Using a man-made intelligence algorithm, researchers at MIT and McMaster University have identified a recent antibiotic that may kill a variety of bacteria that’s chargeable for many drug-resistant infections.

If developed to be used in patients, the drug could help to combat , a species of bacteria that is commonly present in hospitals and may result in pneumonia, meningitis, and other serious infections. The microbe can also be a number one reason for infections in wounded soldiers in Iraq and Afghanistan.

“ can survive on hospital doorknobs and equipment for long periods of time, and it may take up antibiotic resistance genes from its environment. It’s really common now to search out isolates which can be immune to nearly every antibiotic,” says Jonathan Stokes, a former MIT postdoc who’s now an assistant professor of biochemistry and biomedical sciences at McMaster University.

The researchers identified the brand new drug from a library of nearly 7,000 potential drug compounds using a machine-learning model that they trained to guage whether a chemical compound will inhibit the expansion of .

“This finding further supports the premise that AI can significantly speed up and expand our seek for novel antibiotics,” says James Collins, the Termeer Professor of Medical Engineering and Science in MIT’s Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering. “I’m excited that this work shows that we will use AI to assist combat problematic pathogens corresponding to .”

Collins and Stokes are the senior authors of the brand new study, which appears today in . The paper’s lead authors are McMaster University graduate students Gary Liu and Denise Catacutan and up to date McMaster graduate Khushi Rathod.

Drug discovery

Over the past several a long time, many pathogenic bacteria have turn out to be increasingly immune to existing antibiotics, while only a few recent antibiotics have been developed.

Several years ago, Collins, Stokes, and MIT Professor Regina Barzilay (who can also be an writer on the brand new study), got down to combat this growing problem through the use of machine learning, a variety of artificial intelligence that may learn to acknowledge patterns in vast amounts of knowledge. Collins and Barzilay, who co-direct MIT’s Abdul Latif Jameel Clinic for Machine Learning in Health, hoped this approach could possibly be used to discover recent antibiotics whose chemical structures are different from any existing drugs.

Of their initial demonstration, the researchers trained a machine-learning algorithm to discover chemical structures that would inhibit growth of . In a screen of greater than 100 million compounds, that algorithm yielded a molecule that the researchers called halicin, after the fictional artificial intelligence system from “2001: A Space Odyssey.” This molecule, they showed, could kill not only but several other bacterial species which can be immune to treatment.

“After that paper, after 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 ,” Stokes says.

To acquire training data for his or her computational model, the researchers first exposed grown in a lab dish to about 7,500 different chemical compounds to see which of them could inhibit growth of the microbe. Then they fed the structure of every molecule into the model. Additionally they told the model whether each structure could inhibit bacterial growth or not. This allowed the algorithm to learn chemical features related to growth inhibition.

Once the model was trained, the researchers used it to investigate a set of 6,680 compounds it had not seen before, which got here from the Drug Repurposing Hub on the Broad Institute. This evaluation, which took lower than two hours, yielded a couple of hundred top hits. Of those, the researchers selected 240 to check experimentally within the lab, specializing in compounds with structures that were different from those of existing antibiotics or molecules from the training data.

Those tests yielded nine antibiotics, including one which was very potent. This compound, which was originally explored as a possible diabetes drug, turned out to be extremely effective at killing but had no effect on other species of bacteria including , , and carbapenem-resistant .

This “narrow spectrum” killing ability is a desirable feature for antibiotics since it minimizes the danger of bacteria rapidly spreading resistance against the drug. One other advantage is that the drug would likely spare the helpful bacteria that live within the human gut and help to suppress opportunistic infections corresponding to .

“Antibiotics often need to be administered systemically, and the very last thing you ought to do is cause significant dysbiosis and open up these already sick patients to secondary infections,” Stokes says.

A novel mechanism

In studies in mice, the researchers showed that the drug, which they named abaucin, could treat wound infections attributable to . Additionally they showed, in lab tests, that it really works against a wide range of drug-resistant strains isolated from human patients.

Further experiments revealed that the drug kills cells by interfering with a process generally known as lipoprotein trafficking, which cells use to move proteins from the inside of the cell to the cell envelope. Specifically, the drug appears to inhibit LolE, a protein involved on this process.

All Gram-negative bacteria express this enzyme, so the researchers were surprised to search out that abaucin is so selective in targeting . They hypothesize that slight differences in how performs this task might account for the drug’s selectivity.

“We haven’t finalized the experimental data acquisition yet, but we predict it’s because does lipoprotein trafficking slightly bit in another way than other Gram-negative species. We imagine that’s why we’re getting this narrow spectrum activity,” Stokes says.

Stokes’ lab is now working with other researchers at McMaster to optimize the medicinal properties of the compound, in hopes of developing it for eventual use in patients.

The researchers also plan to make use of their modeling approach to discover potential antibiotics for other varieties of drug-resistant infections, including those attributable to and .

The research was funded by the David Braley Center for Antibiotic Discovery, the Weston Family Foundation, the Audacious Project, the C3.ai Digital Transformation Institute, the Abdul Latif Jameel Clinic for Machine Learning in Health, the DTRA Discovery of Medical Countermeasures Against Recent and Emerging Threats program, the DARPA Accelerated Molecular Discovery program, the Canadian Institutes of Health Research, Genome Canada, the Faculty of Health Sciences of McMaster University, the Boris Family, a Marshall Scholarship, and the Department of Energy Biological and Environmental Research program.