The microbes found in an animal's gut, collectively called the microbiome, may play an important role in evolution by reducing the viability of the offspring born from a male and female of different species, a new study published in the journal Science reports.

"It was a high-risk proposition," Seth Bordenstien, associate professor of biological sciences at Vanderbilt University, said in a press release. "The expectation in the field was that the origin of species is principally driven by genetic changes in the nucleus. Our study demonstrates that both the nuclear genome and the microbiome must be considered in a unified framework of speciation."

Bordenstein, along with graduate student Robert Brucker, conducted the research using three species of the jewel wasp Nasonia.

Two of the species they used, N. giraulti and N. longicornis, only diverged about 400,000 years ago, meaning they are still closely related genetically -- a fact reflected in their microbiomes.

The third species (N. vitripennis), on the other hand, diverged about a million years ago, resulting in greater differences in both its genome and microbiome.

Sure enough, the researchers found that the mortality of hybrid offspring from the two closely related species was relatively low, about 8 percent, while the mortality rate of hybrid offspring between either of them and N. vitripennis reached over 90 percent.

"The microbiomes of viable hybrids looked extremely similar to those of their parents, but the microbiomes of those that did not survive looked chaotic and totally different," Brucker reported.

The researchers were able to show that the incompatibilities killing the hybrids had a microbial basis by raising the wasps in a microbe-free environment. They were surprised to find that the germ-free hybrids survived just as well as purebred larvae, but when they gave the germ-free hybrids gut microbes from regular hybrids, their survival rate plummeted.

"Our results move the controversy of hologenomic evolution from an idea to an observed phenomenon," Bordenstein said. "The question is no longer whether the hologenome exists, but how common it is?"