AI zeroes in on an antibiotic drug that can kill a dangerous bacteria

Since the bacteria is multidrug-resistant, discovering new antibiotics against A baumannii has proven challenging through conventional screening approaches. So the team used artificial intelligence (AI) to identify a new drug from 7,500 potential molecular drug compounds. 

They trained a machine-learning model to analyze whether a chemical compound will inhibit the growth of A. baumannii. The team narrowed down on an antibacterial compound called abaucin, effective only against a specific range of organisms, to be a successful antibiotic against A. baumannii.

The team ran tests in an animal model, in mice, to test the efficacy of the antibacterial compound. They found that abaucin could treat wound infections caused by A. baumannii in mice.

“This finding further supports the premise that AI can significantly accelerate and expand our search for novel antibiotics,” says James Collins, a professor at MIT and co-author of the paper, in a press statement. “I’m excited that this work shows that we can use AI to help combat problematic pathogens such as A. baumannii.”

In a preliminary experiment earlier, the team had similarly trained a machine-learning algorithm to identify molecular antibiotics that could inhibit the growth of E. coli. These bacteria typically live in the intestines but may sometimes cause diarrhea and gut infection.

That experiment had the team screen over 100 million antibacterial compounds, of which a molecule that the researchers called ‘halicin’ could kill not only E. coli but also several other antibiotic-resistant bacterial species.

“After that paper, when we showed that these machine-learning approaches can work well for complex antibiotic discovery tasks, we turned our attention to what I perceive to be public enemy No. 1 for multidrug-resistant bacterial infections, which is Acinetobacter,” said Jonathan Stokes, an assistant professor at McMaster University and co-author of the paper.

The new study was published in Nature Chemical Biology. 

Study abstract:

Acinetobacter baumannii is a nosocomial Gram-negative pathogen that often displays multidrug resistance. Discovering new antibiotics against A. baumannii has proven challenging through conventional screening approaches. Fortunately, machine learning methods allow for the rapid exploration of chemical space, increasing the probability of discovering new antibacterial molecules. Here we screened ~7,500 molecules for those that inhibited the growth of A. baumannii in vitro. We trained a neural network with this growth inhibition dataset and performed in silico predictions for structurally new molecules with activity against A. baumannii. Through this approach, we discovered abaucin, an antibacterial compound with narrow-spectrum activity against A. baumannii. Further investigations revealed that abaucin perturbs lipoprotein trafficking through a mechanism involving LolE. Moreover, abaucin could control an A. baumannii infection in a mouse wound model. This work highlights the utility of machine learning in antibiotic discovery and describes a promising lead with targeted activity against a challenging Gram-negative pathogen.