TL;DR:
- Researchers at MIT and McMaster University have harnessed AI to identify a powerful new antibiotic effective against drug-resistant bacteria.
- The antibiotic targets Acinetobacter baumannii, a bacteria commonly found in hospitals and responsible for severe infections.
- Machine learning was used to evaluate thousands of compounds, leading to the discovery of a promising drug candidate.
- The research demonstrates the potential of AI in accelerating antibiotic development and combating multidrug-resistant pathogens.
Main AI News:
MIT and McMaster University researchers have made a groundbreaking discovery in the field of antibiotic development, using artificial intelligence (AI) to identify a potent new drug capable of combating drug-resistant infections. The new antibiotic has demonstrated the ability to effectively eliminate Acinetobacter baumannii, a bacteria species commonly found in hospitals, known to cause severe infections such as pneumonia and meningitis. This microbe also poses a significant threat to wounded soldiers in Iraq and Afghanistan, making the discovery even more critical.
The team of scientists utilized a machine-learning model to evaluate approximately 7,000 potential drug compounds, ultimately pinpointing the novel antibiotic. Acinetobacter baumannii has become increasingly resistant to existing antibiotics, rendering them ineffective in many cases. With few new antibiotics developed in recent years, this discovery presents a ray of hope in the battle against multidrug-resistant bacteria.
The AI algorithm was trained to identify chemical compounds that could inhibit the growth of A. baumannii, leading to the identification of this promising drug candidate. This groundbreaking finding reinforces the notion that AI can significantly expedite and broaden the search for innovative antibiotics, ultimately aiding in the fight against problematic pathogens like A. baumannii.
The senior authors of this breakthrough study are James Collins, the Termeer Professor of Medical Engineering and Science at MIT’s Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering, and Jonathan Stokes, an assistant professor of biochemistry and biomedical sciences at McMaster University. The research was published in Nature Chemical Biology, with the lead authors being McMaster University graduate students Gary Liu and Denise Catacutan, along with recent McMaster graduate Khushi Rathod.
In recent decades, the rise of antibiotic resistance has outpaced the development of new antibiotics, posing a significant global health challenge. To address this urgent issue, Collins, Stokes, and MIT Professor Regina Barzilay embarked on a mission to leverage machine learning, a form of AI, to identify antibiotics with novel chemical structures that differ from existing drugs. Their initial success involved training a machine-learning algorithm to recognize chemical structures inhibiting the growth of E. coli. This groundbreaking research resulted in the discovery of halicin, a molecule capable of eradicating not only E. coli but also several other drug-resistant bacterial species.
Building upon their previous achievements, the researchers turned their attention to Acinetobacter baumannii, which they regard as the primary culprit behind multidrug-resistant bacterial infections. To train their computational model, they exposed A. baumannii to thousands of chemical compounds, identifying those capable of inhibiting bacterial growth. By inputting the chemical structures and their inhibitory effects into the algorithm, they enabled it to learn the features associated with growth inhibition.
Once trained, the model analyzed over 6,000 compounds it had not encountered before, sourced from the Drug Repurposing Hub at the Broad Institute. Within a remarkably short timeframe of less than two hours, the analysis yielded several hundred top hits. Subsequently, the researchers selected 240 compounds for experimental testing, focusing on those with distinct structures compared to existing antibiotics and training data molecules.
Through rigorous testing, the team identified nine antibiotics, including one with exceptional potency. Originally investigated for its potential as a diabetes drug, this compound proved highly effective at eliminating A. baumannii while exhibiting no impact on other bacteria strains such as Pseudomonas aeruginosa, Staphylococcus aureus, and carbapenem-resistant Enterobacteriaceae. This narrow spectrum of killing ability is advantageous, as it minimizes the risk of bacteria developing rapid resistance to the drug. Additionally, the drug is likely to spare beneficial bacteria in the human gut, which play a crucial role in preventing opportunistic infections.
In subsequent mouse studies, the researchers demonstrated the efficacy of the drug, named abaucin, in treating wound infections caused by A. baumannii. Furthermore, lab tests confirmed its effectiveness against a variety of drug-resistant strains isolated from human patients.
Intriguingly, the drug’s mechanism of action involves interfering with a process called lipoprotein trafficking, essential for protein transportation within cells. The researchers discovered that abaucin inhibits a protein called LolE, responsible for this process. Although all Gram-negative bacteria express this enzyme, abaucin exhibits selective targeting towards A. baumannii. The scientists hypothesize that subtle differences in how A. baumannii performs lipoprotein trafficking could account for this specificity.
Moving forward, Stokes’ laboratory is collaborating with other researchers at McMaster to optimize the medicinal properties of abaucin, aiming to develop it for clinical use. The researchers also plan to employ their modeling approach to identify potential antibiotics for other types of drug-resistant infections caused by Staphylococcus aureus and Pseudomonas aeruginosa.
Conlcusion:
The successful application of AI in antibiotic discovery represents a significant breakthrough for the market. This groundbreaking research showcases the ability of machine learning algorithms to identify novel compounds capable of combatting drug-resistant infections. The potential impact on the market is immense, offering hope in the fight against antibiotic resistance and paving the way for accelerated drug development processes. The integration of AI into the pharmaceutical industry has the potential to revolutionize antibiotic discovery, leading to improved treatment options and better outcomes for patients worldwide.
