NEW YORK – A team led by investigators at Stanford University and the University of Pennsylvania has demonstrated the potential benefits of "mining" the human microbiome to unearth peptides or proteins with potential antibiotic activity.
"[W]e focused entirely on mining for new antimicrobial microproteins from the human microbiome — ranging from the gut to the skin," co-corresponding author Ami Bhatt, a medicine and genetics researcher at Stanford University, said in an email.
In work outlined Monday in Cell, the researchers analyzed 1,773 human metagenome collections generated with mouth, skin, gut, or vagina samples from participants in the National Institutes of Health Human Microbiome Project. They computationally searched for members of more than 444,000 small protein families proposed in the past, narrowing in on 323 candidate small protein family members encoded by small open reading frames (smORFs).
"While traditionally scientists searched for new antibiotics from all of the small molecules (chemicals) that are made by microbes," Bhatt explained, "emerging studies like this one suggest that proteins and peptides might be exciting antimicrobials."
The team selected and synthesized a subset of 78 peptides from the computationally predicted set, using antimicrobial activity screens, detecting activity against pathogenic microbes in vitro for 55 (70.5 percent) of the candidate active peptides tested. A handful of top candidate compounds were subsequently tested in mouse infection models involving skin abscesses or infected thigh tissue.
In the latter experiments, the authors explained, a lead candidate compound known as prevotellin-2 "cleared bacterial loads at a comparable level to the current gold standard [antibiotic], polymyxin B, and without notable toxicity to the mammalian host in either infection model."
The candidate compounds, which the researchers dubbed "smORF-encoded peptides" (SEPs), appeared to target bacterial membrane features. Along with their activity against pathogenic microbes, the SEPs seemed to shape gut microbiome patterns, including the activity of commensal gut bugs.
"We think that these might be naturally produced antimicrobials that bacteria use to fight one another in the body," Bhatt noted. "This type of 'warfare' can enable specific bacteria to make space for themselves in crowded and competitive environments."
"Our report supports the existence of hundreds of antimicrobials in the human microbiome amenable to clinical translation," the authors wrote.
In particular, Bhatt noted that clinicians have started turning to antibiotics based on peptides in "situations of last resort," when conventional antibiotics cannot effectively fight an infection, suggesting that additional peptide and protein candidates could expand the repertoire of antibiotic options in the future.
"[W]e hope that studies like ours will help identify new types of peptide- and protein-based medications," she explained, adding that the approach "might help us combat the rising tide of multidrug-resistant bacterial infections."