In a groundbreaking development, researchers at the Massachusetts Institute of Technology (MIT) have harnessed the power of artificial intelligence to design novel antibiotics capable of tackling two formidable infections: drug-resistant Neisseria gonorrhoeae and multi-drug-resistant Staphylococcus aureus (MRSA). This innovative approach, utilizing generative AI algorithms, has resulted in the creation of over 36 million potential compounds, with the most promising candidates exhibiting unique structures and mechanisms that disrupt bacterial cell membranes.
The announcement comes as antibiotic resistance continues to pose a significant global health threat, with drug-resistant bacterial infections estimated to cause nearly 5 million deaths annually. The MIT team’s findings, published today in the journal Cell, represent a significant leap forward in the fight against antibiotic-resistant bacteria.
Revolutionizing Antibiotic Discovery
The research, led by James Collins, the Termeer Professor of Medical Engineering and Science at MIT, underscores the transformative potential of AI in drug discovery. “We’re excited about the new possibilities that this project opens up for antibiotics development,” Collins stated. “Our work shows the power of AI from a drug design standpoint, and enables us to exploit much larger chemical spaces that were previously inaccessible.”
Historically, the development of new antibiotics has been slow, with only a few dozen approved by the FDA over the past 45 years, most of which are derivatives of existing drugs. Meanwhile, bacterial resistance to these medications has been steadily increasing. In response, Collins and his colleagues at MIT’s Antibiotics-AI Project have employed AI to screen vast libraries of chemical compounds, yielding promising candidates like halicin and abaucin.
Exploring Uncharted Chemical Space
In their latest study, the researchers ventured beyond known chemical libraries, using AI to generate hypothetical molecules that had not been previously discovered. This approach allowed them to explore a broader diversity of potential drug compounds. The team employed two distinct strategies: designing molecules based on specific chemical fragments with known antimicrobial activity and allowing AI algorithms to freely generate novel molecules.
For the fragment-based approach, the researchers targeted N. gonorrhoeae, assembling a library of about 45 million known chemical fragments. Using machine-learning models trained to predict antibacterial activity, they narrowed this down to approximately 1 million candidates. “We wanted to get rid of anything that would look like an existing antibiotic, to help address the antimicrobial resistance crisis in a fundamentally different way,” explained Aarti Krishnan, a lead author of the study.
“By venturing into underexplored areas of chemical space, our goal was to uncover novel mechanisms of action,” Krishnan added.
Through rigorous computational analysis, the team identified a promising fragment, F1, which served as the basis for generating additional compounds using two generative AI algorithms. This process yielded about 7 million candidates, ultimately leading to the discovery of NG1, a compound effective against drug-resistant N. gonorrhoeae in both lab and animal models.
Unconstrained Molecular Design
In a parallel effort, the researchers explored the potential of using generative AI to design molecules targeting Gram-positive bacteria, specifically S. aureus. By applying AI algorithms without constraints, they generated over 29 million compounds, eventually synthesizing and testing 22 molecules. Among these, six exhibited strong antibacterial activity against multi-drug-resistant S. aureus, with the top candidate, DN1, proving effective in a mouse model of MRSA infection.
Phare Bio, a nonprofit organization involved in the Antibiotics-AI Project, is now working to further modify NG1 and DN1 for additional testing. “In collaboration with Phare Bio, we are exploring analogs, as well as working on advancing the best candidates preclinically, through medicinal chemistry work,” Collins noted.
Implications and Future Directions
This development follows significant investment from various organizations, including the U.S. Defense Threat Reduction Agency and the National Institutes of Health, highlighting the critical need for innovative solutions to combat antibiotic resistance. The success of MIT’s AI-driven approach could pave the way for similar strategies targeting other bacterial pathogens, such as Mycobacterium tuberculosis and Pseudomonas aeruginosa.
As the research progresses, the implications for global health are profound. By unlocking new chemical spaces and uncovering novel mechanisms of action, AI has the potential to revolutionize antibiotic discovery, offering hope in the battle against drug-resistant infections. The next steps will involve advancing these promising candidates through clinical trials, with the ultimate goal of bringing new, effective antibiotics to market.
