Researchers at Keio University School of Medicine in Tokyo and the Broad Institute of MIT and Harvard have isolated 18 bacterial strains from stool from healthy people that could potentially be a more effective treatment. The team found that these strains suppress the growth of Enterobacteriaceae and alleviate inflammation in the guts of mice by competing with the harmful bacteria for carbohydrates and preventing them from colonizing the intestine.

The findings, which appear today in Nature, could lead to the development of a microbial transplant for patients that manages antibiotic-resistant bacteria in a more targeted way and with fewer side effects than current treatments.

“Despite two decades of microbiome research, we are just beginning to understand how to define health-promoting features of the gut microbiome,” said Marie-Madlen Pust, a computational postdoctoral researcher at Broad and co-first author on the paper.

“Part of the challenge is that each person’s microbiome is unique. This collaborative effort allowed us to functionally characterize the different mechanisms of action these bacteria use to reduce pathogen load and gut inflammation,” she said.

“Microbiome studies can often consist of analyzing collections of genetic sequences, without understanding what each gene does or why certain microbes are beneficial,” said Ramnik Xavier, co-senior author on the study and a core institute member at Broad. “Trying to uncover that function is the next frontier, and this is a nice first step towards figuring out how microbial metabolites influence health and inflammation.”

Pust is in the lab of Xavier, who is co-director of its Infectious Disease and Microbiome Program. Xavier is the Kurt J. Isselbacher Professor of Medicine at Harvard Medical School; director of the Center for Computational and Integrative Biology at Massachusetts General Hospital (MGH); and co-director of the Center for Microbiome Informatics and Therapeutics at MIT.

Kenya Honda of the Keio University School of Medicine is co-senior author of the study. Munehiro Furuichi, Takaaki Kawaguchi, and Keiko Yasuma-Mitobe, all researchers at Keio University, are co-first authors. In this work, the Honda lab used specialized culture techniques and animal models to analyze bacterial infections, while the Xavier lab developed software to analyze unknown microbial metabolites.

Bacterial balances

Antibiotic-resistant Enterobacteriaceae such as E. coli and Klebsiella bacteria are common in hospitals, where they can proliferate in the gut of patients and cause dangerous systemic infections that are difficult to treat. Some research suggests that Enterobacteriaceae also perpetuates inflammation in the intestine and infection by other microbes. Honda, Xavier, and their colleagues wanted to understand which specific bacteria in fecal microbiota transplants could help protect the intestinal microbiome against Enterobacteriaceae. Honda’s team isolated about 40 strains of bacteria from each stool sample from five healthy donors and used them to treat mice infected with E. coli or Klebsiella. They tested different combinations of strains and identified a group of 18 strains that suppressed the Enterobacteriaceae the most.

The Keio University researchers found that in Klebsiella-infected mice treated with the 18 beneficial strains, Klebsiella altered the expression of genes involved in carbohydrate uptake and metabolism. This included downregulating gluconate kinase and transporter genes — indicating increased competition among the gut microbes for nutrients.

Xavier’s team wanted to study samples from patients with and without gut inflammation. In partnership with the Broad’s Metabolomics Platform, led by senior director and study co-author Clary Clish, they analyzed samples from pediatric patients with ulcerative colitis, looking for the presence of alternate gluconate pathway genes of gut microbes and fecal gluconate levels. They found higher levels of gluconate linked to more gluconate-consuming Enterobacteriaceae in samples from pediatric patients with ongoing inflammation, indicated by high levels of the protein calprotectin.

Together, the findings suggest that Enterobacteriaceae processes gluconate as a key nutrient and contributes to inflammation in patients. But when a gut microbiome includes the 18 helpful strains, they likely compete with Enterobacteriaceae for gluconate and other nutrient sources, limiting the proliferation of the harmful bacteria.

The 18 strains also did not disrupt the growth of other healthy bacteria in animals with gut microbes from patients with Crohn’s disease and ulcerative colitis, further underscoring their therapeutic promise.

Although more work will be needed to shed light on the precise mechanisms underlying how different bacteria compete with each other, the findings suggest that microbial therapeutics could be used to tweak the ecology of the gut and suppress harmful bacterial infections with fewer negative side effects than typical antibiotic treatments.

In the meantime, the team aims to uncover the identity and function of unknown metabolites that contribute to gut health and inflammation.