Scientists Have Made a Human Microbiome From Scratch

Our bodies are home to hundreds or thousands of species of microbes — nobody is sure quite how many. That’s just one of many mysteries about the so-called human microbiome.

Our inner ecosystem fends off pathogens, helps digest food and may even influence behavior. But scientists have yet to figure out exactly which microbes do what or how. Many studies suggest that many species have to work together to do each of the microbiome’s jobs.

To better understand how microbes affect our health, scientists have for the first time created a synthetic human microbiome, combining 119 species of bacteria naturally found in the human body. When the researchers gave the concoction to mice that did not have a microbiome of their own, the bacterial strains established themselves and remained stable — even when the scientists introduced other microbes.

The new synthetic microbiome can even withstand aggressive pathogens and cause mice to develop a healthy immune system, as a full microbiome does. The findings were published on Tuesday in the journal Cell.

A better understanding of the microbiome could potentially lead to a powerful way to treat a host of diseases. Already, doctors can use the microbiome to treat life-threatening gut infections of the bacteria Clostridium difficile. They just have to transplant stool from a healthy donor, and the infection usually goes away.

“It works shockingly well,” said Dr. Alice Cheng, a gastroenterologist at Stanford University who led the new study.

Dr. Cheng and her colleagues can now use the new synthetic microbiome to learn about the role of each individual microbe, knowledge that could help doctors tackle other disorders. For example, the scientists could mix a cocktail of 118 of the 119 species in the lab and then see how the modified microbiome affects the health of mice.

“It’s something that has been badly needed for some period of time,” said Dr. Gary Wu, a gastroenterologist at the University of Pennsylvania School of Medicine who was not involved in the research.

Each of us harbors about 30 trillion microbes, roughly the same number as our own cells. But since bacteria are much smaller, they make up only a few pounds of our weight.

Before the 21st century, most of what was known about the human microbiome came from the few species that researchers managed to grow in a petri dish. In the early-2000s, scientists made a major advance by fishing DNA from samples of human spit, stool and skin. With those genetic sequences in hand, they created a catalog of species that live in our bodies.

The list was startlingly long, and many species were new to microbiologists. Making matters more confusing, most species live in some people but not others. There is no one human microbiome.

A number of researchers turned to mice to get better acquainted with some of these unfamiliar organisms. They reared germ-free animals in sterile cages and then put a broth made from human feces into the animals’ intestines. The microbes in that fecal transplant then started replicating in the animals.

These experiments have delivered some tantalizing results. For example, in some experiments, germ-free mice that received microbiomes from obese people put on more weight than did mice transplanted with microbiomes from people of average weight.

But pinpointing why these changes happen has proved harder. There’s no way to manipulate the microbiome in a stool sample, species by species. “It’s completely mixed up, and you can’t alter it,” Dr. Cheng said.

Some researchers have taken on this challenge by giving germ-free mice a single species of microbe and observing its effect. But those experiments have their own limits, since many microbes don’t work properly without ecological partners to help them.

Scientists have tried giving germ-free mice combinations of microbes. But so far, even the best efforts have left mice transplanted with fewer than 20 species — not the hundreds that live in humans. These miniature microbiomes leave the mice with poorly developed immune systems and metabolisms. “You get a mouse that doesn’t work,” said Lora Hooper, an immunologist at the University of Texas Southwestern Medical Center who was not involved in the new study.

In 2017, Dr. Cheng and her colleague at Stanford, Michael Fischbach, had long conversations about how to overcome the shortcomings of previous studies. “We needed to build an ecosystem from scratch,” Dr. Fischbach said.

They knew it would be difficult to grow a wide variety of microbes in the lab. And it was entirely possible that once in a mouse, their ecosystem would crash. “At the time, we could not have expected this to work,” Dr. Fischbach said.

First, Dr. Cheng and her colleagues drew up a list of 166 species that have been found in a sizable fraction of people. When they reached out to labs and companies, they managed to get hold of 104 of them.

Each microbe came with its own instructions for staying alive. To Dr. Fischbach’s surprise, Dr. Cheng figured out how to satisfy each of their fussy requirements to produce colonies in the lab.

Dr. Cheng mixed the 104 species together and put them into germ-free mice. Then she gave the microbes time to settle in — or die off. To see how her makeshift microbiome fared, she had to collect mouse droppings and work with colleagues to sequence all the DNA contained within.

Dr. Cheng found that the 104 species created a stable ecosystem inside the mice. Not only did the microbes endure in the animals, but the ecosystem’s structure didn’t change. Some microbes quickly became abundant and stayed that way. Others became rare but never disappeared. And the same ecosystem came into existence over and over again in different mice.

“It is remarkable how a hundred-plus human gut strains form a stable and resilient community,” said Kiran Patil, a University of Cambridge biologist who was not involved in the study. “It’s like a hundred-piece puzzle that looks daunting, but then you just mix and shake the pieces, and presto! The puzzle solves itself.”

Next Dr. Cheng and her colleagues put their microbiome to a test: They gave the mice stool transplants from human volunteers. Would the animals’ synthetic microbiome be resilient enough to withstand the onslaught?

It was indeed. Only seven of the original species disappeared. Some of the new species found empty places in the ecosystem and became a stable part of the microbiome.

“I lay down on a couch and was looking at the skylight,” Dr. Cheng said. “I had that ‘I can’t believe that worked’ kind of feeling.”

From that second experiment, Dr. Cheng and her colleagues perfected their microbiome. They picked out the 22 most successful newcomer species and added them to their microbial zoo, for a total of 119 species.

This new microbiome, which they have dubbed hCom2, is even more resilient than the first version. When the scientists gave hCom2 mice a stool transplant, none of the newcomers could establish themselves in the animals.

The researchers also tested how well the mice could cope with a potentially lethal strain of E. coli. In previous experiments, scientists have found that this strain can explode in the intestines of mice that have a miniature microbiome of 12 species.

Dr. Cheng and her colleagues gave their hCom2 mice a dose of E. coli and found that they resisted the invaders just as well as did mice that received a full human stool sample.

The hCom2 microbiome also had the same kind of influence on its hosts as did a full microbiome. The mice produced healthy levels of digestive fluids in the gut and developed full-fledged immune systems not found in germ-free mice.

Dr. Cheng and her colleagues have already started running experiments in which they leave out certain microbes from the cocktail to better understand how their microbiome works. They are also providing their bank of microbes to other researchers who want to run experiments of their own.

When asked if she intended to use the synthetic microbiome for her own research, Dr. Hooper responded succinctly: “Heck yeah.”

She hopes to use hCom2 mice to understand how the microbiome influences obesity. Part of the answer clearly lies in how the microbes help our intestines absorb fatty lipids from our foods. But studies on mice have not shed much light on which microbes are helping and which are getting in the way.

“We’ve had a really hard time getting to the answer to this question,” she said. “So this type of experimental system will give us a path forward.”