Infections cause approximately 1.27 million deaths each year worldwide, a figure that, due to increased resistance to antibiotics, could reach 10 million by 2050. This discouraging picture is compounded by the fact that no … There have been no new classes of antibiotics in decades.
Spanish biotechnologist César de la Fuente, leader of the Machine Biology Group at the University of Pennsylvania (USA), believes that the solution to the problem is to bring back to life molecules that microorganisms do not know how to deal with, that belong to extinct organisms. He carries out this technique, called molecular de-extinction, with the help of deep learning and has already produced preclinical antibiotic candidates by ‘resurrecting’ Neanderthal molecules (neanderthalin-1). This Tuesday he presents in the journal ‘Nature Biomedical Engineering’ new hopeful molecules recovered from the mammoth and other animals of the past, such as the straight-tusked elephant, the ancient sea cow, the giant sloth and the giant elk.
The genomes express proteins with natural antimicrobial properties, produced and selected through evolution. Molecular decay hypothesizes that these molecules could be prime candidates for new safe drugs. In the new paper, the team uses deep learning to extract the proteomes – a group of proteins made by an organism – of all the extinct organisms available for the discovery of antibiotic peptides.
To do this, researchers have developed an artificial intelligence called APEX (Antibiotic Peptide Deextinction), which uses a multi-task learning architecture to predict the antimicrobial activity of peptides. The algorithm extracted 10,311,899 peptides. The models predicted 37,176 sequences with antimicrobial activity, 11,035 of which were not found in extant organisms. Finally, the team synthesized 69 peptides and experimentally confirmed their activity against bacterial pathogens. Most of the peptides killed the bacteria by depolarizing their cytoplasmic membrane, unlike known antimicrobial peptides, which tend to attack the outer membrane.
In particular, lead compounds (including mammuthusin-2 from the woolly mammoth, elephasin-2 from the straight-tusked elephant, hydrodamin-1 from the ancient sea cow, mylodonin-2 from the giant sloth, and megalocerin-1 from the extinct giant elk) showed anti-infective activity in mice with skin abscesses or thigh infections.
“Antimicrobial resistance is one of the greatest threats of our time and new antibiotics are urgently needed,” warns in his X De la Fuente account, who just a few days ago published in the journal Cell, together with scientists from the Queensland University of Technology (Australia), research that identified, with the help of AI, almost a million potential sources of antibiotics in nature. In his opinion, molecular analysis aided by deep learning can accelerate the discovery of therapeutic molecules and offer a completely new framework for drug discovery.
