Community Jameel congratulates Professor Jim Collins, faculty lead for life sciences at the MIT Jameel Clinic, on the breakthrough use of generative artificial intelligence (AI) to design two novel antibiotics that could kill drug-resistant gonorrhoea and the superbug multi-drug-resistant Staphylococcus aureus (MRSA).
The findings have been published in the journal Cell. Professor Collins is the senior author of the paper.
A novel approach to antibiotic design
Bacterial resistance to antibiotics – known as antimicrobial resistance (AMR) – is a major global health threat that is associated with causing nearly 5 million deaths each year.
To accelerate the discovery and development of drugs that can kill resistant pathogens, Professor Collins and colleagues employed generative AI design in novel ways.
For Neisseria gonorrhoeae, the bacteria that causes gonorrhoea, the team used AI algorithms to design new molecules based on a chemical fragment that showed antimicrobial activity.
For MRSA, the algorithms were free to generate molecules de novo, without reference to a specific fragment.
After screening and filtering candidate molecules outputted by the algorithms, the team synthesised antiobiotic compounds that were tested in the lab dish and in mouse models.
The team found that NG1, a compound designed using the fragment-based approach, was effective in killing N. gonorrhoeae, apparently by lethally interfering with a protein involved in the synthesis of the bacteria's outer membrane.
DN1, a compound developed by the team using the de novo approach, was successfully able to clear an MRSA skin infection in a mouse model also by, it appears, interfering with bacterial cell membranes.
Speaking to the BBC, Professor Jim Collins, faculty lead for life sciences, MIT Jameel Clinic, said: "AI can enable us to come up with molecules, cheaply and quickly and in this way, expand our arsenal and really give us a leg up in the battle of our wits against the genes of superbugs."
Tackling two deadly drug-resistant bacteria
Gonorrhoea
Caused by the bacteria Neisseria gonorrhoeae, gonorrhoea is a common sexually-transmitted infection that is normally curable with antibiotics and can be prevented in most cases by regular and correct condom use.
Untreated, gonorrhoea can cause pregnancy complications and infertility and increases the risk of contracting the human immunodeficiency virus (HIV).
In 2020, the World Health Orgnisation (WHO) estimated that there were 82.4 million new infections of N. gonorrhoeae among adults aged 15 to 49 years.
The emergence of N. gonorrhoeae resistant to antibiotics is making treatment increasingly challenging, with the risk that gonorrhoea could become untreatable.
The WHO has called the development of new antibiotics to tackle gonorrhoea "critical to decrease this eminent threat".
MRSA
MRSA is one of the six most deadly superbugs and caused more than 100,000 deaths attributable to AMR in 2019, according to a study in The Lancet.
The paper found that, in 2019, the six pathogens – including MRSA – were together responsible for 929,000 deaths attributable to AMR.
AI for drug discovery
The generative AI project builds on earlier work by Professor Collins and colleagues, including Professor Regina Barzilay, faculty lead for AI at the MIT Jameel Clinic, that used deep learning approaches to discover two novel antibiotic compounds – halicin and abaucin – which could both kill drug-resistant bacteria.
Efforts to advance halicin and abaucin and now to modify NG1 and DN1 to make them suitable for additional testing, are being led by Professor Collins through his lab and Phare Bio, a non-profit company he co-founded.