Scientists are using artificial intelligence to search for new antibiotics within Archaea, a group of ancient, single-celled microbes known for their ability to survive in extreme environments.
As antibiotic resistance continues to make infections harder to treat, researchers are exploring whether these resilient microbes might offer new compounds to help address this challenge.
A team at the University of Pennsylvania recently reported in Nature Microbiology that they used artificial intelligence to identify antibiotic compounds within proteins from Archaea. Prior to this study, these microbes had not been explored for their potential to contribute to new medical treatments.
Unique Qualifications
Although Archaea and bacteria share some basic features, they differ in their genetic makeup, cellular structure, and chemistry. These differences enable Archaea to persist in environments such as hydrothermal vents, acidic hot springs, and salt flats, where most other organisms are unable to survive.
The evolution of Archaea has been unusually diverse due to their ability to thrive in extreme habitats. “We were drawn to Archaea because they’ve had to evolve biochemical defenses in unusual environments,” explained Marcelo Torres, co-first author of the study. “If they’ve survived for billions of years under those conditions, maybe they’ve developed unique ways to fight off microbial competitors.”
Despite their unique adaptations, Archaea have received little attention in antibiotic discovery efforts, which have traditionally focused on fungi, bacteria, and animals.
AI as a Discovery Engine
To assess the potential of Archaea, researchers in César de la Fuente’s laboratory at the University of Pennsylvania employed an updated version of APEX, an artificial intelligence platform originally developed to identify antimicrobial peptides in ancient DNA and animal venoms.
The updated AI system, APEX 1.1, was trained using thousands of peptides with known antibacterial activity, as well as data on how different pathogens respond to these amino acid molecules. The AI then analyzed protein sequences from 233 Archaea species to identify peptide molecules with potential antibiotic properties.
This analysis identified over 12,000 potential antibiotic candidates, named archaeasins. These peptides differ in electrical charge from many known antimicrobial peptides, indicating a potentially different biological mechanism.
“Trying to find new antibiotics one molecule at a time is like looking for needles in a haystack,” said co-author Fangping Wan. “AI speeds up the process by showing us where the needles are likely to be.”
unconventional Methods
From the pool of predicted candidates, the team synthesized and tested 80 archaeasins against drug-resistant bacteria. Laboratory results showed that 93 percent of these peptides were active against at least one bacterial strain.
These archaeasins seemed to disable bacteria in an unconventional way. Instead of puncturing membranes, blocking protein production, or relying on traditional methods utilized in many current antibiotics, the peptides disrupted the electrical signals bacteria need to survive.
The researchers selected three of the most promising archaeasins for tests in animal models. In infections with a dangerous hospital-acquired bacterium, a single dose halted bacterial spread within four days. One archaeasin’s performance matched polymyxin B, a last-resort antibiotic often used against resistant infections.
“This research shows that there are potentially many antibiotics waiting to be discovered in Archaea,” said de la Fuente. “With bacteria developing resistance faster than we can replace antibiotics, it’s critical to look in unconventional places.”
Refining the Approach
The discovery of these archaeasins is a good example of how AI can accelerate the search for new antibiotics by combining computer analysis with the unique biology of ancient microbes.
“This is only the beginning,” de la Fuente emphasized. “Archaea is one of the oldest forms of life, and clearly has much to teach us about how to outsmart the pathogens we face today.”
The team now plans to refine APEX to predict antibiotic activity based not only on amino acid sequences but also on three-dimensional structures, potentially improving accuracy. Long-term studies will be necessary to evaluate archaeasins’ safety and effectiveness before human clinical trials can commence.
An Antibiotic Arms Race?
The need for new antibiotics is intensifying, as drug-resistant bacteria are evolving at an unprecedented rate. The World Health Organization has cautioned that if we don’t find better treatments soon, even simple infections and everyday surgeries could become dangerous.
By utilizing the power of AI to explore Archaea, researchers at the University of Pennsylvania have discovered an innovative approach to combating antibiotic resistance. If archaeasins prove to be safe and effective antibiotics, they could mark a breakthrough in the fight against evolving bacteria.
Austin Burgess is a writer and researcher with a background in sales, marketing, and data analytics. He holds a Master of Business Administration and a Bachelor of Science in Business Administration, along with a certification in Data Analytics. His work combines analytical training with a focus on emerging science, aerospace, and astronomical research.