bacterial DNA in swabs of skin, mouths, feces and other sources, some human genetic material is shed in the samples too. The researchers combed the bacterial DNA data for traces of human DNA contamination, and found enough to reconstruct genetic profiles of 100 people.
Comparing the human and bacterial data revealed 51 different human genetic variants that are associated with the relative abundance of certain bacteria living in or on 15 body sites. Some of those genetic variants and the microbes they were associated with have also been linked to diseases. People with a genetic variant near the PCSK2 gene, which is involved in producing insulin, have more Bacteroides bacteria in their intestines, Blekhman reported May 9 at the Biology of Genomes meeting. That same genetic variant has been linked to type 2 diabetes. So has an overabundance of Bacteroides.
People who have a version of the CXCL12 gene previously associated with inflammatory diseases also carry more Granulicatella bacteria on their skin. Those bacteria have previously been linked to skin inflammation.
The findings present a chicken-versus-egg problem, Blekhman says. Still undetermined is if the bacteria are triggering disease in people who carry certain genetic variants, or if the diseases caused by genetic variants lead to more growth of some types of bacteria.
Doctors might be able to use bacterial mixes as markers that patients are at risk of getting certain diseases, says Benjamin Voight, a geneticist at the University of Pennsylvania. But first the researchers will need to establish a convincing statistical argument that genes, diseases and microbes are linked. “There are arrows pointing in the right direction,” Voight says. “It’s an interesting observation.