A. baumannii is frequently found in hospitals and is responsible for many drug-resistant infections.

Scientists at MIT and McMaster University have used artificial intelligence (AI) to identify a new antibiotic that can combat drug-resistant infections caused by Acinetobacter baumannii. The bacterium is commonly found in hospitals and is responsible for pneumonia, meningitis, and other severe infections.

The researchers trained a machine-learning algorithm to evaluate potential drug compounds and determine their effectiveness in inhibiting the growth of A. baumannii. They tested nearly 7,000 compounds and identified a new drug that showed promising results in killing the bacterium. (MIT)

The development of new antibiotics is crucial as many pathogenic bacteria have become resistant to existing drugs. The team previously used machine learning to identify a molecule called halicin, which demonstrated the ability to kill drug-resistant bacterial species. Building on this success, they focused on Acinetobacter baumannii, known for its resistance to multiple antibiotics. By exposing the bacterium to thousands of different chemical compounds and feeding the data into the computational model, the researchers identified several potential antibiotics.

Testing and experiments resulted in the discovery of nine antibiotics, with one compound proving highly potent against A. baumannii. This compound, named abaucin, was originally explored as a potential diabetes drug but demonstrated exceptional efficacy against the bacterium. Importantly, abaucin displayed a “narrow spectrum” killing ability, meaning it targeted A. baumannii specifically without affecting other bacterial species. This specificity minimizes the risk of resistance development and reduces disruption to beneficial gut bacteria.

Further investigations revealed that abaucin interferes with lipoprotein trafficking, a process involved in protein transportation within cells. The researchers found that the drug inhibits the LolE protein, which plays a role in this process. Interestingly, abaucin selectively targets A. baumannii, despite the presence of the targeted enzyme in all Gram-negative bacteria. The researchers speculate that subtle differences in A. baumannii’s lipoprotein trafficking mechanism explain the drug’s selectivity.

The team aims to optimize abaucin’s medicinal properties and develop it for use in patients. They also plan to apply their AI modeling approach to identify potential antibiotics for other drug-resistant infections caused by bacteria like Staphylococcus aureus and Pseudomonas aeruginosa.

Related Story:  AI Could Simulate 40,000 Hypothetical Bioweapons In A Mere Six Hours

Image Credit: Flickr