Scientists Resurrected a 160-Million-Year-Old Antimicrobial Peptide

Bacteria have never met this molecule before, because it hasn't existed for 160 million years — until a lab in Oregon rebuilt it from scratch and pointed it at drug-resistant bacteria.

In a study published August 25 in PLOS Biology, University of Oregon researchers reconstructed a series of extinct versions of lactoferricin, an antimicrobial peptide tucked inside the mammalian protein lactoferrin. Tested in the lab, one of those extinct versions rivaled or outperformed the modern human peptide against several bacteria — even though the modern peptide had 160 million more years to specialize.

Digging a peptide out of deep time

Lactoferrin is a protein mammals make in milk, tears, saliva and other secretions, and it carries a smaller antimicrobial fragment, lactoferricin, that punches holes in bacterial membranes. Lead author Titas Sil and senior author Matt Barber wanted to know how that antimicrobial activity evolved. So they used a technique called ancestral sequence reconstruction, pioneered by former University of Oregon scientist Joseph Thornton: compare the lactoferrin gene across many living species, map the family tree, and statistically infer what the ancestral gene sequence at each branch point most likely looked like.

That let the team reconstruct lactoferricin as it probably existed at four points in mammalian history, tracing back to roughly 160 million years ago, when lactoferrin first arose from a duplicated copy of a related gene, transferrin, near the end of the Jurassic period. The researchers then had cells manufacture each reconstructed ancestral peptide and tested it directly against modern bacteria, including Pseudomonas aeruginosa, Staphylococcus aureus, E. coli and several Streptococcus species — pathogens that had no evolutionary history with these particular molecules at all.

One amino acid, an outsized effect

The results were uneven across deep time, which is itself notable: potency did not simply climb in a straight line from ancient to modern. The very first reconstructed ancestor after the gene duplication, labeled AncLFcin1, was weakly antimicrobial, needing a relatively large amount to cut P. aeruginosa growth by half. But the next ancestor in the lineage, AncLFcin2 — the shared ancestor of human and cattle lactoferrin — was far more potent against the same bacterium at a much lower concentration, and against S. aureus it produced roughly a 100,000-fold drop in viable bacteria in testing. Modern cattle lactoferricin, for comparison, was the single most potent version tested, clearing detectable P. aeruginosa colonies within two hours — evolution kept improving the peptide in some lineages even after that early jump.

The authors traced most of that jump to single point mutations: a substitution that added a positively charged amino acid (arginine) at one position in the peptide chain was, on its own, enough to significantly boost antimicrobial activity against P. aeruginosa, and a second, independent substitution found later in the great-ape lineage further increased potency against several major pathogens. Both changes increased the peptide's positive charge, which helps it latch onto and rupture negatively charged bacterial membranes — all versions tested, ancient and modern, breached bacterial membranes within about 30 minutes, but the modern and near-modern peptides caused irreversible collapse of the cell envelope, while the oldest ancestral version caused only transient damage bacteria could repair. "What was surprising and unexpected," Barber said, "was how small changes in these domains could have such large effects." As Barber put it, evolution has effectively been running "a science experiment billions of years old."

What's solid, and what isn't yet

The potency numbers come from a peer-reviewed paper with a named methodology, specific bacterial strains, and reproducible lab measurements — that's a real strength. But this is bench science, not a therapy. Every experiment described in the paper was run in low-nutrient laboratory culture media, which the authors themselves note doesn't fully represent the chemistry of a living body; pH and ionic conditions inside an animal can change how well these peptides work. Toxicity testing was limited to checking whether the peptides ruptured bovine red blood cells (they didn't, in that assay), which is far short of a full safety profile. And the ancestral peptides, like most natural antimicrobial peptides, are structurally less stable than a conventional small-molecule drug and break down quickly in the body — a limitation Barber's team readily acknowledges.

That gap between a clean laboratory result and a usable treatment is a recurring theme in peptide research generally. A University of California, Los Angeles review covered on this journal in “UCLA reviewed 565 peptide studies. Two-thirds were in animals” found that most published research on popular peptides never leaves cell cultures or animal models before public interest runs ahead of the evidence. This study is squarely in that earlier category: cells and lab culture, not even mice yet, let alone people.

What this means for researchers

The immediate value of this work isn't a new drug candidate — it's a map. By showing exactly which amino acid changes made lactoferricin more effective against which pathogens, and when those changes arose in mammalian evolution, the researchers have identified specific, testable design targets for engineering better synthetic antimicrobial peptides, rather than screening for them by trial and error. The paper's authors describe the approach as a way to "uncover unique variants of extant antimicrobial peptides that could serve as safe and effective pathogen treatment options," while being explicit that this is a starting point for that work, not an endpoint.

The broader pattern is also worth sitting with: the researchers found that antimicrobial potency didn't rise steadily through mammalian evolution but fluctuated by lineage, apparently in response to which pathogens a given species' ancestors actually encountered. For a field grappling with rising antibiotic resistance, that suggests evolutionary history itself is an underused source of leads — asking not just "what antimicrobial peptides exist in nature today" but "what antimicrobial peptides existed in nature's past, and why did some of them work better than others." Turning that map into an actual candidate molecule that survives contact with a real infection is the work Barber's lab says is still ahead.

Coverage of the study, including researcher interviews, was also published by Phys.org and ScienceDaily.