Stanford scientists build first synthetic human microbiome from scratch

A team of researchers from Stanford University has constructed the first synthetic microbiome model, built entirely from scratch and encompassing more than 100 different bacterial species. It's hoped the achievement will revolutionize gut microbiome research by offering scientists a consistent working model for future experiments.

Trillions of microbes live inside our guts. Perhaps one of the most significant discoveries in medical science over recent decades has been how deeply these microbes influence our general health. From affecting how well drugs we consume work, to modulating our immune systems, the gut microbiome plays a powerful role in all aspects of our health.

It’s also mind-bendingly complex. No two people share exactly the same gut microbiome composition. And while researchers frequently home in on ways particular bacteria influence metabolic mechanisms, it has been difficult to translate these findings into actual clinical therapies for humans.

Michael Fischbach, corresponding author on the new study, said the foundation for this research was the realization that science needs some kind of objective gut microbiome model so studies can better understand what particular interventions lead to beneficial health outcomes. Fischbach said there were two specific motivational factors underpinning this research that has spanned over five years.

“First, we were intrigued by experiments in which a (complete, undefined) fecal sample was transplanted human->mouse & a phenotype came along for the ride (e.g. response to anti-PD1). Fascinating but hard to figure out which strains/genes are involved,” he explained on Twitter. “Second, we're interested in the chemistry of the microbiome, with an emphasis on mechanisms. We became unsatisfied with experiments in which we colonize mice with defined but incomplete communities (to test mechanism of a molecule); they often fail to recapitulate normal physiology.”

So the first step was trawling through masses of prior human microbiome research to come up with a shortlist of the most prevalent bacteria found in most people. The research team homed in on 104 bacterial species and dubbed this first microbiome iteration hCom1.

After growing each bacterial species individually, and then mixing them all together, the researchers introduced hCom1 to germ-free mice, animals developed to harbor no natural microbiome. Incredibly, hCom1 was a stable microbial ecosystem when transplanted into the mice. While some bacterial species became more prevalent than others, the 100ish species found a relatively stable balance and the animals were found to be metabolically normal.

The next step was filling the bacterial gaps that the original microbial composition was likely lacking. To do this the researchers challenged hCom1 mice with a human fecal sample. Based on a theory called colonization resistance, the researchers hypothesized any unfilled bacterial niches in hCom1 would be filled by these new invaders.

But Fischbach noted, not everyone thought this part of the experiment would work. Some believed the human fecal sample would completely overtake this artificial community of bacteria the researchers had collected.

"The bacterial species in hCom1 had lived together for only a few weeks," explained Fischbach. "Here we were, introducing a community that had coexisted for a decade. Some people thought they would decimate our colony."

The challenge was successful. For the most part, the bacterial community assembled by the researchers withstood the battle with a human microbiome.

Around 20 new bacterial species were found to have successfully colonized hCom1, and a small handful of previously selected bacteria died off. Ultimately the researchers cataloged 119 bacterial strains, dubbing this second generation of microbiome hCom2. This hCom2 microbiome community was found to function as effectively as any general microbial composition in mice.

“Mice colonized by hCom2 look normal immunologically, have similar microbiome-derived metabolites, and exert colonization resistance against E. coli,” said Fischbach. “There are improvements to make, but we think hCom2 (in its current form) is a good model system of the microbiome.”

So what now? Well, Fischbach and his team are keen to get their microbiome model out to as many researchers as possible. They believe the real impact from this work will be come from other scientists’ research, allowing, for the first time, a consistent microbiome model for studies to build off.

Further down the line the researchers envision a future where patients receive transplants of engineered communities of bacteria. Working toward this, Fischbach is director of the recently founded Stanford Microbiome Therapies Initiative (MITI). The initiative will improve on their microbiome models.

The new study was published in Cell.

Source: Stanford University