For patients with inflammatory bowel disease, antibiotics is usually a double-edged sword. The broad-spectrum drugs often prescribed for gut flare-ups can kill helpful microbes alongside harmful ones, sometimes worsening symptoms over time. When fighting gut inflammation, you don’t all the time need to bring a sledgehammer to a knife fight.
Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and McMaster University have identified a brand new compound that takes a more targeted approach. The molecule, called enterololin, suppresses a bunch of bacteria linked to Crohn’s disease flare-ups while leaving the remainder of the microbiome largely intact. Using a generative AI model, the team mapped how the compound works, a process that sometimes takes years but was accelerated here to simply months.
“This discovery speaks to a central challenge in antibiotic development,” says Jon Stokes, senior creator of a latest paper on the work, assistant professor of biochemistry and biomedical sciences at McMaster, and research affiliate at MIT’s Abdul Latif Jameel Clinic for Machine Learning in Health. “The issue isn’t finding molecules that kill bacteria in a dish — we’ve been in a position to do this for a very long time. A significant hurdle is determining what those molecules actually do inside bacteria. Without that detailed understanding, you may’t develop these early-stage antibiotics into secure and effective therapies for patients.”
Enterololin is a stride toward precision antibiotics: treatments designed to knock out only the bacteria causing trouble. In mouse models of Crohn’s-like inflammation, the drug zeroed in on , a gut-dwelling bacterium that may worsen flares, while leaving most other microbial residents untouched. Mice given enterololin recovered faster and maintained a healthier microbiome than those treated with vancomycin, a standard antibiotic.
Pinning down a drug’s mechanism of motion, the molecular goal it binds inside bacterial cells, normally requires years of painstaking experiments. Stokes’ lab discovered enterololin using a high-throughput screening approach, but determining its goal would have been the bottleneck. Here, the team turned to DiffDock, a generative AI model developed at CSAIL by MIT PhD student Gabriele Corso and MIT Professor Regina Barzilay.
DiffDock was designed to predict how small molecules fit into the binding pockets of proteins, a notoriously difficult problem in structural biology. Traditional docking algorithms search through possible orientations using scoring rules, often producing noisy results. DiffDock as an alternative frames docking as a probabilistic reasoning problem: a diffusion model iteratively refines guesses until it converges on the most probably binding mode.
“In only a few minutes, the model predicted that enterololin binds to a protein complex called LolCDE, which is important for transporting lipoproteins in certain bacteria,” says Barzilay, who also co-leads the Jameel Clinic. “That was a really concrete lead — one that might guide experiments, somewhat than replace them.”
Stokes’ group then put that prediction to the test. Using DiffDock predictions as an experimental GPS, they first evolved enterololin-resistant mutants of within the lab, which revealed that changes within the mutant’s DNA mapped to lolCDE, precisely where DiffDock had predicted enterololin to bind. Additionally they performed RNA sequencing to see which bacterial genes switched on or off when exposed to the drug, in addition to used CRISPR to selectively knock down expression of the expected goal. These laboratory experiments all revealed disruptions in pathways tied to lipoprotein transport, exactly what DiffDock had predicted.
“If you see the computational model and the wet-lab data pointing to the identical mechanism, that’s while you begin to consider you’ve figured something out,” says Stokes.
For Barzilay, the project highlights a shift in how AI is utilized in the life sciences. “Numerous AI use in drug discovery has been about searching chemical space, identifying latest molecules that could be lively,” she says. “What we’re showing here is that AI can even provide mechanistic explanations, that are critical for moving a molecule through the event pipeline.”
That distinction matters because mechanism-of-action studies are sometimes a significant rate-limiting step in drug development. Traditional approaches can take 18 months to 2 years, or more, and price thousands and thousands of dollars. On this case, the MIT–McMaster team cut the timeline to about six months, at a fraction of the price.
Enterololin continues to be within the early stages of development, but translation is already underway. Stokes’ spinout company, Stoked Bio, has licensed the compound and is optimizing its properties for potential human use. Early work can be exploring derivatives of the molecule against other resistant pathogens, corresponding to . If all goes well, clinical trials could begin inside the following few years.
The researchers also see broader implications. Narrow-spectrum antibiotics have long been sought as a technique to treat infections without collateral damage to the microbiome, but they’ve been difficult to find and validate. AI tools like DiffDock could make that process more practical, rapidly enabling a brand new generation of targeted antimicrobials.
For patients with Crohn’s and other inflammatory bowel conditions, the prospect of a drug that reduces symptoms without destabilizing the microbiome could mean a meaningful improvement in quality of life. And in the larger picture, precision antibiotics may help tackle the growing threat of antimicrobial resistance.
“What excites me will not be just this compound, but the concept that we are able to start serious about the mechanism of motion elucidation as something we are able to do more quickly, with the correct combination of AI, human intuition, and laboratory experiments,” says Stokes. “That has the potential to alter how we approach drug discovery for a lot of diseases, not only Crohn’s.”
“One in every of the best challenges to our health is the rise of antimicrobial-resistant bacteria that evade even our greatest antibiotics,” adds Yves Brun, professor on the University of Montreal and distinguished professor emeritus at Indiana University Bloomington, who wasn’t involved within the paper. “AI is becoming a crucial tool in our fight against these bacteria. This study uses a robust and chic combination of AI methods to find out the mechanism of motion of a brand new antibiotic candidate, a crucial step in its potential development as a therapeutic.”
Corso, Barzilay, and Stokes wrote the paper with McMaster researchers Denise B. Catacutan, Vian Tran, Jeremie Alexander, Yeganeh Yousefi, Megan Tu, Stewart McLellan, and Dominique Tertigas, and professors Jakob Magolan, Michael Surette, Eric Brown, and Brian Coombes. Their research was supported, partially, by the Weston Family Foundation; the David Braley Centre for Antibiotic Discovery; the Canadian Institutes of Health Research; the Natural Sciences and Engineering Research Council of Canada; M. and M. Heersink; Canadian Institutes for Health Research; Ontario Graduate Scholarship Award; the Jameel Clinic; and the U.S. Defense Threat Reduction Agency Discovery of Medical Countermeasures Against Latest and Emerging Threats program.
The researchers posted sequencing data in public repositories and released the DiffDock-L code openly on GitHub.
