Antibiotics have prevented many infections and saved many lives, and most of the antibiotics prescribed today are broad-spectrum — designed to kill a wide variety of bacteria. The problem is, they kill not only the “bad” bacteria that cause disease, they also kill the useful “good” bacteria that live in the gut and keep people healthy.

Loss of this good bacteria makes it possible for drug-resistant bacteria like adherent invasive Escherichia coli or AIEC to grow in the gut. This is a particular problem for patients with inflammatory bowel diseases, which are made worse by AIEC. Patients with IBS need a narrow spectrum antibiotic that spares healthy gut bacteria.

When researchers at McMaster University in Canada screened almost 11,000 small molecules for antibacterial activity, they identified a molecule that targets AIEC and named it enterololin.

Using AI shortened by 18 months the time it usually takes to determine how a molecule works.

They used an AI platform developed at MIT called DiffDock to predict how enterololin works. It predicted the molecule attacks a protein complex called LolCDE that AIEC and other types of bacteria called Enterobacteriaceae need to survive. Finding a way to stop Enterobacteriaciae infections, for which we currently have no treatments, could help people suffering from IBS and Crohns disease.

Within a few months, the researchers, led by graduate student Denise Catacutan, found that DiffDock was right. “We did our standard workup to validate the prediction, and the experiments backed up the AI model,” Catacutan said in a statement. Using AI took shortened by 18 months the time it usually takes to determine how a molecule works.

Using AI for quicker turnaround in drug discovery could save researchers and companies millions of dollars in the future.

Enterololin showed low toxicity and prevented AIEC infection in mouse models. The experiments serve as a good proof of concept, Jon Stokes, lead investigator of the study, told TheDoctor, but more work needs to be done.

Patients with IBS need a narrow spectrum antibiotic that spares healthy gut bacteria.

Two issues that remain to be worked out are 1) that Enterololin is not as soluble in water as researchers would like it to be, and 2) it has a difficult time getting through the outer membrane of bacteria like AIEC. “By optimizing its outer membrane permeability, enterololin could be given alone as a monotherapy rather than a combination therapy,” said Stokes, an assistant professor of biochemistry and biomedical sciences at McMaster. Stokes has licensed enterololin from McMaster and is currently developing it for use in humans.

If development of a form that is useable in humans goes well, enterololin could be ready for human trials in about three years.

The study was published in Nature Microbiology.