Researchers at MIT have claimed that with the use of AI they have identified a powerful new antibiotic compound hiding within a small-molecule drug that had been explored as a potential diabetes treatment.
The compound was able to kill most of the drug-resistant bacteria using a unique mechanism compared to other antibiotics. The new antibiotics can treat drug-resistant diseases by killing 35 powerful bacteria.
The drug-resistant bacteria are a large and growing problem, it has caused 2.8 million infections and 35,000 deaths in the U.S. each year and more in developing countries. The computer learning model developed at MIT, described in the journal Cell, has the potential to identify many new types of antibiotics.
The molecule has named halicin after HAL, the murderous sentient computer in 2001: A Space Odyssey. Halicin was tested against dozens of bacterial strains isolated from patients and grown in lab dishes and it was able to kill many that are resistant to treatment, including Clostridium difficile & Mycobacterium tuberculosis.
In 2017, the World Health Organization (WHO) named this bacteria that poses the greatest threat to human health, because of its resistance to antibiotics. The discovery of halicin may be able to change that.
James Collins, the Termeer Professor of Medical Engineering and Science in MIT’s Institute for Medical Engineering and Science and Department of Biological Engineering, said: “We wanted to develop a platform that would allow us to harness the power of AI to usher in a new age of antibiotic drug discovery.”
“Our approach revealed this amazing molecule which is arguably one of the more powerful antibiotics that has been discovered,” said Collins, who served as a co-senior author of a paper published in the journal Cell.
And, in the face of growing antibiotic resistance, Collins pointed to the need for new screening methods that could help revitalize the industry’s pipeline for developing new antibiotics, a costly process that has seen few successes in recent decades compared to other pharmaceutical research fields.
Researchers also discovered 8 more promising antibacterial compounds, 2 of which appear very powerful.
To find new antibiotics, the researchers first trained the model to identify the sorts of molecules that kill bacteria. To do this, they fed the program information on the atomic and molecular features of nearly 2,500 drugs and natural compounds, and how well or not the substance blocked the growth of the bug E. coli.
After the algorithm learned what molecular features made for good antibiotics, the scientists set it to screen a library of about 6,000 compounds from the Drug Repurposing Hub at the Broad Institute of MIT and Harvard.
Rather than looking for any potential antimicrobials, MIT’s machine learning model focused on compounds that looked effective but unlike existing antibiotics. This boosted the chances that the drugs would work in radical new ways that bugs had yet to develop resistance to.
In the lab, researchers found that E. coli was unable to develop any resistance to halicin after 30 days, compared to the 24 to 72 hours needed to defend against the antibiotic ciprofloxacin. The team said it plans to study the drug further and work with nonprofit or industry partners to develop it for use in humans.
The researchers said they identified several other molecules they plan to test further. A separate model was also used to gauge halicin’s toxicity against human cells before it was synthesized for testing. To test halicin’s effectiveness in living animals, the researchers used it to treat mice infected with A. baumannii
The strain of A. baumannii that they used is resistant to all known antibiotics, but the application of a halicin-containing ointment completely cleared the infections within 24 hours. Preliminary studies suggest that halicin kills bacteria by disrupting their ability to maintain an electrochemical gradient across their cell membranes.
The algorithm is developed by computer scientist Regina Barzilay, an MIT professor who specializes in applications of deep learning to chemistry and oncology.
“There is still a question of whether machine-learning tools are really doing something intelligent in healthcare, and how we can develop them to be workhorses in the pharmaceuticals industry,” Barzilay said, as per the Financial Times. “This shows how far you can adapt this tool.”
“When you’re dealing with a molecule that likely associates with membrane components, a cell can’t necessarily acquire a single mutation or a couple of mutations to change the chemistry of the outer membrane,” first author Jonathan Stokes, an MIT and Broad Institute postdoc, said to MIT News.
“Mutations like that tend to be far more complex to acquire evolutionarily.”
“The work really is remarkable,” said Jacob Durrant, who works on computer-aided drug design at the University of Pittsburgh. “Their approach highlights the power of computer-aided drug discovery. It would be impossible to physically test over 100m compounds for antibiotic activity.”
“Given typical drug-development costs, in terms of both time and money, any method that can speed early-stage drug discovery has the potential to make a big impact,” he added.
More in AI