New Antibiotic Compounds from Mining the Human Gut Microbiome
MOLECULAR & COMPUTATIONAL BIOLOGY
8/23/2024
Antibiotic-resistant pathogenic microorganisms have become a pressing global health issue, making the search for alternative antibiotics critically important. As one potential solution, researchers from the University of Pennsylvania have utilized artificial intelligence (AI) to mine peptides with the potential to serve as alternative antibiotics, using data from human microbiome metagenomes. In a study published in Cell in August 2024, the researchers screened 444,054 peptides from 1,773 human microbiome metagenomes and successfully identified 323 antimicrobial peptide candidates encoded in small open reading frames (smORFs). Of the 78 peptides tested in vitro, 70.5% demonstrated antimicrobial activity by targeting the membranes of pathogenic bacteria. These peptides, now termed smORF-encoded peptides (SEPs), also showed the ability to synergize with each other and modulate the gut microbiota to suppress pathogens. The study further confirmed that prevotellin-2, a peptide from Prevotella copri, exhibited activity comparable to polymyxin B, a commonly used traditional antibiotic. These findings open new avenues for the development of alternative antibiotics or other therapeutic compounds derived from microbiome peptides in the future.
