Antimicrobial resistance is one of the greatest threats facing humanity, which makes the need for new antibiotics more important than ever. While most antibiotics stem from bacteria and fungi, Archaea provides a largely undesirable tank to discover antibiotics. In a new study, researchers at the University of Pennsylvania benefited from deep learning to systematically explore living organisms; By protein mining of 233 types of wealthy, they identified 12623 molecules with possible antimicrobial activity.
Torres And others. 80 Archaeasins, 93 % of which showed an antimicrobial activity In the laboratory against Acineetobacter Baumanniiand Show the coldand Klebsiella pneumoniaand Pseudomonas aeruginosaand Staphylococci and Intestinal SPP. Image credit: Torres And others. , Two: 10.1038/S41564-025-02061-0.
“Previous efforts to find new antibiotics mostly have seen fungi, bacteria and animals,” said Dr. Cisar de la Fuente, a researcher at the University of Pennsylvania.
“In the past, we used artificial intelligence models to identify candidates for antibiotics in a group of unbearable sources, from the DNA of extinct organisms to chemicals in animal poison.”
“Now, we apply these tools to a new set of data: hundreds of old microbes proteins.”
“There is another room for life waiting to explore it.”
It differs from both bacteria and from the realities of the cores (which includes plants, animals and fungi), Archaea operates their own branch on the tree of life.
Although it resembles bacteria under a microscope, Archaea is essentially different in genetics, cell membranes and biochemistry.
These differences allow them to survive in some of the most extreme earth environments, starting from the heated bottom holes to hot springs like those in Yellowstone National Park.
Since Archaea often flourishes as a few other organisms can develop – crushing pressure, toxic chemicals and maximum temperatures – their biology has evolved in unusual ways.
This makes them a promising but largely exploitative source of new molecular tools, including compounds that may work like antibiotics but work differently from those currently used.
“I was attracted to Archaea because we had to develop biochemical defenses in unusual environments,” said Dr. Marcelo Torres, of the University of Pennsylvania.
“We thought, if they survived billions of years under these circumstances, may have developed unique ways to combat microbial competitors, and we may learn from that.”
To detect potential antibiotic compounds hidden in Archaea, researchers have turned into artificial intelligence.
They took advantage of an updated version of Apex, an Amnesty International tool they originally developed to identify candidates for antibiotics in ancient biology, including extinct animal proteins such as the Sufi mammoth.
After seeing thousands of peptides – short chains of amino acids – with well -known antimicrobial properties, APEX can predict the possibility of a specific series of amino acids similar effects.
By recruiting APEX 1.1 on thousands of peptides and additional information about bacteria that cause human diseases, scientists have prepared a tool to predict the peptides in Archaea that may prevent bacteria growth.
Surveying 233 species of archaeological, which resulted in more than 12,000 antibiotic candidates.
The authors called these ARCHEAASINS molecules, which have revealed the chemical analysis of well -known antimicrobial peptides (amps), especially in the distribution of the electric charge.
Then they chose 80 Archaeasins for a test against actual bacteria.
“Trying to find new antibiotics, one molecule at one time, similar to the search for needles in a straw pile,” says Fangping Wan, a post -PhD researcher at the University of Pennsylvania.
“Amnesty International increases the process by determining the location of the needles.”
Antibiotics work in several ways. Some punch holes in bacterial membranes, while others close the ability of living organisms to make proteins.
Researchers have found that unlike most known amperes, which attack the external defenses of bacteria, it seems that Archaeasins pull the plug from the inside, leading to a scrambled by electrical signals that keep the cell alive.
In tests against a group of pathogenic and drug -resistant bacteria, 93 % of 80 archaeasins showed by the survey was an anti -microbial activity against at least one bacteria.
Then the team chose three Archaeasins to test animal models.
Four days after one dose, Archaeasins arrested the spread of drug -resistant bacteria often obtained in hospitals.
One of the three vehicles has shown a similar activity for the polycsin B, a antibiotic that is usually used as a last line of defense against drug -resistant infections.
“This research shows that there are many antibiotics pending its discovery in Archaea,” said Dr. De L. Fuente.
“With more bacteria that develop current antibiotics resistance, it is important to find new antibiotics in unconventional places to replace them.”
A paper The results were published today in the magazine Microbiology.
_____
MDT Torres And others. Deep learning reveals antibiotics in archaeological proteins. Nat MicrobolIt was published on the Internet on August 12, 2025; DOI: 10.1038/S41564-025-02061-0