(CNN) — A group of researchers has found a new type of antibiotic that works specifically against dangerous and drug-resistant bacteria, thanks to artificial intelligence.
When they experimentally tested the antibiotic on the skin of mice infected with the superbug, bacterial growth was controlled, suggesting that the method could be used to create antibiotics to fight other drug-resistant pathogens. Is.
The researchers also tested against 41 different types of antibiotic. Acinetobacter baumannii Resistant to antibiotics. The drug worked in all of them, although it needs to be proven and tested in human clinical trials before it can be used in patients.
What’s more, the compound identified by the artificial intelligence acted in a way that only blocked the offending pathogen. It does not appear to kill many other species of beneficial bacteria living in the gut or on the skin, making it a rare and limited-acting agent.
According to the researchers, if there were more antibiotics that worked with this precision, bacteria could be prevented from becoming resistant.
The study was published in the academic journal Nature Chemical Biology.
An associate professor at the University of Pennsylvania Perlman School of Medicine, Dr. “It’s incredibly promising,” says César de la Fuente, who is also using AI to find new treatments but was not involved in the new research.
De la Fuente says this type of approach to finding new drugs is an emerging area that researchers have been testing since about 2018. This dramatically reduces the time taken to screen thousands of promising compounds.
“I think AI, as we’ve seen, can be successfully applied in many domains, and I think drug discovery is kind of the next frontier.”
For the study, the researchers focused on the bacterium acinetobacter baumannii, This bacteria is present in hospitals and other health facilities and tends to stick to surfaces such as doorknobs and counters. Because he’s able to pick up fragments of DNA from other organisms he comes into contact with, he can incorporate his best weapons: genes that help him resist the agents doctors use to treat him.
“This is what we call a professional pathogen in the lab,” says John Stokes, one of the researchers and an assistant professor of biochemistry and biomedical sciences at McMaster University in Hamilton, Ontario.
This species causes skin, blood or respiratory infections that are difficult to treat. The United States Centers for Disease Control and Prevention (CDC) noted in 2019 that infection by acinetobacter baumannii They were the ones who “most needed” new types of antibiotics to treat them.
Recent study of hospitalized patients with infection acinetobacter baumannii Resistant even to powerful carbapenem antibiotics were found to cause 1 in 4 deaths within a month of diagnosis.
For the new study, Stokes and his lab collaborated with researchers from MIT’s Broad Institute and Harvard. First, they used a technique called high-throughput drug screening to cultivate acinetobacter baumannii in laboratory dishes and spent weeks exposing these colonies to more than 7,500 agents: drugs and active pharmaceutical ingredients. They found 480 compounds that inhibited the growth of bacteria.
They fed that information into a computer and used it to train an artificial intelligence algorithm.
“Once our model was trained, all we could do was start showing it new images of chemicals, right? And based on what it had learned during training, it would predict that those molecules were antibacterial. Or not.” Stokes explains.
They then had a model analysis of more than 6,000 molecules, which Stokes said the AI was able to do in a matter of hours.
They narrowed the search down to 240 chemicals, which they tested in the lab. Laboratory tests helped them narrow down a list of the nine best inhibitors of bacteria. From there, they took a closer look at the composition of each one they thought might be dangerous or related to known antibiotics.
They were left with a compound called RS102895, which Stokes believes was originally developed as a potential treatment for diabetes.
According to Stokes, it appears to work in an entirely new way, by preventing components of the bacterium from moving from inside the cell to its surface.
“It’s quite an interesting mechanism that, to my knowledge, is not seen among clinical antibiotics,” he says.
Furthermore, RS102895 – which the researchers have named abousin – only works against acinetobacter baumannii,
According to Stokes, most antibiotics are broad-spectrum and work against many species of bacteria. Broad-spectrum antibiotics exert enormous selection pressure on many types of bacteria, causing many to evolve rapidly and share genes that help them resist the drug and survive.
“In the case of this molecule, since it acts only with great force nonsenseIt doesn’t exert that universal selective pressure, so it’s not going to spread resistance as quickly,” he explains.