A new type of antibiotic, discovered with artificial intelligence, can defeat a dangerous superbug

Matt Clarke/McMaster University

Denise Catacutan, a graduate student at McMaster University, helped identify the new antibacterial compound.


Using artificial intelligence, researchers say, they’ve found a new type of antibiotic that works against a particularly threatening, drug-resistant bacteria.

When they tested the antibiotic on the skin of mice experimentally infected with the superbug, it controlled the growth of the bacteria, suggesting the method could be used to create antibiotics tailored to combat other drug-resistant pathogens.

The researchers also tested the antibiotic against 41 different antibiotic-resistant strains Acinetobacter baumannii. The drug worked in all of them, although it would need further refinement and testing in human clinical trials before it could be used in patients.

In addition, the compound identified by AI worked in a way that only hindered the problem pathogen. It didn’t seem to kill the many other types of beneficial bacteria that live in the gut or on the skin, making it a rare, narrowly targeted remedy.

If more antibiotics worked just like that, the researchers said, it could prevent bacteria from becoming resistant in the first place.

The study was published in the journal Nature Chemical Biology.

“It’s incredibly promising,” said Dr. Cesar de la Fuente, an assistant professor at the University of Pennsylvania Perlman School of Medicine who also uses AI to find new treatments but was not involved in the new research.

De la Fuente says this kind of approach to finding new drugs is an emerging area that researchers have been testing since about 2018. It drastically reduces the time it takes to search thousands of promising connections.

“I think, as we’ve seen, AI can be successfully applied to many domains, and I think drug discovery is sort of the next step.”

For the study, the researchers focused on the bacterium Actinetobacter baumanii. It hangs out in hospitals and other healthcare facilities, clinging to surfaces such as doorknobs and counters. Because it’s able to grab bits of DNA from other organisms it comes into contact with, it can incorporate their best weapons: genes that help them resist drugs doctors use to treat them.

“It’s what we call a professional pathogen in the lab,” said Jon Stokes, one of the researchers and an assistant professor of biochemistry and biomedical sciences at McMaster University in Hamilton, Ontario.

This species causes difficult to treat skin, blood or respiratory infections. The U.S. Centers for Disease Control and Prevention said in 2019 that Acinetobacter baumanii infections were “the greatest need” for new types of antibiotics to treat them.

A recent study of hospitalized patients with Actinetobacter baumanii infections resistant even to potent carbapenem antibiotics found that 1 in 4 died within a month of their diagnosis.

For the new study, Stokes and lab teamed up with researchers from MIT and Harvard’s Broad Institute. First, they used a technique called high-throughput drug screening to grow Acinetobacter baumanii in lab dishes and spent weeks exposing these colonies to more than 7,500 agents: drugs and the active ingredients of drugs. They found 480 compounds that blocked the growth of the bacteria.

They fed that information into a computer and used it to train an artificial intelligence algorithm.

“Once we trained our model, we could show that model brand new pictures of chemicals it had never seen, right? And based on what it learned in training, it would predict for us whether those molecules were antibacterial or not,” Stokes said.

They then had the model screen more than 6,000 molecules, which Stokes said the AI ​​could do over the course of a few hours.

They narrowed the search down to 240 chemicals, which they tested in the lab. The lab tests helped them narrow the list down to nine of the best inhibitors of the bacteria. From there, they took a closer look at the structure of each, eliminating those they thought were dangerous or related to known antibiotics.

They were left with one compound, called RS102895, which Stokes believes was originally developed as a possible treatment for diabetes.

He says it appears to work in an entirely new way, by preventing components of the bacteria from traveling to the surface from inside the cell.

“It’s a rather interesting mechanism and one that, to my knowledge, is not observed in clinical antibiotics,” he said.

In addition, he said, RS102895 — which the researchers renamed abaucin — works only on Actinetobacter baumanii.

Stokes says most antibiotics are broad-spectrum and work against many types of bacteria. Broad-spectrum antibiotics exert a lot of selective pressure on many types of bacteria, causing many to quickly develop and share genes that help them resist and survive the drug.

“With this molecule, because it’s only very potent against Actinetobacter, it doesn’t impose that universal selective pressure, so it won’t spread resistance as quickly,” he said.

Leave a comment