Scientists at Stanford University and the University of Pennsylvania have discovered a new antibiotic candidate in a surprising place: the human gut.
In mice, the antibiotic — a peptide known as prevotellin-2 — showed antimicrobial potency on par with polymyxin B, an antibiotic medication used to treat multidrug-resistant infections. Meanwhile, the peptide mainly left commensal, or beneficial, bacteria alone. The study, published in Cell, also identified several other potent antibiotic peptides with the potential to combat antimicrobial-resistant infections.
The research is part of a larger quest to find new antibiotics that can fight drug-resistant infections, a critical public health threat with more than 2.8 million cases and 35,000 deaths annually in the United States. That quest is urgent, said study author César de la Fuente, PhD, professor of bioengineering at the University of Pennsylvania, Philadelphia.
“The main pillars that have enabled us to almost double our lifespan in the last 100 years or so have been antibiotics, vaccines, and clean water,” said Dr. de la Fuente. “Imagine taking out one of those. I think it would be pretty dramatic.” (Dr. De la Fuente’s lab has become known for finding antibiotic candidates in unusual places, like ancient genetic information of Neanderthals and woolly mammoths.)
The first widely used antibiotic, penicillin, was discovered in 1928, when a physician studying Staphylococcus bacteria returned to his lab after summer break to find mold growing in one of his petri dishes. But many other antibiotics — like streptomycin, tetracycline, and erythromycin — were discovered from soil bacteria, which produce variations of these substances to compete with other microorganisms.
By looking in the gut microbiome, the researchers hoped to identify peptides that the trillions of microbes use against each other in the fight for limited resources — ideally, peptides that wouldn’t broadly kill off the entire microbiome.
Kill the Bad, Spare the Good
Many traditional antibiotics are small molecules. This means they can wipe out the good bacteria in your body, and because each targets a specific bacterial function, bad bacteria can become resistant to them.
Peptide antibiotics, on the other hand, don’t diffuse into the whole body. If taken orally, they stay in the gut; if taken intravenously, they generally stay in the blood. And because of how they kill bacteria, targeting the membrane, they’re also less prone to bacterial resistance.
The microbiome is like a big reservoir of pathogens, said Ami Bhatt, MD, PhD, hematologist at Stanford University in California and one of the study’s authors. Because many antibiotics kill healthy gut bacteria, “what you have left over,” Dr. Bhatt said, “is this big open niche that gets filled up with multidrug-resistant organisms like E coli [Escherichia coli] or vancomycin-resistant Enterococcus.”
Dr. Bhatt has seen cancer patients undergo successful treatment only to die of a multidrug-resistant infection, because current antibiotics fail against those pathogens. “That’s like winning the battle to lose the war.”
By investigating the microbiome, “we wanted to see if we could identify antimicrobial peptides that might spare key members of our regular microbiome, so that we wouldn’t totally disrupt the microbiome the way we do when we use broad-spectrum, small molecule–based antibiotics,” Dr. Bhatt said.
The researchers used artificial intelligence to sift through 400,000 proteins to predict, based on known antibiotics, which peptide sequences might have antimicrobial properties. From the results, they chose 78 peptides to synthesize and test.
“The application of computational approaches combined with experimental validation is very powerful and exciting,” said Jennifer Geddes-McAlister, PhD, professor of cell biology at the University of Guelph in Ontario, Canada, who was not involved in the study. “The study is robust in its approach to microbiome sampling.”