Researchers examined the gut microbiomes of nearly 2,000 individuals and identified numerous potential new antibiotics. The standout candidate, prevotellin-2, exhibited anti-infective properties comparable to polymyxin B, an FDA-approved antibiotic used to combat multidrug-resistant infections. This suggests that the human gut microbiome could hold antibiotics that may eventually be used in clinical settings.
The human gut is home to approximately 100 trillion microbes, which are in continuous competition for scarce resources. “It’s a highly competitive environment,” states César de la Fuente, a Presidential Assistant Professor in various fields at the School of Engineering and Applied Science, Psychiatry and Microbiology in the Perelman School of Medicine, and Chemistry in the School of Arts & Sciences. “These bacteria coexist while also battling each other. This type of setting could spur innovation.”
Amidst this struggle, de la Fuente’s team sees a chance to find new antibiotics that could help bolster our defenses against drug-resistant bacteria. If gut bacteria must develop new ways to survive against each other, then we might utilize their own defenses against them.
In a recent publication in Cell, the research teams of de la Fuente and Ami S. Bhatt, a professor in Medicine (Hematology) and Genetics at Stanford, explored the gut microbiomes of almost 2,000 participants, uncovering many potential antibiotic candidates. “We regard biology as a reservoir of information,” notes de la Fuente. “Everything is encoded, and by creating algorithms to process this data, we can significantly speed up antibiotic discovery.”
De la Fuente’s lab has recently gained attention for uncovering antibiotic candidates from various sources, including the genetic data of extinct species like Neanderthals and woolly mammoths, as well as extensive bacteria samples analyzed through artificial intelligence. “Our main objective is to extract biological information worldwide for antibiotics and other beneficial molecules,” de la Fuente adds. “Instead of relying on traditional methods like collecting soil or water samples and isolating active compounds, we use the abundant biological data in genomes, metagenomes, and proteomes to identify new antibiotics swiftly.”
Considering that bacteria rapidly evolve, de la Fuente and his colleagues theorized that a competitive environment like that of the human gut could harbor many undiscovered antimicrobial substances. “When resources are scarce,” de la Fuente explains, “this is when biological innovation flourishes.”
The team focused on peptides, which are short amino acid chains known for their potential as novel antibiotics. “We computationally analyzed over 400,000 proteins,” de la Fuente says, referring to how AI interprets genetic codes and predicts which sequences might possess antimicrobial qualities based on a library of known antibiotics.
“Interestingly, the compounds we found differ from traditionally recognized antimicrobials,” states Marcelo D.T. Torres, a research associate in de la Fuente’s lab and the paper’s lead author. “The substances we identified represent a new category, and their unique characteristics will enhance our understanding and broaden the antimicrobial sequence diversity.”
Of course, these predictions need to be confirmed through experiments; after identifying a few hundred potential antibiotic candidates, the researchers chose 78 for testing against actual bacteria. They synthesized these peptides and exposed bacterial cultures to each one, observing results after 20 hours to determine which peptides effectively inhibited bacterial growth. They further evaluated the antibiotic candidates in animal models.
More than half of the tested peptides inhibited the growth of either beneficial or harmful bacteria, and the leading candidate, prevotellin-2, showed anti-infective effects similar to those of polymyxin B. This suggests that antibiotics in the human gut microbiome could eventually have clinical uses.
“Finding prevotellin-2, which performs similarly to polymyxin B, one of our last-resort antibiotics, came as a significant surprise to me,” says Bhatt. “This finding indicates that exploring the human microbiome for new and innovative classes of antimicrobial peptides is a hopeful direction for researchers and healthcare professionals, especially for patients.”
This research was conducted at the University of Pennsylvania’s School of Engineering and Applied Science, the Perelman School of Medicine, and Stanford University. Professor de la Fuente holds a Presidential Professorship at the University of Pennsylvania and receives support from various funding sources, including Procter & Gamble, United Therapeutics, a Brain & Behavior Research Foundation Young Investigator Grant, the Nemirovsky Prize, Penn Health-Tech Accelerator Award, and the Defense Threat Reduction Agency. Professor Bhatt was funded by a Paul Allen Distinguished Investigator Award and the NIH. Additional financial support was provided by the Langer Prize (American Institute of Chemical Engineers Foundation), the National Institutes of Health, and the Defense Threat Reduction Agency. Other co-authors include Angela Cesaro at Penn and several researchers at Stanford.
